This report performs market basket analysis on the Online Retail dataset using: *Transaction construction from invoices *Apriori association rules (with redundancy and significance filtering) *Targeted rules with a fixed RHS item *Cross-table exploration for top items *Jaccard-based similarity and hierarchical clustering (transactions and items) *Category aggregation and rule mining *CBA-style classification workflow (prepared features) This analysis was prepared as part of the course Unsupervised Machine Learning at the University of Warsaw.

This report explores Principal Component Analysis (PCA), a method introduced during the course, as a simple and transparent approach to image compression. The idea is to represent an image with fewer degrees of freedom while keeping the reconstruction visually acceptable. Two settings are considered: - PCA applied to a grayscale version of the image, - PCA applied separately to each RGB channel. The report focuses on the practical trade-off between compression strength and visual quality, as observed both numerically and visually. Reconstruction quality is evaluated using MSE and PSNR, and the results are compared against standard JPEG compression at different quality settings.

The aim of this analysis was to explore whether Atlantic hurricanes can be meaningfully grouped based on selected characteristics describing their intensity, duration, and basic spatial properties, such as the location of origin and first landfall. Rather than attempting to classify hurricanes according to predefined categories, the analysis follows an exploratory approach. The objective was to examine whether recurring patterns emerge directly from historical data and whether these patterns remain stable across different clustering methods. This analysis was prepared as part of the course Unsupervised Machine Learning at the University of Warsaw.