Collaborative Filtering Recommendation Algorithm for User Interest and Relationship Based on Score Matrix

DOI

10.2991/mmsa-18.2018.49How to use a DOI?

Keywords

collaborative filtering; trust; user interest; sparse data; matrix filling

Abstract

An improved collaborative filtering recommendation algorithm is proposed to solve the problem of sparse and low recommendation accuracy of traditional collaborative filtering recommendation algorithm. User preferences and user trust relationships are used to calculate the user's preferences for the project, and the user ratings are used to fill the scoring matrix with unrated items. Considering the change of user interest and user relationship, we introduce time based interest weight function and preference degree to the project similarity computation and recommendation process, and identify the nearest neighbor set, so as to achieve the best recommendation. User preferences and user trust relationships are used to calculate the user's preferences for the project, and the user ratings are used to fill the scoring matrix with unrated items.

Copyright

© 2018, 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  - Kejia Xue
AU  - Junyi Wang
PY  - 2018/03
DA  - 2018/03
TI  - Collaborative Filtering Recommendation Algorithm for User Interest and Relationship Based on Score Matrix
BT  - Proceedings of the 2018 International Conference on Mathematics, Modelling, Simulation and Algorithms (MMSA 2018)
PB  - Atlantis Press
SP  - 217
EP  - 221
SN  - 1951-6851
UR  - https://doi.org/10.2991/mmsa-18.2018.49
DO  - 10.2991/mmsa-18.2018.49
ID  - Xue2018/03
ER  -