Credit Card Clustering with Machine Learning

📌 Objective

• Understand customer behavior from credit card usage.
• Segment customers into clusters with similar patterns.
• Help financial institutions create targeted marketing strategies.

📊 Dataset

• Source: Aman Kharwal’s GitHub Dataset
• Contains features like:
  • BALANCE: Average balance
  • PURCHASES: Total purchases
  • CREDIT_LIMIT: Assigned credit limit
  • PAYMENTS: Amount paid
  • TENURE: Months as a customer
  • ONEOFF_PURCHASES, INSTALLMENTS_PURCHASES, etc.

🧹 Data Preprocessing

• Checked for null values and handled them
• Dropped irrelevant columns (e.g., CUST_ID)
• Scaled data using StandardScaler

🧠 Clustering Algorithm

• Used KMeans algorithm
• Determined optimal number of clusters using:
  • Elbow Method
  • Silhouette Score

📉 Dimensionality Reduction

• Applied PCA for visualizing clusters in 2D space

📈 Results & Analysis

• Clusters represent different types of customers:
  • High spenders
  • Low activity users
  • Customers using mostly installments
• Visualized clusters using matplotlib and seaborn

📦 Libraries Used

• pandas
• numpy
• matplotlib, seaborn
• scikit-learn

🔍 Future Improvements

• Try alternative clustering algorithms like DBSCAN, GMM
• Add deeper feature engineering
• Include time-based features for trend analysis

💻 How to Run

1. Clone the repo:
   git clone https://github.com/handecrkc/credit-card-clustering.git
2. Install requirements:
   pip install -r requirements.txt
3. Run the notebook: Open credit_card_clustering.ipynb in Jupyter Notebook or VS Code

🧑‍💻 Author

• Hande Çarkcı
• GitHub: github.com/handecrkc

💳 Credit Card Clustering – Streamlit App

🎯 Proje Amacı

• Kredi kartı kullanıcılarını benzer davranış gruplarına ayırmak
• Finansal kurumlara hedefli pazarlama stratejileri sağlamak
• Kullanıcıya ait segmenti gerçek zamanlı olarak tahmin etmek

🧠 Kullanılan Yöntem

• KMeans Clustering
• StandardScaler ile veri ölçekleme
• Streamlit ile web uygulaması

🗃️ Kullanılan Veri Seti

• Kaynak: CC GENERAL.csv
• Sütunlar: BALANCE, PURCHASES, CREDIT_LIMIT, PAYMENTS, TENURE, vb.

🚀 Uygulamayı Çalıştırmak