News Topic Recommendation using an extended Bayesian Personalized Ranking
Bayesian Personalized Ranking (BPR) is a famous recommendation approach which learns to rank items based on one-class implicit feedback. In this study I use an extended version of BPR using consumption behavior of users on news article to recommend news topic. The extended version performs better compare to the original version of BPR.
Alireza Gharahighehi is a PhD student in Kulak (KU Leuven at Kortrijk). He started his PhD in September 2018 with focus on recommender systems. He received his first M.S in Industrial Engineering form Bu-ali Sina university and two other Masters in Business Economics and Artificial Intelligence from Ku Leuven. Now he is working on a news recommendation system project.