Policy Optimization Recommendation Algorithm Based on Mapping Network for Behavior Enhancement

Linlin Shan, Guisong Jiang, Shuang Li, Shuai Zhao, Kunjie Luo, Long Zhang, Yi Li


The algorithm of policy optimization with learning behavior enhancement based on mapping network technology was proposed, aiming to address the issues of lack and sparsity of learning behavior data and weak generalization ability of the model in AI education. Based on the basic recommendation algorithm and the framework of rein- forcement learning, and model introduces the correlation mapping network to realize the transformation of strong and weak correlation, so as to optimize the input agent policy to improve the performance model of course recommendation. Experiment on MOOC da- tasets show that the proposed algorithm model has a stable improvement compared with the baseline models, and can effectively improve the accuracy of course recommendation.

Article Metrics

Abstract: 49 Viewers PDF: 28 Viewers


Strong/Weak Correlation; Mapping Network; Policy Optimization; Reinforcement Learning; Course Recommendation

Full Text:



  • There are currently no refbacks.


Journal of Applied Data Sciences

ISSN : 2723-6471 (Online)
Organized by : Departement of Information System, Universitas Amikom Purwokerto, Indonesia; Computer Science and Systems Information Technology, King Abdulaziz University, Kingdom of Saudi Arabia.
Website : http://bright-journal.org/JADS
Email : taqwa@amikompurwokerto.ac.id (principal contact)
    husniteja@uinjkt.ac.id (managing editor)
    support@bright-journal.org (technical issues)

 This work is licensed under a Creative Commons Attribution-ShareAlike 4.0