IMPROVEMENT QUALITY OF THE RECOMMENDATION SYSTEM USING THE INTRINSIC CONTEXT

摘要

The traditional recommendation systems provide a solution to the problem of information overload. They provide users with the information and content which are the most relevant for them. These systems ignore the fact that users interact with systems in a particular context. Context plays an important role in determining users' behavior by providing additional information that can be exploited in building predictive models. Context-aware recommendation systems take this information into account to make predictions in order to improve the performance of the filtering process. Most existing Context-aware systems use the extrinsic context. In this paper, we propose an intrinsic contextual recommendation system that we can apply to the recommendation of contents in general (i.e. book, Url, item, product, movie, song, restaurant, etc.). The context in our approach is extracted from the set of attributes for the object itself Our system use a contextual pre-filtering technique based on implicit user feedback. To show the performance of the recommendation process, we consider the movie domain as a case study.