A Multi-criteria Collaborative Filtering Approach for Research Paper Recommendation in Papyres

Graduate students, professors and researchers regularly access, review, and use large amounts of literature. In previous work, we presented Papyres, a Research Paper Management Systems, which combines bibliography functionalities along with paper recommender techniques and document management tools, in order to provide a set of functionalities to locate research papers, handle and maintain the bibliographies, and to manage and share knowledge about the research literature. In this work we detail Papyres’ paper recommendation technique. Specifically, Papyres employs a Hybrid recommender system that combines both Content-based and Collaborative filtering to help researchers locate research material. Particularly, in this work special attention is given to the Collaborative filtering process, were a multi-criteria approach is used to evaluate the articles, allowing researchers to denote their interest in specific parts of articles. Moreover, we propose, test and compare several approaches to determine the neighbourhood in the Collaborative filtering process such as to increase the accuracy of the recommendation.