Context-Aware Ranking with Factorization Models

More about the book

This book explores context-aware ranking, crucial for applications like recommender systems and search engines, emphasizing that rankings depend on context (e.g., user or query). It presents a novel factorization method based on pairwise interactions and introduces the 'Bayesian Context-aware Ranking' framework. The book also applies this theory to item, tag, and sequential-set recommendations, addressing time-variant factors and one-class problems. It builds on award-winning research.