Data Preprocessing System of Depression Recognition

Abstract

The system grabs data packets from campus network routers, divides the collected basic data into seven basic databases, preprocesses all the basic data and preprocesses the text. Through the statistics and calculation of the data, it extracts the frequency of user browsing web pages, the frequency of using application categories, the time period of surfing the Internet, the time of surfing the Internet, the total flow of surfing the Internet, the amount of Posts and the mail. Based on the emotional dictionary, the key words of emotional features and topic features are extracted from the text information such as the content of browsing web pages and the content of posting, and the user’s emotional orientation values are obtained. The obtained behavioral, emotional and topic features are stored in the feature database. Finally, through the functional test and performance test, the system can extract the daily characteristic quantity according to the daily data, get the characteristic data table, and use the characteristic data to judge the depression tendency of college students by the depression recognition and warning model.