Customer segmentation is the practice of separating customers into groups that reflect similarities among customers in each cluster. I will divide customers into segments to optimize the significance of each customer to the business.
The Dataset was collected from Kaggle.
-
'Avg Cred Limit' and Total visits online had outliers and were removed using IQR.
-
Avg_Credit_Limit is skewed
-
Each customer has an average of 5 credit cards, visits bank twice a both online and in person, also makes 2 calls to the bank
-
Average Credit Limit is proportional to the number of credit cards
-
Customers with higher credit limit on credit cards make lesser calls
-
Customers who makes calls to the bank prefer visiting the bank in person rather than online.
-
Loyal Customers have a credit limit higher than around 60k
Finding the number of clusters and data labels for a given value of K is essential to the K-Means algorithm. The ideal number of clusters for K-means clustering is chosen using the elbow approach. The value of the cost function generated by various K values is plotted using the elbow approach. We should cease further clustering the data at the elbow, or the value of K at which improvement in distortion drops the most. For this problem, K should be set to 3.