diff --git a/.DS_Store b/.DS_Store new file mode 100644 index 0000000..9249650 Binary files /dev/null and b/.DS_Store differ diff --git a/.gitignore b/.gitignore new file mode 100644 index 0000000..7076c91 --- /dev/null +++ b/.gitignore @@ -0,0 +1 @@ +consumer_complaints.csv \ No newline at end of file diff --git a/.ipynb_checkpoints/consumer_analysis-checkpoint.ipynb b/.ipynb_checkpoints/consumer_analysis-checkpoint.ipynb new file mode 100644 index 0000000..7f1f005 --- /dev/null +++ b/.ipynb_checkpoints/consumer_analysis-checkpoint.ipynb @@ -0,0 +1,1742 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "2532726f", + "metadata": {}, + "source": [ + "# Analyzing Consumer Complaints from Financial Corporations/Institutions" + ] + }, + { + "cell_type": "markdown", + "id": "5f270b27", + "metadata": {}, + "source": [ + "## Introduction:\n", + "In this notebook I will be analyzing the counts and percentage of **disputed** customer complaints with regard to financial institutions, states and financial products along with a quick and dirty machine learning model to give a rough prediction of when the disputes will occur" + ] + }, + { + "cell_type": "markdown", + "id": "dd4e748c", + "metadata": {}, + "source": [ + "## Getting Required Packages:" + ] + }, + { + "cell_type": "code", + "execution_count": 55, + "id": "0060a11a", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Requirement already satisfied: xgboost in /Users/aryanvakharia/opt/anaconda3/lib/python3.9/site-packages (1.6.2)\n", + "Requirement already satisfied: numpy in /Users/aryanvakharia/opt/anaconda3/lib/python3.9/site-packages (from xgboost) (1.21.5)\n", + "Requirement already satisfied: scipy in /Users/aryanvakharia/opt/anaconda3/lib/python3.9/site-packages (from xgboost) (1.7.3)\n", + "\n", + "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m A new release of pip available: \u001b[0m\u001b[31;49m22.1.2\u001b[0m\u001b[39;49m -> \u001b[0m\u001b[32;49m22.2.2\u001b[0m\n", + "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m To update, run: \u001b[0m\u001b[32;49mpip install --upgrade pip\u001b[0m\n", + "Note: you may need to restart the kernel to use updated packages.\n" + ] + } + ], + "source": [ + "%pip install xgboost\n", + "\n", + "import pandas as pd\n", + "import numpy as np\n", + "import matplotlib.pyplot as plt\n", + "from sklearn.preprocessing import MinMaxScaler\n", + "from sklearn.impute import SimpleImputer\n", + "from sklearn.linear_model import LogisticRegression\n", + "from sklearn.tree import DecisionTreeClassifier\n", + "import xgboost as xgb\n", + "from sklearn.model_selection import train_test_split, StratifiedShuffleSplit\n", + "from sklearn.metrics import accuracy_score, plot_confusion_matrix, classification_report\n", + "\n", + "%matplotlib inline" + ] + }, + { + "cell_type": "markdown", + "id": "53fc5391", + "metadata": {}, + "source": [ + "## Importing and Cleaning Data" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "f92ee33a", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
date_receivedproductsub_productissuesub_issueconsumer_complaint_narrativecompany_public_responsecompanystatezipcodetagsconsumer_consent_providedsubmitted_viadate_sent_to_companycompany_response_to_consumertimely_responseconsumer_disputed?complaint_id
02013-08-30MortgageOther mortgageLoan modification,collection,foreclosureNaNNaNNaNU.S. BancorpCA95993NaNNaNReferral2013-09-03Closed with explanationYesYes511074
12013-08-30MortgageOther mortgageLoan servicing, payments, escrow accountNaNNaNNaNWells Fargo & CompanyCA91104NaNNaNReferral2013-09-03Closed with explanationYesYes511080
22013-08-30Credit reportingNaNIncorrect information on credit reportAccount statusNaNNaNWells Fargo & CompanyNY11764NaNNaNPostal mail2013-09-18Closed with explanationYesNo510473
32013-08-30Student loanNon-federal student loanRepaying your loanRepaying your loanNaNNaNNavient Solutions, Inc.MD21402NaNNaNEmail2013-08-30Closed with explanationYesYes510326
42013-08-30Debt collectionCredit cardFalse statements or representationAttempted to collect wrong amountNaNNaNResurgent Capital Services L.P.GA30106NaNNaNWeb2013-08-30Closed with explanationYesYes511067
\n", + "
" + ], + "text/plain": [ + " date_received product sub_product \\\n", + "0 2013-08-30 Mortgage Other mortgage \n", + "1 2013-08-30 Mortgage Other mortgage \n", + "2 2013-08-30 Credit reporting NaN \n", + "3 2013-08-30 Student loan Non-federal student loan \n", + "4 2013-08-30 Debt collection Credit card \n", + "\n", + " issue \\\n", + "0 Loan modification,collection,foreclosure \n", + "1 Loan servicing, payments, escrow account \n", + "2 Incorrect information on credit report \n", + "3 Repaying your loan \n", + "4 False statements or representation \n", + "\n", + " sub_issue consumer_complaint_narrative \\\n", + "0 NaN NaN \n", + "1 NaN NaN \n", + "2 Account status NaN \n", + "3 Repaying your loan NaN \n", + "4 Attempted to collect wrong amount NaN \n", + "\n", + " company_public_response company state zipcode tags \\\n", + "0 NaN U.S. Bancorp CA 95993 NaN \n", + "1 NaN Wells Fargo & Company CA 91104 NaN \n", + "2 NaN Wells Fargo & Company NY 11764 NaN \n", + "3 NaN Navient Solutions, Inc. MD 21402 NaN \n", + "4 NaN Resurgent Capital Services L.P. GA 30106 NaN \n", + "\n", + " consumer_consent_provided submitted_via date_sent_to_company \\\n", + "0 NaN Referral 2013-09-03 \n", + "1 NaN Referral 2013-09-03 \n", + "2 NaN Postal mail 2013-09-18 \n", + "3 NaN Email 2013-08-30 \n", + "4 NaN Web 2013-08-30 \n", + "\n", + " company_response_to_consumer timely_response consumer_disputed? \\\n", + "0 Closed with explanation Yes Yes \n", + "1 Closed with explanation Yes Yes \n", + "2 Closed with explanation Yes No \n", + "3 Closed with explanation Yes Yes \n", + "4 Closed with explanation Yes Yes \n", + "\n", + " complaint_id \n", + "0 511074 \n", + "1 511080 \n", + "2 510473 \n", + "3 510326 \n", + "4 511067 " + ] + }, + "execution_count": 2, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "df = pd.read_csv('consumer_complaints.csv', parse_dates=['date_received', 'date_sent_to_company'], low_memory=False)\n", + "df.head()" + ] + }, + { + "cell_type": "markdown", + "id": "88e39fd7", + "metadata": {}, + "source": [ + "Having a clean dataset allows for higher accuracy in visualizing, analyzing and in this case predicting on the dataset. In order to make the most efficient use of the above data, I am checking for the number of empty/NaN rows:" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "73bf52d6", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "date_received 0\n", + "product 0\n", + "sub_product 158322\n", + "issue 0\n", + "sub_issue 343335\n", + "consumer_complaint_narrative 489151\n", + "company_public_response 470833\n", + "company 0\n", + "state 4887\n", + "zipcode 4505\n", + "tags 477998\n", + "consumer_consent_provided 432499\n", + "submitted_via 0\n", + "date_sent_to_company 0\n", + "company_response_to_consumer 0\n", + "timely_response 0\n", + "consumer_disputed? 0\n", + "complaint_id 0\n", + "dtype: int64" + ] + }, + "execution_count": 3, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "#Check NaNs\n", + "df.isna().sum()" + ] + }, + { + "cell_type": "markdown", + "id": "e8db998a", + "metadata": {}, + "source": [ + "From the above output it is evident that the dataset if populated with many NaNs. However, removing all NaN values can cause overfitting in the prediction model and inaccuracies in the data analysis.\n", + "\n", + "So, keeping threshold of *column contains >= 10% NaNs* we will clean each column" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "45eebff3", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "sub_product 0.284774\n", + "sub_issue 0.617557\n", + "consumer_complaint_narrative 0.879836\n", + "company_public_response 0.846887\n", + "tags 0.859775\n", + "consumer_consent_provided 0.777936\n", + "dtype: float64" + ] + }, + "execution_count": 4, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "thresh = 0.10\n", + "col_nans = df.isna().mean(axis=0)\n", + "col_w_thresh = col_nans[col_nans >= thresh]\n", + "col_w_thresh" + ] + }, + { + "cell_type": "markdown", + "id": "dee366ce", + "metadata": {}, + "source": [ + "The above columns will now be dropped as they have too many NaNs to have any use in our dataset.\n", + "\n", + "*NOTE:* Along with these, the *complaint_id* column will aso be dropped for not being relevant in our analysis" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "252d522a", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
date_receivedproductissuecompanystatezipcodesubmitted_viadate_sent_to_companycompany_response_to_consumertimely_responseconsumer_disputed?
02013-08-30MortgageLoan modification,collection,foreclosureU.S. BancorpCA95993Referral2013-09-03111
12013-08-30MortgageLoan servicing, payments, escrow accountWells Fargo & CompanyCA91104Referral2013-09-03111
22013-08-30Credit reportingIncorrect information on credit reportWells Fargo & CompanyNY11764Postal mail2013-09-18110
32013-08-30Student loanRepaying your loanNavient Solutions, Inc.MD21402Email2013-08-30111
42013-08-30Debt collectionFalse statements or representationResurgent Capital Services L.P.GA30106Web2013-08-30111
\n", + "
" + ], + "text/plain": [ + " date_received product issue \\\n", + "0 2013-08-30 Mortgage Loan modification,collection,foreclosure \n", + "1 2013-08-30 Mortgage Loan servicing, payments, escrow account \n", + "2 2013-08-30 Credit reporting Incorrect information on credit report \n", + "3 2013-08-30 Student loan Repaying your loan \n", + "4 2013-08-30 Debt collection False statements or representation \n", + "\n", + " company state zipcode submitted_via \\\n", + "0 U.S. Bancorp CA 95993 Referral \n", + "1 Wells Fargo & Company CA 91104 Referral \n", + "2 Wells Fargo & Company NY 11764 Postal mail \n", + "3 Navient Solutions, Inc. MD 21402 Email \n", + "4 Resurgent Capital Services L.P. GA 30106 Web \n", + "\n", + " date_sent_to_company company_response_to_consumer timely_response \\\n", + "0 2013-09-03 1 1 \n", + "1 2013-09-03 1 1 \n", + "2 2013-09-18 1 1 \n", + "3 2013-08-30 1 1 \n", + "4 2013-08-30 1 1 \n", + "\n", + " consumer_disputed? \n", + "0 1 \n", + "1 1 \n", + "2 0 \n", + "3 1 \n", + "4 1 " + ] + }, + "execution_count": 5, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "drops_l = list(col_w_thresh.index) + ['complaint_id']\n", + "df = df.drop(drops_l, axis=1)\n", + "\n", + "#Tokenizing Yes and No to 1 and 0 respectively\n", + "df = df.replace(['Yes', 'No'], [1, 0])\n", + "\n", + "#Also tokenizing company response for easier NLP classification\n", + "df.loc[df['company_response_to_consumer'].str.contains(\"Closed\"), 'company_response_to_consumer'] = 1\n", + "df.loc[df['company_response_to_consumer'] != 1, 'company_response_to_consumer'] = 0\n", + "df.head()" + ] + }, + { + "cell_type": "markdown", + "id": "0e9dee79", + "metadata": {}, + "source": [ + "## Data Analysis and Visualization:\n", + "Since I will be mainly checking the count AND rate of disputed issues, the following functions will be used to get these metrics as a DataFrame:\n", + "\n", + "(*NOTE: Since the data being analysed is categorical I will mostly be using bar graphs*)" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "08b7d908", + "metadata": {}, + "outputs": [], + "source": [ + "def is_disputed(row, cat):\n", + " return df[(df['consumer_disputed?'] == 1) & (df[cat] == row[cat])]['consumer_disputed?'].sum()\n", + "\n", + "def get_count(cat, data=df):\n", + " df_by_cat = df[[cat, 'issue', 'consumer_disputed?']].groupby([cat]).count().reset_index()\n", + " \n", + " df_by_cat['consumer_disputed?'] = df_by_cat.apply(lambda r: is_disputed(r, cat), axis=1)\n", + " \n", + " df_by_cat['dispute_rate'] = df_by_cat.apply(lambda r: (r['consumer_disputed?'] / r['issue']) * 100, axis=1)\n", + " \n", + " return df_by_cat" + ] + }, + { + "cell_type": "markdown", + "id": "6693cced", + "metadata": {}, + "source": [ + "### Disputes according to State:" + ] + }, + { + "cell_type": "code", + "execution_count": 29, + "id": "d6ad5d11", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "State is CA with max disputes: 17615\n" + ] + } + ], + "source": [ + "#State with most disputes using above function:\n", + "disp_st = get_count('state')\n", + "st_c = disp_st['consumer_disputed?']\n", + "m_st = disp_st['state'][pd.Series.argmax(st_c)]\n", + "m_c = disp_st['consumer_disputed?'][pd.Series.argmax(st_c)]\n", + "print(\"State is\", m_st, \"with max disputes:\", m_c)" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "6f0729b6", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "Text(0.5, 1.0, 'Dispute counts per state')" + ] + }, + "execution_count": 8, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "fig1, ax1 = plt.subplots(figsize=(20, 9))\n", + "ax1.bar(disp_st['state'], disp_st['consumer_disputed?'])\n", + "ax1.set_xlabel(\"State\")\n", + "ax1.set_ylabel(\"Disputed issue counts\")\n", + "ax1.set_title(\"Dispute counts per state\")" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "6dfa6225", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "Text(0.5, 1.0, 'Dispute rates per state')" + ] + }, + "execution_count": 9, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "fig2, ax2 = plt.subplots(figsize=(20, 9))\n", + "ax2.bar(disp_st['state'], disp_st['dispute_rate'])\n", + "ax2.set_xlabel(\"State\")\n", + "ax2.set_ylabel(\"Disputed issue rate(disputed count / issue count)\")\n", + "ax2.set_title(\"Dispute rates per state\")" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "17ec819c", + "metadata": {}, + "outputs": [], + "source": [ + "#Getting count of issues per state with products\n", + "state_issue_count = df[['state', 'product', 'issue']].groupby(['state', 'product', 'issue']).size().reset_index()\n", + "state_issue_count.columns = ['state', 'product', 'issue', 'count']" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "id": "15c92542", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
stateproductissuecount
501CAMortgageLoan modification,collection,foreclosure19433
942FLMortgageLoan modification,collection,foreclosure11337
502CAMortgageLoan servicing, payments, escrow account8957
4028TXCredit reportingIncorrect information on credit report8712
485CACredit reportingIncorrect information on credit report8585
\n", + "
" + ], + "text/plain": [ + " state product issue count\n", + "501 CA Mortgage Loan modification,collection,foreclosure 19433\n", + "942 FL Mortgage Loan modification,collection,foreclosure 11337\n", + "502 CA Mortgage Loan servicing, payments, escrow account 8957\n", + "4028 TX Credit reporting Incorrect information on credit report 8712\n", + "485 CA Credit reporting Incorrect information on credit report 8585" + ] + }, + "execution_count": 11, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "#States with 5 highest dispute counts, issues and product\n", + "state_issue_count.sort_values(by = 'count', ascending = False).head()" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "id": "63fa2a3a", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
stateproductissuecount
3083NVPrepaid cardOverdraft, savings or rewards features1
347AZConsumer LoanPayment to acct not credited1
344AZConsumer LoanLender repossessed or sold the vehicle1
1AACredit cardIdentity theft / Fraud / Embezzlement1
0AABank account or serviceAccount opening, closing, or management1
\n", + "
" + ], + "text/plain": [ + " state product issue \\\n", + "3083 NV Prepaid card Overdraft, savings or rewards features \n", + "347 AZ Consumer Loan Payment to acct not credited \n", + "344 AZ Consumer Loan Lender repossessed or sold the vehicle \n", + "1 AA Credit card Identity theft / Fraud / Embezzlement \n", + "0 AA Bank account or service Account opening, closing, or management \n", + "\n", + " count \n", + "3083 1 \n", + "347 1 \n", + "344 1 \n", + "1 1 \n", + "0 1 " + ] + }, + "execution_count": 12, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "#States with 5 lowest dispute counts with related issues and products\n", + "state_issue_count.sort_values(by = 'count', ascending = False).tail()" + ] + }, + { + "cell_type": "markdown", + "id": "eb748a29", + "metadata": {}, + "source": [ + "### **Analysis**: \n", + "As the above console output and bar graphs show, states with higher populations such as CA, NY, TX etc tend to have more disputes over financial issues. Moreover, West Coast states have a higher tendency to dipute complaints. Finally, Mortgage modifications and Credit reporting issues have caused the most disputed among the states \n", + "\n", + "Laws and operations regarding financial products and the companies headquartered in these states are areas which can be further analysed for why they yield such high issue counts for the a certain range of products." + ] + }, + { + "cell_type": "markdown", + "id": "f45775c1", + "metadata": {}, + "source": [ + "### Disputes according to Product:" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "id": "1b73be97", + "metadata": {}, + "outputs": [], + "source": [ + "disp_prod = get_count('product')" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "id": "95b32810", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
productissuecount
75MortgageLoan modification,collection,foreclosure97191
59Credit reportingIncorrect information on credit report66718
76MortgageLoan servicing, payments, escrow account60375
62Debt collectionCont'd attempts collect debt not owed42285
0Bank account or serviceAccount opening, closing, or management26661
\n", + "
" + ], + "text/plain": [ + " product issue count\n", + "75 Mortgage Loan modification,collection,foreclosure 97191\n", + "59 Credit reporting Incorrect information on credit report 66718\n", + "76 Mortgage Loan servicing, payments, escrow account 60375\n", + "62 Debt collection Cont'd attempts collect debt not owed 42285\n", + "0 Bank account or service Account opening, closing, or management 26661" + ] + }, + "execution_count": 14, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "#Highest issues per product\n", + "product_issue_count = df.groupby(['product', 'issue']).size().reset_index()\n", + "product_issue_count.columns = ['product', 'issue', 'count']\n", + "product_issue_count.sort_values(by = 'count', ascending=False).head()" + ] + }, + { + "cell_type": "code", + "execution_count": 15, + "id": "cd258ff1", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "Text(0, 0.5, 'Product')" + ] + }, + "execution_count": 15, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "fig3, ax3 = plt.subplots(figsize=(16, 9))\n", + "ax3.barh(disp_prod['product'], disp_prod['consumer_disputed?']) \n", + "\n", + "ax3.set_title(\"Disputed issue counts per product\")\n", + "ax3.set_xlabel('Disputed issue counts')\n", + "ax3.set_ylabel('Product')" + ] + }, + { + "cell_type": "code", + "execution_count": 16, + "id": "40b666b1", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "Text(0.5, 0, 'Disputed issue rate')" + ] + }, + "execution_count": 16, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "fig4, ax4 = plt.subplots(figsize=(16, 9))\n", + "ax4.barh(disp_prod['product'], disp_prod['dispute_rate']) \n", + "\n", + "ax4.set_title(\"Disputed issue rates per product\")\n", + "ax4.set_ylabel('Product')\n", + "ax4.set_xlabel('Disputed issue rate')" + ] + }, + { + "cell_type": "markdown", + "id": "61b80521", + "metadata": {}, + "source": [ + "### Analysis:\n", + "Using the same visualization it can be concluded that loan and credit related issues like *Mortgages, Credit Reporting, Student loans* etc are more likely to be disputed, meaning our company should be wary and prepared to efficiently face complicated customer service with such products" + ] + }, + { + "cell_type": "markdown", + "id": "32296070", + "metadata": {}, + "source": [ + "### Disputes according to submission medium:" + ] + }, + { + "cell_type": "code", + "execution_count": 17, + "id": "fd3459f9", + "metadata": {}, + "outputs": [], + "source": [ + "disp_med = get_count('submitted_via')" + ] + }, + { + "cell_type": "code", + "execution_count": 18, + "id": "abdbe069", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "Text(0.5, 17.200000000000003, 'Disputed issue count')" + ] + }, + "execution_count": 18, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "fig5, ax5 = plt.subplots(figsize=(16, 9))\n", + "ax5.bar(disp_med['submitted_via'], disp_med['consumer_disputed?'])\n", + "\n", + "ax4.set_title(\"Disputed issue counts per submission medium\")\n", + "ax4.set_xlabel('Medium')\n", + "ax4.set_xlabel('Disputed issue count')" + ] + }, + { + "cell_type": "code", + "execution_count": 19, + "id": "5c033a69", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "Text(0, 0.5, 'Disputed issue rate')" + ] + }, + "execution_count": 19, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "fig6, ax6 = plt.subplots(figsize=(16, 9))\n", + "ax6.bar(disp_med['submitted_via'], disp_med['dispute_rate'])\n", + "\n", + "ax6.set_title(\"Disputed issue rates per submission medium\")\n", + "ax6.set_xlabel('Medium')\n", + "ax6.set_ylabel('Disputed issue rate')" + ] + }, + { + "cell_type": "code", + "execution_count": 21, + "id": "52069de3", + "metadata": {}, + "outputs": [], + "source": [ + "df_by_med = df[['submitted_via', 'timely_response']].groupby(['submitted_via', 'timely_response']).size().reset_index()\n", + "df_by_med = df_by_med[df_by_med['timely_response'] == 1]\n", + "df_by_med.columns = ['submitted_via', 'was timely_response', 'count']" + ] + }, + { + "cell_type": "code", + "execution_count": 22, + "id": "06157da5", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "Text(0, 0.5, 'Timely response count')" + ] + }, + "execution_count": 22, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "fig7, ax7 = plt.subplots(figsize=(16, 9))\n", + "ax7.bar(df_by_med['submitted_via'], df_by_med['count'])\n", + "\n", + "ax7.set_title(\"Number of timely responses per submission medium\")\n", + "ax7.set_xlabel('Medium')\n", + "ax7.set_ylabel('Timely response count')" + ] + }, + { + "cell_type": "markdown", + "id": "86c3e653", + "metadata": {}, + "source": [ + "### Analysis:\n", + "Quicker, digital forms of communication such as *Email* and *Web* clearly have a higher dispute rate than *Postal mail* or *Referral*, however it should be noted that timeliness of the response does not hold much weight for digital forms. This could highlight the possible complacency/bureaucracy that these companies may have that causes them to give a clearly untimely response, and can also warrant further delving into the structure of customer service in such companies." + ] + }, + { + "cell_type": "markdown", + "id": "a6eae2cf", + "metadata": {}, + "source": [ + "## Sentiment Analysis/Prediction:" + ] + }, + { + "cell_type": "markdown", + "id": "05f5c8e9", + "metadata": {}, + "source": [ + "### Setting up data:\n", + "Since the key data is categorical in nature we can use *one-hot encoding* for it as there is no ordinal relationship in any of the data that is being used." + ] + }, + { + "cell_type": "code", + "execution_count": 30, + "id": "c380bd0b", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
date_receivedcompanyzipcodedate_sent_to_companycompany_response_to_consumertimely_responseconsumer_disputed?product_Bank account or serviceproduct_Consumer Loanproduct_Credit card...state_WAstate_WIstate_WVstate_WYsubmitted_via_Emailsubmitted_via_Faxsubmitted_via_Phonesubmitted_via_Postal mailsubmitted_via_Referralsubmitted_via_Web
0735110U.S. Bancorp95993735114111000...0000000010
1735110Wells Fargo & Company91104735114111000...0000000010
2735110Wells Fargo & Company11764735129110000...0000000100
3735110Navient Solutions, Inc.21402735110111000...0000100000
4735110Resurgent Capital Services L.P.30106735110111000...0000000001
\n", + "

5 rows × 181 columns

\n", + "
" + ], + "text/plain": [ + " date_received company zipcode \\\n", + "0 735110 U.S. Bancorp 95993 \n", + "1 735110 Wells Fargo & Company 91104 \n", + "2 735110 Wells Fargo & Company 11764 \n", + "3 735110 Navient Solutions, Inc. 21402 \n", + "4 735110 Resurgent Capital Services L.P. 30106 \n", + "\n", + " date_sent_to_company company_response_to_consumer timely_response \\\n", + "0 735114 1 1 \n", + "1 735114 1 1 \n", + "2 735129 1 1 \n", + "3 735110 1 1 \n", + "4 735110 1 1 \n", + "\n", + " consumer_disputed? product_Bank account or service product_Consumer Loan \\\n", + "0 1 0 0 \n", + "1 1 0 0 \n", + "2 0 0 0 \n", + "3 1 0 0 \n", + "4 1 0 0 \n", + "\n", + " product_Credit card ... state_WA state_WI state_WV state_WY \\\n", + "0 0 ... 0 0 0 0 \n", + "1 0 ... 0 0 0 0 \n", + "2 0 ... 0 0 0 0 \n", + "3 0 ... 0 0 0 0 \n", + "4 0 ... 0 0 0 0 \n", + "\n", + " submitted_via_Email submitted_via_Fax submitted_via_Phone \\\n", + "0 0 0 0 \n", + "1 0 0 0 \n", + "2 0 0 0 \n", + "3 1 0 0 \n", + "4 0 0 0 \n", + "\n", + " submitted_via_Postal mail submitted_via_Referral submitted_via_Web \n", + "0 0 1 0 \n", + "1 0 1 0 \n", + "2 1 0 0 \n", + "3 0 0 0 \n", + "4 0 0 1 \n", + "\n", + "[5 rows x 181 columns]" + ] + }, + "execution_count": 30, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "#Converting dates to numerical values instead of datetime\n", + "date_cols = ['date_received', 'date_sent_to_company']\n", + "for date_col in date_cols:\n", + " df[date_col] = df[date_col].apply(pd.Timestamp.toordinal)\n", + "\n", + "analysis_cols = ['product', 'issue', 'state', 'submitted_via']\n", + "encoded = df\n", + "for c in analysis_cols:\n", + " c_encoded = pd.get_dummies(encoded[c], prefix=c)\n", + " encoded = pd.concat((encoded, c_encoded), axis=1).drop(c, axis=1)\n", + "encoded.head() " + ] + }, + { + "cell_type": "code", + "execution_count": 31, + "id": "44062878", + "metadata": {}, + "outputs": [], + "source": [ + "X = encoded.drop(['company', 'zipcode', 'consumer_disputed?'], axis=1)\n", + "y = encoded['consumer_disputed?']\n", + "X_tr, X_t, y_tr, y_t = train_test_split(X, y, test_size=0.2, random_state=0)" + ] + }, + { + "cell_type": "markdown", + "id": "b0de514c", + "metadata": {}, + "source": [ + "**First using Logistic Regression to classify one-hot-encoded features:**" + ] + }, + { + "cell_type": "code", + "execution_count": 33, + "id": "25a11427", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Classifier accuracy is 0.7980160443197353\n" + ] + } + ], + "source": [ + "classifier1 = LogisticRegression()\n", + "classifier1.fit(X_tr, y_tr)\n", + "y_pred = classifier1.predict(X_t)\n", + "\n", + "print(\"Classifier accuracy is\", accuracy_score(y_t, y_pred))" + ] + }, + { + "cell_type": "code", + "execution_count": 40, + "id": "a3ce0650", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/Users/aryanvakharia/opt/anaconda3/lib/python3.9/site-packages/sklearn/utils/deprecation.py:87: FutureWarning: Function plot_confusion_matrix is deprecated; Function `plot_confusion_matrix` is deprecated in 1.0 and will be removed in 1.2. Use one of the class methods: ConfusionMatrixDisplay.from_predictions or ConfusionMatrixDisplay.from_estimator.\n", + " warnings.warn(msg, category=FutureWarning)\n" + ] + }, + { + "data": { + "text/plain": [ + "" + ] + }, + "execution_count": 40, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "plot_confusion_matrix(classifier1, X_t, y_t, cmap=plt.cm.Greens)" + ] + }, + { + "cell_type": "markdown", + "id": "141b8875", + "metadata": {}, + "source": [ + "### Analysis:\n", + "As we can see the classifier needs a great deal of work to accurately predict a disputed complaint as the above confusion matrix shows that a true disputed complaint is not recognized by the classifier." + ] + }, + { + "cell_type": "markdown", + "id": "730fb665", + "metadata": {}, + "source": [ + "**Using Random Forest Classifier to better prediction:**" + ] + }, + { + "cell_type": "code", + "execution_count": 52, + "id": "fc0a33cc", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Classifier accuracy is 0.7119846751564861\n" + ] + } + ], + "source": [ + "classifier2 = DecisionTreeClassifier(random_state=42)\n", + "classifier2.fit(X_tr, y_tr)\n", + "y_pred = classifier2.predict(X_t)\n", + "\n", + "print(\"Classifier accuracy is\", accuracy_score(y_t, y_pred))" + ] + }, + { + "cell_type": "code", + "execution_count": 53, + "id": "0435bfbf", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/Users/aryanvakharia/opt/anaconda3/lib/python3.9/site-packages/sklearn/utils/deprecation.py:87: FutureWarning: Function plot_confusion_matrix is deprecated; Function `plot_confusion_matrix` is deprecated in 1.0 and will be removed in 1.2. Use one of the class methods: ConfusionMatrixDisplay.from_predictions or ConfusionMatrixDisplay.from_estimator.\n", + " warnings.warn(msg, category=FutureWarning)\n" + ] + }, + { + "data": { + "text/plain": [ + "" + ] + }, + "execution_count": 53, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "plot_confusion_matrix(classifier2, X_t, y_t, cmap=plt.cm.Reds)" + ] + }, + { + "cell_type": "code", + "execution_count": 57, + "id": "2862af72", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + " precision recall f1-score support\n", + "\n", + " 0 0.81 0.84 0.82 88733\n", + " 1 0.26 0.23 0.24 22459\n", + "\n", + " accuracy 0.71 111192\n", + " macro avg 0.53 0.53 0.53 111192\n", + "weighted avg 0.70 0.71 0.70 111192\n", + "\n" + ] + } + ], + "source": [ + "print(classification_report(y_t, y_pred))" + ] + }, + { + "cell_type": "markdown", + "id": "52f413a8", + "metadata": {}, + "source": [ + "### Analysis:\n", + "The decision tree model has a lower accuracy but because it deals with the categorical data better it manages to correctly predict some of the disputed issues." + ] + }, + { + "cell_type": "markdown", + "id": "c04cfc62", + "metadata": {}, + "source": [ + "## Conclusion:" + ] + }, + { + "cell_type": "markdown", + "id": "ce1b2eb9", + "metadata": {}, + "source": [ + "Since both models are very quick and dirty, detailed parsing of the data using the *nltk* library would be ideal in getting a higher recall and higher number of true positive recognition in the future.\n", + "\n", + "However, the visualizations and insights provided by the dataset do inform of key areas for our company to be wary and cautious of, and with constant improvement of data preparation and more robust training models there lies a higher scope of predicting disputed issues and even company responses." + ] + } + ], + "metadata": { + "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": "3.9.12" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/README.md b/README.md index 4519ef1..3c3034a 100644 --- a/README.md +++ b/README.md @@ -1,6 +1,3 @@ -# data-oa -Online Assessments for the Data Team - # Data Science: Exploratory Data Analysis on a Consumer Finance Dataset Exploratory data analysis plays a huge part in quantitative finance. Often, in a variety of scenarios, consumer interests, expectations and beliefs play a huge role in deciding major moves for organisations. @@ -32,30 +29,3 @@ Link to dataset: https://drive.google.com/file/d/1Vnny6HVGBgF_mdDtIo2ddOTtOiGF5j Log in with your GApps@Illinois account to access this dataset. -## **Deliverables** -1. A Jupyter Notebook with all the code and your thought process. (Remember the thought process is what we're looking for) -2. A generated, dynamic PDF File with the insights that you have gathered. -3. A list of future hypotheses that you may not have yet implemented / couldn't figure out how to, but feel that could be interesting. Ideally, this would be structured in LaTex. - -## **Grading** -We will be looking at your project and grading it under these five criteria: -1. Code - - If it works - - Modular - - Follows best practices (ie. OOP) -2. Documentation - - Concise and exact. - - Follows popular conventions. - - Visuality. (Lots of graphs and statistics) -3. Styling - - Human readable - - Can quickly glance to receive all relevant information - - Follows Google Style Guide (preferred if it exists) or most popular convention (ie. PEP8) -4. Robustness - - Customizable - - No technical debt (future proof) - - Handles bad inputs and errors -5. Git - - [Good commit messages](https://cbea.ms/git-commit/#seven-rules) - - Commits are properly sized - diff --git a/consumer_analysis.ipynb b/consumer_analysis.ipynb new file mode 100644 index 0000000..5e10f2a --- /dev/null +++ b/consumer_analysis.ipynb @@ -0,0 +1,1744 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "2532726f", + "metadata": {}, + "source": [ + "# Analyzing Consumer Complaints from Financial Corporations/Institutions" + ] + }, + { + "cell_type": "markdown", + "id": "5f270b27", + "metadata": {}, + "source": [ + "## Introduction:\n", + "In this notebook I will be analyzing the counts and percentage of **disputed** customer complaints with regard to financial institutions, states and financial products along with a quick and dirty machine learning model to give a rough prediction of when the disputes will occur" + ] + }, + { + "cell_type": "markdown", + "id": "dd4e748c", + "metadata": {}, + "source": [ + "## Getting Required Packages:" + ] + }, + { + "cell_type": "code", + "execution_count": 55, + "id": "0060a11a", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Requirement already satisfied: xgboost in /Users/aryanvakharia/opt/anaconda3/lib/python3.9/site-packages (1.6.2)\n", + "Requirement already satisfied: numpy in /Users/aryanvakharia/opt/anaconda3/lib/python3.9/site-packages (from xgboost) (1.21.5)\n", + "Requirement already satisfied: scipy in /Users/aryanvakharia/opt/anaconda3/lib/python3.9/site-packages (from xgboost) (1.7.3)\n", + "\n", + "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m A new release of pip available: \u001b[0m\u001b[31;49m22.1.2\u001b[0m\u001b[39;49m -> \u001b[0m\u001b[32;49m22.2.2\u001b[0m\n", + "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m To update, run: \u001b[0m\u001b[32;49mpip install --upgrade pip\u001b[0m\n", + "Note: you may need to restart the kernel to use updated packages.\n" + ] + } + ], + "source": [ + "%pip install xgboost\n", + "\n", + "import pandas as pd\n", + "import numpy as np\n", + "import matplotlib.pyplot as plt\n", + "from sklearn.preprocessing import MinMaxScaler\n", + "from sklearn.impute import SimpleImputer\n", + "from sklearn.linear_model import LogisticRegression\n", + "from sklearn.tree import DecisionTreeClassifier\n", + "import xgboost as xgb\n", + "from sklearn.model_selection import train_test_split, StratifiedShuffleSplit\n", + "from sklearn.metrics import accuracy_score, plot_confusion_matrix, classification_report\n", + "\n", + "%matplotlib inline" + ] + }, + { + "cell_type": "markdown", + "id": "53fc5391", + "metadata": {}, + "source": [ + "## Importing and Cleaning Data" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "f92ee33a", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
date_receivedproductsub_productissuesub_issueconsumer_complaint_narrativecompany_public_responsecompanystatezipcodetagsconsumer_consent_providedsubmitted_viadate_sent_to_companycompany_response_to_consumertimely_responseconsumer_disputed?complaint_id
02013-08-30MortgageOther mortgageLoan modification,collection,foreclosureNaNNaNNaNU.S. BancorpCA95993NaNNaNReferral2013-09-03Closed with explanationYesYes511074
12013-08-30MortgageOther mortgageLoan servicing, payments, escrow accountNaNNaNNaNWells Fargo & CompanyCA91104NaNNaNReferral2013-09-03Closed with explanationYesYes511080
22013-08-30Credit reportingNaNIncorrect information on credit reportAccount statusNaNNaNWells Fargo & CompanyNY11764NaNNaNPostal mail2013-09-18Closed with explanationYesNo510473
32013-08-30Student loanNon-federal student loanRepaying your loanRepaying your loanNaNNaNNavient Solutions, Inc.MD21402NaNNaNEmail2013-08-30Closed with explanationYesYes510326
42013-08-30Debt collectionCredit cardFalse statements or representationAttempted to collect wrong amountNaNNaNResurgent Capital Services L.P.GA30106NaNNaNWeb2013-08-30Closed with explanationYesYes511067
\n", + "
" + ], + "text/plain": [ + " date_received product sub_product \\\n", + "0 2013-08-30 Mortgage Other mortgage \n", + "1 2013-08-30 Mortgage Other mortgage \n", + "2 2013-08-30 Credit reporting NaN \n", + "3 2013-08-30 Student loan Non-federal student loan \n", + "4 2013-08-30 Debt collection Credit card \n", + "\n", + " issue \\\n", + "0 Loan modification,collection,foreclosure \n", + "1 Loan servicing, payments, escrow account \n", + "2 Incorrect information on credit report \n", + "3 Repaying your loan \n", + "4 False statements or representation \n", + "\n", + " sub_issue consumer_complaint_narrative \\\n", + "0 NaN NaN \n", + "1 NaN NaN \n", + "2 Account status NaN \n", + "3 Repaying your loan NaN \n", + "4 Attempted to collect wrong amount NaN \n", + "\n", + " company_public_response company state zipcode tags \\\n", + "0 NaN U.S. Bancorp CA 95993 NaN \n", + "1 NaN Wells Fargo & Company CA 91104 NaN \n", + "2 NaN Wells Fargo & Company NY 11764 NaN \n", + "3 NaN Navient Solutions, Inc. MD 21402 NaN \n", + "4 NaN Resurgent Capital Services L.P. GA 30106 NaN \n", + "\n", + " consumer_consent_provided submitted_via date_sent_to_company \\\n", + "0 NaN Referral 2013-09-03 \n", + "1 NaN Referral 2013-09-03 \n", + "2 NaN Postal mail 2013-09-18 \n", + "3 NaN Email 2013-08-30 \n", + "4 NaN Web 2013-08-30 \n", + "\n", + " company_response_to_consumer timely_response consumer_disputed? \\\n", + "0 Closed with explanation Yes Yes \n", + "1 Closed with explanation Yes Yes \n", + "2 Closed with explanation Yes No \n", + "3 Closed with explanation Yes Yes \n", + "4 Closed with explanation Yes Yes \n", + "\n", + " complaint_id \n", + "0 511074 \n", + "1 511080 \n", + "2 510473 \n", + "3 510326 \n", + "4 511067 " + ] + }, + "execution_count": 2, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "df = pd.read_csv('consumer_complaints.csv', parse_dates=['date_received', 'date_sent_to_company'], low_memory=False)\n", + "df.head()" + ] + }, + { + "cell_type": "markdown", + "id": "88e39fd7", + "metadata": {}, + "source": [ + "Having a clean dataset allows for higher accuracy in visualizing, analyzing and in this case predicting on the dataset. In order to make the most efficient use of the above data, I am checking for the number of empty/NaN rows:" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "73bf52d6", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "date_received 0\n", + "product 0\n", + "sub_product 158322\n", + "issue 0\n", + "sub_issue 343335\n", + "consumer_complaint_narrative 489151\n", + "company_public_response 470833\n", + "company 0\n", + "state 4887\n", + "zipcode 4505\n", + "tags 477998\n", + "consumer_consent_provided 432499\n", + "submitted_via 0\n", + "date_sent_to_company 0\n", + "company_response_to_consumer 0\n", + "timely_response 0\n", + "consumer_disputed? 0\n", + "complaint_id 0\n", + "dtype: int64" + ] + }, + "execution_count": 3, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "#Check NaNs\n", + "df.isna().sum()" + ] + }, + { + "cell_type": "markdown", + "id": "e8db998a", + "metadata": {}, + "source": [ + "From the above output it is evident that the dataset if populated with many NaNs. However, removing all NaN values can cause overfitting in the prediction model and inaccuracies in the data analysis.\n", + "\n", + "So, keeping threshold of *column contains >= 10% NaNs* we will clean each column" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "45eebff3", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "sub_product 0.284774\n", + "sub_issue 0.617557\n", + "consumer_complaint_narrative 0.879836\n", + "company_public_response 0.846887\n", + "tags 0.859775\n", + "consumer_consent_provided 0.777936\n", + "dtype: float64" + ] + }, + "execution_count": 4, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "thresh = 0.10\n", + "col_nans = df.isna().mean(axis=0)\n", + "col_w_thresh = col_nans[col_nans >= thresh]\n", + "col_w_thresh" + ] + }, + { + "cell_type": "markdown", + "id": "dee366ce", + "metadata": {}, + "source": [ + "The above columns will now be dropped as they have too many NaNs to have any use in our dataset.\n", + "\n", + "*NOTE:* Along with these, the *complaint_id* column will aso be dropped for not being relevant in our analysis" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "252d522a", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
date_receivedproductissuecompanystatezipcodesubmitted_viadate_sent_to_companycompany_response_to_consumertimely_responseconsumer_disputed?
02013-08-30MortgageLoan modification,collection,foreclosureU.S. BancorpCA95993Referral2013-09-03111
12013-08-30MortgageLoan servicing, payments, escrow accountWells Fargo & CompanyCA91104Referral2013-09-03111
22013-08-30Credit reportingIncorrect information on credit reportWells Fargo & CompanyNY11764Postal mail2013-09-18110
32013-08-30Student loanRepaying your loanNavient Solutions, Inc.MD21402Email2013-08-30111
42013-08-30Debt collectionFalse statements or representationResurgent Capital Services L.P.GA30106Web2013-08-30111
\n", + "
" + ], + "text/plain": [ + " date_received product issue \\\n", + "0 2013-08-30 Mortgage Loan modification,collection,foreclosure \n", + "1 2013-08-30 Mortgage Loan servicing, payments, escrow account \n", + "2 2013-08-30 Credit reporting Incorrect information on credit report \n", + "3 2013-08-30 Student loan Repaying your loan \n", + "4 2013-08-30 Debt collection False statements or representation \n", + "\n", + " company state zipcode submitted_via \\\n", + "0 U.S. Bancorp CA 95993 Referral \n", + "1 Wells Fargo & Company CA 91104 Referral \n", + "2 Wells Fargo & Company NY 11764 Postal mail \n", + "3 Navient Solutions, Inc. MD 21402 Email \n", + "4 Resurgent Capital Services L.P. GA 30106 Web \n", + "\n", + " date_sent_to_company company_response_to_consumer timely_response \\\n", + "0 2013-09-03 1 1 \n", + "1 2013-09-03 1 1 \n", + "2 2013-09-18 1 1 \n", + "3 2013-08-30 1 1 \n", + "4 2013-08-30 1 1 \n", + "\n", + " consumer_disputed? \n", + "0 1 \n", + "1 1 \n", + "2 0 \n", + "3 1 \n", + "4 1 " + ] + }, + "execution_count": 5, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "drops_l = list(col_w_thresh.index) + ['complaint_id']\n", + "df = df.drop(drops_l, axis=1)\n", + "\n", + "#Tokenizing Yes and No to 1 and 0 respectively\n", + "df = df.replace(['Yes', 'No'], [1, 0])\n", + "\n", + "#Also tokenizing company response for easier NLP classification\n", + "df.loc[df['company_response_to_consumer'].str.contains(\"Closed\"), 'company_response_to_consumer'] = 1\n", + "df.loc[df['company_response_to_consumer'] != 1, 'company_response_to_consumer'] = 0\n", + "df.head()" + ] + }, + { + "cell_type": "markdown", + "id": "0e9dee79", + "metadata": {}, + "source": [ + "## Data Analysis and Visualization:\n", + "Since I will be mainly checking the count AND rate of disputed issues, the following functions will be used to get these metrics as a DataFrame:\n", + "\n", + "(*NOTE: Since the data being analysed is categorical I will mostly be using bar graphs*)" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "08b7d908", + "metadata": {}, + "outputs": [], + "source": [ + "def is_disputed(row, cat):\n", + " return df[(df['consumer_disputed?'] == 1) & (df[cat] == row[cat])]['consumer_disputed?'].sum()\n", + "\n", + "def get_count(cat, data=df):\n", + " df_by_cat = df[[cat, 'issue', 'consumer_disputed?']].groupby([cat]).count().reset_index()\n", + " \n", + " df_by_cat['consumer_disputed?'] = df_by_cat.apply(lambda r: is_disputed(r, cat), axis=1)\n", + " \n", + " df_by_cat['dispute_rate'] = df_by_cat.apply(lambda r: (r['consumer_disputed?'] / r['issue']) * 100, axis=1)\n", + " \n", + " return df_by_cat" + ] + }, + { + "cell_type": "markdown", + "id": "6693cced", + "metadata": {}, + "source": [ + "### Disputes according to State:" + ] + }, + { + "cell_type": "code", + "execution_count": 29, + "id": "d6ad5d11", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "State is CA with max disputes: 17615\n" + ] + } + ], + "source": [ + "#State with most disputes using above function:\n", + "disp_st = get_count('state')\n", + "st_c = disp_st['consumer_disputed?']\n", + "m_st = disp_st['state'][pd.Series.argmax(st_c)]\n", + "m_c = disp_st['consumer_disputed?'][pd.Series.argmax(st_c)]\n", + "print(\"State is\", m_st, \"with max disputes:\", m_c)" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "6f0729b6", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "Text(0.5, 1.0, 'Dispute counts per state')" + ] + }, + "execution_count": 8, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "fig1, ax1 = plt.subplots(figsize=(20, 9))\n", + "ax1.bar(disp_st['state'], disp_st['consumer_disputed?'])\n", + "ax1.set_xlabel(\"State\")\n", + "ax1.set_ylabel(\"Disputed issue counts\")\n", + "ax1.set_title(\"Dispute counts per state\")" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "6dfa6225", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "Text(0.5, 1.0, 'Dispute rates per state')" + ] + }, + "execution_count": 9, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "fig2, ax2 = plt.subplots(figsize=(20, 9))\n", + "ax2.bar(disp_st['state'], disp_st['dispute_rate'])\n", + "ax2.set_xlabel(\"State\")\n", + "ax2.set_ylabel(\"Disputed issue rate(disputed count / issue count)\")\n", + "ax2.set_title(\"Dispute rates per state\")" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "17ec819c", + "metadata": {}, + "outputs": [], + "source": [ + "#Getting count of issues per state with products\n", + "state_issue_count = df[['state', 'product', 'issue']].groupby(['state', 'product', 'issue']).size().reset_index()\n", + "state_issue_count.columns = ['state', 'product', 'issue', 'count']" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "id": "15c92542", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
stateproductissuecount
501CAMortgageLoan modification,collection,foreclosure19433
942FLMortgageLoan modification,collection,foreclosure11337
502CAMortgageLoan servicing, payments, escrow account8957
4028TXCredit reportingIncorrect information on credit report8712
485CACredit reportingIncorrect information on credit report8585
\n", + "
" + ], + "text/plain": [ + " state product issue count\n", + "501 CA Mortgage Loan modification,collection,foreclosure 19433\n", + "942 FL Mortgage Loan modification,collection,foreclosure 11337\n", + "502 CA Mortgage Loan servicing, payments, escrow account 8957\n", + "4028 TX Credit reporting Incorrect information on credit report 8712\n", + "485 CA Credit reporting Incorrect information on credit report 8585" + ] + }, + "execution_count": 11, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "#States with 5 highest dispute counts, issues and product\n", + "state_issue_count.sort_values(by = 'count', ascending = False).head()" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "id": "63fa2a3a", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
stateproductissuecount
3083NVPrepaid cardOverdraft, savings or rewards features1
347AZConsumer LoanPayment to acct not credited1
344AZConsumer LoanLender repossessed or sold the vehicle1
1AACredit cardIdentity theft / Fraud / Embezzlement1
0AABank account or serviceAccount opening, closing, or management1
\n", + "
" + ], + "text/plain": [ + " state product issue \\\n", + "3083 NV Prepaid card Overdraft, savings or rewards features \n", + "347 AZ Consumer Loan Payment to acct not credited \n", + "344 AZ Consumer Loan Lender repossessed or sold the vehicle \n", + "1 AA Credit card Identity theft / Fraud / Embezzlement \n", + "0 AA Bank account or service Account opening, closing, or management \n", + "\n", + " count \n", + "3083 1 \n", + "347 1 \n", + "344 1 \n", + "1 1 \n", + "0 1 " + ] + }, + "execution_count": 12, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "#States with 5 lowest dispute counts with related issues and products\n", + "state_issue_count.sort_values(by = 'count', ascending = False).tail()" + ] + }, + { + "cell_type": "markdown", + "id": "eb748a29", + "metadata": {}, + "source": [ + "### **Analysis**: \n", + "As the above console output and bar graphs show, states with higher populations such as CA, NY, TX etc tend to have more disputes over financial issues. Moreover, West Coast states have a higher tendency to dipute complaints. Finally, Mortgage modifications and Credit reporting issues have caused the most disputed among the states \n", + "\n", + "Laws and operations regarding financial products and the companies headquartered in these states are areas which can be further analysed for why they yield such high issue counts for the a certain range of products." + ] + }, + { + "cell_type": "markdown", + "id": "f45775c1", + "metadata": {}, + "source": [ + "### Disputes according to Product:" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "id": "1b73be97", + "metadata": {}, + "outputs": [], + "source": [ + "disp_prod = get_count('product')" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "id": "95b32810", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
productissuecount
75MortgageLoan modification,collection,foreclosure97191
59Credit reportingIncorrect information on credit report66718
76MortgageLoan servicing, payments, escrow account60375
62Debt collectionCont'd attempts collect debt not owed42285
0Bank account or serviceAccount opening, closing, or management26661
\n", + "
" + ], + "text/plain": [ + " product issue count\n", + "75 Mortgage Loan modification,collection,foreclosure 97191\n", + "59 Credit reporting Incorrect information on credit report 66718\n", + "76 Mortgage Loan servicing, payments, escrow account 60375\n", + "62 Debt collection Cont'd attempts collect debt not owed 42285\n", + "0 Bank account or service Account opening, closing, or management 26661" + ] + }, + "execution_count": 14, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "#Highest issues per product\n", + "product_issue_count = df.groupby(['product', 'issue']).size().reset_index()\n", + "product_issue_count.columns = ['product', 'issue', 'count']\n", + "product_issue_count.sort_values(by = 'count', ascending=False).head()" + ] + }, + { + "cell_type": "code", + "execution_count": 15, + "id": "cd258ff1", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "Text(0, 0.5, 'Product')" + ] + }, + "execution_count": 15, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "fig3, ax3 = plt.subplots(figsize=(16, 9))\n", + "ax3.barh(disp_prod['product'], disp_prod['consumer_disputed?']) \n", + "\n", + "ax3.set_title(\"Disputed issue counts per product\")\n", + "ax3.set_xlabel('Disputed issue counts')\n", + "ax3.set_ylabel('Product')" + ] + }, + { + "cell_type": "code", + "execution_count": 59, + "id": "40b666b1", + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "data": { + "text/plain": [ + "Text(0.5, 0, 'Disputed issue rate')" + ] + }, + "execution_count": 59, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "fig4, ax4 = plt.subplots(figsize=(16, 9))\n", + "ax4.barh(disp_prod['product'], disp_prod['dispute_rate']) \n", + "\n", + "ax4.set_title(\"Disputed issue rates per product\")\n", + "ax4.set_ylabel('Product')\n", + "ax4.set_xlabel('Disputed issue rate')" + ] + }, + { + "cell_type": "markdown", + "id": "61b80521", + "metadata": {}, + "source": [ + "### Analysis:\n", + "Using the same visualization it can be concluded that loan and credit related issues like *Mortgages, Credit Reporting, Student loans* etc are more likely to be disputed, meaning our company should be wary and prepared to efficiently face complicated customer service with such products" + ] + }, + { + "cell_type": "markdown", + "id": "32296070", + "metadata": {}, + "source": [ + "### Disputes according to submission medium:" + ] + }, + { + "cell_type": "code", + "execution_count": 17, + "id": "fd3459f9", + "metadata": {}, + "outputs": [], + "source": [ + "disp_med = get_count('submitted_via')" + ] + }, + { + "cell_type": "code", + "execution_count": 58, + "id": "abdbe069", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "Text(0.5, 0, 'Disputed issue count')" + ] + }, + "execution_count": 58, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "fig5, ax5 = plt.subplots(figsize=(16, 9))\n", + "ax5.bar(disp_med['submitted_via'], disp_med['consumer_disputed?'])\n", + "\n", + "ax5.set_title(\"Disputed issue counts per submission medium\")\n", + "ax5.set_xlabel('Medium')\n", + "ax5.set_xlabel('Disputed issue count')" + ] + }, + { + "cell_type": "code", + "execution_count": 19, + "id": "5c033a69", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "Text(0, 0.5, 'Disputed issue rate')" + ] + }, + "execution_count": 19, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "fig6, ax6 = plt.subplots(figsize=(16, 9))\n", + "ax6.bar(disp_med['submitted_via'], disp_med['dispute_rate'])\n", + "\n", + "ax6.set_title(\"Disputed issue rates per submission medium\")\n", + "ax6.set_xlabel('Medium')\n", + "ax6.set_ylabel('Disputed issue rate')" + ] + }, + { + "cell_type": "code", + "execution_count": 21, + "id": "52069de3", + "metadata": {}, + "outputs": [], + "source": [ + "df_by_med = df[['submitted_via', 'timely_response']].groupby(['submitted_via', 'timely_response']).size().reset_index()\n", + "df_by_med = df_by_med[df_by_med['timely_response'] == 1]\n", + "df_by_med.columns = ['submitted_via', 'was timely_response', 'count']" + ] + }, + { + "cell_type": "code", + "execution_count": 22, + "id": "06157da5", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "Text(0, 0.5, 'Timely response count')" + ] + }, + "execution_count": 22, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "fig7, ax7 = plt.subplots(figsize=(16, 9))\n", + "ax7.bar(df_by_med['submitted_via'], df_by_med['count'])\n", + "\n", + "ax7.set_title(\"Number of timely responses per submission medium\")\n", + "ax7.set_xlabel('Medium')\n", + "ax7.set_ylabel('Timely response count')" + ] + }, + { + "cell_type": "markdown", + "id": "86c3e653", + "metadata": {}, + "source": [ + "### Analysis:\n", + "Quicker, digital forms of communication such as *Email* and *Web* clearly have a higher dispute rate than *Postal mail* or *Referral*, however it should be noted that timeliness of the response does not hold much weight for digital forms. This could highlight the possible complacency/bureaucracy that these companies may have that causes them to give a clearly untimely response, and can also warrant further delving into the structure of customer service in such companies." + ] + }, + { + "cell_type": "markdown", + "id": "a6eae2cf", + "metadata": {}, + "source": [ + "## Sentiment Analysis/Prediction:" + ] + }, + { + "cell_type": "markdown", + "id": "05f5c8e9", + "metadata": {}, + "source": [ + "### Setting up data:\n", + "Since the key data is categorical in nature we can use *one-hot encoding* for it as there is no ordinal relationship in any of the data that is being used." + ] + }, + { + "cell_type": "code", + "execution_count": 30, + "id": "c380bd0b", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
date_receivedcompanyzipcodedate_sent_to_companycompany_response_to_consumertimely_responseconsumer_disputed?product_Bank account or serviceproduct_Consumer Loanproduct_Credit card...state_WAstate_WIstate_WVstate_WYsubmitted_via_Emailsubmitted_via_Faxsubmitted_via_Phonesubmitted_via_Postal mailsubmitted_via_Referralsubmitted_via_Web
0735110U.S. Bancorp95993735114111000...0000000010
1735110Wells Fargo & Company91104735114111000...0000000010
2735110Wells Fargo & Company11764735129110000...0000000100
3735110Navient Solutions, Inc.21402735110111000...0000100000
4735110Resurgent Capital Services L.P.30106735110111000...0000000001
\n", + "

5 rows × 181 columns

\n", + "
" + ], + "text/plain": [ + " date_received company zipcode \\\n", + "0 735110 U.S. Bancorp 95993 \n", + "1 735110 Wells Fargo & Company 91104 \n", + "2 735110 Wells Fargo & Company 11764 \n", + "3 735110 Navient Solutions, Inc. 21402 \n", + "4 735110 Resurgent Capital Services L.P. 30106 \n", + "\n", + " date_sent_to_company company_response_to_consumer timely_response \\\n", + "0 735114 1 1 \n", + "1 735114 1 1 \n", + "2 735129 1 1 \n", + "3 735110 1 1 \n", + "4 735110 1 1 \n", + "\n", + " consumer_disputed? product_Bank account or service product_Consumer Loan \\\n", + "0 1 0 0 \n", + "1 1 0 0 \n", + "2 0 0 0 \n", + "3 1 0 0 \n", + "4 1 0 0 \n", + "\n", + " product_Credit card ... state_WA state_WI state_WV state_WY \\\n", + "0 0 ... 0 0 0 0 \n", + "1 0 ... 0 0 0 0 \n", + "2 0 ... 0 0 0 0 \n", + "3 0 ... 0 0 0 0 \n", + "4 0 ... 0 0 0 0 \n", + "\n", + " submitted_via_Email submitted_via_Fax submitted_via_Phone \\\n", + "0 0 0 0 \n", + "1 0 0 0 \n", + "2 0 0 0 \n", + "3 1 0 0 \n", + "4 0 0 0 \n", + "\n", + " submitted_via_Postal mail submitted_via_Referral submitted_via_Web \n", + "0 0 1 0 \n", + "1 0 1 0 \n", + "2 1 0 0 \n", + "3 0 0 0 \n", + "4 0 0 1 \n", + "\n", + "[5 rows x 181 columns]" + ] + }, + "execution_count": 30, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "#Converting dates to numerical values instead of datetime\n", + "date_cols = ['date_received', 'date_sent_to_company']\n", + "for date_col in date_cols:\n", + " df[date_col] = df[date_col].apply(pd.Timestamp.toordinal)\n", + "\n", + "analysis_cols = ['product', 'issue', 'state', 'submitted_via']\n", + "encoded = df\n", + "for c in analysis_cols:\n", + " c_encoded = pd.get_dummies(encoded[c], prefix=c)\n", + " encoded = pd.concat((encoded, c_encoded), axis=1).drop(c, axis=1)\n", + "encoded.head() " + ] + }, + { + "cell_type": "code", + "execution_count": 31, + "id": "44062878", + "metadata": {}, + "outputs": [], + "source": [ + "X = encoded.drop(['company', 'zipcode', 'consumer_disputed?'], axis=1)\n", + "y = encoded['consumer_disputed?']\n", + "X_tr, X_t, y_tr, y_t = train_test_split(X, y, test_size=0.2, random_state=0)" + ] + }, + { + "cell_type": "markdown", + "id": "b0de514c", + "metadata": {}, + "source": [ + "**First using Logistic Regression to classify one-hot-encoded features:**" + ] + }, + { + "cell_type": "code", + "execution_count": 33, + "id": "25a11427", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Classifier accuracy is 0.7980160443197353\n" + ] + } + ], + "source": [ + "classifier1 = LogisticRegression()\n", + "classifier1.fit(X_tr, y_tr)\n", + "y_pred = classifier1.predict(X_t)\n", + "\n", + "print(\"Classifier accuracy is\", accuracy_score(y_t, y_pred))" + ] + }, + { + "cell_type": "code", + "execution_count": 40, + "id": "a3ce0650", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/Users/aryanvakharia/opt/anaconda3/lib/python3.9/site-packages/sklearn/utils/deprecation.py:87: FutureWarning: Function plot_confusion_matrix is deprecated; Function `plot_confusion_matrix` is deprecated in 1.0 and will be removed in 1.2. Use one of the class methods: ConfusionMatrixDisplay.from_predictions or ConfusionMatrixDisplay.from_estimator.\n", + " warnings.warn(msg, category=FutureWarning)\n" + ] + }, + { + "data": { + "text/plain": [ + "" + ] + }, + "execution_count": 40, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUUAAAEGCAYAAADyuIefAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAAgeklEQVR4nO3deZhcVZ3/8fenu7MRFgkJWxCIECN7EAhhD+APwurggARQQcMEAUGccWTRkWFmYPCZcRAUkAwgILKETREkglEI8LAkhB1EIkQS9oRlAoGQ7nx/f9xTnepOd3XdpCpdVf159VNPV52659Sp9MOXc8+593wVEZiZWaaptztgZlZLHBTNzIo4KJqZFXFQNDMr4qBoZlakpbc7UEz9m4KBNdUl68HnP7t1b3fBcvjbnFeYP3++VqYNDR0YfLK0vIMXLvl9RIxfmc9b1WorAg1sgZ3X7e1eWA4PTn2gt7tgOey28+4r38gnS2HseuUde8+8oSv/gatWbQVFM6t9oqEn3hwUzSw/rdQZeE1zUDSz/Bo3JjoomllOEjQ3blR0UDSz/Hz6bGZWpHFjooOimeUkoKlxo6KDopnl17gx0UHRzFaA5xTNzBLh1Wczsw4aNyY6KJpZXvLps5lZO68+m5l10rgx0UHRzFaAR4pmZolPn83MOmncmNjIW0WaWdVI5T16bEbfkfSspGckXS9poKQhku6R9GL6vXbR8WdKmi3pBUn7F5XvIOnp9N5FUvbhkgZIujGVPyJp05765KBoZvk1lfkoQdJw4FRgx4jYGmgGJgBnANMiYiQwLb1G0pbp/a2A8cAlkppTc5cCk4CR6VHICzMReDciNgcuAH5UzlczMytfuaPE8q5lbAEGSWoBVgNeA74IXJ3evxr4u/T8i8ANEbE4Il4GZgNjJG0ArBkRD0VEANd0qlNo62Zg38IosjsOimaWX5PKe8BQSTOLHpMKTUTEq8B/A68ArwPvR8TdwHoR8Xo65nWgkM1uODC3qBfzUtnw9LxzeYc6EdEKvA+sU+qreaHFzPIrfzg1PyJ27OqNNFf4RWAE8B5wk6SvlGirqxFelCgvVadbHimaWT6iUqfPXwBejoi3I2IJcCuwK/BmOiUm/X4rHT8P+HRR/Y3ITrfnpeedyzvUSafoawHvlOqUg6KZ5acyH6W9AoyVtFqa59sXeB64HTg2HXMs8Jv0/HZgQlpRHkG2oPJoOsVeKGlsaudrneoU2joc+GOad+yWT5/NLL8KXLwdEY9IuhmYBbQCjwOTgdWBKZImkgXOI9Lxz0qaAjyXjj85ItpScycCVwGDgLvSA+AK4JeSZpONECf01C8HRTPLr0K75ETE2cDZnYoXk40auzr+XODcLspnAlt3Uf4xKaiWy0HRzPKRUJkjxZLnqTXKQdHMcuvhUr92Dopm1ic08B6zDopmlk+2SU55UbGt50NqjoOimeWj8k+f65GDopnlJJqaGvcSZwdFM8utgQeKDopmlk92l1/jRkUHRTPLx3OKZmYdqYHzETgomlluHimamSVCNDubn5nZMh4pmpkVeKHFzKyjBo6J3nnbzPIpXKdYzqNkO9IoSU8UPf5P0mnO+2xmdacSQTEiXoiI0RExGtgBWATchvM+m1ldUXbvczmPHPYF/hoRf8N5n82s3uRI5tdt3udOJgDXp+fO+2xm9SPnvc/d5n1ub0/qDxwKnFnGR3fmvM9m1vsqMadY5ABgVkS8mV4777OZ1ZcmqaxHmY5i2akzOO+zmdUTSTRV6DY/SasB/w84oaj4fJz3uT6ccthxHDf+CCKCZ+f8hUk/PoNRn96Mn55yDgP6D6C1rZXTfnYOM//yFBP2PoTTDj++ve42I0axy7cO46mXnuc3/3E56w9Zl5bmZh58ZianXXwOS5cu5fgDJ3DCIcfQtnQpH368iJMv/AF/fuWvvfiN+567Z07nu5eeS9vSNo4bfwT/fOQJPVfqgyq1S05ELKLTwkdELKAX8z6rh5HkSpE0HrgQaAYuj4jzSx6/Zv9g53VLHdJrNlxnPab9+Dq2n3QgH3+ymGvP+glTZ9zHkeMO4ae3XcXdM6ez/0578Y9HHM/+3/tqh7pbbfpZbjr7Urb8evZ3XmO1wSxc9CEA1//gp9x6/1Ruuu/ODuUHjd2HSQcfzRd/cDy17KOpf+ntLlRMW1sb2xy/H3ee9wuGD12f3U/9e64+4wK22GTz3u5axey28+48NnPWSkW0ARutGRueMqasY+ecMe2xnhZaak3V5hTTRZUXk02ibgkclS6+rFstzS0M6j+Q5qZmBg0YxOsL3iII1lxtdQDWGrw6ry94a7l6Xx53MFPuvaP9dSHwtTS30K+lH4X/MRXKAQYPHEQV/39lXZjxwlNstsEmjNhgY/r3688Rex3EHQ/9obe7VZMqvNBSU6p5+jwGmB0RLwFIuoHsQsrnqviZVfPagjf5yc1X8Jdf3stHixczbdYDTJv1IPPefoPfnnsF//kPp9OkJvb+xyOXq3v4ngdyxDkndii7/dwr2PGz23L3zOnc+sDU9vITDjmGUw/7Ov379WP86V+r9teyIq8teJONhq3f/nr40PV59IUne7FHtatO411Zqrn63N2Flh1ImlS4sJMlS6vYnZXzqdXX5OBd9mWL4/bhM8fszuCBqzFhn0OZdPBRfO+y8xj51b343mXncel3zutQb6dR27Jo8Uc897cXO5Qf+v2JjDh6Nwb068+47ca2l1/221+x1Te+wA+u+C/OOOqkVfLdLNPVVFK9jnaqSWrskWI1g2JZF01GxOSI2DEidqRf7V4htM/2uzLnzXnMf/9dWtta+fWDdzN2i+055guH8esH7wbglvvvYsfPbtuh3hF7HcSUe+/sss3FSz7hjof/yCG7fGG596bcdyeH7Lp8uVXP8KHrM+/tN9pfvzr/DTYcUptz3L2rKrf51Yxq9rq7Cy3r0ty3XmPM50YzaMBAAPYevQsvzH2J1xe8xR7bZpPO40bvwuzX5rTXkcSX9jiAm+5bFhQHD1yN9YcMA6C5qZnxO+3FC3NfAmCzDTdpP+6AMeOY/eqytqz6dhy1DbNfm8OcN+byyZJPuOm+OzlobJeLoH1ejtv86k415xRnACPTRZavkl0fdHQVP6+qZrzwFLfd/3se+tmvaW1r5cm/Ps8Vd93Ak399jv/65vdpaW5h8SeL+daF/9JeZ/dtduLV+W8w541lswiDBw7i5n/9Of379aO5qZn7nniY/70zu271xEO/wt7b78qS1lbe++B9/uHHp6/y79mXtTS3cMFJP+SQ70+kbWkbx+53OFtuOrK3u1WT6vXUuBzVviTnQOAnZJfkXJmuMer++Bq+JMe61kiX5PQFlbgkZ9DGa8Wm/7RbWcf++bS76u6SnKpevB0RvwN+V83PMLNVr5FHir6jxcxya+CY6KBoZnmpbleWy+GgaGa5FK5TbFQOimaWWwPHRAdFM8vPI0Uzs2IOimZmSQU3ma1FjbuEZGZVUUhcVYkNISR9StLNkv4s6XlJu0gaIukeSS+m32sXHX9mSmz/gqT9i8p3kPR0eu+iQhrTlLrgxlT+iKRNe+qTg6KZ5VbBXXIuBKZGxOeA7YDngTOAaRExEpiWXpP2Y50AbEWW7P6StG8rwKXAJLK8LSPT+wATgXcjYnPgAuBHPXXIQdHMcqtEUJS0JrAnWR4VIuKTiHiPjgnsr6ZjYvsbImJxRLwMzAbGpIx/a0bEQykp1TWd6hTauhnYVz10zEHRzHLLsUvO0MJ+qekxqaiZzwBvA7+Q9LikyyUNBtZLGfpIvwsbInS3R+vw9LxzeYc6EdEKvE+nnDCdeaHFzPLJt4Hs/BIbQrQAnwdOiYhHJF1IOlXu7pO7KIsS5aXqdMsjRTPLRVCpTWbnAfMi4pH0+mayIPlmOiUm/X6r6Piu9midl553Lu9QR1ILsBZZqtNuOSiaWW6VmFOMiDeAuZJGpaJ9yXI4FSewP5aOie0npBXlEWQLKo+mU+yFksam+cKvdapTaOtw4I/Rw36JPn02s3wqu6v2KcCvJPUHXgK+TjZYmyJpIvAKKW9zRDwraQpZ4GwFTo6IttTOicBVwCDgrvSAbBHnl5Jmk40QJ/TUIQdFM8utUrf5RcQTQFdzjl3mgUgbVS+3WXVEzAS27qL8Y1JQLZeDopnlIuo3U185HBTNLDcHRTOzAtHQ9z47KJpZfh4pmpkt49NnM7NEQAOfPTsomlleXn02M2snQbOz+ZmZLdO4IdFB0cxWQFNfPH2W9FNKbLETEadWpUdmVtMK6QgaVamR4sxV1gszqyPqmyPFiLi6+LWkwRHxYfW7ZGY1TY09UuxxvjRl13qOLKEMkraTdEnVe2ZmNUlAi1TWox6Vs4j0E2B/YAFARDxJlmzGzPqoCmbzqzllraxHxNxORW1dHmhmDS+7o0VlPXpsS5qT8jU/IWlmKqv5vM9zJe0KhKT+kr5LOpU2s75JZT7KtHdEjC5KcFXzeZ+/CZxMlirwVWB0em1mfVJ5o8SVWKHu1bzPPV68HRHzgWPK/jpm1tBy3uY3tHBanEyOiMlFrwO4W1IAl6X3OuR9llSc9/nhorqF/M5LKDPvs6RC3uf53XW4x6Ao6TPAhcDY9AUeAr4TES/1VNfMGlOOUWCpvM8Au0XEaynw3SPpzyWOrZm8z9cBU4ANgA2Bm4Dry6hnZg2o3PnEcsJmRLyWfr8F3AaMoQ7yPisifhkRrelxLT1EWjNrbJWYU5Q0WNIahefAfsAz1GreZ0lD0tM/SToDuIEsGB4J3Fny25pZA6vYbX7rAbeldY8W4LqImCppBjWa9/kxOp6vn1D0XgD/3lPjZtZ4VKHb/NK6xHZdlC+gFvM+R8SIPA2ZWd/RXKd3q5SjrP0UJW0NbAkMLJRFxDXV6pSZ1a7CHS2NqpxLcs4GxpEFxd8BBwAPkF0gaWZ9UCMHxXJWnw8nO79/IyK+TjYHMKCqvTKzGlbeZhD1uiFEOafPH0XEUkmtktYku2boM1Xul5nVKOEcLTMlfQr4X7IV6Q+AR6vZKTOrYQ2+yWw59z6flJ7+XNJUshuvn6put8ysVglo6YspTiV9vtR7ETGrOl0ys1rXV0eKPy7xXgD7VLgvfG7ECK689rJKN2tmFSWa8uyWWGdKXby996rsiJnVj746UjQzW47U2NcpOiiaWW5N6oMLLWZmXVHldsmpSeXkfZakr0j6YXq9saQx1e+amdWqbKml50c9KqfXlwC7AEel1wuBi6vWIzOreVVOXNWryjl93jkiPi/pcYCIeFdS/yr3y8xqWCOvPpczUlyScqsGgKRhwNKq9srMapZy/JTVntQs6XFJd6TXQyTdI+nF9HvtomPPTIntX5C0f1H5DpKeTu9dVEhjmlIX3JjKH5G0aU/9KScoXkSWUGZdSeeSbRt2Xlnf1swaT0pxWs6jTN8Gni96fQYwLSJGAtPSayRtSZZOYCuyZPeXpAEbwKXAJLK8LSPT+wATgXcjYnPgAuBHPXWmx15HxK+A7wH/CbwO/F1E3NRTPTNrTNkuOeX99NiWtBFwEHB5UXFxAvur6ZjY/oaIWBwRLwOzgTEp49+aEfFQSkp1Tac6hbZuBvZVD+f+5WwyuzGwCPhtcVlEvNJTXTNrRLn2ShwqaWbR68kp4X3BT8gGXWsUla2XMvQREa+nnNCQJbZ/uOi4QtL7Jel55/JCnbmprVZJ7wPrAPO763A5Cy13siyB1UBgBPAC2RDWzPqgHEFxfkTs2E0bBwNvRcRjksaV87FdlEWJ8lJ1ulXO1mHbdOhVtnvOCd0cbmZ9QIU2hNgNOFTSgWQDrjUlXQu8KWmDNErcgGxjayhKbJ8Ukt7PS887lxfXmSepBViLLNVpt3JfXZm2DNspbz0zawyCiqQjiIgzI2KjiNiUbAHljxHxFTomsD+WjontJ6QV5RFkCyqPplPthZLGpvnCr3WqU2jr8PQZKzdSlPSPRS+bgM8Db/dUz8walERzde99Ph+YImki8Aopb3NEPCtpCvAc0AqcHBFtqc6JwFXAIOCu9AC4AvilpNlkI8QJPX14OXOKxROgrWRzjLeUUc/MGlCW4rSyQTEi7gXuTc8XkCXL6+q4c4FzuyifCWzdRfnHpKBarpJBMV0DtHpE/HOeRs2ssTXyHS2l0hG0pCXsbtMSmFnfVO7dKvWo1EjxUbL5wyck3Q7cBHxYeDMibq1y38ysJtXvZg/lKGdOcQiwgCwnS+GaoAAcFM36IEG1F1p6VamguG5aeX6G5S+QLLmkbWYNTKA+GhSbgdVZgSvCzayRlb8DTj0qFRRfj4h/W2U9MbO6kF2S0zeDYuN+azNbKX3ykhy6uXjSzKxC9z7XpG6DYkSUvGnazPomIZqamns+sE45xamZ5dYnR4pmZl2R+u6coplZl/rqJTlmZl3IlY6g7jgomllujTyn2Lj36phZVQjRpOayHiXbkQZKelTSk5KelXROKq/5vM9mZh1UIh0BsBjYJyK2A0YD4yWNpdbzPpuZdaYyf0qJzAfpZb/0CHo577ODopnllmOkOFTSzKLHpE7tNEt6gixj3z0R8Qid8j4DxXmf5xZVL+R3Hk6ZeZ+BQt7nbnmhxcxyEbkWWrrN+wyQEk+NlvQp4DZJy+VZ6fTRyzVRorxUnW55pGhm+agyCy3FIuI9ssRV40l5n7OPqljeZ6qW99nMrBILLZKGpREikgYBXwD+TK3nfTYzKyYqdkfLBsDVaQW5CZgSEXdIeogaz/tsZlakMomrIuIpYPsuyms377OZWVd877OZWRHf+2xmlhRu82tUDopmllsjbwjhoGhm+XiTWTOzZSp4SU5NclA0s9w8UjQzayeavdBiZpbx6bMB8OaCBfz7z3/Ogvffp0ni0L335sjx4/nZddfxwOOP06+lheHrrsv3J01ijcGD2+u9MX8+x5x+OhO/9CWOPuggAE7+j/9gwXvvMaB/fwAuOP10hqy1Fq/Pn895kyfz3sKFrDl4MGefeCLrrlNylyOrsLtnTue7l55L29I2jht/BP985Am93aWa5NPnFSDpSuBg4K2IKLUdUF1obmrilKOPZtSIEXz40Ud841/+hTHbbMNO22zDN488kpbmZi6+4Qau+e1vOXnCstsrL/rVrxi73XbLtXf2SSexxWc+06HsZ9ddxwG7786Be+7JzGef5dIpUzj7xBOr/t0s09bWxmkXn8Od5/2C4UPXZ/dT/56Dx+7LFpts3ttdqzE9byBbz6q5S85VLNsSvO4NXXttRo0YAcDgQYPYZMMNefudd9h5m21oac7mV7bebDPefmfZrkT3zZzJhsOGMWL48C7b7GzOq6+y41ZbAbDDllty/2OPVfhbWCkzXniKzTbYhBEbbEz/fv05Yq+DuOOhP/R2t2pShdIR1KSqBcWImE4P+5bVq9fffpsX//Y3ttpssw7ld0yfzthttwXgo48/5to77uAbX/pSl22cO3kyx551Fr+47TYKOxltvvHG/GnGDCALqIs+/pj3Fy6s4jexYq8teJONhq3f/nr40PV5dcGbvdij2pRtMlveTz3q9TnFtD35JID1Nlqvl3vTs0Uff8xZF17It7/yFQavtlp7+VW/+Q3NTU3sv9tuAFx+661MGD+e1QYOXK6Nfz3pJIYNGcKHH33EWRdeyNQHHuCAPfbgW0cfzf9cfTW/u/9+Ro8axbC116a5uXFX+WpNV9vs1etop6okmlSfAa8cvR4UI2IyMBlgi+1Gldz8sbe1trZy1oUXst+uuzJup53ay383fToPPv44Pz3zzPb/iJ6bPZs/PfooF99wAx8sWoQk+vfrx+H77cewIUOA7DR8v1135bmXXuKAPfZg2Npr85+nnQZkwffeGTNYvSjwWnUNH7o+895+o/31q/PfYMMh65ao0Xc18v8sej0o1ouI4LzLL2fTDTfkqAMPbC9/+MknufaOO7j4Bz9g4IAB7eWX/vCH7c8vv+UWVhs4kMP324/WtjY+WLSIT62xBq2trTz4+OPstHW2DlVYdW5qauKa22/n4L32WnVf0Nhx1DbMfm0Oc96Yy4brrMdN993JVaf/T293qyY18kKLg2KZnvrLX5j6wANs9ulPc+xZZwFwwpe/zAXXXMOS1lZOO/98ALbafHO+941vdNvOkiVL+M6PfkRrWxtLly5lx6224tC99wZg1vPP8/Mbb0QSo0eN4p+OO67q38uWaWlu4YKTfsgh359I29I2jt3vcLbcdGRvd6vmVOo6RUmfJktHuj6wFJgcERdKGgLcCGwKzAG+HBHvpjpnkuVybgNOjYjfp/IdWLbz9u+Ab0dESBqQPmMHYAFwZETMKdmvHtIVrDBJ1wPjgKHAm8DZEXFFqTpbbDcqrrz7sqr0x6pj+6FjersLlsNuO+/OYzNnrVRE23L05+KaP1xZ1rE7Ddvtse6y+aWkVBtExCxJawCPkeVrPg54JyLOl3QGsHZEnC5pS+B6YAywIfAH4LMR0SbpUeDbwMNkQfGiiLhL0knAthHxTUkTgMMi4shSfa7m6vNREbFBRPSLiI16CohmVi+6Tnzf1U8pEfF6RMxKzxcCz5PlaS5OYH81HRPb3xARiyPiZWA2MCYF1zUj4qGUlOqaTnUKbd0M7KseJkR9+mxmueVYfR4qaWbR68lpcbUDSZuS5Wt5BFgvZegjIl6XVFjtGk42EiwoJL1fkp53Li/UmZvaapX0PrAOML+7DjsomlluOeYU53d3+tzelrQ6cAtwWkT8X4mBXHeJ7UslvC/1Xpca92IjM6sKUbk7WiT1IwuIv4qIW1Pxm+mUuDDv+FYqb09snxSS3s9LzzuXd6gjqQVYix5uKnFQNLOcKjOnmOb2rgCej4jia5+KE9gfS8fE9hMkDZA0AhgJPJpOtRdKGpva/FqnOoW2Dgf+GD2sLvv02cxyq9B1irsBXwWelvREKjsLOB+YImki8Aopb3NEPCtpCvAc0AqcHBFtqd6JLLsk5670gCzo/lLSbLIR4rLdWrrhoGhm+SjXQku3IuIBup7zA9i3mzrnAud2UT4TWG43roj4mBRUy+WgaGa5FOYUG5WDopnl1Nj7KToomlluDopmZkV8+mxmVsQjRTOzRHiTWTOzTjxSNDPLyHOKZmYdeE7RzKyIg6KZWSLqN6dzORwUzSy3es3pXA4HRTPLzSNFM7MinlM0M0safU6xcScGzKxqKrHzNoCkKyW9JemZorIhku6R9GL6vXbRe2dKmi3pBUn7F5XvIOnp9N5FhYx9aZfuG1P5IylBVkkOimaWW6WCItlu2eM7lZ0BTIuIkcC09JqU93kCsFWqc4mk5lTnUmASWYqCkUVtTgTejYjNgQuAH/XUIQdFM8utUomrImI6yyeS6tW8zw6KZrYCVOYjy/tc9JhURuMd8j4DxXmf5xYdV8jvPJwy8z4DhbzP3fJCi5nllmOZpce8zyv5sc77bGa9rdxR4gqvUDvvs5nVD6lyc4rdcN5nM6svlbp4W9L1wDiyucd5wNk477OZ1ZtKBcWIOKqbt3ot77NPn83MinikaGa5+TY/M7M+wiNFM8up7Fv46pKDopnlkl2B6KBoZtaukecUHRTNbAU4KJqZtWvckOigaGYrpHHDooOimeXU2OkIHBTNLBevPpuZLcdB0cysXeOGRAdFM1sBnlM0M2u3Urtq1zwHRTPLrZEXWrxLjpnlU8F0BJLGp8T2syWdsQp63yMHRTPrFSmR/cXAAcCWwFEp4X2vclA0s1wK1ymW89ODMcDsiHgpIj4BbiBLXt+r1ENiq1VK0tvA33q7H1UwFJjf252wXBr1b7ZJRAxbmQYkTSX79ynHQODjoteTI2JyaudwYHxEHJ9efxXYOSK+tTL9W1k1tdCysn+sWiVpZgUTgtsq4L9Z9yJifIWayp2oflXw6bOZ9Zbuktv3KgdFM+stM4CRkkZI6k+Wk/n2Xu5TbZ0+N7DJvd0By81/syqLiFZJ3wJ+DzQDV0bEs73crdpaaDEz620+fTYzK+KgaGZWxEGximrxFiYrTdKVkt6S9Exv98V6h4NildTqLUzWo6uASl2HZ3XIQbF6avIWJistIqYD7/R2P6z3OChWz3BgbtHreanMzGqYg2L11OQtTGZWmoNi9dTkLUxmVpqDYvXU5C1MZlaag2KVREQrULiF6XlgSi3cwmSlSboeeAgYJWmepIm93SdbtXybn5lZEY8UzcyKOCiamRVxUDQzK+KgaGZWxEHRzKyIg2IdkdQm6QlJz0i6SdJqK9HWVSmbGpIuL7VZhaRxknZdgc+YI2m5rG/dlXc65oOcn/Wvkr6bt49mnTko1pePImJ0RGwNfAJ8s/jNtDNPbhFxfEQ8V+KQcUDuoGhWjxwU69f9wOZpFPcnSdcBT0tqlvRfkmZIekrSCQDK/EzSc5LuBNYtNCTpXkk7pufjJc2S9KSkaZI2JQu+30mj1D0kDZN0S/qMGZJ2S3XXkXS3pMclXUbX9393IOnXkh6T9KykSZ3e+3HqyzRJw1LZZpKmpjr3S/pcRf41zRInrqpDklrI9mmcmorGAFtHxMspsLwfETtJGgA8KOluYHtgFLANsB7wHHBlp3aHAf8L7JnaGhIR70j6OfBBRPx3Ou464IKIeEDSxmR37WwBnA08EBH/JukgoEOQ68Y30mcMAmZIuiUiFgCDgVkR8U+Sfpja/hZZQqlvRsSLknYGLgH2WYF/RrMuOSjWl0GSnkjP7weuIDutfTQiXk7l+wHbFuYLgbWAkcCewPUR0Qa8JumPXbQ/FpheaCsiuttX8AvAllL7QHBNSWukz/hSqnunpHfL+E6nSjosPf906usCYClwYyq/FrhV0urp+95U9NkDyvgMs7I5KNaXjyJidHFBCg4fFhcBp0TE7zsddyA9b12mMo6BbNpll4j4qIu+lH3fqKRxZAF2l4hYJOleYGA3h0f63Pc6/xuYVZLnFBvP74ETJfUDkPRZSYOB6cCENOe4AbB3F3UfAvaSNCLVHZLKFwJrFB13N9mpLOm40enpdOCYVHYAsHYPfV0LeDcFxM+RjVQLmoDCaPdostPy/wNelnRE+gxJ2q6HzzDLxUGx8VxONl84KyVfuozsjOA24EXgaeBS4L7OFSPibbJ5wFslPcmy09ffAocVFlqAU4Ed00LOcyxbBT8H2FPSLLLT+Fd66OtUoEXSU8C/Aw8XvfchsJWkx8jmDP8tlR8DTEz9exaneLAK8y45ZmZFPFI0MyvioGhmVsRB0cysiIOimVkRB0UzsyIOimZmRRwUzcyK/H/msrqeshyIlgAAAABJRU5ErkJggg==\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "plot_confusion_matrix(classifier1, X_t, y_t, cmap=plt.cm.Greens)" + ] + }, + { + "cell_type": "markdown", + "id": "141b8875", + "metadata": {}, + "source": [ + "### Analysis:\n", + "As we can see the classifier needs a great deal of work to accurately predict a disputed complaint as the above confusion matrix shows that a true disputed complaint is not recognized by the classifier." + ] + }, + { + "cell_type": "markdown", + "id": "730fb665", + "metadata": {}, + "source": [ + "**Using Random Forest Classifier to better prediction:**" + ] + }, + { + "cell_type": "code", + "execution_count": 52, + "id": "fc0a33cc", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Classifier accuracy is 0.7119846751564861\n" + ] + } + ], + "source": [ + "classifier2 = DecisionTreeClassifier(random_state=42)\n", + "classifier2.fit(X_tr, y_tr)\n", + "y_pred = classifier2.predict(X_t)\n", + "\n", + "print(\"Classifier accuracy is\", accuracy_score(y_t, y_pred))" + ] + }, + { + "cell_type": "code", + "execution_count": 53, + "id": "0435bfbf", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/Users/aryanvakharia/opt/anaconda3/lib/python3.9/site-packages/sklearn/utils/deprecation.py:87: FutureWarning: Function plot_confusion_matrix is deprecated; Function `plot_confusion_matrix` is deprecated in 1.0 and will be removed in 1.2. Use one of the class methods: ConfusionMatrixDisplay.from_predictions or ConfusionMatrixDisplay.from_estimator.\n", + " warnings.warn(msg, category=FutureWarning)\n" + ] + }, + { + "data": { + "text/plain": [ + "" + ] + }, + "execution_count": 53, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "plot_confusion_matrix(classifier2, X_t, y_t, cmap=plt.cm.Reds)" + ] + }, + { + "cell_type": "code", + "execution_count": 57, + "id": "2862af72", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + " precision recall f1-score support\n", + "\n", + " 0 0.81 0.84 0.82 88733\n", + " 1 0.26 0.23 0.24 22459\n", + "\n", + " accuracy 0.71 111192\n", + " macro avg 0.53 0.53 0.53 111192\n", + "weighted avg 0.70 0.71 0.70 111192\n", + "\n" + ] + } + ], + "source": [ + "print(classification_report(y_t, y_pred))" + ] + }, + { + "cell_type": "markdown", + "id": "52f413a8", + "metadata": {}, + "source": [ + "### Analysis:\n", + "The decision tree model has a lower accuracy but because it deals with the categorical data better it manages to correctly predict some of the disputed issues." + ] + }, + { + "cell_type": "markdown", + "id": "c04cfc62", + "metadata": {}, + "source": [ + "## Conclusion:" + ] + }, + { + "cell_type": "markdown", + "id": "ce1b2eb9", + "metadata": {}, + "source": [ + "Since both models are very quick and dirty, detailed parsing of the data using the *nltk* library would be ideal in getting a higher recall and higher number of true positive recognition in the future.\n", + "\n", + "However, the visualizations and insights provided by the dataset do inform of key areas for our company to be wary and cautious of, and with constant improvement of data preparation and more robust training models there lies a higher scope of predicting disputed issues and even company responses." + ] + } + ], + "metadata": { + "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": "3.9.12" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/main.tex b/main.tex new file mode 100644 index 0000000..8144cc6 --- /dev/null +++ b/main.tex @@ -0,0 +1,34 @@ +\documentclass{report} +\usepackage[utf8]{inputenc} + +\title{Future Hypotheses} +\author{Aryan Vakharia} +\date{September 30th, 2022} + +\setcounter{secnumdepth}{3} + +\begin{document} + +\maketitle + +\chapter{Current Project} + +\section{Optimizing the ML Model} +A Logistic Regression model and Decision Tree was used, with both identifying true positives(actual disputes) with low accuracy. +\subsection{If a package such as XGBoost is optimized for the data set, then the model bias will be less than the Logistic Regression model.} +\subsection{With deeper analysis on the data set, then the ML model will have greater accuracy with respect to consumer disputes.} + +\section{Data Insights} +For the following features–product, state, submission method, and issue: +\subsection{The more commonly used the financial product is within the company, the more likely it is to be disputed} +\subsection{If a consumer is located on the West Coast of the United States, then they are more likely to dispute.} +\subsection{If a consumer uses an digital form of communication, then they are more likely to dispute their resolution.} + + +\chapter{Additional Hypotheses} + +\section{Data Exploration} +\subsection{Interactive map with highlights of different zip codes.} +\subsection{Robust, accurate sentiment analysis on product, sub-product, issue, and sub-issue and even narratives.} + +\end{document} \ No newline at end of file diff --git a/report.pdf b/report.pdf new file mode 100644 index 0000000..40ea274 Binary files /dev/null and b/report.pdf differ