A Model for Predicting Physical Health of College Students Based on Semantic Web and Deep Learning Under Cloud Edge Collaborative Architecture

A Data Analysis System Based On Cloud Edge Collaboration (Cec)

The physical form, function, and quality test results of college students are influenced by various factors and levels, and the fluctuation of physical test data is significant, even with great fluctuations. It is very difficult to predict the various results tested in college students' physical tests (Zhang, 2021; Dedeliuk et al., 2018). To gain a more real-time, efficient prediction of PHCS, a prediction system based on CEC architecture was designed from a data-driven perspective. The architecture is shown in Figure 1.

Figure 1.

Data Analysis System Architecture Based on CEC

First, preprocess the physical examination data of college students is preprocessed; for instance, missing values are processed. Then, the edge layer deploys an equal dimensional dynamic GOM model to extract corresponding features from various physical measurement indicators such as height and weight, and uploads them to the cloud center processing layer. Finally, the cloud center predicts various indicators and related influencing factors on the basis of the deployed prediction model and analyzes the physical and health status of college students, meeting the requirements of accurate analysis of large-scale college student physical test data and future physical and health status prediction.