Performance Analysis of SVM and Random Forest Algorithms in the Case of the Influence of Music on Mental Health

Abstract

Mental health disorders are conditions that impress a person's behavior, mindset, and emotions. According to WHO data, the rate of mental disorders in Asia has increased significantly in the past two decades, with about one-fifth of the world's adolescent population experiencing stress each year. Music has long been known to have a positive influence on mental health, and music therapy is used as one approach to assist individuals in improving social, mental, and physical conditions. In this study, the authors used data mining techniques to identify relevant patterns regarding the influence of music on mental health. Two classification algorithms, namely the Support Vector Machine (SVM) and Random Forest, is used to analyze and characterize the data. SVM is known to excel at managing high-dimensional data, while Random Forest is effective at handling data with missing outliers and features. This study purpose to oppose the performance of the two algorithms in classifying the influence of music on mental health to identify the superior algorithm in this context. The Random Forest algorithm gets 93% accuracy and SVM gets 95% accuracy, the hyperparameter tuning on the SVM algorithm has a better performance than Random Forest with an accuracy score of 97% for SVM, while for Random Forest it gets an accuracy score of 94%. The results of the study are expected to provide insight into the use of music as a mental health therapy tool.