Intelligent classification and prediction of students' mental health in online learning environments using boosting algorithm and LIWC features.

This study aims to enhance the accuracy and stability of classifying students' mental health status in online learning environments using an intelligent model built on the Boosting algorithm and LIWC (Linguistic Inquiry and Word Count) features. The model extracts emotional and psychological features from online learning platforms using the LIWC dictionary and integrates multiple weak classifiers using the Boosting algorithm. The performance of the model is enhanced with the Antlion Optimization Algorithm. Experimental results show that the model's classification accuracy ranges between 98 and 99%, effectively reducing misclassification rates and accurately identifying students experiencing high stress and anxiety. The model enhances mental health status classification and real-time monitoring accuracy, offering critical support for targeted psychological interventions in education.