Data Fusion Model for Psychological Crisis Early Warning System Using Data Mining Techniques

Abstract

The mental well-being of university students has garnered significant attention from various sectors of society. Psychological challenges can disrupt educational institutions' harmonious progress, necessitating early intervention. Psychological disorders have the potential to hinder the smooth functioning of institutions of higher education, and their immediate intervention is necessary. The study proposes a notification system for identifying psychological crises among college students based on a data fusion approach. This research implements a range of machine learning methods, such as decision trees, random forests, logistic regression, and the Apriori algorithm. These methods utilize psychological profile data and survey questionnaires, which are subjected to encoding and feature selection for preprocessing. The suggested data fusion approach enhances the accuracy of prediction through the combination of the various methods, achieving a high F1 score of 82.2% compared to individual methods. The results highlight the effectiveness of data fusion in identifying psychological crises and, hence, providing a holistic platform for the early intervention of students' mental health issues.