Prediction and Analysis of Students’ Behavior Based on Data Mining in Educational Administration

Educational administration is a complex system engineering, which involves students, teachers, courses and many other aspects. By accurately predicting students’ behavior, educational institutions and teachers can adopt corresponding teaching strategies in advance, thus guiding students more effectively. The purpose of this article is to explore the method of predicting and analyzing students’ behavior in educational administration through data mining (DM) technology. To accomplish the objective, this article employs a Convolutional Neural Network (CNN) to construct a predictive model capable of deeply learning the intrinsic characteristics of students’ behavior data. By gathering and preprocessing multivariate data, including students’ academic performance and online learning activities, essential features are extracted and fed into the CNN model for training. Experimental results demonstrate the model’s exceptional performance in terms of training loss, prediction error, and prediction accuracy, with the latter exceeding 90%, significantly outperforming other prevalent algorithms. This indicates that the CNN model, based on DM, holds extensive application potential and practical value in the realm of student behavior prediction.