Avoiding Fake Neighborhoods in e-Commerce Collaborative Recommender Systems: A Semantic Approach

Recommender systems are very useful tools in application domains that suffer from information overload, offering the users suggestions they may be interested in. Owing to its business interest, e-commerce has become a major domain in this research field, since identifying those products that the users will appreciate could increase users´ consumption. However, current e-commerce recommender systems overlook some implications of the great diversity of products and services available in the market, giving rise to form fake neighborhoods in collaborative filtering strategies. In this paper, we propose applying semantic reasoning techniques to solve this problem, thus improving, qualitatively and computationally, the recommendation process.