Application of Neural Network Algorithm Combined with Bee Colony Algorithm in English Course Recommendation.

Area of Science:

• Artificial Intelligence
• Educational Technology
• Computer Science

Background:

• Traditional BP neural networks suffer from local minima and slow convergence.
• These limitations hinder their effectiveness in complex applications like educational recommendations.

Purpose of the Study:

• To enhance BP neural network performance for English curriculum recommendation technology.
• To address the limitations of BP neural networks using an optimization algorithm.

Main Methods:

• An artificial bee colony algorithm (ABC) was employed to cross-optimize BP network parameters (weights and thresholds).
• The combined ABC-BP model was applied to English course recommendation and teaching design.
• The model underwent 4690 iterations to reach target accuracy.

Main Results:

• The optimized model demonstrated improved solution accuracy and accelerated convergence speed.
• The prediction error of the combined model was successfully maintained below 10%.
• The neural network achieved target accuracy after 4690 iterations.

Conclusions:

• The integration of the artificial bee colony algorithm with BP neural networks offers a superior approach for English curriculum recommendation.
• The combined model effectively overcomes the traditional limitations of BP neural networks, showing good prediction performance.