Tag Recommendation Based on High Order Singular Value Decomposition

DOI

10.2991/ncce-18.2018.122How to use a DOI?

Keywords

tag recommendation, tensor decomposition, HOSVD.

Abstract

Social tag recommendation system provides the information sharing platform for users, allowing users to "tag" for the browsing of the items marked information. Tags both describe the semantics of the item and reflect the user's preferences. However, existing tag recommendation systems face the problem that different users may have different tags for the same item due to different interests, or the same tag may have different semantics for different users. To solve this problem, a three-dimensional tensor model is introduced. Three dimensions of the three-dimensional tensor are used to describe three types of entities in the tag recommendation system: users, objects and tags respectively. The high order singular value decomposition (HOSVD) is used to reduce the tensor model and realize the potential semantic association analysis among the three types of entities, so as to improve the accuracy of the tag recommendation system. Experimental results show that this method has significantly improved in accuracy and recall performance.

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  - Yuliang Shi
AU  - Shuo Liu
PY  - 2018/05
DA  - 2018/05
TI  - Tag Recommendation Based on High Order Singular Value Decomposition
BT  - Proceedings of the 2018 International Conference on Network, Communication, Computer Engineering (NCCE 2018)
PB  - Atlantis Press
SP  - 744
EP  - 748
SN  - 1951-6851
UR  - https://doi.org/10.2991/ncce-18.2018.122
DO  - 10.2991/ncce-18.2018.122
ID  - Shi2018/05
ER  -