This repository contains the final project for the Computational Tools for Data Science course. The goal was to analyze gym user behavior using a dataset from Kaggle (link) to provide actionable insights for gym owners.
- Frequent Itemset Mining: Uncovered peak attendance times and common check-in behaviors.
- Clustering: Identified user profiles using DBSCAN and evaluated cluster quality with Silhouette and Davies-Bouldin scores.
- Classification: Predicted workout preferences based on demographic features like age and gender.
- Anomaly Detection: Flagged irregular attendance patterns for retention strategy insights.
data/: Processed datasets used in the analysis.data/_raw: Raw datasets.data/plots: Visualizations and figures generated during the analysis.
scripts/: Jupyter notebooks.
The dataset includes information on gym check-ins, user metadata, and workout sessions.