Semantic recommendation through Reinforcement Learning and weighted meta-paths

Jing Yu; Yue Lang; Xuewen Li; Jin Zhang; Shaojie Zheng; Jibing Gong

doi:10.1117/12.2635683

6 May 2022 Semantic recommendation through Reinforcement Learning and weighted meta-paths

Jing Yu, Yue Lang, Xuewen Li, Jin Zhang, Shaojie Zheng, Jibing Gong

Proceedings Volume 12256, International Conference on Electronic Information Engineering, Big Data, and Computer Technology (EIBDCT 2022); 122561N (2022) https://doi.org/10.1117/12.2635683
Event: 2022 International Conference on Electronic Information Engineering, Big Data and Computer Technology, 2022, Sanya, China

Abstract

Explainable recommendation systems, which can produce high-accuracy recommendations and help users make quick decisions, have become a hotspot in research field. Most of existing research algorithms committed to improving the accuracy of recommendation, but ignored the interpretability of recommendation. We propose a semantic recommendation algorithm through Reinforcement Learning and weighted meta-paths which analyzes and selects the meta-path, and uses Reinforcement Learning network to train the weight of meta-path. This method can effectively solve data sparsity, improve the accuracy of recommendation, and increase interpretability. Compared with the baseline method, indicators of the recommended methods for integrating Reinforcement Learning with Heterogeneous Information Networks have been greatly improved, as verified and compared on a real MovieLens 1M dataset. Experimental results prove that the algorithm can effectively improve the accuracy and interpretability of the recommendation.

Citation Download Citation

Jing Yu, Yue Lang, Xuewen Li, Jin Zhang, Shaojie Zheng, and Jibing Gong "Semantic recommendation through Reinforcement Learning and weighted meta-paths", Proc. SPIE 12256, International Conference on Electronic Information Engineering, Big Data, and Computer Technology (EIBDCT 2022), 122561N (6 May 2022); https://doi.org/10.1117/12.2635683

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available