An Efficient Student Mental Health Analysis Method Based on Recurrent Neural Networks
TL;DR
This analysis uses three types of RNN methods such as long short-term memory (LSTM), gated recurrent neural network (GRU), and simple RNN (SRNN) for analyzing student mental health.
Abstract
Today's college students will undoubtedly experience stress. When under stress, a person may exhibit potent emotional and behavioral reactions. Stress-related mental health problems are among the most common causes of stress for college students worldwide. Research on the effects of particular activities, such as study trips, on students' mental health and how these activities might be tracked using cutting-edge technologies is lacking, though. Deep learning has lately found broad use in college students' mental health (SMH) education and management due to their capacity to evaluate, categorize, and notify psychological data with high quality. The present research work focused on three recurrent neural network (RNN) methods that were used for the analysis of student mental health issues. This analysis uses three types of RNN methods such as long short-term memory (LSTM), gated recurrent neural network (GRU), and simple RNN (SRNN) for analyzing student mental health. The dataset is collected from various engineering college students to analyze their SMH issues. Examining the experimental outcomes offered on test data based on f-score, recall, accuracy, and precision.