The traditional classification and query methods for students’ mental health data have problems such as low operational efficiency. Therefore, a classification and query method for students’ mental health data based on Internet of Things technology is designed. Based on the edge computing technology of the Internet of Things and the deep neural network design data classification mining method, students’ mental health data mining is implemented. Design a data deduplication model based on open channel solid-state disks and implement deduplication processing for mining data. Design a classification and query algorithm based on naive Bayesian algorithm to achieve classification and query of students’ mental health data. The test results show that the accuracy, recall rate, and F1 value of this method are high, and the running time is short.