University Course Recommendation Based on Multi-source Educational Knowledge Graph

Junsheng Peng, Jiang Long, Yangming Guo, Jin Wang, Bo Zheng

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In response to the course recommendation problem in the undergraduate teaching scenario of higher education, a recommendation algorithm based on a multi-source educational knowledge graph is proposed. Firstly, the course correlation is analyzed, and relevant data is extracted from various databases of the university for preprocessing. Then, a multi-path RNN encoding method is proposed to embed multi-source information into the educational knowledge graph. Subsequently, an MLP is utilized to model the interaction between students and courses, aiming to predict student course selections. Finally, a comparative experiment is conducted to validate the feasibility of the university recommendation method based on the multi-source educational knowledge graph.

Original languageEnglish
Title of host publicationProceedings - 2023 4th International Conference on Computer, Big Data and Artificial Intelligence, ICCBD+AI 2023
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages727-732
Number of pages6
ISBN (Electronic)9798350373233
DOIs
StatePublished - 2023
Event4th International Conference on Computer, Big Data and Artificial Intelligence, ICCBD+AI 2023 - Guiyang, China
Duration: 15 Dec 202317 Dec 2023

Publication series

NameProceedings - 2023 4th International Conference on Computer, Big Data and Artificial Intelligence, ICCBD+AI 2023

Conference

Conference4th International Conference on Computer, Big Data and Artificial Intelligence, ICCBD+AI 2023
Country/TerritoryChina
CityGuiyang
Period15/12/2317/12/23

Keywords

  • course recommendation
  • educational data mining
  • higher education
  • knowledge graph

Fingerprint

Dive into the research topics of 'University Course Recommendation Based on Multi-source Educational Knowledge Graph'. Together they form a unique fingerprint.

Cite this