Online learning resource recommendation method based on multi-similarity metric optimization under the COVID-19...

Area of Science:

• Educational Technology
• Information Science
• Computer Science

Background:

• The COVID-19 pandemic accelerated the adoption of online learning globally.
• Online learning environments often suffer from information overload and knowledge maze issues.
• Effective learning resource recommendation is crucial for navigating digital educational content.

Purpose of the Study:

• To propose a novel learning resource recommendation method to address information overload in online learning.
• To enhance the accuracy and effectiveness of learning resource recommendations.
• To improve the overall online learning experience for students.

Main Methods:

• Developed a multi-similarity measure optimization for learning resource recommendations.
• Introduced information entropy to optimize user score similarity.
• Employed particle swarm optimization to determine comprehensive similarity weights.
• Implemented secondary screening for nearest neighbor user identification based on score and interest similarity.

Main Results:

• The proposed method significantly improves recommendation accuracy.
• The algorithm maintains stable recommendation coverage.
• Experimental results validate the effectiveness of the multi-similarity optimization approach.

Conclusions:

• The multi-similarity measure optimization method effectively enhances online learning resource recommendations.
• This approach helps mitigate information overload and knowledge maze problems.
• The findings support the development of more intelligent and effective online learning systems.