To search, Click below search items.


All Published Papers Search Service


Novel method for users’ profiles construction through collaborative filtering


Oumayma BANOUAR†, Said RAGHAY†


Vol. 17  No. 8  pp. 170-176


Current information systems provide transparent access to multiple, distributed, autonomous and potentially redundant data sources. Their users may not know the sources they questioned, nor their description and content. Consequently, their queries reflect no more a need that must be satisfied but an intention that must be refined. The purpose of the personalization is to facilitate the expression of users’ needs. It allows them to obtain relevant information by maximizing the exploitation of their preferences grouped in their respective profiles. In this work, we present a matrix completion approach that minimize the nuclear norm to construct users’ profiles. The initial data matrix corresponds to the ratings provided by users to items. Each row or column contains at least one observation. The proposed approach, based on collaborative filtering concept, starts by a learning process to classify users and preferences. It exploits then these clusters to run a predictive method in the aim to recover the missing or unknown data. Finally, it uses an assignment function to find the ratings of preferences that were not included in the initial data matrix due to the fact of lack of observation.


Personalization, enrichment, user query, user profile, collaborative filtering, bi-clustering, matrix completion, aggregation, assignment function.