The goal of this project is to understand the different types of users through their behavior over time, in order to detect the most likely to make purchases.

• Cleaning and Exploratory analysis
• Feature engineering and clustering
• Classification

• Latent Dirichlet Allocation (LDA)
• Density-Based Spatial Clustering of Applications with Noise (DBSCAN)
• K-means
• Random Forests Clasifier (RFC)
• Gradient Boosting Clasifier (GBC)
• eXtreme Gradiengt Boosting Clasifier (XGBC)

Excellent accuracy score for the final classification model (Gradient Boosting (target variable:Customer_Category_km13 )) : 0.987