Context:
AllLife Bank aims to boost its credit card customer base with personalized campaigns and improved support services based on market research insights.

Objective:
Segment customers using machine learning techniques based on spending patterns and interactions.
Provide recommendations for targeted marketing and service enhancements.
Data:
Customer Key: ID
Avg_Credit_Limit: Average credit limit
Total_Credit_Cards: Number of credit cards
Total_Visits_Bank: Yearly bank visits
Total_Visits_Online: Yearly online visits
Total_Calls_Made: Yearly calls to customer service
Machine Learning Techniques:
Data Preprocessing:

Handle duplicates and missing values.
Scale features using StandardScaler to normalize data.
Clustering Algorithms:

K-Means Clustering:

Determined optimal clusters using the Elbow Method and Silhouette Score.
Finalized 3 clusters based on silhouette coefficient.
Hierarchical Clustering:

Used different distance metrics and linkage methods.
Chose Euclidean distance and average linkage based on cophenetic correlation.
Key Findings:
Cluster 0 (Moderate Users):

Avg Credit Limit: $33,782
Credit Cards: 5.5
Bank Visits: 3.5/year
Online Visits: 1/year
Calls: 2/year
Proportion: ~59%
Cluster 1 (Premium Users):

Avg Credit Limit: $141,040
Credit Cards: 8.7
Bank Visits: 0.6/year
Online Visits: 10.9/year
Calls: 1.1/year
Proportion: ~8%
Cluster 2 (Frequent Callers):

Avg Credit Limit: $12,174
Credit Cards: 2.4
Bank Visits: 0.9/year
Online Visits: 3.6/year
Calls: 6.9/year
Proportion: ~34%
Recommendations:
Marketing:

Cluster 1: Upsell premium services, promote digital-first products.
Cluster 0: Incentivize online banking.
Cluster 2: Promote online banking benefits.
Service Delivery:

Enhance online banking experience.
Automate phone banking with IVR and chatbots.
Retention:

Provide personalized support for Cluster 1.
Improve in-person and phone support efficiency.
Cross-Sell & Up-Sell:

Cluster 0: Cross-sell additional financial products.
Cluster 2: Offer credit card upgrades and increased limits.
Conclusion:
Using clustering algorithms like K-Means and Hierarchical Clustering enables AllLife Bank to identify distinct customer segments. This approach supports tailored marketing strategies and service improvements, enhancing customer satisfaction and driving growth.