A knowledge based recommender system with multigranular linguistic information

Recommender systems are applications that have emerged in the e-commerce area in order to assist users in their searches in elctronic shops. These shops usu-ally offer a wide range of items to satisfy the neccessi-ties of a great variety of users. Nevertheless, search-ing in such a wide range of items could be a very dif-ficult and tedious task. Recommender Systems assist users to find items by means of recommendations based on information provided from different sources such as: other users, experts, etc. Most of the recommender sys-tems force users to provide their preferences or neces-sities using an unique numerical scale of information fixed in advance. Normally, this information is usually related to opinions, tastes and perceptions, and there- fore, it means that it is usually better expressed in a qualitative way, with linguistics terms, than in a quan-titative way, with precise numbers. In this contribution, we propose a Knowledge Based Recommender System that uses the fuzzy linguistic approach to define a flexi-ble framework that captures the uncertainty of the user's preferences. Thus, this framework will allow users to express their necessities in a different scale, closer to their knowledge, from the scale used to describe the items.