PROJECT DESCRIPTION
This project aims to cluster customers based on their use of credit cards. Such analysis could increase bank's revenue by finding suitable marketing campaigns which suited the profiles of each customer. To achieve such purpose, this project applies some unsupervised Machine Learning algorithms, such as K-Means Clustering and Principal Component Analysis (PCA). Here are some questions addressed in this project:
- How can customers be segmented to determine the best marketing strategy aimed at attracting customers and promoting specific products?
- As we have found out the segmentation, which customer should be targeted to increase revenue?
- After finding the prospective customer, what marketing strategy should be implemented to increase revenue?
The dataset being used in this project can be retrieved from Kaggle: https://www.kaggle.com/datasets/arjunbhasin2013/ccdata with the name 'Credit Card Dataset for Clustering'. It summarizes the usage behavior of about 9000 active credit card holders during the last 6 months. The file is at a customer level with 18 behavioral variables.
PROJECT DETAILS
- Exploratory Data Analysis (EDA)
- Treating Missing Value
- Data Scaling
- Outlier Dropping
- Dropping PURCHASES column
- Principal Component Analysis (PCA)
- K-Means Clustering
- Customer Segmentation