Data Mining Based on Student Campus Card Consumption Information

This study focuses on the consumption data of the campus card of students in Qingyuan Polytechnic. By cleaning and analyzing the original data exported from the campus card center, the consumption patterns of students were obtained and provided reference for optimizing the school's management. We exported a total of 2,932,536 original consumption data from the campus card center for one academic year, imported the data into a database, cleaned and analyzed it using SQL statements, and conducted student consumption clustering analysis using the k-means++ algorithm. We used numpy, matplotlib, and other tools to create visualizations of the data. The analysis mainly focused on the overall consumption situation of students, daily consumption patterns, meal consumption patterns, consumption locations, and consumption clustering. We proposed some suggestions for optimizing student services based on the analysis.