A Personalized Recommender System from Probabilistic Relational Model and Users’ Preferences

• DOI: 10.1016/j.procs.2014.08.193
• Publication Date: 2014-09-13T02:46:51Z
• Authors: Rajani Chulyadyo; Philippe Leray
• Keywords: Users’ Preferences; Probabilistic Relational Model (PRM); Hybrid recommender system; 0202 electrical engineering, electronic engineering, information engineering; Personalized Recommender Systems; Cold start problem; [INFO.INFO-LG] Computer Science [cs]/Machine Learning [cs.LG]; 02 engineering and technology

Recommender systems are applications to retrieve useful information from large amount of online data assist users in discovering interesting items/products the system. Collaborative filtering, content-based demographics-based filtering and hybrid approach main approaches realize recommendation systems. Most existing algorithms use a single deal with problems. Besides, traditional mainly dyadic relationships between items whereas real world generally conceptualized terms objects relations them. based on Probabilistic Relational Model (PRM)1,2, framework for learning probabilistic models relational data, have tried address this issue. However, PRM-based do not fit into our context where we struggling contradictory situation real-world application that requires building personalized recommender when no user profile exists. Therefore, propose novel build model help users' preferences decision making criteria. Using approach, content-based, collaborative as well can be achieved same PRM. Applying cold system, show is actually capable personalizing recommendations coldstart situation.