Data - In-Memory Association Rule Mining with Hierarchical Taxonomies

This dataset contains anonymised wishlist interaction data collected from an e-commerce platform. Each row represents a unique (wishlist, product category) pair, derived by joining wishlist events with product and category metadata. The data has been preprocessed as follows: duplicate interactions from the same user–product pair have been removed, products without descriptive content (title or description) have been excluded, and duplicate category entries per wishlist have been dropped. The final dataset contains 5,089,834 rows across two columns: wishlist_id (anonymised wishlist identifier) and category_name (full hierarchical product category path, e.g. Health & Beauty > Personal Care > Cosmetics). The dataset is intended for use in association rule mining on consumer wishlist behaviour, with the aim of discovering co-occurrence patterns between product categories across user wishlists.