Online Education Intelligent Performance Prediction Model Based on Logistics Regression Algorithm

With the rapid development of online education, a large number of students' online learning behavior data is stored. Based on the online learning behavior of students, the prediction of teaching performance can effectively identify students who are at risk of learning, thus helping educators to provide learning assistance for such students. It can improve their learning performance and ensure the quality of online learning. However, at present, the inductive methods of predicting factors of academic performance are relatively simple, and there is a general lack of standardized data selection methods. The perspective of academic performance prediction is often curriculum-centered, and the curriculum often ignores the dominant position of students. Therefore, a deep knowledge tracking model combining a multi-feature model and performance prediction was proposed in the paper. The traditional single prediction model is planned into a feature set. Through integrating a plurality of teaching indexes, a plurality of feature sets is innovatively input into the model for feature processing. The optimized regression algorithm performs correction after establishing the weight to ensure the accuracy of the prediction result. Finally, a simulation experiment environment is set up. The comparative test results indicate that the prediction accuracy of the proposed model is higher than that of the single data prediction algorithm. Compared with the traditional prediction algorithm, the prediction accuracy is 92. 4% and 89. 1%, respectively. The results of this study judge the quality of students' online learning through useful information and provide empirical evidence for online learning platforms to improve students' learning effectiveness. The use of network behavior information and interactive information to improve the learning effect provides auxiliary significance.