Okino5040/pension-contributor-segmentation-ml

Unsupervised machine learning project for segmenting pension contributors to improve customer retention in Nigeria. # pension-contributor-segmentation-ml Unsupervised machine learning project for segmenting pension contributors to improve customer retention in Nigeria. ## Objectives - Segment pension contributors into meaningful groups - Identify behavioural patterns in contributors - Improve retention strategies for pension administrators - Apply unsupervised machine learning techniques for data analysis - ## Technologies Used - Python - Pandas & NumPy - Scikit-learn - Matplotlib & Seaborn - Jupyter Notebook - ## Machine Learning Techniques - K-Means Clustering - Hierarchical Clustering - DBSCAN - Feature Engineering - Data Normalization (StandardScaler) - ## Dataset The dataset used in this project is a publicly available customer churn dataset from Kaggle, adapted to represent pension contributor behaviour. Features include: - Age - Income - Account Balance - Tenure - Activity Level - ## Installation ### 1. Clone the repository git clone github.com ### 2. Move into directory cd pension-contributor-segmentation-ml ### 3. Install dependencies pip install -r requirements.txt ## How to Run 1. Open Jupyter Notebook 2. Run notebooks in this order: - 01_data_exploration.ipynb - 02_preprocessing_feature_engineering.ipynb - 03_kmeans_clustering.ipynb - 04_hierarchical_clustering.ipynb - 05_dbscan_clustering.ipynb - 06_model_evaluation_comparison.ipynb - ## Results The clustering models successfully identified distinct contributor groups based on behaviour and financial attributes. K-Means provided the most stable clusters, while DBSCAN detected outliers representing inactive contributors. ## Project Structure data/ notebooks/ src/ models/ outputs/ reports/ README.md requirements.txt main.py ## Future Work - Integration with real pension contributor datasets - Deployment as a dashboard - Hybrid clustering with deep learning - Integration with churn prediction models - ## Author Developed by: Muhammed Kabiru Okino Departme …