Loading...

Extracting implicit social relation for social recommendation techniques in user rating prediction

Taheri, S. M ; Sharif University of Technology | 2019

497 Viewed
  1. Type of Document: Article
  2. DOI: 10.1145/3041021.3051153
  3. Publisher: International World Wide Web Conferences Steering Committee , 2019
  4. Abstract:
  5. Recommendation plays an increasingly important role in our daily lives. Recommender systems automatically suggest items to users that might be interesting for them. Recent studies illustrate that incorporating social trust in Matrix Factorization methods demonstrably improves accuracy of rating prediction. Such approaches mainly use the trust scores explicitly expressed by users. However, it is often challenging to have users provide explicit trust scores of each other. There exist quite a few works, which propose Trust Metrics to compute and predict trust scores between users based on their interactions. In this paper, first we present how social relation can be extracted from users' ratings to items by describing Hellinger distance between users in recommender systems. Then, we propose to incorporate the predicted trust scores into social matrix factorization models. By analyzing social relation extraction from three well-known real-world datasets, which both: trust and recommendation data available, we conclude that using the implicit social relation in social recommendation techniques has almost the same performance compared to the actual trust scores explicitly expressed by users. Hence, we build our method, called Hell-TrustSVD, on top of the state-of-the-art social recommendation technique to incorporate both the extracted implicit social relations and ratings given by users on the prediction of items for an active user. To the best of our knowledge, this is the first work to extend TrustSVD with extracted social trust information. The experimental results support the idea of employing implicit trust into matrix factorization whenever explicit trust is not available, can perform much better than the state-of-the-art approaches in user rating prediction. © 2017 International World Wide Web Conference Committee (IW3C2), published under Creative Commons CC BY 4.0 License
  6. Keywords:
  7. Matrix Factorization ; Recommender Systems ; Social Networks ; Social Recommendation Techniques ; Data mining ; Factorization ; Forecasting ; Social networking (online) ; World Wide Web ; Hellinger distance ; Implicit trusts ; Matrix factorizations ; Real-world datasets ; Social recommendations ; Social relations ; State of the art ; State-of-the-art approach ; Matrix algebra
  8. Source: 26th International World Wide Web Conference, WWW 2017 Companion, 3 April 2017 through 7 April 2017 ; 2019 , Pages 1343-1351 ; 9781450349147 (ISBN)
  9. URL: https://dl.acm.org/doi/10.1145/3041021.3051153