This project analyses apartment prices in Warsaw using spatial data analysis and machine learning methods. The dataset was cleaned, transformed into a spatial object, and enriched with OpenStreetMap accessibility variables such as distance to subway stations, tram infrastructure, and parks. The analysis includes exploratory maps, spatial clustering, a Spatial Random Forest model, variable-importance interpretation, residual analysis, and kriging interpolation. The results show that apartment prices in Warsaw are strongly influenced by location, accessibility, building characteristics, and spatial-market segmentation.

The goal of this project is to perform a comprehensive Market Basket Analysis (MBA) on an online retail dataset. While standard association rules (Apriori algorithm) are excellent for finding frequent patterns, they often miss high-value relationships that occur less frequently.

A comprehensive unsupervised learning project analyzing credit card customer behavior. By integrating PCA for dimensionality reduction, t-SNE for manifold visualization, and a hybrid clustering approach (K-Means + DBSCAN), this report identifies four distinct behavioral personas: The VIP Shoppers, The Active Transactors, The Revolving Borrowers, and The Inactive users. Includes technical justifications for median imputation, feature scaling, and outlier detection to drive strategic financial recommendations.

This project aims to apply various clustering techniques to the Global Crocodile Species Dataset.