Analysis of a Bank's Data and Segmentation of its Customers into Groups

Tools Used: Python and Tableau

PROJECT OVERIVIEW

The dataset I used for this project is a public bank transaction dataset on Kaggle, containing over 1 million transactions by a bank’s customers from August to October 2016, and can be found here. The bank in question is an unnamed Indian bank and all customer Personal Identifiers (PIDs) had been taken out from the dataset. I performed Exploratory Data Analysis (EDA) of the bank's data and segmentation of its customers into various groups based on shared characteristics. To perform customer segmentation, I used scikit-learn's K-Means clustering algorithm to build a model to cluster the bank's customers into different segments.  

Please see my Google Colab notebook here to see all the steps and python code I used in this project.

I also wrote an article outlining the steps I took in my analysis. This article was published in a Medium publication, Level Up Coding, and can be found here.  

Find below the Tableau dashboard I created to visualize and share the key takeaways from my analysis. Note that the dashboard is best viewed on a PC or tablet.