Credit Card Clustering Project: Project Overview

This project uses k-mean clustering to clusters credit card users into different groups based on their balance and various spending habits.

• Create a tool that clusters credit card users into different groups to help credit card providers identify the right audience when creating marketing campaigns
• Optimized k-mean cluster analysis using the Elbow Method to reach the best model

Code and Resources Used

• Python Version: 3.7
• Packages: pandas, numpy, matplotlib, seaborn, sklearn
• Kaggle data: https://www.kaggle.com/arjunbhasin2013/ccdata

About the data

The data contain the follwing columns

• CUSTID : Identification of Credit Card holder (Categorical)
• BALANCE : Total amount of money that you owe to your credit card company
• BALANCEFREQUENCY : How frequently the Balance is updated, score between 0 and 1 (1 = frequently updated, 0 = not frequently updated)
• PURCHASES : Amount of purchases made from account
• ONEOFFPURCHASES : Maximum purchase amount done in one-go
• INSTALLMENTSPURCHASES : Amount of purchase done in installment
• CASHADVANCE : Cash in advance given by the user
• PURCHASESFREQUENCY : How frequently the Purchases are being made, score between 0 and 1 (1 = frequently purchased, 0 = not frequently purchased)
• ONEOFFPURCHASESFREQUENCY : How frequently Purchases are happening in one-go (1 = frequently purchased, 0 = not frequently purchased)
• PURCHASESINSTALLMENTSFREQUENCY : How frequently purchases in installments are being done (1 = frequently done, 0 = not frequently done)
• CASHADVANCEFREQUENCY : How frequently the cash in advance being paid
• CASHADVANCETRX : Number of Transactions made with "Cash in Advanced"
• PURCHASESTRX : Numbe of purchase transactions made
• CREDITLIMIT : Limit of Credit Card for user
• PAYMENTS : Amount of Payment done by user
• MINIMUM_PAYMENTS : Minimum amount of payments made by user
• PRCFULLPAYMENT : Percent of full payment paid by user
• TENURE : Tenure of credit card service for user

EDA

I looked at the distributions of the data. Below are a few highlights from the pivot tables.

Model Building

First I delete the rows with missing data from the dataset. These rows only make up a very small amount of data compare to the whole dataset and thus could be deleted safely. I then scaled the data using Standard Scaler from sklearn library. This is done so that the variables are bing processed with the same weight. This is a necessary step with working with distance based algorithm such as k-mean clustering. I then optimized the model using the elbow method the find the optimal number of clusters

Model Performance

I found that the optimal number of clusters in this dataset is 4. Below is the visual representation of how the data would clustered for each variables.