Emotional State Analysis of Duzce University Students Using MODUM Application Data with Data Mining Methods

The academic and social performance of students is significantly impacted by their mental health, which is a pivotal component of public health. It is well-documented that students in universities often undergo substantial psychological and emotional changes that can profoundly impact their daily behaviors and established mental health state. These alterations may manifest as a range of symptoms, including decreased focus and academic performance, as well as potential psychological discomfort and feelings of loneliness. The present study examines the psychological well-being of students at Duzce University in this regard. The study aims to categorize students' emotional states and identify those who could benefit from prompt psychological assistance. To this end, the study employs data mining and text mining techniques. This study was based on data previously collected from MODUM, an application used by Duzce University. To achieve this goal, the study integrates a range of data mining and text mining techniques for the classification of emotional states. The machine learning algorithms employed include Naive Bayes, Random Forest, Decision Tree (J48), Support Vector Machine (SVM), Artificial Neural Networks (ANN), and Logistic Regression. These algorithms were implemented across multiple software environments, such as Python, MATLAB, and R. Additionally, a variety of natural language processing techniques, including Bag of Words, TF-IDF, Word2Vec, and FastText, were used for effective text representation and preprocessing.