Visualization Method of Key Knowledge Points of Nursing Teaching Management System Based on SOM Algorithm and...

Area of Science:

• Nursing Education
• Computer Science
• Information Retrieval

Background:

• Traditional collaborative filtering and matrix decomposition algorithms struggle with data sparsity and scalability in nursing knowledge point recommendations.
• Existing methods often yield low accuracy due to recommendations solely based on prediction scores.

Purpose of the Study:

• To propose an improved nursing teaching knowledge point recommendation method.
• To enhance recommendation accuracy and efficiency by addressing limitations of traditional algorithms.

Main Methods:

• Utilized a Self-Organizing Map (SOM) neural network for user clustering based on academic background.
• Constructed partial order relationships of knowledge points using explicit and implicit user web behavior.
• Employed a Factor Decomposition Machine (FDM) as a ranking function, incorporating user background, web access, and nursing teaching text data.
• Applied a peer-to-peer ranking learning algorithm for precise knowledge point recommendations.

Main Results:

• The proposed method effectively alleviates the issue of data sparsity in recommendation systems.
• Demonstrated significant improvements in both the accuracy and efficiency of nursing teaching knowledge point recommendations.
• The integration of SOM and FDM proved superior to traditional approaches.

Conclusions:

• The novel recommendation method offers a robust solution for personalized nursing education.
• This approach enhances the learning experience by providing more relevant and accurate knowledge point suggestions.
• The findings suggest a promising direction for applying advanced machine learning techniques in educational technology.