Main Article Content


In this work, we tend to propose a userservice rating prediction model supported probabilistic matrix factorization by exploring rating behaviors. Usually, users are seemingly to participate in services within which they are interested and revel in sharing experiences with their friends by description and ratings. Social users with similar interests tend to possess similar behaviors. It's the idea for the cooperative filtering primarily based recommendation model. Social users’ rating behaviors may well be well-mined from the subsequent four factors: personal interest, social interest similarity, social rating behavior similarity, and social rating behavior diffusion. By considering these four factors, the rating behavior in recommender system may well be embodied in these aspects: once user rated the item, what the rating is, what the item is, what the user interest we tend to might dig from his/her rating records is, and the way user’s rating behavior diffuse among his/her social friends. During this paper, we tend to propose a userservice rating prediction approach by exploring social users’ rating behaviors in a very unified matrix factorisation framework. we tend to found that users high on Openness tend to rate a lot of things than needed, whereas low Conscientiousness could be a essential issue that provokes users to rate things in an explosive method. Our findings are helpful for researchers curious about user modeling, preference stimulant, recommender systems and on-line promoting.

Article Details

How to Cite
Kavya.v, & Arumugam.S. (2017). Exploring rating of product using collaborative filtering approach . International Journal of Intellectual Advancements and Research in Engineering Computations, 5(1), 728–733. Retrieved from