From 8232e11d576db0f9e7d19b2634f2088ff8286ebf Mon Sep 17 00:00:00 2001
From: Konrad Firley <konrad.firley@student.reutlingen-university.de>
Date: Wed, 26 Jun 2024 15:05:44 +0000
Subject: [PATCH] Delete notebook_1.ipynb -> redundant
---
.../notebook_1.ipynb | 4885 -----------------
1 file changed, 4885 deletions(-)
delete mode 100644 Insurance/Insurance Fraud detection/notebook_1.ipynb
diff --git a/Insurance/Insurance Fraud detection/notebook_1.ipynb b/Insurance/Insurance Fraud detection/notebook_1.ipynb
deleted file mode 100644
index 6d9ac59..0000000
--- a/Insurance/Insurance Fraud detection/notebook_1.ipynb
+++ /dev/null
@@ -1,4885 +0,0 @@
-{
- "cells": [
- {
- "attachments": {},
- "cell_type": "markdown",
- "id": "6e39bc3c",
- "metadata": {
- "editable": true,
- "slideshow": {
- "slide_type": ""
- },
- "tags": []
- },
- "source": [
- "# 1. Business Understanding"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "a526e65c-4cd2-42af-b98f-f711b45e0274",
- "metadata": {
- "editable": true,
- "include": true,
- "paragraph": "Title",
- "slideshow": {
- "slide_type": ""
- },
- "tags": []
- },
- "source": [
- "Insurance Fraud detection - Versicherungs Betrugserkennung "
- ]
- },
- {
- "attachments": {},
- "cell_type": "markdown",
- "id": "135c407f",
- "metadata": {
- "editable": true,
- "include": false,
- "paragraph": "BusinessUnderstanding",
- "slideshow": {
- "slide_type": ""
- },
- "tags": []
- },
- "source": [
- "Versicherungen verfügen über eine Vielzahl von Daten und kreiren täglich neue. Unter diesen Daten sind auch sehr sensible Informationen wie Name, Geburtsdaten, Adressen und Kontoverbindung ihrer Versicherten. Diese Daten werden von den Versicherungsunternehmen zunehmend automatisiert verarbeitet, ausgewertet und für weitere Versicherungsprozesse genutzt. Dabei geht es natürlich nach wie vor darum, für bestehende Versicherungsprodukte das aktuelle Risiko zu berechnen und darauf aufbauend die Prämie und die mögliche Schadenshöhe zu ermitteln. Die Schaffung neuer, bedarfsgerechter Versicherungsprodukte, die kurzfristig abgeschlossen werden können und eine sehr kurze Laufzeit haben, ist ein weiterer Trend, der durch Daten unterstützt werden kann. Die zentralen Fragen hierbei sind natürlich: Welche Daten sind für die Aufdeckung von Versicherungsbetrug relevant? Wie müssen diese Daten strukturiert sein? Welches Modell ist am besten geeignet, um Versicherungsbetrug im Schadenfall vorherzusagen? Nach welchen Kriterien sollte man verschiedene ML-Modelle vergleichen? Wie zuverlässig funktioniert die Vorhersage von Versicherungsbetrug?"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "4ca514ca-c655-462d-9b0c-3b36d1e0f95f",
- "metadata": {
- "editable": true,
- "include": true,
- "paragraph": "Business",
- "slideshow": {
- "slide_type": ""
- },
- "tags": []
- },
- "source": [
- "Für jedes Versicherungs-Unternehmen ist wichtig nur begründete und gerechtfertigte Summen aus zuzahlen. Für Versicherungen ist daher eine Einschätzung der Legitimität von Schadensansprüchen, essentiell. Problematisch sind hierbei die gezielten Betrugsversuche (= Fraud). Soweit sich das Risiko Betrugs erkennen lässt, können Gegenmaßnahmen eingeleitet werden. Die Abschätzung der Wahrscheinlichkeit, mit der ein Kunde einen Betrug versucht, ist hierbei essentiell. Darüber hinaus stellt sich die Frage, anhand welcher Merkmale Betrugsversuche zu erkennen sind. Mit dieser Demo kann erkannt werden ob ein Versicherungsvorfall ein Betrug ist oder nicht. "
- ]
- },
- {
- "attachments": {},
- "cell_type": "markdown",
- "id": "93491ec8",
- "metadata": {
- "editable": true,
- "slideshow": {
- "slide_type": ""
- },
- "tags": []
- },
- "source": [
- "# 2. Datenverständnis"
- ]
- },
- {
- "attachments": {},
- "cell_type": "markdown",
- "id": "a20002bf",
- "metadata": {
- "editable": true,
- "include": false,
- "paragraph": "DataUnderstanding",
- "slideshow": {
- "slide_type": ""
- },
- "tags": []
- },
- "source": [
- "Der verwendete Datensatz besteht aus 1000 Sätzen und hat 40 verschiedene sogenannte Features, d.h. gesammelte Datenkategorien. Das bedeutet, dass die Datenbasis nicht sehr groß ist, dafür sind die Möglichkeiten, verschiedene Merkmale zu untersuchen, umso größer. Es werden Informationen zu Versicherungsnehmern, Vertragsdaten zu Versicherungsnehmern und deren Kraftfahrzeugen, sowie Unfälle und die Höhe der Schäden angezeigt. Da der Datensatz so viele Merkmale enthält, werden sie und ihre Beschreibungen in der folgenden Tabelle erläutert. Die Zielvariable zeigt an, ob ein Versicherungsbetrug vorliegt (\"fraud_reported\"). Der Datensatz ist ein gutes Beispiel für Klassifizierungsmodelle, da es sich um eine binäre Zielvariable handelt (später auch als Ziel bezeichnet)."
- ]
- },
- {
- "cell_type": "markdown",
- "id": "0eda98ab-e58f-4df2-93ee-8cc621dcb4a1",
- "metadata": {
- "editable": true,
- "include": true,
- "paragraph": "Daten",
- "slideshow": {
- "slide_type": ""
- },
- "tags": []
- },
- "source": [
- "Der verwendete Datensatz besteht aus 1000 Sätzen und hat 40 verschiedene sogenannte Features, d.h. gesammelte Datenkategorien. Das bedeutet, dass die Datenbasis nicht sehr groß ist, dafür sind die Möglichkeiten, verschiedene Merkmale zu untersuchen, umso größer. Sie zeigt Informationen zu den Versicherungsnehmern, zu den Versicherungsdaten der Versicherten und ihrer Kraftfahrzeuge sowie zu Unfällen und Schadenshöhen. Die Zielvariable zeigt an, ob ein Versicherungsbetrug vorliegt (\"fraud_reported\"). Der Datensatz ist aufgrund der binären Zielvariable (später auch Ziel genannt) ein gutes Beispiel für Klassifikationsmodelle. "
- ]
- },
- {
- "attachments": {},
- "cell_type": "markdown",
- "id": "0c9b55e3",
- "metadata": {
- "editable": true,
- "slideshow": {
- "slide_type": ""
- },
- "tags": []
- },
- "source": [
- "## 2.1 Import von relevanten Modulen"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 1,
- "id": "cde07ec9",
- "metadata": {},
- "outputs": [],
- "source": [
- "import pandas as pd\n",
- "import numpy as np\n",
- "import matplotlib.pyplot as plt\n",
- "import seaborn as sns\n",
- "\n",
- "sns.set()\n",
- "\n",
- "%matplotlib inline"
- ]
- },
- {
- "attachments": {},
- "cell_type": "markdown",
- "id": "e6ea7872",
- "metadata": {},
- "source": [
- "## 2.2 Daten einlesen"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 2,
- "id": "a564973b",
- "metadata": {},
- "outputs": [],
- "source": [
- "raw_data = pd.read_csv('https://storage.googleapis.com/ml-service-repository-datastorage/Insurance_Fraud_detection_dataset.csv')"
- ]
- },
- {
- "attachments": {},
- "cell_type": "markdown",
- "id": "576e2df5",
- "metadata": {},
- "source": [
- "## 2.3 Deskriptive Analyse"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 3,
- "id": "91fb0caa",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "(1000, 40)"
- ]
- },
- "execution_count": 3,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "raw_data.shape"
- ]
- },
- {
- "attachments": {},
- "cell_type": "markdown",
- "id": "70f940c9",
- "metadata": {
- "editable": true,
- "slideshow": {
- "slide_type": ""
- },
- "tags": []
- },
- "source": [
- "--> 1000 samples and 40 columns"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "ff4755b1-f8db-4f01-a8fa-575d8f23bdd1",
- "metadata": {
- "editable": true,
- "include": true,
- "paragraph": "Datenvorbereitung",
- "slideshow": {
- "slide_type": ""
- },
- "tags": []
- },
- "source": [
- "Zunächst werden die Daten eingelesen und auf ihre Vollständigkeit überprüft. Danach werden die einzelnen Kundenmerkmale einer deskriptiven Analyse unterzogen. Damit lässt sich der Zusammenhang zur Zielvariable Betrug_Erkannt darstellen. In den Daten werden Untypische Daten ersetzt wie zum Beispiel „?“ mit „NaN“. Auf der Grundlage einer Korrelationsanalyse werden die Zusammenhänge zwischen Vorfalls-/Kundendaten und der Zielvariable untersucht. Merkmale die keinen Mehrwert bieten werden entfernt ( Alter, Adresse, usw.). Die Ausgewogenheit des Datensatzes in Bezug auf die Zielvariable wird grafisch dargestellt. Die Fälle in denen kein Betrug herscht machen etwa 75 % des gesamten Datensatzes aus, während Betrugsfälle etwa 25% ausmachen. Somit liegt ein unausgewogener Datensatz vor. Anschließend werden die kategorialen Werte umgewandelt (= Bildung von Dummy Variablen). Schließlich werden alle Kundenmerkmale auf ein gemeinsames Messniveau gebracht (= Standardisierung). Durch ein Undersampling wird die Unausgewogenheit des Datensatzes ausgeglichen. Abschließend werden Trainings- und Testdaten gebildet."
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 4,
- "id": "bb87fd51",
- "metadata": {
- "editable": true,
- "slideshow": {
- "slide_type": ""
- },
- "tags": []
- },
- "outputs": [
- {
- "data": {
- "text/html": [
- "<div>\n",
- "<style scoped>\n",
- " .dataframe tbody tr th:only-of-type {\n",
- " vertical-align: middle;\n",
- " }\n",
- "\n",
- " .dataframe tbody tr th {\n",
- " vertical-align: top;\n",
- " }\n",
- "\n",
- " .dataframe thead th {\n",
- " text-align: right;\n",
- " }\n",
- "</style>\n",
- "<table border=\"1\" class=\"dataframe\">\n",
- " <thead>\n",
- " <tr style=\"text-align: right;\">\n",
- " <th></th>\n",
- " <th>months_as_customer</th>\n",
- " <th>age</th>\n",
- " <th>policy_number</th>\n",
- " <th>policy_bind_date</th>\n",
- " <th>policy_state</th>\n",
- " <th>policy_csl</th>\n",
- " <th>policy_deductable</th>\n",
- " <th>policy_annual_premium</th>\n",
- " <th>umbrella_limit</th>\n",
- " <th>insured_zip</th>\n",
- " <th>...</th>\n",
- " <th>police_report_available</th>\n",
- " <th>total_claim_amount</th>\n",
- " <th>injury_claim</th>\n",
- " <th>property_claim</th>\n",
- " <th>vehicle_claim</th>\n",
- " <th>auto_make</th>\n",
- " <th>auto_model</th>\n",
- " <th>auto_year</th>\n",
- " <th>fraud_reported</th>\n",
- " <th>_c39</th>\n",
- " </tr>\n",
- " </thead>\n",
- " <tbody>\n",
- " <tr>\n",
- " <th>0</th>\n",
- " <td>328</td>\n",
- " <td>48</td>\n",
- " <td>521585</td>\n",
- " <td>2014-10-17</td>\n",
- " <td>OH</td>\n",
- " <td>250/500</td>\n",
- " <td>1000</td>\n",
- " <td>1406.91</td>\n",
- " <td>0</td>\n",
- " <td>466132</td>\n",
- " <td>...</td>\n",
- " <td>YES</td>\n",
- " <td>71610</td>\n",
- " <td>6510</td>\n",
- " <td>13020</td>\n",
- " <td>52080</td>\n",
- " <td>Saab</td>\n",
- " <td>92x</td>\n",
- " <td>2004</td>\n",
- " <td>Y</td>\n",
- " <td>NaN</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>1</th>\n",
- " <td>228</td>\n",
- " <td>42</td>\n",
- " <td>342868</td>\n",
- " <td>2006-06-27</td>\n",
- " <td>IN</td>\n",
- " <td>250/500</td>\n",
- " <td>2000</td>\n",
- " <td>1197.22</td>\n",
- " <td>5000000</td>\n",
- " <td>468176</td>\n",
- " <td>...</td>\n",
- " <td>?</td>\n",
- " <td>5070</td>\n",
- " <td>780</td>\n",
- " <td>780</td>\n",
- " <td>3510</td>\n",
- " <td>Mercedes</td>\n",
- " <td>E400</td>\n",
- " <td>2007</td>\n",
- " <td>Y</td>\n",
- " <td>NaN</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>2</th>\n",
- " <td>134</td>\n",
- " <td>29</td>\n",
- " <td>687698</td>\n",
- " <td>2000-09-06</td>\n",
- " <td>OH</td>\n",
- " <td>100/300</td>\n",
- " <td>2000</td>\n",
- " <td>1413.14</td>\n",
- " <td>5000000</td>\n",
- " <td>430632</td>\n",
- " <td>...</td>\n",
- " <td>NO</td>\n",
- " <td>34650</td>\n",
- " <td>7700</td>\n",
- " <td>3850</td>\n",
- " <td>23100</td>\n",
- " <td>Dodge</td>\n",
- " <td>RAM</td>\n",
- " <td>2007</td>\n",
- " <td>N</td>\n",
- " <td>NaN</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>3</th>\n",
- " <td>256</td>\n",
- " <td>41</td>\n",
- " <td>227811</td>\n",
- " <td>1990-05-25</td>\n",
- " <td>IL</td>\n",
- " <td>250/500</td>\n",
- " <td>2000</td>\n",
- " <td>1415.74</td>\n",
- " <td>6000000</td>\n",
- " <td>608117</td>\n",
- " <td>...</td>\n",
- " <td>NO</td>\n",
- " <td>63400</td>\n",
- " <td>6340</td>\n",
- " <td>6340</td>\n",
- " <td>50720</td>\n",
- " <td>Chevrolet</td>\n",
- " <td>Tahoe</td>\n",
- " <td>2014</td>\n",
- " <td>Y</td>\n",
- " <td>NaN</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>4</th>\n",
- " <td>228</td>\n",
- " <td>44</td>\n",
- " <td>367455</td>\n",
- " <td>2014-06-06</td>\n",
- " <td>IL</td>\n",
- " <td>500/1000</td>\n",
- " <td>1000</td>\n",
- " <td>1583.91</td>\n",
- " <td>6000000</td>\n",
- " <td>610706</td>\n",
- " <td>...</td>\n",
- " <td>NO</td>\n",
- " <td>6500</td>\n",
- " <td>1300</td>\n",
- " <td>650</td>\n",
- " <td>4550</td>\n",
- " <td>Accura</td>\n",
- " <td>RSX</td>\n",
- " <td>2009</td>\n",
- " <td>N</td>\n",
- " <td>NaN</td>\n",
- " </tr>\n",
- " </tbody>\n",
- "</table>\n",
- "<p>5 rows × 40 columns</p>\n",
- "</div>"
- ],
- "text/plain": [
- " months_as_customer age policy_number policy_bind_date policy_state \\\n",
- "0 328 48 521585 2014-10-17 OH \n",
- "1 228 42 342868 2006-06-27 IN \n",
- "2 134 29 687698 2000-09-06 OH \n",
- "3 256 41 227811 1990-05-25 IL \n",
- "4 228 44 367455 2014-06-06 IL \n",
- "\n",
- " policy_csl policy_deductable policy_annual_premium umbrella_limit \\\n",
- "0 250/500 1000 1406.91 0 \n",
- "1 250/500 2000 1197.22 5000000 \n",
- "2 100/300 2000 1413.14 5000000 \n",
- "3 250/500 2000 1415.74 6000000 \n",
- "4 500/1000 1000 1583.91 6000000 \n",
- "\n",
- " insured_zip ... police_report_available total_claim_amount injury_claim \\\n",
- "0 466132 ... YES 71610 6510 \n",
- "1 468176 ... ? 5070 780 \n",
- "2 430632 ... NO 34650 7700 \n",
- "3 608117 ... NO 63400 6340 \n",
- "4 610706 ... NO 6500 1300 \n",
- "\n",
- " property_claim vehicle_claim auto_make auto_model auto_year \\\n",
- "0 13020 52080 Saab 92x 2004 \n",
- "1 780 3510 Mercedes E400 2007 \n",
- "2 3850 23100 Dodge RAM 2007 \n",
- "3 6340 50720 Chevrolet Tahoe 2014 \n",
- "4 650 4550 Accura RSX 2009 \n",
- "\n",
- " fraud_reported _c39 \n",
- "0 Y NaN \n",
- "1 Y NaN \n",
- "2 N NaN \n",
- "3 Y NaN \n",
- "4 N NaN \n",
- "\n",
- "[5 rows x 40 columns]"
- ]
- },
- "execution_count": 4,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "raw_data.head()"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 5,
- "id": "09bf37ac",
- "metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "<class 'pandas.core.frame.DataFrame'>\n",
- "RangeIndex: 1000 entries, 0 to 999\n",
- "Data columns (total 40 columns):\n",
- " # Column Non-Null Count Dtype \n",
- "--- ------ -------------- ----- \n",
- " 0 months_as_customer 1000 non-null int64 \n",
- " 1 age 1000 non-null int64 \n",
- " 2 policy_number 1000 non-null int64 \n",
- " 3 policy_bind_date 1000 non-null object \n",
- " 4 policy_state 1000 non-null object \n",
- " 5 policy_csl 1000 non-null object \n",
- " 6 policy_deductable 1000 non-null int64 \n",
- " 7 policy_annual_premium 1000 non-null float64\n",
- " 8 umbrella_limit 1000 non-null int64 \n",
- " 9 insured_zip 1000 non-null int64 \n",
- " 10 insured_sex 1000 non-null object \n",
- " 11 insured_education_level 1000 non-null object \n",
- " 12 insured_occupation 1000 non-null object \n",
- " 13 insured_hobbies 1000 non-null object \n",
- " 14 insured_relationship 1000 non-null object \n",
- " 15 capital-gains 1000 non-null int64 \n",
- " 16 capital-loss 1000 non-null int64 \n",
- " 17 incident_date 1000 non-null object \n",
- " 18 incident_type 1000 non-null object \n",
- " 19 collision_type 1000 non-null object \n",
- " 20 incident_severity 1000 non-null object \n",
- " 21 authorities_contacted 1000 non-null object \n",
- " 22 incident_state 1000 non-null object \n",
- " 23 incident_city 1000 non-null object \n",
- " 24 incident_location 1000 non-null object \n",
- " 25 incident_hour_of_the_day 1000 non-null int64 \n",
- " 26 number_of_vehicles_involved 1000 non-null int64 \n",
- " 27 property_damage 1000 non-null object \n",
- " 28 bodily_injuries 1000 non-null int64 \n",
- " 29 witnesses 1000 non-null int64 \n",
- " 30 police_report_available 1000 non-null object \n",
- " 31 total_claim_amount 1000 non-null int64 \n",
- " 32 injury_claim 1000 non-null int64 \n",
- " 33 property_claim 1000 non-null int64 \n",
- " 34 vehicle_claim 1000 non-null int64 \n",
- " 35 auto_make 1000 non-null object \n",
- " 36 auto_model 1000 non-null object \n",
- " 37 auto_year 1000 non-null int64 \n",
- " 38 fraud_reported 1000 non-null object \n",
- " 39 _c39 0 non-null float64\n",
- "dtypes: float64(2), int64(17), object(21)\n",
- "memory usage: 312.6+ KB\n"
- ]
- }
- ],
- "source": [
- "raw_data.info()"
- ]
- },
- {
- "attachments": {},
- "cell_type": "markdown",
- "id": "5c7fd07d",
- "metadata": {},
- "source": [
- "# 3. Datenaufbereitung"
- ]
- },
- {
- "attachments": {},
- "cell_type": "markdown",
- "id": "05609119",
- "metadata": {},
- "source": [
- "## 3.1 Datenbereinigung"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 6,
- "id": "66d09246",
- "metadata": {},
- "outputs": [],
- "source": [
- "# replace \"?\" with \"NaN\" in the dataset\n",
- "raw_data.replace('?', np.nan, inplace=True)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 7,
- "id": "edcc3b6f",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "months_as_customer 0\n",
- "age 0\n",
- "policy_number 0\n",
- "policy_bind_date 0\n",
- "policy_state 0\n",
- "policy_csl 0\n",
- "policy_deductable 0\n",
- "policy_annual_premium 0\n",
- "umbrella_limit 0\n",
- "insured_zip 0\n",
- "insured_sex 0\n",
- "insured_education_level 0\n",
- "insured_occupation 0\n",
- "insured_hobbies 0\n",
- "insured_relationship 0\n",
- "capital-gains 0\n",
- "capital-loss 0\n",
- "incident_date 0\n",
- "incident_type 0\n",
- "collision_type 178\n",
- "incident_severity 0\n",
- "authorities_contacted 0\n",
- "incident_state 0\n",
- "incident_city 0\n",
- "incident_location 0\n",
- "incident_hour_of_the_day 0\n",
- "number_of_vehicles_involved 0\n",
- "property_damage 360\n",
- "bodily_injuries 0\n",
- "witnesses 0\n",
- "police_report_available 343\n",
- "total_claim_amount 0\n",
- "injury_claim 0\n",
- "property_claim 0\n",
- "vehicle_claim 0\n",
- "auto_make 0\n",
- "auto_model 0\n",
- "auto_year 0\n",
- "fraud_reported 0\n",
- "_c39 1000\n",
- "dtype: int64"
- ]
- },
- "execution_count": 7,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "# checking missing values\n",
- "raw_data.isnull().sum()"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 8,
- "id": "b71a2875",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "(1000, 39)"
- ]
- },
- "execution_count": 8,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "# delete column _c39, no relevant feature\n",
- "data_no_mv = raw_data.drop('_c39', axis=1)\n",
- "data_no_mv.shape"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 9,
- "id": "22813b8d",
- "metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Rear Collision\n",
- "NO\n",
- "NO\n"
- ]
- }
- ],
- "source": [
- "# since there are relatively few records anyway, the zero values are replaced by the mean value of the respective column\n",
- "print(data_no_mv['collision_type'].mode()[0])\n",
- "print(data_no_mv['property_damage'].mode()[0])\n",
- "print(data_no_mv['police_report_available'].mode()[0])"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 10,
- "id": "7f41969b",
- "metadata": {},
- "outputs": [],
- "source": [
- "data_no_mv['collision_type'] = data_no_mv['collision_type'].fillna(data_no_mv['collision_type'].mode()[0])\n",
- "data_no_mv['property_damage'] = data_no_mv['property_damage'].fillna(data_no_mv['property_damage'].mode()[0])\n",
- "data_no_mv['police_report_available'] = data_no_mv['police_report_available'].fillna(data_no_mv['police_report_available'].mode()[0])"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 11,
- "id": "3c714633",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "months_as_customer 0\n",
- "age 0\n",
- "policy_number 0\n",
- "policy_bind_date 0\n",
- "policy_state 0\n",
- "policy_csl 0\n",
- "policy_deductable 0\n",
- "policy_annual_premium 0\n",
- "umbrella_limit 0\n",
- "insured_zip 0\n",
- "insured_sex 0\n",
- "insured_education_level 0\n",
- "insured_occupation 0\n",
- "insured_hobbies 0\n",
- "insured_relationship 0\n",
- "capital-gains 0\n",
- "capital-loss 0\n",
- "incident_date 0\n",
- "incident_type 0\n",
- "collision_type 0\n",
- "incident_severity 0\n",
- "authorities_contacted 0\n",
- "incident_state 0\n",
- "incident_city 0\n",
- "incident_location 0\n",
- "incident_hour_of_the_day 0\n",
- "number_of_vehicles_involved 0\n",
- "property_damage 0\n",
- "bodily_injuries 0\n",
- "witnesses 0\n",
- "police_report_available 0\n",
- "total_claim_amount 0\n",
- "injury_claim 0\n",
- "property_claim 0\n",
- "vehicle_claim 0\n",
- "auto_make 0\n",
- "auto_model 0\n",
- "auto_year 0\n",
- "fraud_reported 0\n",
- "dtype: int64"
- ]
- },
- "execution_count": 11,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "# checking missing values\n",
- "data_no_mv.isnull().sum()"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 12,
- "id": "8cc2f1b2",
- "metadata": {},
- "outputs": [],
- "source": [
- "# checking duplicates\n",
- "data_no_dup = data_no_mv.copy()\n",
- "data_no_dup['policy_number'] = data_no_dup['policy_number'].duplicated(keep=False)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 13,
- "id": "e666f85b",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "(1000, 39)"
- ]
- },
- "execution_count": 13,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "data_no_dup.shape"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 14,
- "id": "005fa322",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/html": [
- "<div>\n",
- "<style scoped>\n",
- " .dataframe tbody tr th:only-of-type {\n",
- " vertical-align: middle;\n",
- " }\n",
- "\n",
- " .dataframe tbody tr th {\n",
- " vertical-align: top;\n",
- " }\n",
- "\n",
- " .dataframe thead th {\n",
- " text-align: right;\n",
- " }\n",
- "</style>\n",
- "<table border=\"1\" class=\"dataframe\">\n",
- " <thead>\n",
- " <tr style=\"text-align: right;\">\n",
- " <th></th>\n",
- " <th>months_as_customer</th>\n",
- " <th>age</th>\n",
- " <th>policy_number</th>\n",
- " <th>policy_bind_date</th>\n",
- " <th>policy_state</th>\n",
- " <th>policy_csl</th>\n",
- " <th>policy_deductable</th>\n",
- " <th>policy_annual_premium</th>\n",
- " <th>umbrella_limit</th>\n",
- " <th>insured_zip</th>\n",
- " <th>...</th>\n",
- " <th>witnesses</th>\n",
- " <th>police_report_available</th>\n",
- " <th>total_claim_amount</th>\n",
- " <th>injury_claim</th>\n",
- " <th>property_claim</th>\n",
- " <th>vehicle_claim</th>\n",
- " <th>auto_make</th>\n",
- " <th>auto_model</th>\n",
- " <th>auto_year</th>\n",
- " <th>fraud_reported</th>\n",
- " </tr>\n",
- " </thead>\n",
- " <tbody>\n",
- " <tr>\n",
- " <th>count</th>\n",
- " <td>1000.000000</td>\n",
- " <td>1000.000000</td>\n",
- " <td>1000</td>\n",
- " <td>1000</td>\n",
- " <td>1000</td>\n",
- " <td>1000</td>\n",
- " <td>1000.000000</td>\n",
- " <td>1000.000000</td>\n",
- " <td>1.000000e+03</td>\n",
- " <td>1000.000000</td>\n",
- " <td>...</td>\n",
- " <td>1000.000000</td>\n",
- " <td>1000</td>\n",
- " <td>1000.00000</td>\n",
- " <td>1000.000000</td>\n",
- " <td>1000.000000</td>\n",
- " <td>1000.000000</td>\n",
- " <td>1000</td>\n",
- " <td>1000</td>\n",
- " <td>1000.000000</td>\n",
- " <td>1000</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>unique</th>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>1</td>\n",
- " <td>951</td>\n",
- " <td>3</td>\n",
- " <td>3</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>...</td>\n",
- " <td>NaN</td>\n",
- " <td>2</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>14</td>\n",
- " <td>39</td>\n",
- " <td>NaN</td>\n",
- " <td>2</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>top</th>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>False</td>\n",
- " <td>2006-01-01</td>\n",
- " <td>OH</td>\n",
- " <td>250/500</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>...</td>\n",
- " <td>NaN</td>\n",
- " <td>NO</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>Dodge</td>\n",
- " <td>RAM</td>\n",
- " <td>NaN</td>\n",
- " <td>N</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>freq</th>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>1000</td>\n",
- " <td>3</td>\n",
- " <td>352</td>\n",
- " <td>351</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>...</td>\n",
- " <td>NaN</td>\n",
- " <td>686</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>80</td>\n",
- " <td>43</td>\n",
- " <td>NaN</td>\n",
- " <td>753</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>mean</th>\n",
- " <td>203.954000</td>\n",
- " <td>38.948000</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>1136.000000</td>\n",
- " <td>1256.406150</td>\n",
- " <td>1.101000e+06</td>\n",
- " <td>501214.488000</td>\n",
- " <td>...</td>\n",
- " <td>1.487000</td>\n",
- " <td>NaN</td>\n",
- " <td>52761.94000</td>\n",
- " <td>7433.420000</td>\n",
- " <td>7399.570000</td>\n",
- " <td>37928.950000</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>2005.103000</td>\n",
- " <td>NaN</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>std</th>\n",
- " <td>115.113174</td>\n",
- " <td>9.140287</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>611.864673</td>\n",
- " <td>244.167395</td>\n",
- " <td>2.297407e+06</td>\n",
- " <td>71701.610941</td>\n",
- " <td>...</td>\n",
- " <td>1.111335</td>\n",
- " <td>NaN</td>\n",
- " <td>26401.53319</td>\n",
- " <td>4880.951853</td>\n",
- " <td>4824.726179</td>\n",
- " <td>18886.252893</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>6.015861</td>\n",
- " <td>NaN</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>min</th>\n",
- " <td>0.000000</td>\n",
- " <td>19.000000</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>500.000000</td>\n",
- " <td>433.330000</td>\n",
- " <td>-1.000000e+06</td>\n",
- " <td>430104.000000</td>\n",
- " <td>...</td>\n",
- " <td>0.000000</td>\n",
- " <td>NaN</td>\n",
- " <td>100.00000</td>\n",
- " <td>0.000000</td>\n",
- " <td>0.000000</td>\n",
- " <td>70.000000</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>1995.000000</td>\n",
- " <td>NaN</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>25%</th>\n",
- " <td>115.750000</td>\n",
- " <td>32.000000</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>500.000000</td>\n",
- " <td>1089.607500</td>\n",
- " <td>0.000000e+00</td>\n",
- " <td>448404.500000</td>\n",
- " <td>...</td>\n",
- " <td>1.000000</td>\n",
- " <td>NaN</td>\n",
- " <td>41812.50000</td>\n",
- " <td>4295.000000</td>\n",
- " <td>4445.000000</td>\n",
- " <td>30292.500000</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>2000.000000</td>\n",
- " <td>NaN</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>50%</th>\n",
- " <td>199.500000</td>\n",
- " <td>38.000000</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>1000.000000</td>\n",
- " <td>1257.200000</td>\n",
- " <td>0.000000e+00</td>\n",
- " <td>466445.500000</td>\n",
- " <td>...</td>\n",
- " <td>1.000000</td>\n",
- " <td>NaN</td>\n",
- " <td>58055.00000</td>\n",
- " <td>6775.000000</td>\n",
- " <td>6750.000000</td>\n",
- " <td>42100.000000</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>2005.000000</td>\n",
- " <td>NaN</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>75%</th>\n",
- " <td>276.250000</td>\n",
- " <td>44.000000</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>2000.000000</td>\n",
- " <td>1415.695000</td>\n",
- " <td>0.000000e+00</td>\n",
- " <td>603251.000000</td>\n",
- " <td>...</td>\n",
- " <td>2.000000</td>\n",
- " <td>NaN</td>\n",
- " <td>70592.50000</td>\n",
- " <td>11305.000000</td>\n",
- " <td>10885.000000</td>\n",
- " <td>50822.500000</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>2010.000000</td>\n",
- " <td>NaN</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>max</th>\n",
- " <td>479.000000</td>\n",
- " <td>64.000000</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>2000.000000</td>\n",
- " <td>2047.590000</td>\n",
- " <td>1.000000e+07</td>\n",
- " <td>620962.000000</td>\n",
- " <td>...</td>\n",
- " <td>3.000000</td>\n",
- " <td>NaN</td>\n",
- " <td>114920.00000</td>\n",
- " <td>21450.000000</td>\n",
- " <td>23670.000000</td>\n",
- " <td>79560.000000</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>2015.000000</td>\n",
- " <td>NaN</td>\n",
- " </tr>\n",
- " </tbody>\n",
- "</table>\n",
- "<p>11 rows × 39 columns</p>\n",
- "</div>"
- ],
- "text/plain": [
- " months_as_customer age policy_number policy_bind_date \\\n",
- "count 1000.000000 1000.000000 1000 1000 \n",
- "unique NaN NaN 1 951 \n",
- "top NaN NaN False 2006-01-01 \n",
- "freq NaN NaN 1000 3 \n",
- "mean 203.954000 38.948000 NaN NaN \n",
- "std 115.113174 9.140287 NaN NaN \n",
- "min 0.000000 19.000000 NaN NaN \n",
- "25% 115.750000 32.000000 NaN NaN \n",
- "50% 199.500000 38.000000 NaN NaN \n",
- "75% 276.250000 44.000000 NaN NaN \n",
- "max 479.000000 64.000000 NaN NaN \n",
- "\n",
- " policy_state policy_csl policy_deductable policy_annual_premium \\\n",
- "count 1000 1000 1000.000000 1000.000000 \n",
- "unique 3 3 NaN NaN \n",
- "top OH 250/500 NaN NaN \n",
- "freq 352 351 NaN NaN \n",
- "mean NaN NaN 1136.000000 1256.406150 \n",
- "std NaN NaN 611.864673 244.167395 \n",
- "min NaN NaN 500.000000 433.330000 \n",
- "25% NaN NaN 500.000000 1089.607500 \n",
- "50% NaN NaN 1000.000000 1257.200000 \n",
- "75% NaN NaN 2000.000000 1415.695000 \n",
- "max NaN NaN 2000.000000 2047.590000 \n",
- "\n",
- " umbrella_limit insured_zip ... witnesses \\\n",
- "count 1.000000e+03 1000.000000 ... 1000.000000 \n",
- "unique NaN NaN ... NaN \n",
- "top NaN NaN ... NaN \n",
- "freq NaN NaN ... NaN \n",
- "mean 1.101000e+06 501214.488000 ... 1.487000 \n",
- "std 2.297407e+06 71701.610941 ... 1.111335 \n",
- "min -1.000000e+06 430104.000000 ... 0.000000 \n",
- "25% 0.000000e+00 448404.500000 ... 1.000000 \n",
- "50% 0.000000e+00 466445.500000 ... 1.000000 \n",
- "75% 0.000000e+00 603251.000000 ... 2.000000 \n",
- "max 1.000000e+07 620962.000000 ... 3.000000 \n",
- "\n",
- " police_report_available total_claim_amount injury_claim \\\n",
- "count 1000 1000.00000 1000.000000 \n",
- "unique 2 NaN NaN \n",
- "top NO NaN NaN \n",
- "freq 686 NaN NaN \n",
- "mean NaN 52761.94000 7433.420000 \n",
- "std NaN 26401.53319 4880.951853 \n",
- "min NaN 100.00000 0.000000 \n",
- "25% NaN 41812.50000 4295.000000 \n",
- "50% NaN 58055.00000 6775.000000 \n",
- "75% NaN 70592.50000 11305.000000 \n",
- "max NaN 114920.00000 21450.000000 \n",
- "\n",
- " property_claim vehicle_claim auto_make auto_model auto_year \\\n",
- "count 1000.000000 1000.000000 1000 1000 1000.000000 \n",
- "unique NaN NaN 14 39 NaN \n",
- "top NaN NaN Dodge RAM NaN \n",
- "freq NaN NaN 80 43 NaN \n",
- "mean 7399.570000 37928.950000 NaN NaN 2005.103000 \n",
- "std 4824.726179 18886.252893 NaN NaN 6.015861 \n",
- "min 0.000000 70.000000 NaN NaN 1995.000000 \n",
- "25% 4445.000000 30292.500000 NaN NaN 2000.000000 \n",
- "50% 6750.000000 42100.000000 NaN NaN 2005.000000 \n",
- "75% 10885.000000 50822.500000 NaN NaN 2010.000000 \n",
- "max 23670.000000 79560.000000 NaN NaN 2015.000000 \n",
- "\n",
- " fraud_reported \n",
- "count 1000 \n",
- "unique 2 \n",
- "top N \n",
- "freq 753 \n",
- "mean NaN \n",
- "std NaN \n",
- "min NaN \n",
- "25% NaN \n",
- "50% NaN \n",
- "75% NaN \n",
- "max NaN \n",
- "\n",
- "[11 rows x 39 columns]"
- ]
- },
- "execution_count": 14,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "data_no_dup.describe(include='all')"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 15,
- "id": "87b9ba9d",
- "metadata": {},
- "outputs": [],
- "source": [
- "# map the target from \"no\" and \"yes\" to 0 and 1\n",
- "data_no_dup.fraud_reported = data_no_dup.fraud_reported.map({'N': 0, 'Y': 1})"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 16,
- "id": "786725ec",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/html": [
- "<div>\n",
- "<style scoped>\n",
- " .dataframe tbody tr th:only-of-type {\n",
- " vertical-align: middle;\n",
- " }\n",
- "\n",
- " .dataframe tbody tr th {\n",
- " vertical-align: top;\n",
- " }\n",
- "\n",
- " .dataframe thead th {\n",
- " text-align: right;\n",
- " }\n",
- "</style>\n",
- "<table border=\"1\" class=\"dataframe\">\n",
- " <thead>\n",
- " <tr style=\"text-align: right;\">\n",
- " <th></th>\n",
- " <th>months_as_customer</th>\n",
- " <th>age</th>\n",
- " <th>policy_number</th>\n",
- " <th>policy_bind_date</th>\n",
- " <th>policy_state</th>\n",
- " <th>policy_csl</th>\n",
- " <th>policy_deductable</th>\n",
- " <th>policy_annual_premium</th>\n",
- " <th>umbrella_limit</th>\n",
- " <th>insured_zip</th>\n",
- " <th>...</th>\n",
- " <th>witnesses</th>\n",
- " <th>police_report_available</th>\n",
- " <th>total_claim_amount</th>\n",
- " <th>injury_claim</th>\n",
- " <th>property_claim</th>\n",
- " <th>vehicle_claim</th>\n",
- " <th>auto_make</th>\n",
- " <th>auto_model</th>\n",
- " <th>auto_year</th>\n",
- " <th>fraud_reported</th>\n",
- " </tr>\n",
- " </thead>\n",
- " <tbody>\n",
- " <tr>\n",
- " <th>0</th>\n",
- " <td>328</td>\n",
- " <td>48</td>\n",
- " <td>False</td>\n",
- " <td>2014-10-17</td>\n",
- " <td>OH</td>\n",
- " <td>250/500</td>\n",
- " <td>1000</td>\n",
- " <td>1406.91</td>\n",
- " <td>0</td>\n",
- " <td>466132</td>\n",
- " <td>...</td>\n",
- " <td>2</td>\n",
- " <td>YES</td>\n",
- " <td>71610</td>\n",
- " <td>6510</td>\n",
- " <td>13020</td>\n",
- " <td>52080</td>\n",
- " <td>Saab</td>\n",
- " <td>92x</td>\n",
- " <td>2004</td>\n",
- " <td>1</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>1</th>\n",
- " <td>228</td>\n",
- " <td>42</td>\n",
- " <td>False</td>\n",
- " <td>2006-06-27</td>\n",
- " <td>IN</td>\n",
- " <td>250/500</td>\n",
- " <td>2000</td>\n",
- " <td>1197.22</td>\n",
- " <td>5000000</td>\n",
- " <td>468176</td>\n",
- " <td>...</td>\n",
- " <td>0</td>\n",
- " <td>NO</td>\n",
- " <td>5070</td>\n",
- " <td>780</td>\n",
- " <td>780</td>\n",
- " <td>3510</td>\n",
- " <td>Mercedes</td>\n",
- " <td>E400</td>\n",
- " <td>2007</td>\n",
- " <td>1</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>2</th>\n",
- " <td>134</td>\n",
- " <td>29</td>\n",
- " <td>False</td>\n",
- " <td>2000-09-06</td>\n",
- " <td>OH</td>\n",
- " <td>100/300</td>\n",
- " <td>2000</td>\n",
- " <td>1413.14</td>\n",
- " <td>5000000</td>\n",
- " <td>430632</td>\n",
- " <td>...</td>\n",
- " <td>3</td>\n",
- " <td>NO</td>\n",
- " <td>34650</td>\n",
- " <td>7700</td>\n",
- " <td>3850</td>\n",
- " <td>23100</td>\n",
- " <td>Dodge</td>\n",
- " <td>RAM</td>\n",
- " <td>2007</td>\n",
- " <td>0</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>3</th>\n",
- " <td>256</td>\n",
- " <td>41</td>\n",
- " <td>False</td>\n",
- " <td>1990-05-25</td>\n",
- " <td>IL</td>\n",
- " <td>250/500</td>\n",
- " <td>2000</td>\n",
- " <td>1415.74</td>\n",
- " <td>6000000</td>\n",
- " <td>608117</td>\n",
- " <td>...</td>\n",
- " <td>2</td>\n",
- " <td>NO</td>\n",
- " <td>63400</td>\n",
- " <td>6340</td>\n",
- " <td>6340</td>\n",
- " <td>50720</td>\n",
- " <td>Chevrolet</td>\n",
- " <td>Tahoe</td>\n",
- " <td>2014</td>\n",
- " <td>1</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>4</th>\n",
- " <td>228</td>\n",
- " <td>44</td>\n",
- " <td>False</td>\n",
- " <td>2014-06-06</td>\n",
- " <td>IL</td>\n",
- " <td>500/1000</td>\n",
- " <td>1000</td>\n",
- " <td>1583.91</td>\n",
- " <td>6000000</td>\n",
- " <td>610706</td>\n",
- " <td>...</td>\n",
- " <td>1</td>\n",
- " <td>NO</td>\n",
- " <td>6500</td>\n",
- " <td>1300</td>\n",
- " <td>650</td>\n",
- " <td>4550</td>\n",
- " <td>Accura</td>\n",
- " <td>RSX</td>\n",
- " <td>2009</td>\n",
- " <td>0</td>\n",
- " </tr>\n",
- " </tbody>\n",
- "</table>\n",
- "<p>5 rows × 39 columns</p>\n",
- "</div>"
- ],
- "text/plain": [
- " months_as_customer age policy_number policy_bind_date policy_state \\\n",
- "0 328 48 False 2014-10-17 OH \n",
- "1 228 42 False 2006-06-27 IN \n",
- "2 134 29 False 2000-09-06 OH \n",
- "3 256 41 False 1990-05-25 IL \n",
- "4 228 44 False 2014-06-06 IL \n",
- "\n",
- " policy_csl policy_deductable policy_annual_premium umbrella_limit \\\n",
- "0 250/500 1000 1406.91 0 \n",
- "1 250/500 2000 1197.22 5000000 \n",
- "2 100/300 2000 1413.14 5000000 \n",
- "3 250/500 2000 1415.74 6000000 \n",
- "4 500/1000 1000 1583.91 6000000 \n",
- "\n",
- " insured_zip ... witnesses police_report_available total_claim_amount \\\n",
- "0 466132 ... 2 YES 71610 \n",
- "1 468176 ... 0 NO 5070 \n",
- "2 430632 ... 3 NO 34650 \n",
- "3 608117 ... 2 NO 63400 \n",
- "4 610706 ... 1 NO 6500 \n",
- "\n",
- " injury_claim property_claim vehicle_claim auto_make auto_model auto_year \\\n",
- "0 6510 13020 52080 Saab 92x 2004 \n",
- "1 780 780 3510 Mercedes E400 2007 \n",
- "2 7700 3850 23100 Dodge RAM 2007 \n",
- "3 6340 6340 50720 Chevrolet Tahoe 2014 \n",
- "4 1300 650 4550 Accura RSX 2009 \n",
- "\n",
- " fraud_reported \n",
- "0 1 \n",
- "1 1 \n",
- "2 0 \n",
- "3 1 \n",
- "4 0 \n",
- "\n",
- "[5 rows x 39 columns]"
- ]
- },
- "execution_count": 16,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "data_no_dup.head()"
- ]
- },
- {
- "attachments": {},
- "cell_type": "markdown",
- "id": "7a3ad030",
- "metadata": {},
- "source": [
- "## 3.2 Test auf Korrelation"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 17,
- "id": "a8491fbe",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "image/png": "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",
- "text/plain": [
- "<Figure size 1296x864 with 2 Axes>"
- ]
- },
- "metadata": {
- "needs_background": "light"
- },
- "output_type": "display_data"
- }
- ],
- "source": [
- "# checking correlation\n",
- "plt.figure(figsize=(18, 12))\n",
- "\n",
- "feature_corr = data_no_dup.corr()\n",
- "mask = np.triu(np.ones_like(feature_corr, dtype = bool))\n",
- "sns.heatmap(feature_corr, mask=mask, annot=True, cmap='coolwarm')\n",
- "\n",
- "plt.show()"
- ]
- },
- {
- "attachments": {},
- "cell_type": "markdown",
- "id": "410675dc",
- "metadata": {},
- "source": [
- "--> hohe Korrelation zwischen *Alter* und *Monate_als_Kunde*, auch zwischen *Gesamtschadenshöhe* und *Verletzungsschaden, Sachschaden und Fahrzeugschaden*. "
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 18,
- "id": "7826e0db",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "(1000, 29)"
- ]
- },
- "execution_count": 18,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "# drop features because of the correlation \n",
- "cols_to_drop = ['age', 'total_claim_amount', 'policy_number', 'policy_bind_date', 'policy_state', \n",
- " 'incident_state', 'incident_city', 'incident_location', 'incident_hour_of_the_day', \n",
- " 'insured_zip'] \n",
- "data01 = data_no_dup.drop(cols_to_drop, axis=1)\n",
- "data01.shape"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 19,
- "id": "6faa309f",
- "metadata": {},
- "outputs": [],
- "source": [
- "# seperate numeric and categorical feature\n",
- "data_cat = data01.select_dtypes(include='object')\n",
- "data_num = data01.select_dtypes(exclude='object')"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 20,
- "id": "f9c6db85",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/html": [
- "<div>\n",
- "<style scoped>\n",
- " .dataframe tbody tr th:only-of-type {\n",
- " vertical-align: middle;\n",
- " }\n",
- "\n",
- " .dataframe tbody tr th {\n",
- " vertical-align: top;\n",
- " }\n",
- "\n",
- " .dataframe thead th {\n",
- " text-align: right;\n",
- " }\n",
- "</style>\n",
- "<table border=\"1\" class=\"dataframe\">\n",
- " <thead>\n",
- " <tr style=\"text-align: right;\">\n",
- " <th></th>\n",
- " <th>policy_csl</th>\n",
- " <th>insured_sex</th>\n",
- " <th>insured_education_level</th>\n",
- " <th>insured_occupation</th>\n",
- " <th>insured_hobbies</th>\n",
- " <th>insured_relationship</th>\n",
- " <th>incident_date</th>\n",
- " <th>incident_type</th>\n",
- " <th>collision_type</th>\n",
- " <th>incident_severity</th>\n",
- " <th>authorities_contacted</th>\n",
- " <th>property_damage</th>\n",
- " <th>police_report_available</th>\n",
- " <th>auto_make</th>\n",
- " <th>auto_model</th>\n",
- " </tr>\n",
- " </thead>\n",
- " <tbody>\n",
- " <tr>\n",
- " <th>0</th>\n",
- " <td>250/500</td>\n",
- " <td>MALE</td>\n",
- " <td>MD</td>\n",
- " <td>craft-repair</td>\n",
- " <td>sleeping</td>\n",
- " <td>husband</td>\n",
- " <td>2015-01-25</td>\n",
- " <td>Single Vehicle Collision</td>\n",
- " <td>Side Collision</td>\n",
- " <td>Major Damage</td>\n",
- " <td>Police</td>\n",
- " <td>YES</td>\n",
- " <td>YES</td>\n",
- " <td>Saab</td>\n",
- " <td>92x</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>1</th>\n",
- " <td>250/500</td>\n",
- " <td>MALE</td>\n",
- " <td>MD</td>\n",
- " <td>machine-op-inspct</td>\n",
- " <td>reading</td>\n",
- " <td>other-relative</td>\n",
- " <td>2015-01-21</td>\n",
- " <td>Vehicle Theft</td>\n",
- " <td>Rear Collision</td>\n",
- " <td>Minor Damage</td>\n",
- " <td>Police</td>\n",
- " <td>NO</td>\n",
- " <td>NO</td>\n",
- " <td>Mercedes</td>\n",
- " <td>E400</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>2</th>\n",
- " <td>100/300</td>\n",
- " <td>FEMALE</td>\n",
- " <td>PhD</td>\n",
- " <td>sales</td>\n",
- " <td>board-games</td>\n",
- " <td>own-child</td>\n",
- " <td>2015-02-22</td>\n",
- " <td>Multi-vehicle Collision</td>\n",
- " <td>Rear Collision</td>\n",
- " <td>Minor Damage</td>\n",
- " <td>Police</td>\n",
- " <td>NO</td>\n",
- " <td>NO</td>\n",
- " <td>Dodge</td>\n",
- " <td>RAM</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>3</th>\n",
- " <td>250/500</td>\n",
- " <td>FEMALE</td>\n",
- " <td>PhD</td>\n",
- " <td>armed-forces</td>\n",
- " <td>board-games</td>\n",
- " <td>unmarried</td>\n",
- " <td>2015-01-10</td>\n",
- " <td>Single Vehicle Collision</td>\n",
- " <td>Front Collision</td>\n",
- " <td>Major Damage</td>\n",
- " <td>Police</td>\n",
- " <td>NO</td>\n",
- " <td>NO</td>\n",
- " <td>Chevrolet</td>\n",
- " <td>Tahoe</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>4</th>\n",
- " <td>500/1000</td>\n",
- " <td>MALE</td>\n",
- " <td>Associate</td>\n",
- " <td>sales</td>\n",
- " <td>board-games</td>\n",
- " <td>unmarried</td>\n",
- " <td>2015-02-17</td>\n",
- " <td>Vehicle Theft</td>\n",
- " <td>Rear Collision</td>\n",
- " <td>Minor Damage</td>\n",
- " <td>None</td>\n",
- " <td>NO</td>\n",
- " <td>NO</td>\n",
- " <td>Accura</td>\n",
- " <td>RSX</td>\n",
- " </tr>\n",
- " </tbody>\n",
- "</table>\n",
- "</div>"
- ],
- "text/plain": [
- " policy_csl insured_sex insured_education_level insured_occupation \\\n",
- "0 250/500 MALE MD craft-repair \n",
- "1 250/500 MALE MD machine-op-inspct \n",
- "2 100/300 FEMALE PhD sales \n",
- "3 250/500 FEMALE PhD armed-forces \n",
- "4 500/1000 MALE Associate sales \n",
- "\n",
- " insured_hobbies insured_relationship incident_date \\\n",
- "0 sleeping husband 2015-01-25 \n",
- "1 reading other-relative 2015-01-21 \n",
- "2 board-games own-child 2015-02-22 \n",
- "3 board-games unmarried 2015-01-10 \n",
- "4 board-games unmarried 2015-02-17 \n",
- "\n",
- " incident_type collision_type incident_severity \\\n",
- "0 Single Vehicle Collision Side Collision Major Damage \n",
- "1 Vehicle Theft Rear Collision Minor Damage \n",
- "2 Multi-vehicle Collision Rear Collision Minor Damage \n",
- "3 Single Vehicle Collision Front Collision Major Damage \n",
- "4 Vehicle Theft Rear Collision Minor Damage \n",
- "\n",
- " authorities_contacted property_damage police_report_available auto_make \\\n",
- "0 Police YES YES Saab \n",
- "1 Police NO NO Mercedes \n",
- "2 Police NO NO Dodge \n",
- "3 Police NO NO Chevrolet \n",
- "4 None NO NO Accura \n",
- "\n",
- " auto_model \n",
- "0 92x \n",
- "1 E400 \n",
- "2 RAM \n",
- "3 Tahoe \n",
- "4 RSX "
- ]
- },
- "execution_count": 20,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "data_cat.head()"
- ]
- },
- {
- "attachments": {},
- "cell_type": "markdown",
- "id": "78a33f86",
- "metadata": {},
- "source": [
- "## 3.2 Überprüfung der Merkmale in Bezug auf das Ziel"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 21,
- "id": "4190f23f",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/html": [
- "<div>\n",
- "<style scoped>\n",
- " .dataframe tbody tr th:only-of-type {\n",
- " vertical-align: middle;\n",
- " }\n",
- "\n",
- " .dataframe tbody tr th {\n",
- " vertical-align: top;\n",
- " }\n",
- "\n",
- " .dataframe thead th {\n",
- " text-align: right;\n",
- " }\n",
- "</style>\n",
- "<table border=\"1\" class=\"dataframe\">\n",
- " <thead>\n",
- " <tr style=\"text-align: right;\">\n",
- " <th></th>\n",
- " <th>policy_csl</th>\n",
- " <th>insured_sex</th>\n",
- " <th>insured_education_level</th>\n",
- " <th>insured_occupation</th>\n",
- " <th>insured_hobbies</th>\n",
- " <th>insured_relationship</th>\n",
- " <th>incident_date</th>\n",
- " <th>incident_type</th>\n",
- " <th>collision_type</th>\n",
- " <th>incident_severity</th>\n",
- " <th>authorities_contacted</th>\n",
- " <th>property_damage</th>\n",
- " <th>police_report_available</th>\n",
- " <th>auto_make</th>\n",
- " <th>auto_model</th>\n",
- " <th>fraud_reported</th>\n",
- " </tr>\n",
- " </thead>\n",
- " <tbody>\n",
- " <tr>\n",
- " <th>0</th>\n",
- " <td>250/500</td>\n",
- " <td>MALE</td>\n",
- " <td>MD</td>\n",
- " <td>craft-repair</td>\n",
- " <td>sleeping</td>\n",
- " <td>husband</td>\n",
- " <td>2015-01-25</td>\n",
- " <td>Single Vehicle Collision</td>\n",
- " <td>Side Collision</td>\n",
- " <td>Major Damage</td>\n",
- " <td>Police</td>\n",
- " <td>YES</td>\n",
- " <td>YES</td>\n",
- " <td>Saab</td>\n",
- " <td>92x</td>\n",
- " <td>1</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>1</th>\n",
- " <td>250/500</td>\n",
- " <td>MALE</td>\n",
- " <td>MD</td>\n",
- " <td>machine-op-inspct</td>\n",
- " <td>reading</td>\n",
- " <td>other-relative</td>\n",
- " <td>2015-01-21</td>\n",
- " <td>Vehicle Theft</td>\n",
- " <td>Rear Collision</td>\n",
- " <td>Minor Damage</td>\n",
- " <td>Police</td>\n",
- " <td>NO</td>\n",
- " <td>NO</td>\n",
- " <td>Mercedes</td>\n",
- " <td>E400</td>\n",
- " <td>1</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>2</th>\n",
- " <td>100/300</td>\n",
- " <td>FEMALE</td>\n",
- " <td>PhD</td>\n",
- " <td>sales</td>\n",
- " <td>board-games</td>\n",
- " <td>own-child</td>\n",
- " <td>2015-02-22</td>\n",
- " <td>Multi-vehicle Collision</td>\n",
- " <td>Rear Collision</td>\n",
- " <td>Minor Damage</td>\n",
- " <td>Police</td>\n",
- " <td>NO</td>\n",
- " <td>NO</td>\n",
- " <td>Dodge</td>\n",
- " <td>RAM</td>\n",
- " <td>0</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>3</th>\n",
- " <td>250/500</td>\n",
- " <td>FEMALE</td>\n",
- " <td>PhD</td>\n",
- " <td>armed-forces</td>\n",
- " <td>board-games</td>\n",
- " <td>unmarried</td>\n",
- " <td>2015-01-10</td>\n",
- " <td>Single Vehicle Collision</td>\n",
- " <td>Front Collision</td>\n",
- " <td>Major Damage</td>\n",
- " <td>Police</td>\n",
- " <td>NO</td>\n",
- " <td>NO</td>\n",
- " <td>Chevrolet</td>\n",
- " <td>Tahoe</td>\n",
- " <td>1</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>4</th>\n",
- " <td>500/1000</td>\n",
- " <td>MALE</td>\n",
- " <td>Associate</td>\n",
- " <td>sales</td>\n",
- " <td>board-games</td>\n",
- " <td>unmarried</td>\n",
- " <td>2015-02-17</td>\n",
- " <td>Vehicle Theft</td>\n",
- " <td>Rear Collision</td>\n",
- " <td>Minor Damage</td>\n",
- " <td>None</td>\n",
- " <td>NO</td>\n",
- " <td>NO</td>\n",
- " <td>Accura</td>\n",
- " <td>RSX</td>\n",
- " <td>0</td>\n",
- " </tr>\n",
- " </tbody>\n",
- "</table>\n",
- "</div>"
- ],
- "text/plain": [
- " policy_csl insured_sex insured_education_level insured_occupation \\\n",
- "0 250/500 MALE MD craft-repair \n",
- "1 250/500 MALE MD machine-op-inspct \n",
- "2 100/300 FEMALE PhD sales \n",
- "3 250/500 FEMALE PhD armed-forces \n",
- "4 500/1000 MALE Associate sales \n",
- "\n",
- " insured_hobbies insured_relationship incident_date \\\n",
- "0 sleeping husband 2015-01-25 \n",
- "1 reading other-relative 2015-01-21 \n",
- "2 board-games own-child 2015-02-22 \n",
- "3 board-games unmarried 2015-01-10 \n",
- "4 board-games unmarried 2015-02-17 \n",
- "\n",
- " incident_type collision_type incident_severity \\\n",
- "0 Single Vehicle Collision Side Collision Major Damage \n",
- "1 Vehicle Theft Rear Collision Minor Damage \n",
- "2 Multi-vehicle Collision Rear Collision Minor Damage \n",
- "3 Single Vehicle Collision Front Collision Major Damage \n",
- "4 Vehicle Theft Rear Collision Minor Damage \n",
- "\n",
- " authorities_contacted property_damage police_report_available auto_make \\\n",
- "0 Police YES YES Saab \n",
- "1 Police NO NO Mercedes \n",
- "2 Police NO NO Dodge \n",
- "3 Police NO NO Chevrolet \n",
- "4 None NO NO Accura \n",
- "\n",
- " auto_model fraud_reported \n",
- "0 92x 1 \n",
- "1 E400 1 \n",
- "2 RAM 0 \n",
- "3 Tahoe 1 \n",
- "4 RSX 0 "
- ]
- },
- "execution_count": 21,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "data_cat = pd.concat([data_cat, data_num.fraud_reported], axis=1)\n",
- "data_cat.head()"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 22,
- "id": "c4059c89",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "0 2015-01-25\n",
- "1 2015-01-21\n",
- "3 2015-01-10\n",
- "5 2015-01-02\n",
- "14 2015-01-15\n",
- " ... \n",
- "974 2015-02-08\n",
- "977 2015-02-21\n",
- "982 2015-01-01\n",
- "986 2015-02-19\n",
- "987 2015-01-13\n",
- "Name: incident_date, Length: 247, dtype: object"
- ]
- },
- "execution_count": 22,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "data_cat.incident_date.loc[data_cat.fraud_reported == 1]"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 23,
- "id": "6fe725f2",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "image/png": "iVBORw0KGgoAAAANSUhEUgAACGMAAAGdCAYAAAC8FywNAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAABVbUlEQVR4nO39e5xdWV0n/H+SqnNS3U1IuiHQICDOYC1RURlEh9FxmJfK/SJ0Iw4oeAEBcVAR1EdRW2Wc+TEOKoqgYjPIxUH7InJXUUdweHREGFRwjT6PPsjQQEuTEKDTVUnq90edQJJO0rWr1qla59T7/XrVK6lTWZ/z3WvvvfY+K6vO2bO2thYAAAAAAAAAANrYu9MFAAAAAAAAAADME4sxAAAAAAAAAAAashgDAAAAAAAAAKAhizEAAAAAAAAAABqyGAMAAAAAAAAAoCGLMQAAAAAAAAAAGrIYAwDmWCnlJaWUvy+l/Icp5f9iKeWqaWRvRSnld0spdxzY5stLKf+wxee9uJTymlLK+0sptZTyDVvJAwAAgN3EPMagNluexzgt60tLKTe0yAIAPmtxpwsAAKbqaUnuUWv94E4Xss2+foee96okn6y13ruUco8k7yyl/Pku7H8AAADYDPMY26iUspjk3yf5wSS324kaAGCeWYwBAHOqlPL2JHuSvLmU8l1JXpnkT5N8SZIfTrI6+XOc5E5JXlFr/dFSygOT/GKt9YsnOZ/5vpRy+yQvS/KlSW5IcjzJO26jjgcm+fkkn8r6C/v7J3lQkudNnvvTSZ5Ta33n5LdT7pXk7knukuQ9SZ5Sa/1EKeWLkvxikjskWUvyX2qtv36O/HdNnvoPSykPS3Jy0u4eSUZJ/lut9acntT0jyfclOZLkL89T/xcmec05fvTztdaXn/XYY5I8IUlqrR8opfxekm9M8sIL9REAAADsduYxdmQe418kuU/W5zN+70L9AgAMZzEGAMypWuu/LqWsJfm3tdZ/KqUkyV/VWh9fStmT5A+SPLnW+rellLsm+UAp5edvI/Ynktyc5AuS3DHJX+Q2JjEmvjjJP6u1/n+llM9P8tNJHlhr/dhkcuL3Syn3mvzbf5Pky5PcmORVSX6slPJDSX4nyXNrrddN6v2zUsrfnp2fJKWUbz1tu/8gyc/WWl9fSllK8qZSyt8l+d9ZfyeLL621friU8tLz9OP7knzZBrYxWZ98+cfTvv9gkrttsC0AAADsWuYxtn8eo9b6Z5O67rmRfw8ADGMxBgDsLm9PklrrWinlkUkeUUp5QpJ7Z/23Ty65jfZfl+R7a61rSW4spVy/wef9x1MTDFl/6827JHnbZGIlWf+tj1OTGL9Va/1IkpRSfi3JzyW5OslSrfW6Sf0fKqVcm+QhSf7wrPzPKKVckvVJkctKKT81efh2WZ+UuHuS3621fnjy+K9M8s7OGPIbJXuz/tsup+xJcuIcbQEAAIDbZh5juvMYAMAUWYwBALvLJ5PPvLh/d5Lrsz6xcXWSb8j6RMba5M9TxmdlnP6z40Oed2IhydtqrY8/9UAp5e5JPpT1t8U8PXNv1hczLOTMRQ6nfjY6R/7pFib1/qta66cnz3XHJMey/jm0t7ktA98Z4wNJ7prkI5Pv75r1tygFAAAAhjOPMd15DABgivbudAEAwI74/CS3T/K8Wuvrkzwwyb6sv+i/Mck9Sil3mrwN6Ded1u7NSb6jlLK3lHJpkkdv4rnfluRBpZQvSJLJ56G+N8lFk58/upRyoJSyN8lTk7w+yd8kWS2lPHbS5q5Jrsj5P8/0RJJRrfUTSf7vJM+etDuY5E8mdf/upI5THyPyrZvYlrO9Lsl3Tp7rbln/DZU3NMgFAACA3cw8xnTmMQCAKbIYAwB2p/dmfZHA35RS3p/kkUnel+Rek9+g+OUkf571CYC/P63dVUlWsz6p8Pokfzn0iSf535nkv5VS/leSn0ryqFrrqd8K+UiSNyV5f5IjSX661rqa9d94+Z5SynuT/H6Sn6y1/uF5nua3kvz3UsoXJ3lCkn9ZSvnLJH+a5Ddqra+utf5lkh/I+tuM/nmSpaHbcg4/nuR2pZS/ntT43Frr/9MgFwAAAHYz8xjTmccAAKZoz9ra2e+UBQCwM0opVyW5Y631u3e6FgAAAIALMY8BAFzI4k4XAADMvlLKa5OU8/z48bXWup31AAAAAJyPeQwAYDt4ZwwAAAAAAAAAgIb27nQBAAAAAAAAAADzxGIMAAAAAAAAAICGZmkxxmKSe07+BAAAAOiVOQwAAADY5WZpUuBuSf7+Yx/7ZE6eXNvpWgAAAIApOHRo/56drqEBcxgAAACwC1xoHmOW3hkDAAAAAAAAAKB7FmMAAAAAAAAAADRkMQYAAAAAAAAAQEMWYwAAAAAAAAAANGQxBgAAAAAAAABAQxZjAAAAAAAAAAA0ZDEGAAAAAAAAAEBDFmMAAAAAAAAAADRkMQYAAAAAAAAAQEMWYwAAAAAAAAAANGQxBgAAAAAAAABAQxZjAAAAAAAAAAA0ZDEGAAAAAAAAAEBDiztdAFt32YF9WRiPB7c7sbKSm47cMoWKAAAAAAAAAGD3shhjDiyMx/nQi589uN1dn/nCJBZjAAAAAAAAAEBLPqYEAAAAAAAAAKAhizEAAAAAAAAAABqyGAMAAAAAAAAAoCGLMQAAAAAAAAAAGrIYAwAAAAAAAACgIYsxAAAAAAAAAAAashgDAAAAAAAAAKAhizEAAAAAAAAAABqyGAMAAAAAAAAAoCGLMQAAAAAAAAAAGrIYAwAAAAAAAACgIYsxAAAAAAAAAAAashgDAAAAAAAAAKAhizEAAAAAAAAAABqyGAMAAAAAAAAAoCGLMQAAAAAAAAAAGrIYAwAAAAAAAACgIYsxAAAAAAAAAAAashgDAAAAAAAAAKAhizEAAAAAAAAAABqyGAMAAAAAAAAAoCGLMQAAAAAAAAAAGrIYAwAAAAAAAACgIYsxAAAAAAAAAAAashgDAAAAAAAAAKAhizEAAAAAAAAAABqyGAMAAAAAAAAAoCGLMQAAAAAAAAAAGrIYAwAAAAAAAACgIYsxAAAAAAAAAAAashgDAAAAAAAAAKAhizEAAAAAAAAAABqyGAMAAAAAAAAAoKHFaYaXUn48yTdOvn1jrfUHSilfl+SFSS5K8tpa6/OmWQMAAAAAAAAAwHaa2jtjTBZdPCjJfZN8WZL7lVL+XZKrkzw6yb2T3L+U8tBp1QAAAAAAAAAAsN2m+TElNyT5/lrrSq11Ncn7kywn+dta69/XWo8neVWSx02xBgAAAAAAAACAbTW1jymptf71qb+XUj4/6x9X8gtZX6Rxyg1J7jYk9w53uF2T+lh36ND+nS4BoJmVE6sZL4y2rR0AAFyIOQwAAADYvaa2GOOUUsoXJXljkucmOZ71d8c4ZU+Sk0PyPvaxT+bkybV2Bc6BrSyouPHGow0rAdhZhw7tz8N++4cHt3vTN/y08RAAoBPz9EsD5jAAAABgvl1oHmOaH1OSUspXJXlbkh+qtb4iyQeT3OW0f3J5kg9NswYAAAAAAAAAgO00tXfGKKXcPclvJ3l8rfUPJg//6fqPyr2S/H2SJyS5elo1AAAAAAAAAABst2l+TMlzkiwleWEp5dRjL03yrUmunfzsTUmumWINAAAAAAAAAADbamqLMWqt35Pke87z4y+d1vMCAAAAAAAAAOykvTtdAAAAAAAAAADAPLEYAwAAAAAAAACgIYsxAAAAAAAAAAAashgDAAAAAAAAAKAhizEAAAAAAAAAABqyGAMAAAAAAAAAoCGLMQAAAAAAAAAAGlrc6QIAAGBe3P7gxdk3Whjc7pbVE/nE4U9PoSIAAAAAAHaCxRgAANDIvtFCnnX9Pw5u96LH3H0K1QAAAAAAsFN8TAkAAAAAAAAAQEMWYwAAAAAAAAAANGQxBgAAAAAAAABAQxZjAAAAAAAAAAA0ZDEGAAAAAAAAAEBDFmMAAAAAAAAAADRkMQYAAAAAAAAAQEMWYwAAAAAAAAAANGQxBgAAAAAAAABAQxZjAAAAAAAAAAA0ZDEGAAAAAAAAAEBDiztdAAAwHfsPLmVpNBrc7tjqao4ePjaFigAAAAAAAHYHizEAYE4tjUZ5+HUvGdzujY99Ro7GYgwAAAAAAIDN8jElAAAAAAAAAAANWYwBAAAAAAAAANCQxRgAAAAAAAAAAA1ZjAEAAAAAAAAA0JDFGAAAAAAAAAAADVmMAQAAAAAAAADQkMUYAAAAAAAAAAANWYwBAAAAAAAAANCQxRgAAAAAAAAAAA1ZjAEAAAAAAAAA0NDiThcwyy47sJSF8WhQmxMrq7npyLEpVQQAAAAAAAAA7DSLMbZgYTzKR1/6okFt7vT0ZyWxGAMAAAAAAAAA5pWPKQEAAAAAAAAAaMhiDAAAAAAAAACAhizGAAAAAAAAAABoyGIMAAAAAAAAAICGLMYAAAAAAAAAAGjIYgwAAAAAAAAAgIYsxgAAAAAAAAAAaMhiDAAAAAAAAACAhizGAAAAAAAAAABoyGIMAAAAAAAAAICGLMYAAAAAAAAAAGhocacLgNYuPTDO4njf4HbHV27Jx4+sNKvj4IFxRpuoY3XllhxuWEcLBw+OMhotDW63unoshw+vTqEimJ79B/dlaTQe3O7Y6kqOHr5lChUB7IyDBy/JaDR87fbq6skcPvypKVQE03PZgYuzMF4Y1ObEyoncdOTTU6po8y47cFEWxsNf6p9YOZ6bjtw8hYoAAAAAdieLMZg7i+N9eeevPGJwuwd85xuStFsEMRrvy1t/7WGD2z34O97UtI4WRqOlvOa/Pnhwuyd861uTWIzBbFkajfOw658/uN2bHvO8HI3FGMD8GI325pXX3Ti43bc89tAUqoHpWhgv5CM/965Bbe78vfebUjVbszBezEd/8XcHt7vTdz9oCtUAAAAA7F4+pgQAAAAAAAAAoCGLMQAAAAAAAAAAGrIYAwAAAAAAAACgIYsxAAAAAAAAAAAashgDAAAAAAAAAKAhizEAAAAAAAAAABqyGAMAAAAAAAAAoKHFaYaXUm6f5H8keUSt9R9KKS9P8tVJPjX5Jz9Ra71+mjUAAAAAAAAAAGynqS3GKKV8ZZJfTbJ82sNfnuRraq03TOt5AQAAAAAAAAB20jQ/puSpSZ6Z5ENJUkq5OMk9klxdSnlvKeUnSik+JgUAAAAAAAAAmCtTe2eMWutTkqSUcuqhy5P8QZLvSnIkyRuSfEfW3z1jw+5wh9u1K3KHHDq0/zN/Xzt+PHsWh+2GzbTZSC1bdfL4SvYujret3TS07I+t6KWOFuZpW5hvLY7VUxkrJ45nvDB8nD69XYuMrXDuMmtWTpzMeGH4Ot/NtpuGeTrv5mlb4ELm7Vift+3pwTzMYQAAAACbM7XFGGertf6/SR5z6vtSyi8keVIGLsb42Mc+mZMn1xpXtzmbnai68cajZ2R8+CXPH9T+8mc871YZm3V6zlYdOrQ/f/PiRw9u9wXPfF3zOjZrHutoYZ62hfnW4lhtlfHw6//z4PZvfMxzz8y47kXDMx77rKbbArPi0KH9edy17x3c7reu+BLX//OYp22B29LitV0v5uHcnadFIT3NYQAAAADtXWgeY9t+DbCUcp9SyhWnPbQnyep2PT8AAAAAAAAAwHbYtnfGyPrii58rpfxBkk8m+c4kr9jG5wcAAAAAAAAAmLpte2eMWut7k/zHJH+S5H1J3lNr/Y3ten4AAAAAAAAAgO0w9XfGqLXe87S//1KSX5r2cwIAAAAAAAAA7JRte2cMAAAAAAAAAIDdwGIMAAAAAAAAAICGLMYAAAAAAAAAAGjIYgwAAAAAAAAAgIYsxgAAAAAAAAAAaMhiDAAAAAAAAACAhizGAAAAAAAAAABoaHGnCwDYbgcOjjIeLQ1qs7J6LEcOrzbN6MXtD46zb7RvUJtbVm/JJw6vTKkierL/4FKWRqPB7Y6trubo4WNTqGjz9h+8KEuj4bc+x1aP5+jhm7eUM42MebL/4MVZGi0ManNs9USOHv70lCqC+XfZgUuyMB62Nv/EysncdORTU6po8y47cHEWxsPGkBMrJ3LTEWMIAAAAANNjMQaw64xHS3nRqx88qM2znvjWJJ9dSDEeLeUFvzEs4wf+3ZkZvdg32pdvu/4hg9q8/DFvSWIxxm6wNBrl4de+bHC7N17xlBxNX4sxlkaLecQ1rx3c7g1XPj5Hz8p55DXXDcp4/ZWPvVXGo655/aCM37nykWdkzJOl0UKuuPbPBrW59oqvmNv+gO2wMN6bf/i5Dw9qc8/vvXxK1WzNwnghH37hXw9qc/mzv2hK1QAAAADAOh9TAgAAAAAAAADQkMUYAAAAAAAAAAANWYwBAAAAAAAAANCQxRgAAAAAAAAAAA1ZjAEAAAAAAAAA0JDFGAAAAAAAAAAADVmMAQAAAAAAAADQkMUYAAAAAAAAAAANWYwBAAAAAAAAANDQhhZjlFI+5xyPfWH7cgAAAAAAAAAAZtvihX5YSrls8tc3lVIemGTP5PtRkuuSfMH0SgMAAAAAAAAAmD0XXIyR5DeSfP3k7x877fHjSa6ZSkXAZxw8MM5ovG9wu9WVW3L4yMoUKoLp2X9wX5ZG48Htjq2u5OjhW6ZQEa3sP7iUpdFocLtjq6s5evjYFCqaD/sPXpSl0W3dyp3p2OrxHD1885Qqguk4ePCSjEbDP11xdfVkDh/+1BQqgum57MBFWRgPG9uT5MTK8dx0xPgOAAAA0JMLzvLUWh+cJKWUq2ut3749JQGnjMb78vqrHzq43SO//c1JLMZgtiyNxnno6545uN2bH/3iHI3FGD1bGo3yiGtfMbjdG654co7GYozzWRot5tHXvGVQm9dd+ZAcnVI9MC2j0d68+bX/NLjdQx9/xylUA9O1MF7MR170R4Pb3flZD2xeCwAAAABbs6Ffuam1fnsp5XOTXJbPflRJaq1/Ma3CAAAAAAAAAABm0YYWY5RSfiLJc5N8NMna5OG1JP9sSnUBAAAAAAAAAMykjX4Y7ZOS3KvW+qFpFgMAAAAAAAAAMOv2bvDf/aOFGAAAAAAAAAAAt22j74zxtlLKC5K8LsnNpx6stf7FVKoCAAAAAAAAAJhRG12M8a2TPx932mNrSf5Z02oAAAAAAAAAAGbchhZj1Fo/b9qFAAAAAAAAAADMgw0txiilPPtcj9daX9i2HAAAAAAAAACA2bbRjym5z2l/Hyf5N0ne1r4cAAAAAAAAAIDZttGPKfm2078vpdw1ya9NpSIAAAAAAAAAgBm20XfGOEOt9UOllHs2rmVbXXZgKQvj0aA2J1ZWc9ORY1OqiCS59MA4i+N9g9sdX7klHz+yMoWKZt/BA+OMNtGnqyu35LA+BQBgii47cHEWxguD2pxYOZGbjnx6ShVt3mUHLsrCePhL7BMrx3PTkZunUBEAAAAAO2lDM0WllGef9u2eJF+e5KNTqWibLIxHufGlLxvU5tDTn5LEYoxpWhzvy7tf+sjB7e779NcnsXDgXEbjfbnm5Q8Z3O7Kb3tL9CkAANO0MF7Ih3/m7wa1ufw595pSNVuzMF7MR37+nYPb3fl7HjCFagAAAADYaRv9tZ37nPb3tSQfSPLc9uUAAAAAAAAAAMy2DS3GqLV+W5KUUj43yajWOuxXlwAAAAAAAAAAdomNfkzJvZK8Lsldk+wtpfxTkkfUWt8/zeIAAAAAAAAAAGbN3g3+u19M8oJa66W11gNJnp/kxdMrCwAAAAAAAABgNm10Mcada62vOPVNrfXlSQ5NpyQAAAAAAAAAgNm10cUYi6WUy059U0q5Y5K16ZQEAAAAAAAAADC7Fjf4734hyf9dSnlt1hdhfFOSn51aVQAAAAAAAAAAM2qj74zxpqwvwhgn+cIkn5Pk+mkVBQAAAAAAAAAwqza6GOO/JnlxrfUHk3xzkh9JcvW0igIAAAAAAAAAmFUbXYxxx1rri5Kk1nqs1vpzSe4ytaoAAAAAAAAAAGbURhdjLJZS7nrqm1LKnZPsmU5JAAAAAAAAAACza3GD/+6FSd5TSnlLkrUkX5fkuVOrCoCZsv/gOEujfYPbHVu9JUcPr0yhIoDh9h+8OEujhcHtjq2eyNHDn55CRfTk0oOXZHG00bXsn3V89WQ+fvhTbWs5cEkWx8NqOb5yMh8/0raOyw5cnIXxsHPmxMqJ3HTE+bIbXHbgoiyMNzrl8FknVo5PoRoAAACA7behmZFa69WllD9P8rVJjif5z7XWv5pqZQDMjKXRvjz0dVcMbvfmR1+bo7EYA+jD0mghj7/u7wa3e+1j75WjU6iHviyO9uadr7hxcLsHPPlQ+1rGe/P+l3xkUJt7P+POzetYGC/khhf8n0Ft7vIDn9O8Dvq0MF7MR3/p2sHt7vRdw+8pAQAAAHq04V9TqbW+N8l7p1gLAAAAAAAAAMDMG/4+uwAAAAAAAAAAnJfFGAAAAAAAAAAADVmMAQAAAAAAAADQkMUYAAAAAAAAAAANWYwBAAAAAAAAANCQxRgAAAAAAAAAAA0tTjO8lHL7JP8jySNqrf9QSvm6JC9MclGS19ZanzfN5wcAAAAAAAAA2G5Te2eMUspXJnlHkuXJ9xcluTrJo5PcO8n9SykPndbzAwAAAAAAAADshGl+TMlTkzwzyYcm339Fkr+ttf59rfV4klcledwUnx8AAAAAAAAAYNtN7WNKaq1PSZJSyqmH7prkhtP+yQ1J7jat5wcAAAAAAAAA2AlTW4xxDnuTrJ32/Z4kJ4eG3OEOt2tW0GYcOrR/bjJa5mxVL32iP6aXcfzEShYXxoPbb7bdNPTSp73oaRySMf8ZKydOZLywMLj9Ztudr455y9hM/7Tq0xbmdRw6fmItiwt7BrffbLtp6OV6p44z9Xasy2ib00sGZ9rpOQwAAJgFm5nT6GkeBOB8tnMxxgeT3OW07y/PZz/CZMM+9rFP5uTJtdv+h7dhs5NMN954dG4yzs7ZqhZ19JLRQi/b0lPGy1/xoMHtv+3Jv9t0v5yqZTNan7u96GUckiHjtjIecc2rB7d/w5VPbH596GUMaZXxDde8bVD7377ya+dqXD49p6eM519/w23861t73mPu0t29jDra15H0NYbImM9xqAULO9a1msMAAIB5dujQ/rz4+o8MavPMx9y5y/l2YPe50BzIdi7G+NMkpZRyryR/n+QJSa7exucHAAAAAAAAAJi6vdv1RLXWY0m+Ncm1Sd6X5G+SXLNdzw8AAAAAAAAAsB2m/s4YtdZ7nvb3tyX50mk/JwAAAAAAAADATtm2d8YAAAAAAAAAANgNLMYAAAAAAAAAAGjIYgwAAAAAAAAAgIYsxgAAAAAAAAAAaMhiDAAAAAAAAACAhizGAAAAAAAAAABoyGIMAAAAAAAAAICGFne6AIDd6sDBUcajpUFtVlaP5cjh1SlVBDDc/oMXZWk07Jby2OrxHD1885QqAgCm5bID+7IwHg9ud2JlJTcduWUKFcH828zcQWL+AICNuf3Bi7NvtDCozS2rJ/KJw5+eUkUA88ViDIAdMh4t5arffPCgNld941uTmEwB+rE0Wsxjrn3HoDbXX/HVOTqlegCA6VkYj/Phlzx/cLvLn/G8JBZjwGaMR0t50auHzR0kybOeaP4AgNu2b7SQZ13/j4PavOgxd59SNQDzx8eUAAAAAAAAAAA0ZDEGAAAAAAAAAEBDFmMAAAAAAAAAADRkMQYAAAAAAAAAQEMWYwAAAAAAAAAANGQxBgAAAAAAAABAQxZjAAAAAAAAAAA0ZDEGAAAAAAAAAEBDFmMAAAAAAAAAADRkMQYAAAAAAAAAQEMWYwAAAAAAAAAANLS40wVsxmUHlrIwHg1qc2JlNTcdOTaligAAgLMdPHhJRqPh679XV0/m8OFPTaEiAACA6dt/cClLo2H/h5Ekx1ZXc/Sw/8cAYPe47MBFWRgPW7JwYuV4bjpy85QqamsmF2MsjEe58SWvGtTm0DO+OYmbGAAA2C6j0d5cd80/DW732CvvOIVqAAAAtsfSaJRHXPPqwe3ecOUTc9T/YwCwiyyMF/PRX/zdQW3u9N0PmlI17fmYEgAAAAAAAACAhizGAAAAAAAAAABoyGIMAAAAAAAAAICGLMYAAAAAAAAAAGjIYgwAAAAAAAAAgIYsxgAAAAAAAAAAaMhiDAAAAAAAAACAhizGAAAAAAAAAABoyGIMAAAAAAAAAICGLMYAAAAAAAAAAGhocacLoA+XHRhnYbxvcLsTK7fkpiMrU6gIzu3AwVHGo6VBbVZWj+XI4dUpVbSzbn9wnH2jYefuLau35BOHnbcA8+zAwUsyHg1fd72yejJHDn9qChXtrIMHL8loE/2xunoyh+ewP4DZc9mBpSyMR4PbnVhZzU1Hjk2hIgAAdtL+gxdnabQwuN2x1RM5evjTU6gIgHOxGIMkycJ4Xz7woisHt7vHs65J4j912T7j0VJ++ZUPHtTmad/y1iTzuRhj32hfnnPNQwa1+Zkr3xLnLcB8G4/25sXXf2Rwu2c+5s5TqGbnjUZ784evvnFwu3/7xENTqAZguIXxKDe+9GWD2x16+lOSWIwBADBvlkYLefx1fze43Wsfe68cnUI9AJybjykBAAAAAAAAAGjIYgwAAAAAAAAAgIYsxgAAAAAAAAAAaMhiDAAAAAAAAACAhizGAAAAAAAAAABoyGIMAAAAAAAAAICGLMYAAAAAAAAAAGjIYgwAAAAAAAAAgIYsxgAAAAAAAAAAaMhiDAAAAAAAAACAhizGAAAAAAAAAABoaHGnCwAAANgNLj1wSRbHw9fDH185mY8f+dQUKoJzu+zARVkYD58uOLFyPDcduXkKFcF0XHpgnMXxvkFtjq/cko8fWZlSRbRy8OAoo9HS4Harq8dy+PDqFCqiJ/sPjrM0GnbuJ8mx1Vty9LDzf5r2H1zK0mg0qM2x1dUcPXxsShUBDHfw4CUZjYa99l9dPZnDh73uP5/LDlyShU3Mp5xYOZmbGs6nXHbg4iyMFzZRx4ncdOTTzepgtliMAQAAsA0Wx3vz7pd9dHC7+z7lTlOoBs5vYbyYj7749YPb3emZj5xCNTA9i+N9+ZsXP3pQmy945uuS+M/Y3o1GS3nNf33w4HZP+Na3JrEYY94tjfbloa+7YnC7Nz/62hx1/k/V0miUh1/3kkFt3vjYZ+RoLMYA+jEa7c2bX/tPg9o89PF3nFI182FhvDf/8HMfHtzunt97eeM6FvLhn/m7we0uf869mtbBbPExJQAAAAAAAAAADVmMAQAAAAAAAADQkMUYAAAAAAAAAAANWYwBAAAAAAAAANCQxRgAAAAAAAAAAA1ZjAEAAAAAAAAA0JDFGAAAAAAAAAAADS3uxJOWUv4wyZ2SrE4eelqt9U93ohYAAAAAAAAAgJa2fTFGKWVPkuUkn1trPb7dzw8AAAAAAAAAME078TElZfLn75ZS/lcp5bt3oAYAAAAAAAAAgKnYicUYlyZ5W5LHJPnaJE8vpXz9DtQBAAAAAAAAANDctn9MSa31nUneeer7UsqvJXlYkt/bSPs73OF2m37uQ4f2b7rtPGa0ypHRXi/bIkPGtDNa5ciQIUPGTufIkCFDxk7nzFPGPNnKHEYrveyTXupoofW2nDy+kr2L46m3YWNO7d/jJ1ayuDC8jzfbbhrmaVt60fr8XzmxmvHCaOpttsPKieMZLwz774bNtDmXebrGJJ/dnpUTJzJeWBjcfrPt2JiVEyczXhj+e86nt2uRsRXzdN/fSx0t9LgtJ4+vZe/inm1rNw299GsvdcyTWenTbV+MUUr56iT7aq1vmzy0J8nqRtt/7GOf3PRkxo03Hv3M3ze7g+Yp4/QcGWdmtNDLtsxTxlZyZPSZcXqODBmzkLGVHBl9ZpyeI0PGLGRsJUdGnxmn58iYPx/72Cdz8uTalnPm6XVuT1qc+63qeOevPGJQmwd85xu67NNetBrHXv6KBw1u/21P/t3uzrtDh/bnBb/x4MHtf+DfvXUuj7OexrJDh/bnYb/9w4PavOkbfrrL/XLo0P48/LoXDWrzxsc+q/k9WS9anbuPuOa1g9u/4crHd9kn8+LQof254to/G9zu2iu+4ox9+7hr3zs447eu+JIuX1+20EsdLczbtrz7ZR8d3O6+T7lTd/dD81THvJmHc+ZC27DtizGSHEzyk6WUf5VklOTJSZ6+A3UAAAAAAAAAADS39fczGqjW+oYkb0zy7iTvSnL15KNLAAAAAAAAAABm3k68M0ZqrT+a5Ed34rkBAAAAAAAAAKZp298ZAwAAAAAAAABgnlmMAQAAAAAAAADQkMUYAAAAAAAAAAANWYwBAAAAAAAAANCQxRgAAAAAAAAAAA1ZjAEAAAAAAAAA0JDFGAAAAAAAAAAADS3udAHQo0sPjLM43je43fGVW/LxIytTqAgAAABIkssOjLOwidfsJ1ZuyU1es7ONDhwcZTxaGtxuZfVYjhxenUJFMP/2H1zK0mg0qM2x1dUcPXxsShXB/Dtw8OKMRwuD2qysnsiRw5+eUkWbd/DgJRmNhv0e++rqyRw+/KkpVQTTcdmBi7IwHr5M4MTK8dx05OYpVLQ1m9me7doWizHgHBbH+/JHv/rwwe0e+NQ3JjGxAwAAANOyMN6XD7zoysHt7vGsa+I1O9tpPFrKL7/ywYPbPe1b3prEYgzYjKXRKA+/9mWD2rzxiqfkaCzGgM0ajxby/OtvGNTmeY+5y5Sq2ZrRaG+uu+afBrV57JV3nFI1MD0L48V85OffObjdnb/nAVOoZusWxov56ItfP6jNnZ75yClVcyYfUwIAAAAAAAAA0JDFGAAAAAAAAAAADVmMAQAAAAAAAADQkMUYAAAAAAAAAAANWYwBAAAAAAAAANCQxRgAAAAAAAAAAA1ZjAEAAAAAAAAA0JDFGAAAAAAAAAAADVmMAQAAAAAAAADQkMUYAAAAAAAAAAANWYwBAAAAAAAAANDQ4k4XAAAAAMB0XXZgKQvj0eB2J1ZWc9ORYw3r2JeF8XgTdazkpiO3NKtj3lx6YJzF8b5BbY6v3JKPH1lpWsfBA+OMBtaxunJLDjeug/l1+4Pj7BsNO8aS5JbVW/KJw46z89l/cClLo2HXiGOrqzl6uN31oSeb6Y9kfvtk/8GLsjQa/l9Jx1aP5+jhm+euDvp18OAlGY2G/Q766urJHD78qSlVtHmb2Zak3+1hPl124KIsjIePyydWjuemI7tnXLYYAwAAAGDOLYxH+ehLXzS43Z2e/qwk7f5jaWE8zode/OzB7e76zBcmsRjjfBbH+/JHv/rwQW0e+NQ3Jmn7n9Oj8b68/uqHDmrzyG9/c/M6mF/7Rvvybdc/ZHC7lz/mLXGcnd/SaJSHX/+fB7V542Oem6MNrw89WRqN8ohrXzG43RuuePJc9snSaDGPuub1g9v9zpWPzNHGdXzDNW8b3O63r/zapnXQr9Fob1553Y2D2nzLYw9NqZqtGY325g9fPWxbkuTfPrHP7WE+LYwX85EX/dHgdnd+1gOb19IzH1MCAAAAAAAAANCQxRgAAAAAAAAAAA1ZjAEAAAAAAAAA0JDFGAAAAAAAAAAADVmMAQAAAAAAAADQkMUYAAAAAAAAAAANWYwBAAAAAAAAANCQxRgAAAAAAAAAAA1ZjAEAAAAAAAAA0JDFGAAAAAAAAAAADVmMAQAAAAAAAADQ0OJOFwAAAAAwry47sJSF8WhwuxMrq7npyLEpVERLlx4YZ3G8b1Cb4yu35ONHVqZU0ew7eGCc0cA+TZLVlVtyWL927fYHx9k3Gr5vb1m9JZ843Ne+3X9wX5ZG48Htjq2u5OjhW3a8lmnUARey/+BFWRoN/++oY6vHc/TwzVOoCObfpQcvyeJo+O/kH189mY8f/lS7Og5cksXxJupYOZmPH2lXx2UHLs7CeGFwuxMrJ3LTkU93VUcv28LGWIwBAAAAMCUL41FufMmrBrc79IxvTmIxRu8Wx/vy7pc+clCb+z799Un6+o/lnozG+3LNyx8yuN2V3/aW6Ne+7Rvty3OuGb5vf+bK/vbt0mich77umYPbvfnRL87RtF0EsTQa52HXP39Qmzc95nnN64ALWRot5tHXvGVwu9dd+ZAcnUI9sBssjvbmna+4cXC7Bzz5UNs6xnvz/pd8ZHC7ez/jzk3rWBgv5IYX/J/B7e7yA5/TvI4Pv/CvB7e7/NlfdEbGR37uXYMz7vy99xvchq3zMSUAAAAAAAAAAA1ZjAEAAAAAAAAA0JDFGAAAAAAAAAAADVmMAQAAAAAAAADQkMUYAAAAAAAAAAANWYwBAAAAAAAAANCQxRgAAAAAAAAAAA1ZjAEAAAAAAAAA0JDFGAAAAAAAAAAADVmMAQAAAAAAAADQ0OJOFwAAAAAA7KyDB8YZjfcNarO6cksOH1mZUkX05MDBUcajpcHtVlaP5cjh1SlUBGzE/oMXZWk0/L+Bjq0ez9HDN0+hop2lPwA4n8sOXJSF8bBrxImV47npyIWvDxZjAAAAAMAuNxrvy1t/7WGD2jz4O96UxGKM3WA8WspVv/ngwe2u+sa3JrEYA3bK0mgxj7zmusHtXn/lY3N0CvXstKXRYh5z7TsGt7v+iq+ey/4A4LMWxov56C9dO6jNnb7ritv8Nz6mBAAAAAAAAACgIYsxAAAAAAAAAAAashgDAAAAAAAAAKAhizEAAAAAAAAAABqyGAMAAAAAAAAAoCGLMQAAAAAAAAAAGrIYAwAAAAAAAACgocWdeNJSyhOSPC/JKMnP1VpfvBN1AAAAAAAAAAC0tu3vjFFK+Zwk/yHJVyf5siTfWUr5wu2uAwAAAAAAAABgGnbiY0q+Lskf1FpvqrV+Ksk1Sa7cgToAAAAAAAAAAJrbiY8puWuSG077/oYkX7GBdgtJsnfvniTJ3v2XDH7iU20/8/3+2zXI2L/ljIX9BxpkXDo44+ychf2Htpwx2n+nLWeMG2Tsu93WM5Y6ybioQcbFt7vzljMuaZBxu0u2nrHZnLMz9jfIuH2DjIMXbz3j0gYZd2iQcaeLtj6G3Omiy7aecfHBBhnDx+VbZ9y+Qcbwa8x0MoZfL2+dMfy6PZ2Mi7ecsdmcW2dc1EnG0pYzDl28r0HGuEHGaHDG2TmHLt7c7fHpGZddvLDljAMNMvZfvLl112dc/xtkXNwg46IGGUuXbD1jX4OM8e22npEko/3Dc87OWLz98OPsVq9DmmQMP+9unTH8/L/Va7vbb30c2rt/+Hh4ds7e/cPH5elkDL/G3Dpjc9fdJPdM8sEkxzcb0IEtz2Gc3n49Y+v3ZJuZwzg7YzNzGLfO6GMOI9ncPMbZGZuZxzg7YzPzGGdnbGYOYhoZm5nHODtjnuYxWmRsZg7j7IzNzGGcnbGZOYyzMzYzh3F2xmbmMM7O6GUOI9ncPMatM7Y+v7yZeYxbZ2x9rrxNxtbn/edpHqNNxtbvDdtkbP0+dzNzGLfOGP7a4dYZfcxhJJubxzg7YzPzGLeas9/Ea/+zMzYzj3Gr+5AGGZuZx7jVfdkczWNsZg7j7IzNzGGcnbGZOYxbZ2z9vNvMHMbZGZuZw7hVRidzGJvNuXXG1ufKNzOPMcm4Z84zj7FnbW1tcOhWlFJ+JMlSrfVHJ98/Ncn9aq1Pv42mX53k7dOuDwAAANhxn5fkH3a6iC0whwEAAAC7xznnMXbinTE+mORfn/b95Uk+tIF2/3PS7oYkJ6ZQFwAAANCHD+50AVtkDgMAAAB2j3POY+zEO2N8TpJ3ZP2jST6V5H8k+c5a659tayEAAAAAAAAAAFOwuQ/p2YJa6/9J8iNJ/jDJe5K8xkIMAAAAAAAAAGBebPs7YwAAAAAAAAAAzLNtf2cMAAAAAAAAAIB5ZjEGAAAAAAAAAEBDFmMAAAAAAAAAADRkMQYAAAAAAAAAQEMWYwAAAAAAAAAANGQxBgAAAAAAAABAQxZjAFtSSrnDDj73qJRyeSnlsp3MmFellEs22a5pn+7kMXa6FnV0tC2b2re96mV7etm/cCHzdJxu5dx3D8Fu1NH1sos6gN3JHMb82ul7w7PyurjnNo/Rp162pZd9C7dlno7VnZxvdw/BLOromrmlOhZbFbLdSik/VWv90VLKwSS/mOShSVaTXJ/kB2qtR7epjrsn+YUkd58893+stZ6Y/OwNtdZHbCDj4iQ/luRxST4nyckkH0ry5iTPq7Ue2UDGJ5I8tdb62h3elgNJnpvk40n+W5LfTHKfJO9I8pRa64e2aVta9OmWj7FSyucm+U9JnpdkJcmvJ7l/kncl+fZa6/+zHXVMclrs37sn+ekkn07y/0vy20kun+yzK2qt/2s7tqeUcqckv5rkIVkfxz5WSllI8ltJvr/W+qntyGihxbE6RW9P8i82+o8b7ZcWx1ir8XCrdbTK2NJ5ex5D9+2Wx/ZJzpbHxPMYtD0tNNq/La4RLY73LWe00Og61aJPW9xD9HJ/2OI4bXL+T8ngc7+Xe4hGx2oX+6ajOlrcY/Zy/z+tPh16/Z/W9WHbr9s9MIdxzhzzGGdmmMc4M8McRuOMFuZpDiMxjzHFjLmYxzCHcasMcxhn1tHkWDePcc46zGNM9HIP0fnr3EE6qqPV//v1cv9vHuMCZnYxRpKHJ/nRrF8gPpDkn2f9nT6+O+sHymNuK6CU8mMX+nmt9Sc3UMfVSV6T5L1Jrkry+lLKo2qtx7N+wdmIV2f9oP43SW6YPHaXJE9K8htJHraBjH9K8rRSyjcn+cFa6/s2+Nyna7Etr0jy/iRfmuR7kjw/yauSPD7JLyd55AYyWmxLiz7d8jGW9W1/ZZJ/zPrF7VWT535U1vvqq7epjqTd/r0uySVJ3pnkGbXW60opX5X1gf5fb9P2vCzr/fpNSf5dkttnfZ9/f5Jfmzw+9YxGY8iWj9VSysuTrF2gjm/fQMaxJKPJt3smeXtKKSeTrNVaF24rI232S4tjrMUY0qKOFhlbPm8b7dsWY3vSYExssT2Nzt0W+7fFNaLF8b7ljBbjUNpcp1r0aYt7iF7uD1uNZVs6/1ucc43GsqSTe4i0OVa72Dcd1dHiHrOX+/8WfdrinGlxfWh17s4Dcxi3Zh7jTOYxzmQOo3GGOYxzMo/RPmOe5jHMYZzJHMaZWlynEvMYZzOPcaYu7iHSz+tccxjTyell/87tPMYsL8Y45Utrrd9y2vc/WUr56w22HSX5viT/Jeur9DbjDrXWlydJKeVRWb9gvSobG8BOKbXWs0+IDyb56VLKX20w43CSr03y7Vm/UP7vrK8+enuSD9ZaVzaQ0WJbPq/W+g2llFGSf6y1/srk8ZeXUv79BjMOZ+vb0qJPT9nKMXbJqT4opdyl1vprk8dfW0p53jbWkbTZv5fWWn9xkvHva63XJUmt9U9KKfsH5CRb25571Fp/a/L3q0sp76q1/lySHyqlbHRwbZHRYgxpcay+I8nPZn3l4bFN1vGAJC9K8rOn9msp5d211vsOyGjRpy2OscPZ+hjSoo4WGS3O2xb7tsXYnrQZE1tsT4tzt8X+bdEfh7P1471FRotxqMXx3qJPW4zLvdwftjhOW5z/Lc65Fud+0s89RItjtZd900sdp2z1nnmrGb3s2xbnzOFs/frQ6tydJ+YwPutwzGOczjzGmcxhtM8wh3Fr5jHaZ8zTPIY5jDOZwzhTi2M9MY9xNvMYZ+rlHqKX17nmMKaT08v+ndt5jL2bbdiBy0spj0/ywVLKvzz1YCnl/tngBbTW+qNZ3wmfqrX+xNlfG6xjtZTyRZO8tayv8jtUSnlpNr7Y5cZSyuNKKZ/ZH6WUPaWUb0py4wYzUmtdm5wkn5/1VVBfleR1SW7aYESLbVktpZRa62qSrzv1YCnlvhkwKDXYlhZ9uuVjLMmHSilPnfz97aWUh04yHpz1FVrbVUfSZv9+opTytFLKc5MsllIeManlqwbU0mJ7VkopXzNp93VJPjH5+5cnuXm7MhqNIVs+Vifnyi9l/YL3irO/Npjx7iQPSvJ1pZSrSym3ywVWiJ9Hi/3S4hhrMYa0qKNFxpbP20b7tsnYngZjYovtaXTutti/La4RLY73LWe0GIfS5jrVok9b3EP0cn/Yahza0vnf4pxrNJYlndxDpM2x2sW+6aiOFveYvdz/t+jTJudMg+tDq3N3HpjDOAfzGGcwj3EmcxiNM8xhnJN5jPYZ8zSPYQ7jTOYwztTiOpWYxzibeYwzdXEPkX5e55rDmE5OL/t3bucxZvmdMX4o6x14adZXMF5RSvneyd+/cUDOs5M8egt1PDvJG0opP1JrfU2tdbWsrxx8dZIv3GDGN2f94v+yUsqRrO/UA1lfqfPkDWbsOfWXWuvJJG+afA3RYlu+L+urje5da/2rJCmlPDrJi7Px/dJiW87u02S9T/84G+/T8x1jP5D1zz3biKckeWUp5flZf4uf7y3rn1n0wSSP3WIdmznWt7p/n5Tkp7K+kOtfJXlFWX9bt5uzvdvz3Ul+q5RyUZJPTjLuk+SlSZ4+MGMpyafOynjaBjOSrY8hLc7/JPnxJF+zhTpSa705yXdNjou3JRn6m0Jn75crN9GnFzrGrthgRosxpMWx3mJbWpy3LfZti7E9aTMmnr49j8zmtifZ+rnb4hi5UH9s5/HeIiPZ+jjU4nhvcYy1GJd7uT9scZx+32RbvmCL5/9Wz7lW536L63+L612L87+XfdPqGrHVOlrcu7e4T20xDjXp0wbX/ybXhwZ1zAtzGLdmHuNM5jHOZA7j3BnmMCYaXV/MY2w8YzfOY5jDOJM5jDM1OdZjHuNs5jHOZA7j1sxhbCzHPEZH8xh71tbm55dSSim3T/LJSQdv93OPJit+Tn/sy2qt7xmQsZjkjlm/0Hy0rn8e10bbHqq1DvoNlAtkbXlbzmq7L8nqRvdL423ZdJ+eJ29/1le7DTrGSil3yPpnNS0m+XCt9f/dYh2bPtansH/vWGvd8Orn82RsansaPfeWM1pofaw2qOfyJA+vn31LqiFtm/bp0LyWY8hW6miZUUoZ17PeQmuz5+1W9u3ZNSU5vslxqNmYWEq5c5JHbHV7WtjC/t10f7Q43qd1zmxWi+O9xTHWYlze6fvD8+S1GMsG3dtNw1bP/V7uIaZwj7jj+2ZSx6avEY3r2NS9+1kZm71P7WrfTs6Zh9darx7Qpvn1YTN1zLPdPIcxaW8e49xZ5jHObGcOo3FGC/M0hzFpbx6jccY8zWOYw7hVO3MYp2l1rJvHuGDerp/H6OUeorfXua3M0xzGJMc8RkMt5jFmdjFGKeVAkquS3CPJ9bXWV532s1+ptX7nBjN+Isndt5ixpTp6UdbfPuqpWV91dbesv3XMh7K+augXzr4IT7GOFvtlMcm3Zv1tZ34v66v87pP1VZg/VGs9uoGMLfdHL306qaXFOdPF9vRy3l2gP96c5EXb2B9PutDPa62/voGMFuddi3Om5XH6jUk+Jzu3X3o5Tlvu27P7dNvHshZa9EnDOq7KfBwjWx6HWuhwHFLHhevY1nG5lY76tUXGqXvmjyf53Xz2nvkdSX5w4D3zpq8RLY6RXvZLL1rcD/WSMS/MYUxHZ69RzWM01Mu9UAu9nHe93JP1Mocx+bdd3D/0cr3s6Fg1j3EacxjnrGMu5jAmtfQ2DqnjwnXM3OuYjvrUHMbGMgZdp3oZm1voZQ5iGvdli0MbdOTlSf4yyWuS/FAp5WtOO6i+fGDG2xtkbLqORi9CWtxAvDTrqxWvSnJD1t/O5fIk35L17fzmbaqjxX75lSQXJblTkh9L8sYk/yHrA9tLkzxxAxlb7o8WGQ1vDlucM71sTxfnXfrpj6/N+tuB/WZOexumibUk23XetThnWh6nP77ZOubsOG25bzfdp0lX17st94lj5Fa2PA71Mi6nk+ulOs7U6n5ozo6zFhmn3zP/aLZ2z7yVa0QX993pZFxufJy22C87nTEvzGFMJ6eXa4x5jNP0cq+cOdqWOTtfepnDSDq5f4h5jPPVsaPzGB3d15nDOHcd8zCHkfQ3DqmjUR0djSG99Kk5jPYZSSdjs3mMC5vlxRifV2t9bJKUUt6U9c8a+i+11u/PrS+gvWe0eBHSIuNraq1fcNZjf5fkHaWU922gfas6WvTp/Wut9ymlXJzkA7XWH5k8/hOllHdvMKNFf/TSp0mbfu1le3o577roj1rrk0splyZ5R938WyX1cnz0Usc8Hae99GnSz/VunvZNF3U0God6GZd7OWfUcaZW90PzdJy5Z26f0cu43Mtx2kvGvOjlvO/lOG+V08s1xph8pl76dJ62ZW7Ol47mMJJ+7h+62Dfp51idpz7tpT/UcZqO5jCS+RqH1HGmXsaQXvrU/XL7jKSfsbmXY7WXjDPM8mKMlFIur7V+uNZ6cynlsUn+uJTyw1nfsTOT0eLi3+gG4mgp5f611v95+oOllAckObqNdbTYLyfL5DOwSimfWaVUSrlb1lc0bcSW+6NFRqs+nTzvVvu1m+3p4bxLR/2R5GnZ2MrP8+rh+Oiljjk7Trvo06Sr693c7Jte6pjY0jjUy7g8+fddnDPq+KxWx/qcHWfumdtndDEud3Sc9pIxNzo577s4zhvmdHONMSZ/Vkd9OjfbMm/nS/qYw0g6uX9oUcecHatz06e99Ic6zqmHOYxkjsYhdZypozGklz51v9w+41SbHR+bOzpWe8k4wywvxrgqybtKKd9Va31drfVIKeUhSV6f5EtmLCNp8CKkQcZTk7yylLKU9bdeWUty1yQ3D8zdah1Xpc1+eXcp5Z611rckSSnl65O8KuvbuREt+qOXPk3a9Gsv23NV+jjveumP1FpvSPIzW4i4Kn0cH73UkczPcdqijlZ9mvRxvbsq87NvuqmjwTiU9DEuX5U+zhl13FqL8aNFTi/92ipjXu6Ze9kvSR/Xul76tOU9xKy7Kv2c970c5y1yernGXBVj8tl66NN52pYWdfTSH73MYST93D/0sm+uSh/Haos6eunTFhlXZX72Szd1dDKHkczXOKSOW+thDOmlT1tlzMv9cqtj7Kr0MTa3yOilX5vPY+xZWxuyWLAvpZT9SUa11ptOe2xvkkfVWn97ljJ6Ukq5R9YPrL1JPlhr/cAO1NBiv1xca/30ad8fTLL39MwN5my5P3ro00kdTY7VHranp/Ouh/44WynlDbXWRwxs08Xx0UsdLfRynM5Tn7Yyb/umxzo2Mw61Mi/jkDr61kO/umdun9HLuNxKD33aKmMedHTez9VxnvRxjBmT25une5Cezrse+uNsOzmHMWnXxf1DD/uml2N1nvq0hXnbLz3WsZNzGJPnn4txSB396qFP3S9PLaOLsbmVjvq13fm/trY2N1/Ly8tvkCFDxuzVImMqGX/RSR0yZMxELTKmkmEcktF9Rk+1yJAhY/d99dKfvWT0VIsMGTK2PaOL1w6d9YkMGTK2N8M4JEOGjJnP6KkWGZ/92rvpVRx9uqsMGTK2PUdGnxl7GmT0si0y5jejVY6MPjOMQzJmIaNVjgwZMrY3Y1700p+9ZLTKkSFDxuxl9PLaoVWODBkyZi/DOCRDhox5yGiVI6NhxrwtxmhxwZQhY7dktMqR0WfGwxpk9LItMuY3o1WOjD4zjEMyZiGjVY4MGTK2N2Ne9NKfvWS0ypEhQ8bsZfTy2qFVjgwZMmYvwzgkQ4aMecholSOjYcZigwJ60uKCKUPGbslolSOjw4xa6w091CFDxjblyNjhjFLK3ZP8QpK7J7k+yX88NQ5t8XNXZ7I/ZGzIQxtkGIdkyNi9GfOil/7sJaNVjgwZMmYso6M5jFY5MmTI6Dij8zmMVjky2mf0Mo8hQ8Z25MhomLFnbW2tQQ2713ku3CcmP9vKhXsm62hhnralBf0x31rs31LKj13o57XWn9yOOjhTo31rv5yll37tJaMXpZTfS/KaJO9NclWShSSPrrWullLeXWu97zbV0UWf9nJ8dJRxIMlzk3w8yX9L8ptJ7pPkHUmeUmv90G1ltNJLn8yTXvpURp8Z9Ken/dpTLVs1T9vSgv6YX+Yw5pv7h/Z66dNeMnrRyxzGpJYu+rWXY6SjjC7mMXrpj3nSS5+22i+91DJPGdMws++M0ejGfcsZSa7OmRfu15dSHl1rXU3yORtoP1d12Jb2GWnQH61qkdE+I2327yjJ9yX5L0lObrBN8zp66dNeMtJm387NfmmY00W/9pLR0f69Q6315ZO8R2V9u16Z5Js20LZlHV30aYs65izjFUnen+RLk3xPkucneVWSxyf55SSPvK2AVuNQOumTXs5dx7uMaWc0PHdn3jydsz2Nyfq1z4zoj7nNyBzNYST99GsvGZmj+4deMtJJn/aS0dF+6WUOI+mkX1vUMWcZvcxjdNEfvZy789SnjTJ6qmVuMqYxj7F3aIOOjJL8QNZXLe45x9d2Zdyh1vryWuu7kjwqyZGsX7iHmKc6bEv7jBb90aoWGe0ztrx/a60/mvUVup+qtf7E2V/bVUf66dNeMlr06Tztl1Y5vfRrLxm97N/VUsoXJUmtdS3Jk5IcKqW8NBtf/NvL8dFLHfOU8Xm11v8ryXckGddaf6XW+unJ5NeQSfsW41AvfdLLuet4lzHtjFbn7jyYp3O2pzFZv/aZoT/mN2Oe5jCSfvq1l4x5un/oJaOXPu0lo5f90sscRtJPv/ZyjPSS0cs8Ri/90cu5O0992upeppda5imj+TzGTH9MSSnlZUn+d631BTuVUUr50yTfXmv968n3oyRvSfK3Sb6q1nqf3VRHi4x52pYWGa36o0UtMtpnNDzeb5/1t9PbzAXbeTeFjBZ9Om/7pUVOL/3aS8ak3Y7v31LKV2X9NwR+pNb6msljlyR5dZJH1loXtqmOLvq0l+Ojo4w/T/LEWmstpXxxrfWvJo/fN8mv1lq//LYyJv++xbHeRZ9M2u34udsio5c+ldFnxqRdk3uIeTAv52yLOjqsZZ76dccz9Mf8ZszbHMak7Y73ay8Z83b/0ENGL33aS8akXQ/7pYs5jElGF/3ayzHSUUYX8xi99EeLbeklo5c+bbhfuqhlnjIm7ZrOY8zyO2MkybOT3LDDGc9O8oZSyhOSpK6/1cmjklye5At3YR0tMuZpW1pktOqPFrXIaJ/RZP/WWj+x2UmMlnWkjz7tJaNFn87bfmmR00u/9pJxKmdH92+t9U9qrZ+X5LdOe+xTtdZvSHK/7aoj/fRpL8dHLxnfl/W3BVw4bQLj0Ulen+RZG8w4VUuLY72HPjmV08PY7HiXMc2MUzkt7iHmwbycsy3q6LGWeerXnc7QH/ObMW9zGKeydrpfe8mYt/uHHjJ66dNeMk7lmMM4M6OHfu3lGOklo5d5jF7641TOTo+pLTJ66dOW+6WHWuYp41ROu3mMtbU1Xw2+lpeXR+d47Mt2ax22RX/42r79u7y8vHd5eflpy8vLv7+8vPw3y8vL75v8/fvPlTutOnxNr0/tl377tZeMnf66wDj07KHjUC/HRy91zFPGWW3Hy8vLe7d7n/TcJ7P+1Uufyugzw1d/Xz3t155qsS36w9f09605jP6/3D/Mb5/2krHTX73NYfTUr70cI71knNV2x+YxeuyPWf/qpU9b7ZdeapmnjJZfG/38q+6UUvYmeWqSxyW5W5KTST6U5M1JXjRZ7bKtGaWU0zPelOQXdmJbeqnDtvTTH71uj4xm+/elWX+no6uyvlpvT9ZX+n1Lkpcn+ebtqKPHPu0lo0Wfzvp+mUYtvfRrLxnZ2f3bdBxqsS299Gkvx0cnGd+Y9c9W/UxGKeUXdmoc6qRPurlWtcjopU9l9JmRLZy782Bez9nN1tFTLfPUrz1m6I/5zZj1OYyzc9JJv/aSMev3Dz1m9NKnvWTEHMatcnrp116OkU4ydnQeo8P+6GZMbZHRS5+2upeZ9e3pMSON5jFmdjFGGlwwZcjYRRk91SKjfcbX1Fq/4KzH/i7JO0op79tA+1Z1yJAxK7XIaJ9hHJKxkYwf3+E6WuXIkCFj+8/dedBLf/aS0VMtMmTI2J2vHVrlyJAhY/YyjEMyNpqx0/MYMmTMSi0yLmSn3pJjq1/Ly8t/c4GfvU+GDBl91iJjKhn/c3l5+f7nePwBy8vLfzpj2yJjTjN6qkXGVDKMQzK6z+ipFhkyZAw7d+fhq5f+7CWjp1pkyJCxO187dNYnMmTIMA7tdJ/IkCFjxjJ6qkXGhb/2Dl690Y+jpZT7n/1gKeUBSY7KkCGj21pktM94apKrSyl/W0r541LKfy+l/G2SX07ylG2sQ4aMWalFRvsM45CMWcjoqRYZMmQMO3fnQS/92UtGT7XIkCFjd752aJUjQ4aM2cswDsmQIWMeMnqqRcYFzPLHlDw1yStLKUtZf5uQtSR3TXJzkifKkCGj21pkNM6otb4nyX1KKfeYtN2b5IO11g9ssIYmdciQMUO1yGicYRySMSMZPdUiQ4aMYefuPOilP3vJ6KkWGTJk7M7XDq1yZMiQMWMZxiEZMmTMSUZPtci4gD1ra2ubadeNLV4wZcjYVRk91SKjfcZZeW+otT5iJ+qQIWNWapHRPuOsPOOQjC4zeqpFhgwZu08v/dlLRk+1yJAhY3szzsrbsdcOrXJkyJAxexln5RmHZMiQMZMZPdUi4zw289kmvX4tLy+/QYYMGbNXi4ypZPxFJ3XIkDETtciYSoZxSEb3GT3VIkOGjN331Ut/9pLRUy0yZMjY9owuXjt01icyZMjY3gzjkAwZMmY+o6daZHz2a++mV3H06a4yZMjY9hwZfWbsaZDRy7bImN+MVjky+swwDsmYhYxWOTJkyNjejHnRS3/2ktEqR4YMGbOX0ctrh1Y5MmTImL0M45AMGTLmIaNVjoyGGfO2GKPFBVOGjN2S0SpHRp8ZD22Q0cu2yJjfjFY5MvrMMA7JmIWMVjkyZMjY3ox50Ut/9pLRKkeGDBmzl9HLa4dWOTJkyJi9DOOQDBky5iGjVY6MhhnzthijxQVThozdktEqR8YOZ5RSfmry58FSyqtKKR9L8p5SyktKKfu3qw4ZMnYwR8YOZxiHZMxwRqscGTJkbG/GvOilP3vJaJUjQ4aMjjM6f+3QKkeGDBkdZxiHZMiQMccZrXJkNMyY2cUYLS6YMmTsloyeapHRPiPJwyd//kKSDyT550m+MMkNSX59lrZFxvxm9FSLjPYZMQ7JmIGMnmqRIUPGlie6Z04v/dlLRk+1yJAhY3e+dmiVI0OGjNnLiHFIhgwZc5DRUy0yLmxmF2OkwQVThoxdlNFTLTLaZ5zypbXWH661Hq613lRr/ckky9tYhwwZs1KLjPYZpxiHZPSc0VMtMmTIGH6NmXW99GcvGT3VIkOGjN352qFVjgwZMmYv4xTjkAwZMmY5o6daZFzI2traTH4tLy//xeTP957jZ38tQ4aMPmuRMZWMDy0vLz9+eXn5TcvLy//ytMfvv7y8/K4Z2xYZc5rRUy0yppJhHJLRfUZPtciQIWPYuTsPX730Zy8ZPdUiQ4aM3fnaobM+kSFDhnFop/tEhgwZM5bRUy0yLvy1d1MrOPpweSnl8Uk+WEr5l6ceLKXcP8kxGTJkdFuLjPYZP5Tkq5JcmuS5k/bfm+S3kzxrG+uQIWNWapHRPsM4JGMWMnqqRYYMGcPO3XnQS3/2ktFTLTJkyNidrx1a5ciQIWP2MoxDMmTImIeMnmqRcQGzvBijxQVThozdktFTLTIaZ9Raf73W+qxa6wNqrVdMHr46yd1rrX+yXXXIkDFDtchonGEckjEjGT3VIkOGjGHn7jzopT97yeipFhkyZOzO1w6tcmTIkDFjGcYhGTJkzElGT7XIuIA9a2trm2nXpVLK7ZN8stZ6UoYMGbNTi4z2GS30si0y5jejp1pktM9ooZdtkTG/GT3VIkOGjN2nl/7sJaOnWmTIkLG9GS0Yh2TIkGEckiFDxm7P6KkWGZ81V4sxAHajUsqTLvTzWuuvb1ctwO5kHAIAADbCawdgpxmHANhOiztdwGa1uGDKkLFbMnqqRUb7jCRfm+SKJL+ZZM9ZP1tLMjPbImN+M3qqRUb7jBiHZMxARk+1yJAhY+MZ86KX/uwlo6daZMiQsb0Z6eS1Q6scGTJkzF5GjEMyZMiYg4yeapFxYTO7GCMNLpgyZOyijJ5qkdE4o9b65FLKpUneUWu9egPPOZU6ZMiYoVpkNM4wDsmYkYyeapEhQ8awc3ce9NKfvWT0VIsMGTJ252uHVjkyZMiYsQzjkAwZMuYko6daZFzI2trazH4tLy//zvLy8rfLkCFjtmqRMZWMuywvLz+ngzpkyJiJWmRMJcM4JKP7jJ5qkSFDxu776qU/e8noqRYZMmRse0YXrx066xMZMmRsb4ZxSIYMGTOf0VMtMs7/tXfw6o2+PC3JZTJkyJi5WmQ0zqi13lBr/ZmdrkOGjBmqRUbjDOOQjBnJ6KkWGTJk7D699GcvGT3VIkOGjG3M6Oi1Q6scGTJkzFiGcUiGDBlzktFTLTLOY8/a2lqrLAB2SCnlwUkel+RuSU4m+VCSN9dar93RwoBdwzgEAABshNcOwE4zDgGwXWZ6MUaLC6YMGbslo6daZLTNKKX8ZJKvSPKqJDdk/XOsLk/yhCTvq7U+Z1a2RcZ8Z/RUi4y2GcYhGbOS0VMtMmTI2H0T3b30Zy8ZPdUiQ4aM3fnaoVWODBkyZivDOCRDhox5yeipFhnnN7OLMVpcMGXI2C0ZPdUiYyoZNcm9a60nz3p8Iclf1VrvPUPbImNOM3qqRcZUMoxDMrrP6KkWGTJkDDt350Ev/dlLRk+1yJAhY3e+dmi4PTJkyJi9DOOQDBkyZj6jp1pk3Ia1tbWZ/FpeXq7Ly8t7z/H4wvLy8vtlyJDRZy0yppLxv5aXl+9xjsf/2fLy8ntmbFtkzGlGT7XImEqGcUhG9xk91SJDhoxh5+48fPXSn71k9FSLDBkydudrh876RIYMGcahne4TGTJkzFhGT7XIuPDX3sGrN/pxLOtvD3K2z01yiwwZMrqtRUb7jO9P8vZSyu+VUn69lPKKUsrvJfnDJN+3jXXIkDErtchon2EckjELGT3VIkOGjGHn7jzopT97yeipFhkyZOzO1w6tcmTIkDF7GcYhGTJkzENGT7XIuIDFzTTqxKkL5v/O+tuErCW5a5LlJN8qQ4aMbmuR0Tij1vr7pZQfnLQ5keTvk3wwyZ8meXLWX0jMxLbImOuMnmqR0TjDOCRjRjJ6qkWGDBnDzt150Et/9pLRUy0yZMjYna8dWuXIkCFjxjKMQzJkyJiTjJ5qkXEBM/vOGLXW30/yg0nenqQmeWuSn8p6Z3y+DBky+qxFRvuMUsp/SvIdSQ4leVaSvbXWP6613pLk6bO0LTLmN6OnWmS0zzAOyZiFjJ5qkSFDxrBzdx700p+9ZPRUiwwZMnbna4dWOTJkyJi9DOOQDBky5iGjp1pkXNjMLsZoccGUIWO3ZPRUi4z2GUkenuShtdZ/n+Srk/xUKeVxk5/tmaVtkTG/GT3VIqN9RoxDMmYgo6daZMiQMezcnQe99GcvGT3VIkOGjN352qFVjgwZMmYvI8YhGTJkzEFGT7XIuLCZXYyRBhdMGTJ2UUZPtchon7En62+VlFrr3yZ5RJKfL6U88NTj21SHDBmzUouM9hnGIRmzkNFTLTJkyBh27s6DXvqzl4yeapEhQ8bufO3QKkeGDBmzl2EckiFDxjxk9FSLjAuY5cUYLS6YMmTsloyeapHRPuO3kvxRKeUrJjl/neRxSX4zyT/fxjpkyJiVWmS0zzAOyZiFjJ5qkSFDxrBzdx700p+9ZPRUiwwZMnbna4dWOTJkyJi9DOOQDBky5iGjp1pkXMAsL8ZoccGUIWO3ZPRUi4zGGbXWn0hyVZKjpz32J0nul+Tl21WHDBkzVIuMxhnGIRkzktFTLTJkyBh27s6DXvqzl4yeapEhQ8bufO3QKkeGDBkzlmEckiFDxpxk9FSLjAuY2cUYLS6YMmTsloyeapHRPmPS5m211vef9dg/1lq/d7vqkCFjVmqR0T5j0sY4JKPrjJ5qkSFDxrBzdx700p+9ZPRUiwwZMnbna4dWOTJkyJi9jEkb45AMGTJmOqOnWmRc2J61tU29owYAAAAAAAAAAOcws++MAQAAAAAAAADQI4sxAAAAAAAAAAAashgDAAAAAAAAAKAhizEAgFsppXx5KeWaTbZ9Tynl4Dkef04p5b9uoaYDpZQ/2ES7DT1vKeVXSyn321RxAAAAwI4whwEA9GpxpwsAAPpTa/3zJFdusu2Xta3mMy5N8hVTyk6Sr0/yy1PMBwAAABozhwEA9MpiDADgVkopD0zyi0n+PMknktwnyd2TvDfJk2qtnyylfGWSFyW5JMlKkufUWv+glLKW5FCSI5Off32Sjyb5yOSxlFIOJPn5Se4oyduSPLfWeryUcizJf0ryoCR3SfKCWutLkrw8yUWllPckuV+t9cR5ah9d4Hn/ZZIXJNk3yf69Wut3lFL+Q5K7Jnl1KeVJSf7mfPVtoVsBAACAxsxhmMMAgF75mBIA4LbcL8lDktw7yT2TPG4yWfDbSX6y1vrFSZ6a5OdLKaffW3xXkuUkX5j1SYV7nPazn03yrlrr/ZLcN8kdkzx78rN9Sf6p1vqvsv6bLT9bSllK8m1Jbq61ftn5JjE28Lzfk+THaq1fOfn5o0op96u1/kiSDyV5Yq31T2+jPgAAAKBP5jDMYQBAN7wzBgBwW95Sa70lSUopf5nksqz/tsWJWusbk6TW+q7JYymlnGr3dUleU2tdSbJSSnl1ki+Z/OwRSb6ilPIdk+8vOus5Xzf58y+yPrFxyYB6L/S8T07ysFLKDyf5gsnz3u4cGbdVHwAAANAfcxjmMACgGxZjAAC35ebT/r6WZE+S45O/f0Yp5Yuz/taYp9tz2t9Pf3vMhSSPq7W+f9L24Fl5NydJrXVtMjFyes5GnO95/zjrb1P6liS/meQrz5N9W/UBAAAA/TGHYQ4DALrhY0oAgM2oSdZKKV+fJKWUf5HkD3LmvcWbkzyplLI0eYvOx5/2s7cm+b5Syp5Syr4kv5Pku2/jOY8nWSil3NakxjmfdzIZcf8kP1hrvS7J3ZLcK+uTFqfyR1uoDwAAAOiPOQwAYEdYjAEADDZ5y8/HJvnxUsp7krw0yWMnb6t5yi8n+fMkf5Xkvyf5+9N+9qysv23nX2b9tzz+MskLbuNpb0jyZ0n+upRyhwv8u3M+b631cJL/mOQvSil/leSHkvxJ1iczkuS6JK8qpTxok/UBAAAAnTGHAQDslD1ra96tCgAAAAAAAACglcWdLgAAYKhSytuT7D/Pj/91rfXodtYDAAAAcC7mMABg9/LOGAAAAAAAAAAADe3d6QIAAAAAAAAAAOaJxRgAAAAAAAAAAA1ZjAEAAAAAAAAA0JDFGAAAAAAAAAAADVmMAQAAAAAAAADQ0P8fJZx0o0fXtKQAAAAASUVORK5CYII=",
- "text/plain": [
- "<Figure size 2160x360 with 2 Axes>"
- ]
- },
- "metadata": {
- "needs_background": "light"
- },
- "output_type": "display_data"
- }
- ],
- "source": [
- "# data_cat.incident_date.values\n",
- "chart = sns.catplot(x=\"incident_date\", col=\"fraud_reported\", data=data_cat, kind=\"count\", aspect=3)\n",
- "chart.set_xticklabels(rotation=90)\n",
- "# plt.savefig('./Daten/VglIncidentVSFraud')\n",
- "plt.show()"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 24,
- "id": "9496b3f1",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "<Figure size 1728x1728 with 0 Axes>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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",
- "text/plain": [
- "<Figure size 720x360 with 2 Axes>"
- ]
- },
- "metadata": {
- "needs_background": "light"
- },
- "output_type": "display_data"
- }
- ],
- "source": [
- "plt.figure(figsize=(24, 24))\n",
- "chart = sns.catplot(x=\"auto_make\", col=\"fraud_reported\", data=data_cat, kind=\"count\")\n",
- "chart.set_xticklabels(rotation=45)\n",
- "# plt.savefig('./Daten/VglIncidentVSFraud')\n",
- "plt.show()"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 25,
- "id": "2987fc69",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "<Figure size 2304x2304 with 0 Axes>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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",
- "text/plain": [
- "<Figure size 1080x360 with 2 Axes>"
- ]
- },
- "metadata": {
- "needs_background": "light"
- },
- "output_type": "display_data"
- }
- ],
- "source": [
- "plt.figure(figsize=(32, 32))\n",
- "chart = sns.catplot(x=\"auto_model\", col=\"fraud_reported\", data=data_cat, kind=\"count\", aspect=1.5)\n",
- "chart.set_xticklabels(rotation=90)\n",
- "# plt.savefig('./Daten/VglIncidentVSFraud')\n",
- "plt.show()"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 26,
- "id": "2c3a012f",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "<Figure size 2304x2304 with 0 Axes>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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",
- "text/plain": [
- "<Figure size 1080x360 with 2 Axes>"
- ]
- },
- "metadata": {
- "needs_background": "light"
- },
- "output_type": "display_data"
- }
- ],
- "source": [
- "plt.figure(figsize=(32, 32))\n",
- "chart = sns.catplot(x=\"policy_csl\", col=\"fraud_reported\", data=data_cat, kind=\"count\", aspect=1.5)\n",
- "chart.set_xticklabels(rotation=90)\n",
- "# plt.savefig('./Daten/VglIncidentVSFraud')\n",
- "plt.show()"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 27,
- "id": "49932ce2",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/html": [
- "<div>\n",
- "<style scoped>\n",
- " .dataframe tbody tr th:only-of-type {\n",
- " vertical-align: middle;\n",
- " }\n",
- "\n",
- " .dataframe tbody tr th {\n",
- " vertical-align: top;\n",
- " }\n",
- "\n",
- " .dataframe thead th {\n",
- " text-align: right;\n",
- " }\n",
- "</style>\n",
- "<table border=\"1\" class=\"dataframe\">\n",
- " <thead>\n",
- " <tr style=\"text-align: right;\">\n",
- " <th></th>\n",
- " <th>policy_csl</th>\n",
- " <th>fraud_reported</th>\n",
- " </tr>\n",
- " </thead>\n",
- " <tbody>\n",
- " <tr>\n",
- " <th>1</th>\n",
- " <td>250/500</td>\n",
- " <td>0.262108</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>0</th>\n",
- " <td>100/300</td>\n",
- " <td>0.257880</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>2</th>\n",
- " <td>500/1000</td>\n",
- " <td>0.216667</td>\n",
- " </tr>\n",
- " </tbody>\n",
- "</table>\n",
- "</div>"
- ],
- "text/plain": [
- " policy_csl fraud_reported\n",
- "1 250/500 0.262108\n",
- "0 100/300 0.257880\n",
- "2 500/1000 0.216667"
- ]
- },
- "execution_count": 27,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "data_cat[['policy_csl', 'fraud_reported']].groupby(['policy_csl'], as_index=False).mean().sort_values(by='fraud_reported', ascending=False)\n"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 28,
- "id": "7e4a21c5",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/html": [
- "<div>\n",
- "<style scoped>\n",
- " .dataframe tbody tr th:only-of-type {\n",
- " vertical-align: middle;\n",
- " }\n",
- "\n",
- " .dataframe tbody tr th {\n",
- " vertical-align: top;\n",
- " }\n",
- "\n",
- " .dataframe thead th {\n",
- " text-align: right;\n",
- " }\n",
- "</style>\n",
- "<table border=\"1\" class=\"dataframe\">\n",
- " <thead>\n",
- " <tr style=\"text-align: right;\">\n",
- " <th></th>\n",
- " <th>insured_hobbies</th>\n",
- " <th>fraud_reported</th>\n",
- " </tr>\n",
- " </thead>\n",
- " <tbody>\n",
- " <tr>\n",
- " <th>5</th>\n",
- " <td>chess</td>\n",
- " <td>0.826087</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>6</th>\n",
- " <td>cross-fit</td>\n",
- " <td>0.742857</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>19</th>\n",
- " <td>yachting</td>\n",
- " <td>0.301887</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>2</th>\n",
- " <td>board-games</td>\n",
- " <td>0.291667</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>14</th>\n",
- " <td>polo</td>\n",
- " <td>0.276596</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>15</th>\n",
- " <td>reading</td>\n",
- " <td>0.265625</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>0</th>\n",
- " <td>base-jumping</td>\n",
- " <td>0.265306</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>10</th>\n",
- " <td>hiking</td>\n",
- " <td>0.230769</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>13</th>\n",
- " <td>paintball</td>\n",
- " <td>0.228070</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>16</th>\n",
- " <td>skydiving</td>\n",
- " <td>0.224490</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>18</th>\n",
- " <td>video-games</td>\n",
- " <td>0.200000</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>17</th>\n",
- " <td>sleeping</td>\n",
- " <td>0.195122</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>8</th>\n",
- " <td>exercise</td>\n",
- " <td>0.192982</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>1</th>\n",
- " <td>basketball</td>\n",
- " <td>0.176471</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>12</th>\n",
- " <td>movies</td>\n",
- " <td>0.163636</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>3</th>\n",
- " <td>bungie-jumping</td>\n",
- " <td>0.160714</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>7</th>\n",
- " <td>dancing</td>\n",
- " <td>0.116279</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>9</th>\n",
- " <td>golf</td>\n",
- " <td>0.109091</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>11</th>\n",
- " <td>kayaking</td>\n",
- " <td>0.092593</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>4</th>\n",
- " <td>camping</td>\n",
- " <td>0.090909</td>\n",
- " </tr>\n",
- " </tbody>\n",
- "</table>\n",
- "</div>"
- ],
- "text/plain": [
- " insured_hobbies fraud_reported\n",
- "5 chess 0.826087\n",
- "6 cross-fit 0.742857\n",
- "19 yachting 0.301887\n",
- "2 board-games 0.291667\n",
- "14 polo 0.276596\n",
- "15 reading 0.265625\n",
- "0 base-jumping 0.265306\n",
- "10 hiking 0.230769\n",
- "13 paintball 0.228070\n",
- "16 skydiving 0.224490\n",
- "18 video-games 0.200000\n",
- "17 sleeping 0.195122\n",
- "8 exercise 0.192982\n",
- "1 basketball 0.176471\n",
- "12 movies 0.163636\n",
- "3 bungie-jumping 0.160714\n",
- "7 dancing 0.116279\n",
- "9 golf 0.109091\n",
- "11 kayaking 0.092593\n",
- "4 camping 0.090909"
- ]
- },
- "execution_count": 28,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "data_cat[['insured_hobbies', 'fraud_reported']].groupby(['insured_hobbies'], as_index=False).mean().sort_values(by='fraud_reported', ascending=False)\n"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 29,
- "id": "e2cbb5c7",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/html": [
- "<div>\n",
- "<style scoped>\n",
- " .dataframe tbody tr th:only-of-type {\n",
- " vertical-align: middle;\n",
- " }\n",
- "\n",
- " .dataframe tbody tr th {\n",
- " vertical-align: top;\n",
- " }\n",
- "\n",
- " .dataframe thead th {\n",
- " text-align: right;\n",
- " }\n",
- "</style>\n",
- "<table border=\"1\" class=\"dataframe\">\n",
- " <thead>\n",
- " <tr style=\"text-align: right;\">\n",
- " <th></th>\n",
- " <th>insured_occupation</th>\n",
- " <th>fraud_reported</th>\n",
- " </tr>\n",
- " </thead>\n",
- " <tbody>\n",
- " <tr>\n",
- " <th>3</th>\n",
- " <td>exec-managerial</td>\n",
- " <td>0.368421</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>4</th>\n",
- " <td>farming-fishing</td>\n",
- " <td>0.301887</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>2</th>\n",
- " <td>craft-repair</td>\n",
- " <td>0.297297</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>13</th>\n",
- " <td>transport-moving</td>\n",
- " <td>0.291667</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>12</th>\n",
- " <td>tech-support</td>\n",
- " <td>0.282051</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>11</th>\n",
- " <td>sales</td>\n",
- " <td>0.276316</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>1</th>\n",
- " <td>armed-forces</td>\n",
- " <td>0.246377</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>6</th>\n",
- " <td>machine-op-inspct</td>\n",
- " <td>0.236559</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>10</th>\n",
- " <td>protective-serv</td>\n",
- " <td>0.222222</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>9</th>\n",
- " <td>prof-specialty</td>\n",
- " <td>0.211765</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>5</th>\n",
- " <td>handlers-cleaners</td>\n",
- " <td>0.203704</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>0</th>\n",
- " <td>adm-clerical</td>\n",
- " <td>0.169231</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>7</th>\n",
- " <td>other-service</td>\n",
- " <td>0.169014</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>8</th>\n",
- " <td>priv-house-serv</td>\n",
- " <td>0.169014</td>\n",
- " </tr>\n",
- " </tbody>\n",
- "</table>\n",
- "</div>"
- ],
- "text/plain": [
- " insured_occupation fraud_reported\n",
- "3 exec-managerial 0.368421\n",
- "4 farming-fishing 0.301887\n",
- "2 craft-repair 0.297297\n",
- "13 transport-moving 0.291667\n",
- "12 tech-support 0.282051\n",
- "11 sales 0.276316\n",
- "1 armed-forces 0.246377\n",
- "6 machine-op-inspct 0.236559\n",
- "10 protective-serv 0.222222\n",
- "9 prof-specialty 0.211765\n",
- "5 handlers-cleaners 0.203704\n",
- "0 adm-clerical 0.169231\n",
- "7 other-service 0.169014\n",
- "8 priv-house-serv 0.169014"
- ]
- },
- "execution_count": 29,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "data_cat[['insured_occupation', 'fraud_reported']].groupby(['insured_occupation'], as_index=False).mean().sort_values(by='fraud_reported', ascending=False)\n"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 30,
- "id": "267dec9b",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/html": [
- "<div>\n",
- "<style scoped>\n",
- " .dataframe tbody tr th:only-of-type {\n",
- " vertical-align: middle;\n",
- " }\n",
- "\n",
- " .dataframe tbody tr th {\n",
- " vertical-align: top;\n",
- " }\n",
- "\n",
- " .dataframe thead th {\n",
- " text-align: right;\n",
- " }\n",
- "</style>\n",
- "<table border=\"1\" class=\"dataframe\">\n",
- " <thead>\n",
- " <tr style=\"text-align: right;\">\n",
- " <th></th>\n",
- " <th>insured_education_level</th>\n",
- " <th>fraud_reported</th>\n",
- " </tr>\n",
- " </thead>\n",
- " <tbody>\n",
- " <tr>\n",
- " <th>6</th>\n",
- " <td>PhD</td>\n",
- " <td>0.264000</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>4</th>\n",
- " <td>MD</td>\n",
- " <td>0.263889</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>1</th>\n",
- " <td>College</td>\n",
- " <td>0.262295</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>3</th>\n",
- " <td>JD</td>\n",
- " <td>0.260870</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>0</th>\n",
- " <td>Associate</td>\n",
- " <td>0.234483</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>2</th>\n",
- " <td>High School</td>\n",
- " <td>0.225000</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>5</th>\n",
- " <td>Masters</td>\n",
- " <td>0.223776</td>\n",
- " </tr>\n",
- " </tbody>\n",
- "</table>\n",
- "</div>"
- ],
- "text/plain": [
- " insured_education_level fraud_reported\n",
- "6 PhD 0.264000\n",
- "4 MD 0.263889\n",
- "1 College 0.262295\n",
- "3 JD 0.260870\n",
- "0 Associate 0.234483\n",
- "2 High School 0.225000\n",
- "5 Masters 0.223776"
- ]
- },
- "execution_count": 30,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "data_cat[['insured_education_level', 'fraud_reported']].groupby(['insured_education_level'], as_index=False).mean().sort_values(by='fraud_reported', ascending=False)\n"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 31,
- "id": "29500b25",
- "metadata": {},
- "outputs": [],
- "source": [
- "# drop not relevant features\n",
- "data_cat.drop(['incident_date', 'auto_make', 'auto_model'], axis=1, inplace=True)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 32,
- "id": "035e1ac9",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/html": [
- "<div>\n",
- "<style scoped>\n",
- " .dataframe tbody tr th:only-of-type {\n",
- " vertical-align: middle;\n",
- " }\n",
- "\n",
- " .dataframe tbody tr th {\n",
- " vertical-align: top;\n",
- " }\n",
- "\n",
- " .dataframe thead th {\n",
- " text-align: right;\n",
- " }\n",
- "</style>\n",
- "<table border=\"1\" class=\"dataframe\">\n",
- " <thead>\n",
- " <tr style=\"text-align: right;\">\n",
- " <th></th>\n",
- " <th>policy_csl</th>\n",
- " <th>insured_sex</th>\n",
- " <th>insured_education_level</th>\n",
- " <th>insured_occupation</th>\n",
- " <th>insured_hobbies</th>\n",
- " <th>insured_relationship</th>\n",
- " <th>incident_type</th>\n",
- " <th>collision_type</th>\n",
- " <th>incident_severity</th>\n",
- " <th>authorities_contacted</th>\n",
- " <th>property_damage</th>\n",
- " <th>police_report_available</th>\n",
- " <th>fraud_reported</th>\n",
- " </tr>\n",
- " </thead>\n",
- " <tbody>\n",
- " <tr>\n",
- " <th>0</th>\n",
- " <td>250/500</td>\n",
- " <td>MALE</td>\n",
- " <td>MD</td>\n",
- " <td>craft-repair</td>\n",
- " <td>sleeping</td>\n",
- " <td>husband</td>\n",
- " <td>Single Vehicle Collision</td>\n",
- " <td>Side Collision</td>\n",
- " <td>Major Damage</td>\n",
- " <td>Police</td>\n",
- " <td>YES</td>\n",
- " <td>YES</td>\n",
- " <td>1</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>1</th>\n",
- " <td>250/500</td>\n",
- " <td>MALE</td>\n",
- " <td>MD</td>\n",
- " <td>machine-op-inspct</td>\n",
- " <td>reading</td>\n",
- " <td>other-relative</td>\n",
- " <td>Vehicle Theft</td>\n",
- " <td>Rear Collision</td>\n",
- " <td>Minor Damage</td>\n",
- " <td>Police</td>\n",
- " <td>NO</td>\n",
- " <td>NO</td>\n",
- " <td>1</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>2</th>\n",
- " <td>100/300</td>\n",
- " <td>FEMALE</td>\n",
- " <td>PhD</td>\n",
- " <td>sales</td>\n",
- " <td>board-games</td>\n",
- " <td>own-child</td>\n",
- " <td>Multi-vehicle Collision</td>\n",
- " <td>Rear Collision</td>\n",
- " <td>Minor Damage</td>\n",
- " <td>Police</td>\n",
- " <td>NO</td>\n",
- " <td>NO</td>\n",
- " <td>0</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>3</th>\n",
- " <td>250/500</td>\n",
- " <td>FEMALE</td>\n",
- " <td>PhD</td>\n",
- " <td>armed-forces</td>\n",
- " <td>board-games</td>\n",
- " <td>unmarried</td>\n",
- " <td>Single Vehicle Collision</td>\n",
- " <td>Front Collision</td>\n",
- " <td>Major Damage</td>\n",
- " <td>Police</td>\n",
- " <td>NO</td>\n",
- " <td>NO</td>\n",
- " <td>1</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>4</th>\n",
- " <td>500/1000</td>\n",
- " <td>MALE</td>\n",
- " <td>Associate</td>\n",
- " <td>sales</td>\n",
- " <td>board-games</td>\n",
- " <td>unmarried</td>\n",
- " <td>Vehicle Theft</td>\n",
- " <td>Rear Collision</td>\n",
- " <td>Minor Damage</td>\n",
- " <td>None</td>\n",
- " <td>NO</td>\n",
- " <td>NO</td>\n",
- " <td>0</td>\n",
- " </tr>\n",
- " </tbody>\n",
- "</table>\n",
- "</div>"
- ],
- "text/plain": [
- " policy_csl insured_sex insured_education_level insured_occupation \\\n",
- "0 250/500 MALE MD craft-repair \n",
- "1 250/500 MALE MD machine-op-inspct \n",
- "2 100/300 FEMALE PhD sales \n",
- "3 250/500 FEMALE PhD armed-forces \n",
- "4 500/1000 MALE Associate sales \n",
- "\n",
- " insured_hobbies insured_relationship incident_type \\\n",
- "0 sleeping husband Single Vehicle Collision \n",
- "1 reading other-relative Vehicle Theft \n",
- "2 board-games own-child Multi-vehicle Collision \n",
- "3 board-games unmarried Single Vehicle Collision \n",
- "4 board-games unmarried Vehicle Theft \n",
- "\n",
- " collision_type incident_severity authorities_contacted property_damage \\\n",
- "0 Side Collision Major Damage Police YES \n",
- "1 Rear Collision Minor Damage Police NO \n",
- "2 Rear Collision Minor Damage Police NO \n",
- "3 Front Collision Major Damage Police NO \n",
- "4 Rear Collision Minor Damage None NO \n",
- "\n",
- " police_report_available fraud_reported \n",
- "0 YES 1 \n",
- "1 NO 1 \n",
- "2 NO 0 \n",
- "3 NO 1 \n",
- "4 NO 0 "
- ]
- },
- "execution_count": 32,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "data_cat.head()"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 33,
- "id": "7d855ead",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/html": [
- "<div>\n",
- "<style scoped>\n",
- " .dataframe tbody tr th:only-of-type {\n",
- " vertical-align: middle;\n",
- " }\n",
- "\n",
- " .dataframe tbody tr th {\n",
- " vertical-align: top;\n",
- " }\n",
- "\n",
- " .dataframe thead th {\n",
- " text-align: right;\n",
- " }\n",
- "</style>\n",
- "<table border=\"1\" class=\"dataframe\">\n",
- " <thead>\n",
- " <tr style=\"text-align: right;\">\n",
- " <th></th>\n",
- " <th>months_as_customer</th>\n",
- " <th>policy_deductable</th>\n",
- " <th>policy_annual_premium</th>\n",
- " <th>umbrella_limit</th>\n",
- " <th>capital-gains</th>\n",
- " <th>capital-loss</th>\n",
- " <th>number_of_vehicles_involved</th>\n",
- " <th>bodily_injuries</th>\n",
- " <th>witnesses</th>\n",
- " <th>injury_claim</th>\n",
- " <th>property_claim</th>\n",
- " <th>vehicle_claim</th>\n",
- " <th>auto_year</th>\n",
- " <th>fraud_reported</th>\n",
- " </tr>\n",
- " </thead>\n",
- " <tbody>\n",
- " <tr>\n",
- " <th>0</th>\n",
- " <td>328</td>\n",
- " <td>1000</td>\n",
- " <td>1406.91</td>\n",
- " <td>0</td>\n",
- " <td>53300</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>1</td>\n",
- " <td>2</td>\n",
- " <td>6510</td>\n",
- " <td>13020</td>\n",
- " <td>52080</td>\n",
- " <td>2004</td>\n",
- " <td>1</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>1</th>\n",
- " <td>228</td>\n",
- " <td>2000</td>\n",
- " <td>1197.22</td>\n",
- " <td>5000000</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>780</td>\n",
- " <td>780</td>\n",
- " <td>3510</td>\n",
- " <td>2007</td>\n",
- " <td>1</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>2</th>\n",
- " <td>134</td>\n",
- " <td>2000</td>\n",
- " <td>1413.14</td>\n",
- " <td>5000000</td>\n",
- " <td>35100</td>\n",
- " <td>0</td>\n",
- " <td>3</td>\n",
- " <td>2</td>\n",
- " <td>3</td>\n",
- " <td>7700</td>\n",
- " <td>3850</td>\n",
- " <td>23100</td>\n",
- " <td>2007</td>\n",
- " <td>0</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>3</th>\n",
- " <td>256</td>\n",
- " <td>2000</td>\n",
- " <td>1415.74</td>\n",
- " <td>6000000</td>\n",
- " <td>48900</td>\n",
- " <td>-62400</td>\n",
- " <td>1</td>\n",
- " <td>1</td>\n",
- " <td>2</td>\n",
- " <td>6340</td>\n",
- " <td>6340</td>\n",
- " <td>50720</td>\n",
- " <td>2014</td>\n",
- " <td>1</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>4</th>\n",
- " <td>228</td>\n",
- " <td>1000</td>\n",
- " <td>1583.91</td>\n",
- " <td>6000000</td>\n",
- " <td>66000</td>\n",
- " <td>-46000</td>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>1300</td>\n",
- " <td>650</td>\n",
- " <td>4550</td>\n",
- " <td>2009</td>\n",
- " <td>0</td>\n",
- " </tr>\n",
- " </tbody>\n",
- "</table>\n",
- "</div>"
- ],
- "text/plain": [
- " months_as_customer policy_deductable policy_annual_premium \\\n",
- "0 328 1000 1406.91 \n",
- "1 228 2000 1197.22 \n",
- "2 134 2000 1413.14 \n",
- "3 256 2000 1415.74 \n",
- "4 228 1000 1583.91 \n",
- "\n",
- " umbrella_limit capital-gains capital-loss number_of_vehicles_involved \\\n",
- "0 0 53300 0 1 \n",
- "1 5000000 0 0 1 \n",
- "2 5000000 35100 0 3 \n",
- "3 6000000 48900 -62400 1 \n",
- "4 6000000 66000 -46000 1 \n",
- "\n",
- " bodily_injuries witnesses injury_claim property_claim vehicle_claim \\\n",
- "0 1 2 6510 13020 52080 \n",
- "1 0 0 780 780 3510 \n",
- "2 2 3 7700 3850 23100 \n",
- "3 1 2 6340 6340 50720 \n",
- "4 0 1 1300 650 4550 \n",
- "\n",
- " auto_year fraud_reported \n",
- "0 2004 1 \n",
- "1 2007 1 \n",
- "2 2007 0 \n",
- "3 2014 1 \n",
- "4 2009 0 "
- ]
- },
- "execution_count": 33,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "data_num.head()"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 34,
- "id": "dfdf6abc",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "image/png": "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",
- "text/plain": [
- "<Figure size 1080x360 with 2 Axes>"
- ]
- },
- "metadata": {
- "needs_background": "light"
- },
- "output_type": "display_data"
- }
- ],
- "source": [
- "chart = sns.catplot(x=\"auto_year\", col=\"fraud_reported\", data=data_num, kind=\"count\", aspect=1.5)\n",
- "chart.set_xticklabels(rotation=45)\n",
- "# plt.savefig('./Daten/VglIncidentVSFraud')\n",
- "plt.show()"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 35,
- "id": "567eda9a",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "image/png": "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",
- "text/plain": [
- "<Figure size 432x288 with 1 Axes>"
- ]
- },
- "metadata": {
- "needs_background": "light"
- },
- "output_type": "display_data"
- }
- ],
- "source": [
- "plt.scatter(data_num.policy_annual_premium, data_num.fraud_reported)\n",
- "\n",
- "plt.show()"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 36,
- "id": "a3405af2",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "image/png": "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",
- "text/plain": [
- "<Figure size 432x288 with 1 Axes>"
- ]
- },
- "metadata": {
- "needs_background": "light"
- },
- "output_type": "display_data"
- }
- ],
- "source": [
- "plt.scatter(data_num.umbrella_limit, data_num.fraud_reported)\n",
- "\n",
- "plt.show()"
- ]
- },
- {
- "attachments": {},
- "cell_type": "markdown",
- "id": "f6069417",
- "metadata": {},
- "source": [
- "## 3.3 Merkmal Technik\n",
- "### neues Merkmal mit prozentualem Anteil des bezahlten Schadens (ohne Selbstbeteiligung)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 37,
- "id": "8d204f0b",
- "metadata": {},
- "outputs": [
- {
- "name": "stderr",
- "output_type": "stream",
- "text": [
- "<ipython-input-37-587d575ccaa7>:3: SettingWithCopyWarning: \n",
- "A value is trying to be set on a copy of a slice from a DataFrame.\n",
- "Try using .loc[row_indexer,col_indexer] = value instead\n",
- "\n",
- "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
- " data_num['total_claims'] = data_num.loc[:, 'injury_claim':'vehicle_claim'].apply(sum, axis=1)\n"
- ]
- },
- {
- "data": {
- "text/html": [
- "<div>\n",
- "<style scoped>\n",
- " .dataframe tbody tr th:only-of-type {\n",
- " vertical-align: middle;\n",
- " }\n",
- "\n",
- " .dataframe tbody tr th {\n",
- " vertical-align: top;\n",
- " }\n",
- "\n",
- " .dataframe thead th {\n",
- " text-align: right;\n",
- " }\n",
- "</style>\n",
- "<table border=\"1\" class=\"dataframe\">\n",
- " <thead>\n",
- " <tr style=\"text-align: right;\">\n",
- " <th></th>\n",
- " <th>months_as_customer</th>\n",
- " <th>policy_deductable</th>\n",
- " <th>policy_annual_premium</th>\n",
- " <th>umbrella_limit</th>\n",
- " <th>capital-gains</th>\n",
- " <th>capital-loss</th>\n",
- " <th>number_of_vehicles_involved</th>\n",
- " <th>bodily_injuries</th>\n",
- " <th>witnesses</th>\n",
- " <th>injury_claim</th>\n",
- " <th>property_claim</th>\n",
- " <th>vehicle_claim</th>\n",
- " <th>auto_year</th>\n",
- " <th>fraud_reported</th>\n",
- " <th>total_claims</th>\n",
- " </tr>\n",
- " </thead>\n",
- " <tbody>\n",
- " <tr>\n",
- " <th>0</th>\n",
- " <td>328</td>\n",
- " <td>1000</td>\n",
- " <td>1406.91</td>\n",
- " <td>0</td>\n",
- " <td>53300</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>1</td>\n",
- " <td>2</td>\n",
- " <td>6510</td>\n",
- " <td>13020</td>\n",
- " <td>52080</td>\n",
- " <td>2004</td>\n",
- " <td>1</td>\n",
- " <td>71610</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>1</th>\n",
- " <td>228</td>\n",
- " <td>2000</td>\n",
- " <td>1197.22</td>\n",
- " <td>5000000</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>780</td>\n",
- " <td>780</td>\n",
- " <td>3510</td>\n",
- " <td>2007</td>\n",
- " <td>1</td>\n",
- " <td>5070</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>2</th>\n",
- " <td>134</td>\n",
- " <td>2000</td>\n",
- " <td>1413.14</td>\n",
- " <td>5000000</td>\n",
- " <td>35100</td>\n",
- " <td>0</td>\n",
- " <td>3</td>\n",
- " <td>2</td>\n",
- " <td>3</td>\n",
- " <td>7700</td>\n",
- " <td>3850</td>\n",
- " <td>23100</td>\n",
- " <td>2007</td>\n",
- " <td>0</td>\n",
- " <td>34650</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>3</th>\n",
- " <td>256</td>\n",
- " <td>2000</td>\n",
- " <td>1415.74</td>\n",
- " <td>6000000</td>\n",
- " <td>48900</td>\n",
- " <td>-62400</td>\n",
- " <td>1</td>\n",
- " <td>1</td>\n",
- " <td>2</td>\n",
- " <td>6340</td>\n",
- " <td>6340</td>\n",
- " <td>50720</td>\n",
- " <td>2014</td>\n",
- " <td>1</td>\n",
- " <td>63400</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>4</th>\n",
- " <td>228</td>\n",
- " <td>1000</td>\n",
- " <td>1583.91</td>\n",
- " <td>6000000</td>\n",
- " <td>66000</td>\n",
- " <td>-46000</td>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>1300</td>\n",
- " <td>650</td>\n",
- " <td>4550</td>\n",
- " <td>2009</td>\n",
- " <td>0</td>\n",
- " <td>6500</td>\n",
- " </tr>\n",
- " </tbody>\n",
- "</table>\n",
- "</div>"
- ],
- "text/plain": [
- " months_as_customer policy_deductable policy_annual_premium \\\n",
- "0 328 1000 1406.91 \n",
- "1 228 2000 1197.22 \n",
- "2 134 2000 1413.14 \n",
- "3 256 2000 1415.74 \n",
- "4 228 1000 1583.91 \n",
- "\n",
- " umbrella_limit capital-gains capital-loss number_of_vehicles_involved \\\n",
- "0 0 53300 0 1 \n",
- "1 5000000 0 0 1 \n",
- "2 5000000 35100 0 3 \n",
- "3 6000000 48900 -62400 1 \n",
- "4 6000000 66000 -46000 1 \n",
- "\n",
- " bodily_injuries witnesses injury_claim property_claim vehicle_claim \\\n",
- "0 1 2 6510 13020 52080 \n",
- "1 0 0 780 780 3510 \n",
- "2 2 3 7700 3850 23100 \n",
- "3 1 2 6340 6340 50720 \n",
- "4 0 1 1300 650 4550 \n",
- "\n",
- " auto_year fraud_reported total_claims \n",
- "0 2004 1 71610 \n",
- "1 2007 1 5070 \n",
- "2 2007 0 34650 \n",
- "3 2014 1 63400 \n",
- "4 2009 0 6500 "
- ]
- },
- "execution_count": 37,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "# Preparation\n",
- "# data_num.loc[:, 'injury_claim':'vehicle_claim'].apply(sum, axis=1)\n",
- "data_num['total_claims'] = data_num.loc[:, 'injury_claim':'vehicle_claim'].apply(sum, axis=1)\n",
- "data_num.head()"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 38,
- "id": "a0213331",
- "metadata": {},
- "outputs": [
- {
- "name": "stderr",
- "output_type": "stream",
- "text": [
- "<ipython-input-38-2d01ef42697c>:2: SettingWithCopyWarning: \n",
- "A value is trying to be set on a copy of a slice from a DataFrame.\n",
- "Try using .loc[row_indexer,col_indexer] = value instead\n",
- "\n",
- "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
- " data_num['pct_paid_insurance'] = (data_num.total_claims - data_num.policy_deductable) / data_num.total_claims\n"
- ]
- }
- ],
- "source": [
- "# add new feature\n",
- "data_num['pct_paid_insurance'] = (data_num.total_claims - data_num.policy_deductable) / data_num.total_claims "
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 39,
- "id": "26aa3f21",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "0 0.986035\n",
- "1 0.605523\n",
- "2 0.942280\n",
- "3 0.968454\n",
- "4 0.846154\n",
- " ... \n",
- "995 0.988532\n",
- "996 0.990782\n",
- "997 0.992593\n",
- "998 0.957429\n",
- "999 0.802372\n",
- "Name: pct_paid_insurance, Length: 1000, dtype: float64"
- ]
- },
- "execution_count": 39,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "data_num.pct_paid_insurance"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 40,
- "id": "d6b1864e",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "image/png": "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",
- "text/plain": [
- "<Figure size 540x360 with 2 Axes>"
- ]
- },
- "metadata": {
- "needs_background": "light"
- },
- "output_type": "display_data"
- }
- ],
- "source": [
- "chart = sns.catplot(x=\"pct_paid_insurance\", col=\"fraud_reported\", data=data_num, kind=\"bar\", aspect=0.75)\n",
- "chart.set_xticklabels(rotation=45)\n",
- "# plt.savefig('./Daten/VglIncidentVSFraud')\n",
- "plt.show()"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 41,
- "id": "2e349d53",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "image/png": "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",
- "text/plain": [
- "<Figure size 1296x864 with 2 Axes>"
- ]
- },
- "metadata": {
- "needs_background": "light"
- },
- "output_type": "display_data"
- }
- ],
- "source": [
- "# checking correlation with the new feature\n",
- "plt.figure(figsize=(18, 12))\n",
- "\n",
- "feature_corr = data_num.corr()\n",
- "mask = np.triu(np.ones_like(feature_corr, dtype = bool))\n",
- "sns.heatmap(feature_corr, mask=mask, annot=True, cmap='coolwarm')\n",
- "\n",
- "plt.show()"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 42,
- "id": "b4800eb6",
- "metadata": {},
- "outputs": [
- {
- "name": "stderr",
- "output_type": "stream",
- "text": [
- "C:\\Users\\eebal\\Anaconda3\\lib\\site-packages\\pandas\\core\\frame.py:4163: SettingWithCopyWarning: \n",
- "A value is trying to be set on a copy of a slice from a DataFrame\n",
- "\n",
- "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
- " return super().drop(\n"
- ]
- }
- ],
- "source": [
- "# drop not relevant features\n",
- "data_num.drop(['auto_year', 'total_claims'], axis=1, inplace=True)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 43,
- "id": "dc6f52a4",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/html": [
- "<div>\n",
- "<style scoped>\n",
- " .dataframe tbody tr th:only-of-type {\n",
- " vertical-align: middle;\n",
- " }\n",
- "\n",
- " .dataframe tbody tr th {\n",
- " vertical-align: top;\n",
- " }\n",
- "\n",
- " .dataframe thead th {\n",
- " text-align: right;\n",
- " }\n",
- "</style>\n",
- "<table border=\"1\" class=\"dataframe\">\n",
- " <thead>\n",
- " <tr style=\"text-align: right;\">\n",
- " <th></th>\n",
- " <th>months_as_customer</th>\n",
- " <th>policy_deductable</th>\n",
- " <th>policy_annual_premium</th>\n",
- " <th>umbrella_limit</th>\n",
- " <th>capital-gains</th>\n",
- " <th>capital-loss</th>\n",
- " <th>number_of_vehicles_involved</th>\n",
- " <th>bodily_injuries</th>\n",
- " <th>witnesses</th>\n",
- " <th>injury_claim</th>\n",
- " <th>property_claim</th>\n",
- " <th>vehicle_claim</th>\n",
- " <th>fraud_reported</th>\n",
- " <th>pct_paid_insurance</th>\n",
- " </tr>\n",
- " </thead>\n",
- " <tbody>\n",
- " <tr>\n",
- " <th>0</th>\n",
- " <td>328</td>\n",
- " <td>1000</td>\n",
- " <td>1406.91</td>\n",
- " <td>0</td>\n",
- " <td>53300</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>1</td>\n",
- " <td>2</td>\n",
- " <td>6510</td>\n",
- " <td>13020</td>\n",
- " <td>52080</td>\n",
- " <td>1</td>\n",
- " <td>0.986035</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>1</th>\n",
- " <td>228</td>\n",
- " <td>2000</td>\n",
- " <td>1197.22</td>\n",
- " <td>5000000</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>780</td>\n",
- " <td>780</td>\n",
- " <td>3510</td>\n",
- " <td>1</td>\n",
- " <td>0.605523</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>2</th>\n",
- " <td>134</td>\n",
- " <td>2000</td>\n",
- " <td>1413.14</td>\n",
- " <td>5000000</td>\n",
- " <td>35100</td>\n",
- " <td>0</td>\n",
- " <td>3</td>\n",
- " <td>2</td>\n",
- " <td>3</td>\n",
- " <td>7700</td>\n",
- " <td>3850</td>\n",
- " <td>23100</td>\n",
- " <td>0</td>\n",
- " <td>0.942280</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>3</th>\n",
- " <td>256</td>\n",
- " <td>2000</td>\n",
- " <td>1415.74</td>\n",
- " <td>6000000</td>\n",
- " <td>48900</td>\n",
- " <td>-62400</td>\n",
- " <td>1</td>\n",
- " <td>1</td>\n",
- " <td>2</td>\n",
- " <td>6340</td>\n",
- " <td>6340</td>\n",
- " <td>50720</td>\n",
- " <td>1</td>\n",
- " <td>0.968454</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>4</th>\n",
- " <td>228</td>\n",
- " <td>1000</td>\n",
- " <td>1583.91</td>\n",
- " <td>6000000</td>\n",
- " <td>66000</td>\n",
- " <td>-46000</td>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>1300</td>\n",
- " <td>650</td>\n",
- " <td>4550</td>\n",
- " <td>0</td>\n",
- " <td>0.846154</td>\n",
- " </tr>\n",
- " </tbody>\n",
- "</table>\n",
- "</div>"
- ],
- "text/plain": [
- " months_as_customer policy_deductable policy_annual_premium \\\n",
- "0 328 1000 1406.91 \n",
- "1 228 2000 1197.22 \n",
- "2 134 2000 1413.14 \n",
- "3 256 2000 1415.74 \n",
- "4 228 1000 1583.91 \n",
- "\n",
- " umbrella_limit capital-gains capital-loss number_of_vehicles_involved \\\n",
- "0 0 53300 0 1 \n",
- "1 5000000 0 0 1 \n",
- "2 5000000 35100 0 3 \n",
- "3 6000000 48900 -62400 1 \n",
- "4 6000000 66000 -46000 1 \n",
- "\n",
- " bodily_injuries witnesses injury_claim property_claim vehicle_claim \\\n",
- "0 1 2 6510 13020 52080 \n",
- "1 0 0 780 780 3510 \n",
- "2 2 3 7700 3850 23100 \n",
- "3 1 2 6340 6340 50720 \n",
- "4 0 1 1300 650 4550 \n",
- "\n",
- " fraud_reported pct_paid_insurance \n",
- "0 1 0.986035 \n",
- "1 1 0.605523 \n",
- "2 0 0.942280 \n",
- "3 1 0.968454 \n",
- "4 0 0.846154 "
- ]
- },
- "execution_count": 43,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "data_num.head()"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 44,
- "id": "37ce51ac",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/html": [
- "<div>\n",
- "<style scoped>\n",
- " .dataframe tbody tr th:only-of-type {\n",
- " vertical-align: middle;\n",
- " }\n",
- "\n",
- " .dataframe tbody tr th {\n",
- " vertical-align: top;\n",
- " }\n",
- "\n",
- " .dataframe thead th {\n",
- " text-align: right;\n",
- " }\n",
- "</style>\n",
- "<table border=\"1\" class=\"dataframe\">\n",
- " <thead>\n",
- " <tr style=\"text-align: right;\">\n",
- " <th></th>\n",
- " <th>policy_csl</th>\n",
- " <th>insured_sex</th>\n",
- " <th>insured_education_level</th>\n",
- " <th>insured_occupation</th>\n",
- " <th>insured_hobbies</th>\n",
- " <th>insured_relationship</th>\n",
- " <th>incident_type</th>\n",
- " <th>collision_type</th>\n",
- " <th>incident_severity</th>\n",
- " <th>authorities_contacted</th>\n",
- " <th>property_damage</th>\n",
- " <th>police_report_available</th>\n",
- " <th>fraud_reported</th>\n",
- " </tr>\n",
- " </thead>\n",
- " <tbody>\n",
- " <tr>\n",
- " <th>count</th>\n",
- " <td>1000</td>\n",
- " <td>1000</td>\n",
- " <td>1000</td>\n",
- " <td>1000</td>\n",
- " <td>1000</td>\n",
- " <td>1000</td>\n",
- " <td>1000</td>\n",
- " <td>1000</td>\n",
- " <td>1000</td>\n",
- " <td>1000</td>\n",
- " <td>1000</td>\n",
- " <td>1000</td>\n",
- " <td>1000.000000</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>unique</th>\n",
- " <td>3</td>\n",
- " <td>2</td>\n",
- " <td>7</td>\n",
- " <td>14</td>\n",
- " <td>20</td>\n",
- " <td>6</td>\n",
- " <td>4</td>\n",
- " <td>3</td>\n",
- " <td>4</td>\n",
- " <td>5</td>\n",
- " <td>2</td>\n",
- " <td>2</td>\n",
- " <td>NaN</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>top</th>\n",
- " <td>250/500</td>\n",
- " <td>FEMALE</td>\n",
- " <td>JD</td>\n",
- " <td>machine-op-inspct</td>\n",
- " <td>reading</td>\n",
- " <td>own-child</td>\n",
- " <td>Multi-vehicle Collision</td>\n",
- " <td>Rear Collision</td>\n",
- " <td>Minor Damage</td>\n",
- " <td>Police</td>\n",
- " <td>NO</td>\n",
- " <td>NO</td>\n",
- " <td>NaN</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>freq</th>\n",
- " <td>351</td>\n",
- " <td>537</td>\n",
- " <td>161</td>\n",
- " <td>93</td>\n",
- " <td>64</td>\n",
- " <td>183</td>\n",
- " <td>419</td>\n",
- " <td>470</td>\n",
- " <td>354</td>\n",
- " <td>292</td>\n",
- " <td>698</td>\n",
- " <td>686</td>\n",
- " <td>NaN</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>mean</th>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>0.247000</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>std</th>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>0.431483</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>min</th>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>0.000000</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>25%</th>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>0.000000</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>50%</th>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>0.000000</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>75%</th>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>0.000000</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>max</th>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>NaN</td>\n",
- " <td>1.000000</td>\n",
- " </tr>\n",
- " </tbody>\n",
- "</table>\n",
- "</div>"
- ],
- "text/plain": [
- " policy_csl insured_sex insured_education_level insured_occupation \\\n",
- "count 1000 1000 1000 1000 \n",
- "unique 3 2 7 14 \n",
- "top 250/500 FEMALE JD machine-op-inspct \n",
- "freq 351 537 161 93 \n",
- "mean NaN NaN NaN NaN \n",
- "std NaN NaN NaN NaN \n",
- "min NaN NaN NaN NaN \n",
- "25% NaN NaN NaN NaN \n",
- "50% NaN NaN NaN NaN \n",
- "75% NaN NaN NaN NaN \n",
- "max NaN NaN NaN NaN \n",
- "\n",
- " insured_hobbies insured_relationship incident_type \\\n",
- "count 1000 1000 1000 \n",
- "unique 20 6 4 \n",
- "top reading own-child Multi-vehicle Collision \n",
- "freq 64 183 419 \n",
- "mean NaN NaN NaN \n",
- "std NaN NaN NaN \n",
- "min NaN NaN NaN \n",
- "25% NaN NaN NaN \n",
- "50% NaN NaN NaN \n",
- "75% NaN NaN NaN \n",
- "max NaN NaN NaN \n",
- "\n",
- " collision_type incident_severity authorities_contacted \\\n",
- "count 1000 1000 1000 \n",
- "unique 3 4 5 \n",
- "top Rear Collision Minor Damage Police \n",
- "freq 470 354 292 \n",
- "mean NaN NaN NaN \n",
- "std NaN NaN NaN \n",
- "min NaN NaN NaN \n",
- "25% NaN NaN NaN \n",
- "50% NaN NaN NaN \n",
- "75% NaN NaN NaN \n",
- "max NaN NaN NaN \n",
- "\n",
- " property_damage police_report_available fraud_reported \n",
- "count 1000 1000 1000.000000 \n",
- "unique 2 2 NaN \n",
- "top NO NO NaN \n",
- "freq 698 686 NaN \n",
- "mean NaN NaN 0.247000 \n",
- "std NaN NaN 0.431483 \n",
- "min NaN NaN 0.000000 \n",
- "25% NaN NaN 0.000000 \n",
- "50% NaN NaN 0.000000 \n",
- "75% NaN NaN 0.000000 \n",
- "max NaN NaN 1.000000 "
- ]
- },
- "execution_count": 44,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "data_cat.describe(include='all')"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 45,
- "id": "0f6823ba",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/html": [
- "<div>\n",
- "<style scoped>\n",
- " .dataframe tbody tr th:only-of-type {\n",
- " vertical-align: middle;\n",
- " }\n",
- "\n",
- " .dataframe tbody tr th {\n",
- " vertical-align: top;\n",
- " }\n",
- "\n",
- " .dataframe thead th {\n",
- " text-align: right;\n",
- " }\n",
- "</style>\n",
- "<table border=\"1\" class=\"dataframe\">\n",
- " <thead>\n",
- " <tr style=\"text-align: right;\">\n",
- " <th></th>\n",
- " <th>months_as_customer</th>\n",
- " <th>policy_deductable</th>\n",
- " <th>policy_annual_premium</th>\n",
- " <th>umbrella_limit</th>\n",
- " <th>capital-gains</th>\n",
- " <th>capital-loss</th>\n",
- " <th>number_of_vehicles_involved</th>\n",
- " <th>bodily_injuries</th>\n",
- " <th>witnesses</th>\n",
- " <th>injury_claim</th>\n",
- " <th>property_claim</th>\n",
- " <th>vehicle_claim</th>\n",
- " <th>fraud_reported</th>\n",
- " <th>pct_paid_insurance</th>\n",
- " </tr>\n",
- " </thead>\n",
- " <tbody>\n",
- " <tr>\n",
- " <th>count</th>\n",
- " <td>1000.000000</td>\n",
- " <td>1000.000000</td>\n",
- " <td>1000.000000</td>\n",
- " <td>1.000000e+03</td>\n",
- " <td>1000.000000</td>\n",
- " <td>1000.000000</td>\n",
- " <td>1000.00000</td>\n",
- " <td>1000.000000</td>\n",
- " <td>1000.000000</td>\n",
- " <td>1000.000000</td>\n",
- " <td>1000.000000</td>\n",
- " <td>1000.000000</td>\n",
- " <td>1000.000000</td>\n",
- " <td>1000.000000</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>mean</th>\n",
- " <td>203.954000</td>\n",
- " <td>1136.000000</td>\n",
- " <td>1256.406150</td>\n",
- " <td>1.101000e+06</td>\n",
- " <td>25126.100000</td>\n",
- " <td>-26793.700000</td>\n",
- " <td>1.83900</td>\n",
- " <td>0.992000</td>\n",
- " <td>1.487000</td>\n",
- " <td>7433.420000</td>\n",
- " <td>7399.570000</td>\n",
- " <td>37928.950000</td>\n",
- " <td>0.247000</td>\n",
- " <td>0.923527</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>std</th>\n",
- " <td>115.113174</td>\n",
- " <td>611.864673</td>\n",
- " <td>244.167395</td>\n",
- " <td>2.297407e+06</td>\n",
- " <td>27872.187708</td>\n",
- " <td>28104.096686</td>\n",
- " <td>1.01888</td>\n",
- " <td>0.820127</td>\n",
- " <td>1.111335</td>\n",
- " <td>4880.951853</td>\n",
- " <td>4824.726179</td>\n",
- " <td>18886.252893</td>\n",
- " <td>0.431483</td>\n",
- " <td>0.639068</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>min</th>\n",
- " <td>0.000000</td>\n",
- " <td>500.000000</td>\n",
- " <td>433.330000</td>\n",
- " <td>-1.000000e+06</td>\n",
- " <td>0.000000</td>\n",
- " <td>-111100.000000</td>\n",
- " <td>1.00000</td>\n",
- " <td>0.000000</td>\n",
- " <td>0.000000</td>\n",
- " <td>0.000000</td>\n",
- " <td>0.000000</td>\n",
- " <td>70.000000</td>\n",
- " <td>0.000000</td>\n",
- " <td>-19.000000</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>25%</th>\n",
- " <td>115.750000</td>\n",
- " <td>500.000000</td>\n",
- " <td>1089.607500</td>\n",
- " <td>0.000000e+00</td>\n",
- " <td>0.000000</td>\n",
- " <td>-51500.000000</td>\n",
- " <td>1.00000</td>\n",
- " <td>0.000000</td>\n",
- " <td>1.000000</td>\n",
- " <td>4295.000000</td>\n",
- " <td>4445.000000</td>\n",
- " <td>30292.500000</td>\n",
- " <td>0.000000</td>\n",
- " <td>0.963948</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>50%</th>\n",
- " <td>199.500000</td>\n",
- " <td>1000.000000</td>\n",
- " <td>1257.200000</td>\n",
- " <td>0.000000e+00</td>\n",
- " <td>0.000000</td>\n",
- " <td>-23250.000000</td>\n",
- " <td>1.00000</td>\n",
- " <td>1.000000</td>\n",
- " <td>1.000000</td>\n",
- " <td>6775.000000</td>\n",
- " <td>6750.000000</td>\n",
- " <td>42100.000000</td>\n",
- " <td>0.000000</td>\n",
- " <td>0.980558</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>75%</th>\n",
- " <td>276.250000</td>\n",
- " <td>2000.000000</td>\n",
- " <td>1415.695000</td>\n",
- " <td>0.000000e+00</td>\n",
- " <td>51025.000000</td>\n",
- " <td>0.000000</td>\n",
- " <td>3.00000</td>\n",
- " <td>2.000000</td>\n",
- " <td>2.000000</td>\n",
- " <td>11305.000000</td>\n",
- " <td>10885.000000</td>\n",
- " <td>50822.500000</td>\n",
- " <td>0.000000</td>\n",
- " <td>0.988895</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>max</th>\n",
- " <td>479.000000</td>\n",
- " <td>2000.000000</td>\n",
- " <td>2047.590000</td>\n",
- " <td>1.000000e+07</td>\n",
- " <td>100500.000000</td>\n",
- " <td>0.000000</td>\n",
- " <td>4.00000</td>\n",
- " <td>2.000000</td>\n",
- " <td>3.000000</td>\n",
- " <td>21450.000000</td>\n",
- " <td>23670.000000</td>\n",
- " <td>79560.000000</td>\n",
- " <td>1.000000</td>\n",
- " <td>0.995548</td>\n",
- " </tr>\n",
- " </tbody>\n",
- "</table>\n",
- "</div>"
- ],
- "text/plain": [
- " months_as_customer policy_deductable policy_annual_premium \\\n",
- "count 1000.000000 1000.000000 1000.000000 \n",
- "mean 203.954000 1136.000000 1256.406150 \n",
- "std 115.113174 611.864673 244.167395 \n",
- "min 0.000000 500.000000 433.330000 \n",
- "25% 115.750000 500.000000 1089.607500 \n",
- "50% 199.500000 1000.000000 1257.200000 \n",
- "75% 276.250000 2000.000000 1415.695000 \n",
- "max 479.000000 2000.000000 2047.590000 \n",
- "\n",
- " umbrella_limit capital-gains capital-loss \\\n",
- "count 1.000000e+03 1000.000000 1000.000000 \n",
- "mean 1.101000e+06 25126.100000 -26793.700000 \n",
- "std 2.297407e+06 27872.187708 28104.096686 \n",
- "min -1.000000e+06 0.000000 -111100.000000 \n",
- "25% 0.000000e+00 0.000000 -51500.000000 \n",
- "50% 0.000000e+00 0.000000 -23250.000000 \n",
- "75% 0.000000e+00 51025.000000 0.000000 \n",
- "max 1.000000e+07 100500.000000 0.000000 \n",
- "\n",
- " number_of_vehicles_involved bodily_injuries witnesses \\\n",
- "count 1000.00000 1000.000000 1000.000000 \n",
- "mean 1.83900 0.992000 1.487000 \n",
- "std 1.01888 0.820127 1.111335 \n",
- "min 1.00000 0.000000 0.000000 \n",
- "25% 1.00000 0.000000 1.000000 \n",
- "50% 1.00000 1.000000 1.000000 \n",
- "75% 3.00000 2.000000 2.000000 \n",
- "max 4.00000 2.000000 3.000000 \n",
- "\n",
- " injury_claim property_claim vehicle_claim fraud_reported \\\n",
- "count 1000.000000 1000.000000 1000.000000 1000.000000 \n",
- "mean 7433.420000 7399.570000 37928.950000 0.247000 \n",
- "std 4880.951853 4824.726179 18886.252893 0.431483 \n",
- "min 0.000000 0.000000 70.000000 0.000000 \n",
- "25% 4295.000000 4445.000000 30292.500000 0.000000 \n",
- "50% 6775.000000 6750.000000 42100.000000 0.000000 \n",
- "75% 11305.000000 10885.000000 50822.500000 0.000000 \n",
- "max 21450.000000 23670.000000 79560.000000 1.000000 \n",
- "\n",
- " pct_paid_insurance \n",
- "count 1000.000000 \n",
- "mean 0.923527 \n",
- "std 0.639068 \n",
- "min -19.000000 \n",
- "25% 0.963948 \n",
- "50% 0.980558 \n",
- "75% 0.988895 \n",
- "max 0.995548 "
- ]
- },
- "execution_count": 45,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "data_num.describe()"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 46,
- "id": "0ba21ca3",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "image/png": "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",
- "text/plain": [
- "<Figure size 1800x1440 with 1 Axes>"
- ]
- },
- "metadata": {
- "needs_background": "light"
- },
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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",
- "text/plain": [
- "<Figure size 360x360 with 1 Axes>"
- ]
- },
- "metadata": {
- "needs_background": "light"
- },
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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",
- "text/plain": [
- "<Figure size 360x360 with 1 Axes>"
- ]
- },
- "metadata": {
- "needs_background": "light"
- },
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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",
- "text/plain": [
- "<Figure size 360x360 with 1 Axes>"
- ]
- },
- "metadata": {
- "needs_background": "light"
- },
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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",
- "text/plain": [
- "<Figure size 360x360 with 1 Axes>"
- ]
- },
- "metadata": {
- "needs_background": "light"
- },
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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",
- "text/plain": [
- "<Figure size 360x360 with 1 Axes>"
- ]
- },
- "metadata": {
- "needs_background": "light"
- },
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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",
- "text/plain": [
- "<Figure size 360x360 with 1 Axes>"
- ]
- },
- "metadata": {
- "needs_background": "light"
- },
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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",
- "text/plain": [
- "<Figure size 360x360 with 1 Axes>"
- ]
- },
- "metadata": {
- "needs_background": "light"
- },
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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",
- "text/plain": [
- "<Figure size 360x360 with 1 Axes>"
- ]
- },
- "metadata": {
- "needs_background": "light"
- },
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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",
- "text/plain": [
- "<Figure size 360x360 with 1 Axes>"
- ]
- },
- "metadata": {
- "needs_background": "light"
- },
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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",
- "text/plain": [
- "<Figure size 360x360 with 1 Axes>"
- ]
- },
- "metadata": {
- "needs_background": "light"
- },
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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",
- "text/plain": [
- "<Figure size 360x360 with 1 Axes>"
- ]
- },
- "metadata": {
- "needs_background": "light"
- },
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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",
- "text/plain": [
- "<Figure size 360x360 with 1 Axes>"
- ]
- },
- "metadata": {
- "needs_background": "light"
- },
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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",
- "text/plain": [
- "<Figure size 360x360 with 1 Axes>"
- ]
- },
- "metadata": {
- "needs_background": "light"
- },
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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",
- "text/plain": [
- "<Figure size 360x360 with 1 Axes>"
- ]
- },
- "metadata": {
- "needs_background": "light"
- },
- "output_type": "display_data"
- }
- ],
- "source": [
- "plt.figure(figsize = (25, 20))\n",
- "plotnumber = 1\n",
- "\n",
- "for col in data_num.columns:\n",
- " if plotnumber <= data_num.shape[1]:\n",
- " ax = plt.subplot(5, 5, plotnumber)\n",
- " sns.displot(data_num[col])\n",
- " plt.xlabel(col, fontsize = 15)\n",
- " \n",
- " plotnumber += 1\n",
- " \n",
- "plt.tight_layout()\n",
- "plt.show()"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 47,
- "id": "610ae82c",
- "metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "0 753\n",
- "1 247\n",
- "Name: fraud_reported, dtype: int64\n"
- ]
- },
- {
- "data": {
- "text/plain": [
- "([<matplotlib.patches.Wedge at 0x13cf911c550>,\n",
- " <matplotlib.patches.Wedge at 0x13cf913b250>],\n",
- " [Text(-0.7704522141128092, -0.7851136132870644, 'No Fraud'),\n",
- " Text(0.8054727308753049, 0.8208006334161048, 'Fraud')],\n",
- " [Text(-0.4202466622433504, -0.42824378906567145, '75.3'),\n",
- " Text(0.45526719571212887, 0.463930792800407, '24.7')])"
- ]
- },
- "execution_count": 47,
- "metadata": {},
- "output_type": "execute_result"
- },
- {
- "data": {
- "image/png": "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",
- "text/plain": [
- "<Figure size 432x288 with 1 Axes>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- }
- ],
- "source": [
- "print(data_num.fraud_reported.value_counts())\n",
- "plt.pie(data_num.fraud_reported.value_counts(), labels=['No Fraud', 'Fraud'], autopct='%.1f', \n",
- " startangle=90, explode=[0, 0.05], colors=['#7ed6df', '#ffbe76'], textprops={'fontsize': 12})"
- ]
- },
- {
- "attachments": {},
- "cell_type": "markdown",
- "id": "ffc37012",
- "metadata": {},
- "source": [
- "--> Es handelt sich um einen unausgewogenen Datensatz."
- ]
- },
- {
- "attachments": {},
- "cell_type": "markdown",
- "id": "da842b3c",
- "metadata": {},
- "source": [
- "## 3.4 Dummy-Variablen erstellen"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 48,
- "id": "024c1724",
- "metadata": {},
- "outputs": [],
- "source": [
- "# create dummy variables\n",
- "data_cat.drop('fraud_reported', axis=1, inplace=True)\n",
- "dummies = pd.get_dummies(data_cat, drop_first=True)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 49,
- "id": "74de27cf",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/html": [
- "<div>\n",
- "<style scoped>\n",
- " .dataframe tbody tr th:only-of-type {\n",
- " vertical-align: middle;\n",
- " }\n",
- "\n",
- " .dataframe tbody tr th {\n",
- " vertical-align: top;\n",
- " }\n",
- "\n",
- " .dataframe thead th {\n",
- " text-align: right;\n",
- " }\n",
- "</style>\n",
- "<table border=\"1\" class=\"dataframe\">\n",
- " <thead>\n",
- " <tr style=\"text-align: right;\">\n",
- " <th></th>\n",
- " <th>policy_csl_250/500</th>\n",
- " <th>policy_csl_500/1000</th>\n",
- " <th>insured_sex_MALE</th>\n",
- " <th>insured_education_level_College</th>\n",
- " <th>insured_education_level_High School</th>\n",
- " <th>insured_education_level_JD</th>\n",
- " <th>insured_education_level_MD</th>\n",
- " <th>insured_education_level_Masters</th>\n",
- " <th>insured_education_level_PhD</th>\n",
- " <th>insured_occupation_armed-forces</th>\n",
- " <th>...</th>\n",
- " <th>collision_type_Side Collision</th>\n",
- " <th>incident_severity_Minor Damage</th>\n",
- " <th>incident_severity_Total Loss</th>\n",
- " <th>incident_severity_Trivial Damage</th>\n",
- " <th>authorities_contacted_Fire</th>\n",
- " <th>authorities_contacted_None</th>\n",
- " <th>authorities_contacted_Other</th>\n",
- " <th>authorities_contacted_Police</th>\n",
- " <th>property_damage_YES</th>\n",
- " <th>police_report_available_YES</th>\n",
- " </tr>\n",
- " </thead>\n",
- " <tbody>\n",
- " <tr>\n",
- " <th>0</th>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>...</td>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>1</td>\n",
- " <td>1</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>1</th>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>...</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>2</th>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>...</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>3</th>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>1</td>\n",
- " <td>...</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>4</th>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>...</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " </tr>\n",
- " </tbody>\n",
- "</table>\n",
- "<p>5 rows × 60 columns</p>\n",
- "</div>"
- ],
- "text/plain": [
- " policy_csl_250/500 policy_csl_500/1000 insured_sex_MALE \\\n",
- "0 1 0 1 \n",
- "1 1 0 1 \n",
- "2 0 0 0 \n",
- "3 1 0 0 \n",
- "4 0 1 1 \n",
- "\n",
- " insured_education_level_College insured_education_level_High School \\\n",
- "0 0 0 \n",
- "1 0 0 \n",
- "2 0 0 \n",
- "3 0 0 \n",
- "4 0 0 \n",
- "\n",
- " insured_education_level_JD insured_education_level_MD \\\n",
- "0 0 1 \n",
- "1 0 1 \n",
- "2 0 0 \n",
- "3 0 0 \n",
- "4 0 0 \n",
- "\n",
- " insured_education_level_Masters insured_education_level_PhD \\\n",
- "0 0 0 \n",
- "1 0 0 \n",
- "2 0 1 \n",
- "3 0 1 \n",
- "4 0 0 \n",
- "\n",
- " insured_occupation_armed-forces ... collision_type_Side Collision \\\n",
- "0 0 ... 1 \n",
- "1 0 ... 0 \n",
- "2 0 ... 0 \n",
- "3 1 ... 0 \n",
- "4 0 ... 0 \n",
- "\n",
- " incident_severity_Minor Damage incident_severity_Total Loss \\\n",
- "0 0 0 \n",
- "1 1 0 \n",
- "2 1 0 \n",
- "3 0 0 \n",
- "4 1 0 \n",
- "\n",
- " incident_severity_Trivial Damage authorities_contacted_Fire \\\n",
- "0 0 0 \n",
- "1 0 0 \n",
- "2 0 0 \n",
- "3 0 0 \n",
- "4 0 0 \n",
- "\n",
- " authorities_contacted_None authorities_contacted_Other \\\n",
- "0 0 0 \n",
- "1 0 0 \n",
- "2 0 0 \n",
- "3 0 0 \n",
- "4 1 0 \n",
- "\n",
- " authorities_contacted_Police property_damage_YES \\\n",
- "0 1 1 \n",
- "1 1 0 \n",
- "2 1 0 \n",
- "3 1 0 \n",
- "4 0 0 \n",
- "\n",
- " police_report_available_YES \n",
- "0 1 \n",
- "1 0 \n",
- "2 0 \n",
- "3 0 \n",
- "4 0 \n",
- "\n",
- "[5 rows x 60 columns]"
- ]
- },
- "execution_count": 49,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "dummies.head()"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 50,
- "id": "07d760ae",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "(1000, 74)"
- ]
- },
- "execution_count": 50,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "data_preprocessed = pd.concat([dummies, data_num], axis=1)\n",
- "data_preprocessed.shape"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 51,
- "id": "220ba08b",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/html": [
- "<div>\n",
- "<style scoped>\n",
- " .dataframe tbody tr th:only-of-type {\n",
- " vertical-align: middle;\n",
- " }\n",
- "\n",
- " .dataframe tbody tr th {\n",
- " vertical-align: top;\n",
- " }\n",
- "\n",
- " .dataframe thead th {\n",
- " text-align: right;\n",
- " }\n",
- "</style>\n",
- "<table border=\"1\" class=\"dataframe\">\n",
- " <thead>\n",
- " <tr style=\"text-align: right;\">\n",
- " <th></th>\n",
- " <th>policy_csl_250/500</th>\n",
- " <th>policy_csl_500/1000</th>\n",
- " <th>insured_sex_MALE</th>\n",
- " <th>insured_education_level_College</th>\n",
- " <th>insured_education_level_High School</th>\n",
- " <th>insured_education_level_JD</th>\n",
- " <th>insured_education_level_MD</th>\n",
- " <th>insured_education_level_Masters</th>\n",
- " <th>insured_education_level_PhD</th>\n",
- " <th>insured_occupation_armed-forces</th>\n",
- " <th>...</th>\n",
- " <th>capital-gains</th>\n",
- " <th>capital-loss</th>\n",
- " <th>number_of_vehicles_involved</th>\n",
- " <th>bodily_injuries</th>\n",
- " <th>witnesses</th>\n",
- " <th>injury_claim</th>\n",
- " <th>property_claim</th>\n",
- " <th>vehicle_claim</th>\n",
- " <th>fraud_reported</th>\n",
- " <th>pct_paid_insurance</th>\n",
- " </tr>\n",
- " </thead>\n",
- " <tbody>\n",
- " <tr>\n",
- " <th>0</th>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>...</td>\n",
- " <td>53300</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>1</td>\n",
- " <td>2</td>\n",
- " <td>6510</td>\n",
- " <td>13020</td>\n",
- " <td>52080</td>\n",
- " <td>1</td>\n",
- " <td>0.986035</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>1</th>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>...</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>780</td>\n",
- " <td>780</td>\n",
- " <td>3510</td>\n",
- " <td>1</td>\n",
- " <td>0.605523</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>2</th>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>...</td>\n",
- " <td>35100</td>\n",
- " <td>0</td>\n",
- " <td>3</td>\n",
- " <td>2</td>\n",
- " <td>3</td>\n",
- " <td>7700</td>\n",
- " <td>3850</td>\n",
- " <td>23100</td>\n",
- " <td>0</td>\n",
- " <td>0.942280</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>3</th>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>1</td>\n",
- " <td>...</td>\n",
- " <td>48900</td>\n",
- " <td>-62400</td>\n",
- " <td>1</td>\n",
- " <td>1</td>\n",
- " <td>2</td>\n",
- " <td>6340</td>\n",
- " <td>6340</td>\n",
- " <td>50720</td>\n",
- " <td>1</td>\n",
- " <td>0.968454</td>\n",
- " </tr>\n",
- " <tr>\n",
- " <th>4</th>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>0</td>\n",
- " <td>...</td>\n",
- " <td>66000</td>\n",
- " <td>-46000</td>\n",
- " <td>1</td>\n",
- " <td>0</td>\n",
- " <td>1</td>\n",
- " <td>1300</td>\n",
- " <td>650</td>\n",
- " <td>4550</td>\n",
- " <td>0</td>\n",
- " <td>0.846154</td>\n",
- " </tr>\n",
- " </tbody>\n",
- "</table>\n",
- "<p>5 rows × 74 columns</p>\n",
- "</div>"
- ],
- "text/plain": [
- " policy_csl_250/500 policy_csl_500/1000 insured_sex_MALE \\\n",
- "0 1 0 1 \n",
- "1 1 0 1 \n",
- "2 0 0 0 \n",
- "3 1 0 0 \n",
- "4 0 1 1 \n",
- "\n",
- " insured_education_level_College insured_education_level_High School \\\n",
- "0 0 0 \n",
- "1 0 0 \n",
- "2 0 0 \n",
- "3 0 0 \n",
- "4 0 0 \n",
- "\n",
- " insured_education_level_JD insured_education_level_MD \\\n",
- "0 0 1 \n",
- "1 0 1 \n",
- "2 0 0 \n",
- "3 0 0 \n",
- "4 0 0 \n",
- "\n",
- " insured_education_level_Masters insured_education_level_PhD \\\n",
- "0 0 0 \n",
- "1 0 0 \n",
- "2 0 1 \n",
- "3 0 1 \n",
- "4 0 0 \n",
- "\n",
- " insured_occupation_armed-forces ... capital-gains capital-loss \\\n",
- "0 0 ... 53300 0 \n",
- "1 0 ... 0 0 \n",
- "2 0 ... 35100 0 \n",
- "3 1 ... 48900 -62400 \n",
- "4 0 ... 66000 -46000 \n",
- "\n",
- " number_of_vehicles_involved bodily_injuries witnesses injury_claim \\\n",
- "0 1 1 2 6510 \n",
- "1 1 0 0 780 \n",
- "2 3 2 3 7700 \n",
- "3 1 1 2 6340 \n",
- "4 1 0 1 1300 \n",
- "\n",
- " property_claim vehicle_claim fraud_reported pct_paid_insurance \n",
- "0 13020 52080 1 0.986035 \n",
- "1 780 3510 1 0.605523 \n",
- "2 3850 23100 0 0.942280 \n",
- "3 6340 50720 1 0.968454 \n",
- "4 650 4550 0 0.846154 \n",
- "\n",
- "[5 rows x 74 columns]"
- ]
- },
- "execution_count": 51,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "data_preprocessed.head()"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "f562caaf-a351-4b63-8155-a49aee7bc7e3",
- "metadata": {},
- "source": [
- "data_preprocessed.to_csv('dataset_dummies', index=False)"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "13b5c9f4-f6d9-479d-be00-a00b9623a0c3",
- "metadata": {
- "editable": true,
- "include": true,
- "paragraph": "Teaser",
- "slideshow": {
- "slide_type": ""
- },
- "tags": []
- },
- "source": [
- "Versicherungs Unternehmen werden häufig zu Zielen von Betrügern, weshalb es sehr wichtig ist solche Betrugsversuche frühzeitig zu erkennen. Die Zeilen des Datensatzes stellen jeweils einen Kunden und Seine Vorfall dar. Die Spalten beschreiben die Merkmale der Kunden und die des Vorfalls für welchen sie ihre Versicherung in anspruch nehmen. Daten wie diese, werden von den Versicherungsunternehmen zunehmend automatisiert verarbeitet, ausgewertet und für weitere Versicherungsprozesse genutzt. Ziel ist es für bestehende Versicherungsprodukte das aktuelle Risiko zu berechnen und darauf aufbauend die Prämie und die mögliche Schadenshöhe zu ermitteln. Anhand dieses Datensatz soll mit „Machine-Learning“ ermittelt werden ob sich bei dem jeweiligen Fall um Betrug oder einen legitiemen Anspruch handelt. Logistische Regression, Entscheidungsbäume, Random Forest und Support Vector Machines werden hierbei genutzt um eine Vorhersage zu Fällen zu treffen. Das Finale Modell erreicht eine Genauigkeit von 95 % und einen Recall von 75 %. Die Mehrheit der Betrugsversuche wird mit diesem Modell erkannt. "
- ]
- },
- {
- "cell_type": "markdown",
- "id": "bd6e4d06-d1a3-4b65-b333-22d35816872a",
- "metadata": {
- "editable": true,
- "include": true,
- "paragraph": "Datenmodell",
- "slideshow": {
- "slide_type": ""
- },
- "tags": []
- },
- "source": [
- "Klassifizierungsmodelle sind vielfältig und umfassen zum Beispiel logistische Regression, Entscheidungsbaum, Random Forest und Support Vector Machines. Alle oben genannten Modelle wurden mit dem Datensatz getestet und anschließend wird das mit der Höchste Präzision genutzt. In diesem Fall ist das die Support Vector Machines "
- ]
- },
- {
- "cell_type": "markdown",
- "id": "25702f9d-9b4c-44d8-94a8-0b3d5c02b0ea",
- "metadata": {
- "editable": true,
- "include": true,
- "paragraph": "Evaluation",
- "slideshow": {
- "slide_type": ""
- },
- "tags": []
- },
- "source": [
- "Für die Bewertung der Qualität einer Klassifikation werden Metriken wir Accuracy (= allgemeine Genauigkeit der Klassifikation), Precision (= Präzision der Vorhersage der Kundenabwanderung) und Recall (= Menge der abwanderungswilligen Kunden die korrekt klassifiziert wurden) genutzt. In einer ersten Modellstufe wird eine Accuracy von 92%, ein Recall von 75% sowie eine Precision von 95% erreicht. Schlussendlich konnten 85% der Betrugsfälle korrekt erkannt werden. "
- ]
- },
- {
- "cell_type": "markdown",
- "id": "0edf9cce-4d87-4831-983b-cafd8f113f4f",
- "metadata": {
- "editable": true,
- "include": true,
- "paragraph": "Umsetzung",
- "slideshow": {
- "slide_type": ""
- },
- "tags": []
- },
- "source": [
- "Die Umsetzung bzw. Einbindung des Datenmodells bietet sich in CRM-Systemen an. Auf Basis von Vorfalls Merkmalen kann automatisiert eine Vorhersage über eine potenziellen Betrugsversuch erstellt werden. Auf diese Weise lassen sich Betrugsfälle identifizieren, in Form von Dashboards visualisieren sowie teil-automatisiert bearbeiten."
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "02d0d7b8-549e-43cf-9fba-44a72a67b2e5",
- "metadata": {
- "editable": true,
- "slideshow": {
- "slide_type": ""
- },
- "tags": []
- },
- "outputs": [],
- "source": []
- }
- ],
- "metadata": {
- "category": "Insurance",
- "kernelspec": {
- "display_name": "Python 3 (ipykernel)",
- "language": "python",
- "name": "python3"
- },
- "language_info": {
- "codemirror_mode": {
- "name": "ipython",
- "version": 3
- },
- "file_extension": ".py",
- "mimetype": "text/x-python",
- "name": "python",
- "nbconvert_exporter": "python",
- "pygments_lexer": "ipython3",
- "version": "3.11.2"
- },
- "skipNotebookInDeployment": true,
- "title": "Insurance Fraud detection - Versicherungs Betrugserkennung "
- },
- "nbformat": 4,
- "nbformat_minor": 5
-}
--
GitLab