Authors: Hubert Wojewoda, Kacper Sokołowski Road safety is a paramount concern for both individuals and authorities, with the impact of traffic accidents ranging from minor to fatal. Understanding the factors that contribute to the severity of these incidents can inform preventive measures and policies. This analysis aims to investigate the influence of the distance from accident sites to traffic hotspots—locations where accidents occur more frequently—on the severity of these accidents. The working hypothesis suggests that accidents occurring closer to hotspots are more likely to result in minor (slight) injuries rather than being fatal.

Association rules are a powerful data mining technique that helps to uncover relationships between variables in datasets. In the context of a grocery dataset, it is possible to identify frequently purchased items and any associations between them. This information can then be used to inform sales and marketing strategies, such as creating product bundles, cross-selling, and up-selling.

The goal of this notebook is to use principal component analysis (PCA) to reduce the dimensionality of the data and identify a smaller number of underlying latent concepts that describe the human interests of the participants. This could provide important insights into human interests and contribute to our understanding of how to better capture and describe the complexity of human preferences.

This notebook serves as a hands-on demonstration of unsupervised learning techniques for landscape image clustering. The objective is to analyze the quality of clustering with respect to the size of the image, and to see how the clustering algorithms perform when applied to a landscape image.