Integration of Athlete Training Monitoring Information based on Deep Learning

Abstract

In order to solve the problem of mining and analyzing athlete training monitoring information, the author proposes a deep learning based integration of athlete training monitoring information. The author proposes a deep learning based method for integrating athlete training monitoring information, deploying agents on various data source nodes, collecting athlete training information from each data source, and implementing denoising and dimensionality reduction on the monitoring information; Building an information integration model based on convolutional neural networks in deep learning; Extracting monitoring information features through convolutional layers, and fusing information with similar features into the same category through output layer classifiers, completing the integration of athlete training volume monitoring information. The experimental results show that as the number of iterations increases, the classification accuracy of the integrated model based on convolutional neural networks is continuously improving, while the error is continuously decreasing and getting closer to zero. When the maximum iteration number is 100, the model accuracy is 99.74%. The average Gini coefficient of the author's research method is higher, indicating a higher integration accuracy of the method.