Modeling Implicit Feedback and Latent Visual Features for Machine-Learning Based Recommendation

DOI

10.2991/eusflat-19.2019.44How to use a DOI?

Keywords

Multi-view information Machine learning Stacked convolutional auto-encoder Topic modeling Information fusion

Abstract

With the rapid accumulation of rich media data on the Internet, this paper proposes a Multi-View Bayesian Personalized Ranking (MVBPR) recommendation model that combines visual and textual contents, along with uncertainty modeling in consumer preferences and in visual representation in forms of implicit feedbacks and latent factors. MVBPR is a machine-leaning framework integral of deep-learning (i.e., SCAE) and topic modeling (i.e., LDA) strategies to fuse images and texts. Moreover, extensive experiments from real data sets demonstrate MVBPR's advantages over baseline models, including its superiority in dealing with the cold start situation.

Copyright

© 2019, 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  - Yue Guan
AU  - Qiang Wei
AU  - Guoqing Chen
AU  - Xunhua Guo
PY  - 2019/08
DA  - 2019/08
TI  - Modeling Implicit Feedback and Latent Visual Features for Machine-Learning Based Recommendation
BT  - Proceedings of the 11th Conference of the European Society for Fuzzy Logic and Technology (EUSFLAT 2019)
PB  - Atlantis Press
SP  - 305
EP  - 312
SN  - 2589-6644
UR  - https://doi.org/10.2991/eusflat-19.2019.44
DO  - 10.2991/eusflat-19.2019.44
ID  - Guan2019/08
ER  -