Context-aware reinforcement learning for course recommendation

Lin, Yuanguo; Lin, Fan; Yang, Lvqing; Zeng, Wenhua; Liu, Yong; Wu, Pengcheng

doi:10.1016/j.asoc.2022.109189

Cited by 13 publications

(2 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the basis of the above work, Lin et al [44] proposed a hierarchical reinforcement learning model with dynamic recurrent mechanism for course recommendation, and it employed a recurrent scheme by context-aware learning to exploit the current knowledge. Lin et al [45] focused on improving the performance of the profile reviser, making the agent obtain the cumulative rewards of the context and adopt better policies. Lin et al [46] used a recommendation model that could capture user preferences by historical courses and improve the accuracy of course recommendations.…”

Section: Rl-based Course Recommendationmentioning

confidence: 99%

An Actor-Critic Hierarchical Reinforcement Learning Model for Course Recommendation

Liang,

Zhang,

Guo

et al. 2023

Electronics

View full text Add to dashboard Cite

Online learning platforms provide diverse course resources, but this often results in the issue of information overload. Learners always want to learn courses that are appropriate for their knowledge level and preferences quickly and accurately. Effective course recommendation plays a key role in helping learners select appropriate courses and improving the efficiency of online learning. However, when a user is enrolled in multiple courses, existing course recommendation methods face the challenge of accurately recommending the target course that is most relevant to the user because of the noise courses. In this paper, we propose a novel reinforcement learning model named Actor-Critic Hierarchical Reinforcement Learning (ACHRL). The model incorporates the actor-critic method to construct the profile reviser. This can remove noise courses and make personalized course recommendations effectively. Furthermore, we propose a policy gradient based on the temporal difference error to reduce the variance in the training process, to speed up the convergence of the model, and to improve the accuracy of the recommendation. We evaluate the proposed model using two real datasets, and the experimental results show that the proposed model significantly outperforms the existing recommendation models (improving 3.77% to 13.66% in terms of HR@5).

show abstract

Section: Rl-based Course Recommendationmentioning

confidence: 99%

An Actor-Critic Hierarchical Reinforcement Learning Model for Course Recommendation

Liang,

Zhang,

Guo

et al. 2023

Electronics

View full text Add to dashboard Cite

show abstract

“…In these Internet platforms, various data, such as user behavior, points of interest, concerns, opinions, etc., form user profiles [10]. User profiles are acquired and processed based on multiple data sources, which are pervasive and iterative and can help companies and individuals better understand and serve their users [11][12]. At the same time, user profiling is also a very accurate tool to gain timely insights into consumer needs [13].…”

Section: Introductionmentioning

confidence: 99%

Research on the Application of Knowledge Graph in Demand-Side Flexible Resource Profiling and Aggregation Techniques

Yang,

Wang,

Feng

et al. 2024

Applied Mathematics and Nonlinear Sciences

View full text Add to dashboard Cite

This paper commences by assessing the current landscape of power system development, focusing on the theory, principles, and structures of demand-side flexible resources and their aggregation technology. Utilizing network crawler technology within a knowledge graph framework, the research data pertinent to demand-side flexible resources and aggregation technology are extracted. These data undergo a meticulous cleaning process before being stored, culminating in the development of a knowledge graph tailored to the imaging and technology of demand-side flexible resources. The findings reveal a response rate of 7.25% ± 1.15%, with an uncertainty interval of 2.33%. Variations in air-conditioning load states appear to exert minimal impact on the response time lag. Following the issuance of a response signal, all systems can rapidly initiate appropriate response actions, demonstrating an uncertainty interval of approximately 52s±11s and 22s. The duration of the responses averages around 75s±11s, with an uncertainty interval of about 30s. This study fulfills the power system criteria for standards of demand-side flexible resources and augments the competitiveness of China’s power market.

show abstract