The application domain is book recommendation. The datasets provided contains both interactions between users and books, and tokens (words) extracted from the book text. The main goal of the project is to discover which item (book) a user will interact with.
The datasets includes around 135k interactions, 7947 users, 25975 items and 20000 tokens. The training-test split is done via random holdout (85% training, 15% test). The goal is to recommend a list of 10 potentially relevant items for each user. MAP@10 is used for evaluation.
- Amedeo Cavallo
- Federico Capaccio
- Maurizio Ferrari Dacrema