Personalized Recommendation of Classification based on Social Relationship and Time Information

DOI

10.2991/icmmita-16.2016.188How to use a DOI?

Keywords

Recommender performance; Social relationship; Time information; Classification

Abstract

Improving recommender performance is beneficial. However, sparsity and scalability of data is a problem faced by recent recommender systems. Customer interests and available products are changing constantly. Social interactions among users are highly influential on the effectiveness of the recommendation. Following these intuitions, this paper proposed a recommendation method incorporating with social and temporal information with probabilistic matrix factorization, which is called PMFST (Probabilistic Matrix Factorization with Social and Temporal information), to solve the problem of data sparsity and achieve real and dynamic recommender performance. The experiment on two real data sets shows that the proposed method outperforms the state-of-the-art methods in terms of minimal error and recommender performance.

Copyright

© 2017, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Mengzi Tang
AU  - Li Li
PY  - 2017/01
DA  - 2017/01
TI  - Personalized Recommendation of Classification based on Social Relationship and Time Information
BT  - Proceedings of the 2016 4th International Conference on Machinery, Materials and Information Technology Applications
PB  - Atlantis Press
SN  - 2352-538X
UR  - https://doi.org/10.2991/icmmita-16.2016.188
DO  - 10.2991/icmmita-16.2016.188
ID  - Tang2017/01
ER  -