The application of AI technology promotes the innovation of education management mode and provides more possibilities for students’ personalized development. Based on AI technology, this study proposes a smart education management model and analyzes the consumption and behavioral data of college students through data mining methods to explore possible paths for their personalized development. The research method mainly includes the construction of student behavioral portrait and the application of K-prototype clustering algorithm. In the data collection phase, 61.84 million consumption records, 31,754 students’ performance data and 1,304,736 book borrowing records were analyzed. It was found that based on the K-prototype clustering algorithm, students were categorized into four consumption groups, in which the average monthly consumption of students in group 3 was $291.61 and the average monthly food and beverage consumption was $128.46, while group 4 had the highest average monthly consumption of $961.56. In addition, the analysis revealed the significant correlation between factors such as age and gender and students’ consumption behavior. The conclusion shows that the smart education management model and data mining methods provide important support for education management and students’ personalized development, especially in the areas of consumption behavior, academic performance and mental health, providing basic data support for personalized education management.
