Extraction of PE Online Teaching Resources With Positive Psychology Based on Advanced Intelligence Algorithm

Area of Science:

• Educational Technology
• Artificial Intelligence
• Sports Science

Background:

• Physical education (PE) teaching resources are crucial for effective instruction.
• The internet provides abundant PE resources, but quality varies significantly.
• Identifying positive psychology content in PE resources is vital to prevent harm to students' values.

Purpose of the Study:

• To develop an AI-driven method for accurately identifying positive psychology in online PE teaching resources.
• To address the challenge of uneven quality and potential negative influences of online PE content.
• To leverage advanced intelligent algorithms for resource classification.

Main Methods:

• A novel text sentiment analysis model, multi-layer-attention convolutional neural network (ACNN)-CNN, was proposed.
• The model integrates hierarchical CNN with an attention mechanism for enhanced text feature extraction.
• Position encoding information was incorporated into the embedding layer to improve classification accuracy.

Main Results:

• The ACNN-CNN model demonstrated superior performance in classifying positive psychology within online PE teaching resources.
• The model effectively extracted text features, proving more suitable for long text emotion classification.
• Experimental results confirmed the model's accuracy and effectiveness compared to other methods.

Conclusions:

• The proposed ACNN-CNN model offers a robust solution for identifying high-quality, positive psychology-focused PE teaching resources.
• This AI approach can help educators curate safer and more beneficial online content for students.
• Accurate classification of online educational resources is essential in the digital age.