Design and Research of an Artificial Intelligence-Based Basketball Teaching System

Basketball contains a variety of technical actions, such as shooting, dribbling, passing, and defense. Long-term dependencies in the data of these action sequences lead to the discrepancy between the final student’s training score and the actual situation. Therefore, an artificial intelligence-based basketball teaching system is designed and researched. After students use the system, they first need to input personal information for the initial learning of basketball, and the data of students’ learning of basketball are saved in the data storage layer and fused with the data of usual training and examination using a neural network. The CNN–BiGRU model is constructed by combining the advantages of CNN in processing image and video data to extract students’ technical movement characteristics and game performance and the ability of BiGRU to capture long-term dependencies in sequential data. The fused data are used as input data to give the intelligent evaluation results of basketball training. According to the evaluation results, the pre-similarity of the student’s training data is calculated again through the feedback mechanism, and the suitable basketball teaching subjects are selected so that a more objective and efficient basketball teaching effect can be obtained by using artificial intelligence technology combined with the student’s own training situation in the whole teaching process. The experimental results show that in the initial personalized subject recommendation, suitable basketball teaching subjects can be recommended according to the initial data; through the training of CNN and BiGRU, CNN convolution kernel size is 7*7; BiGRU batch size is 128, learning rate is 0.05 can ensure that the CNN–BiGRU has the best training and evaluation accuracy; After a certain period of learning, the artificial intelligence can accurately evaluate the students’ training results and give the score, and through the feedback mechanism, it can give the best training program for students with different scores.