Exploring collaborative annotations in hibrid Recommender systems

Abstract

Recommender systems are an important tool to cope with the information overload problem. However, a common problem among the approaches is how to obtain meaningful information about the content and about the users' preferences. This problem is known as semantic gap, and the related problems have been studied over the years. On the other hand, with the Web 2.0 advent, and the possibility of users to produce new content and enrich existent data, new research possibilities have been created to reduce the effects of the semantic gap problem. This plan aims at investigating some of the challenges of using collaborative user annotations to improve recommender systems. To do that, we propose the development of a unified recommender model which will analyze data produced by the interaction between users and system, in order to obtain richer metadata about the content and personal interests of users. As a result, we hope to develop new ways to integrate techniques from different fields, such as information retrieval, machine learning and natural language processing, in the context of recommender systems. (AU)