This project applies a pure dimension reduction approach using t-SNE to the Wine Quality dataset, which consists of physicochemical measurements of red wine samples and corresponding quality ratings assigned by expert tasters. The objective of the analysis is to explore whether wines with similar quality scores exhibit similar physicochemical characteristics when projected into a two-dimensional space

USL Association Rules Project

This project applies unsupervised learning to identify distinct and interpretable player importance profiles in the Spanish ACB league in the 2024-25 season, using engineered performance features that go beyond traditional box-score evaluations. By combining clustering and principal component analysis, the study examines whether dimensionality reduction clarifies and stabilizes the structure of player importance without altering the underlying groupings.

rastercontourntif_3rPlot

USL Clustering and Dimension Reduction Project

This report explores the use of association rule mining techniques: Apriori, FP-Growth style pattern mining, and ECLAT to identify co-occurring health risk factors across countries. Using 2015 data from the WHO Global Health Observatory, the analysis treats each country as a transaction and examines patterns involving BMI, cholesterol, depression, and alcohol consumption.

rasterpolygonotif4_shp <- st_read("rasterxpolygonoRasterT_tif4.shp") plot(st_geometry(rasterpolygonotif4_shp), axes=TRUE) rasterpolygonotif4_shp <- st_transform(rasterpolygonotif4_shp, crs = 4324)