Generalist Recommender System Kernel


Every day, new data appears on the Web. Everyone browsing on the Internet can have the perception of the huge amount of information available, which can lead to a situation of information overload. In this case, techniques to retrieve useful information become more and more important. The usefulness of information depends on the users and their objectives, so retrieval systems have to try to understand the purpose of a user search in order to propose information he could be interested in.

A Recommender System (RS) is a widely used mechanism for providing advice to people in order to select a set of items, activities or any other kind of products. Typically, RSs are designed to provide recommendations for a single user considering the user's interests and tastes. A RS infers the user's preferences by analyzing the available user data, information about other users and information about the environment.

However, many daily activities such as watching a movie or going to a restaurant involve a group of users, in which case recommendations must take into account the preferences of all users in the group. This type of systems is called Group Recommender System (GRS). The main issue in group recommendation is to identify the items which are likely to match the group's tastes or preferences. By taking into account the individuals' profiles, and the preferences of the group as a whole, GRSs are capable of finding a compromise that is accepted by all the members in the group.

GRSK (Generalist Recommender System Kernel) is a RS based in a semantic description of the domain that uses a hybrid recommendation technique, fed by the recommendations obtained from different algorithms. The task of GRSK is to generate the list of the top N items that will be of interest to the user or to a group of users. GRSK can be parameterized to adjust the system working model, i.e. to use the desired recommendation techniques. Besides, it is prepared to include as many techniques as desired by simply developing new modules. On the other hand, it is a domain-independent engine, able to work with different catalogs of items to recommend.


This work has been partially supported by the GVA projects “Sistema recomendador de turismo en la ciudad de Valencia vía web” [GVPRE/2008/384] and “Generación online de rutas turísticas en la Comunidad Valenciana adaptadas al perfil de usuario” [GV04A-388].