Forecast of Students’ Mental Health Combining an Artificial Intelligence Technique and Fuzzy Inference System

Mental health prediction is one of the most necessary parts of decreasing the probability of severe mental illness. In the meantime, mental health prediction could present a hypothetical base for the Department of Public Health to work out mental interference tactics for students. This paper intends to forecast students’ mental health using Artificial Intelligence (AI). This paper uses the Survey Student Responses dataset from the Kaggle repository, 19 factors of 1182 students of different age groups from different educational institutions in Delhi National Capital Region (NCR) through a cross-sectional survey, with the outcomes of the Self-reporting Inventory, was used to distinguish mental health. This paper proposes a new forecasting technique using the Mamdani Fuzzy Inference System based on the Min-Max Method (for data labelling) and Pearson’s correlation coefficient (for feature selection), which could choose and order the very significant features that impact the student’s mental health. In addition, we utilize Multinomial Logistic Regression (MLR) and Multilayer Perceptron (MLP) (for ML) algorithm and the Dl4jMlpClassifier (for DL) algorithm to forecast the mental health of students. The outcomes demonstrate that the prediction accuracy of the proposed technique is high, which is superior to the previous algorithms. Therefore, this technique could be used to forecast the student’s mental health efficiently.