Improving e-Commerce Collaborative Recommendations by Semantic Inference of Neighbors' Practical Expertise

Abstract

E-commerce has become a major application domain for recommender systems due to its business interest. These tools aim to identify the products each user may like or find useful, which can boost users' consumption. Particularly, collaborative recommender systems rely on a set of like-minded users to select the products to offer. Taking into account the expertise of the users who drive such decision can increase the accuracy of the process. However, current approaches require extra data, that is not often available, to obtain expertise measures. In this paper, we apply a semantic approach to get a measure of practical expertise by exploiting the data available in any e-commerce recommender system-the consumption histories of the users. This way, we improve recommendation results transparently to the users.