Predicting Generalized Anxiety Disorder Among Female Students Using Random Forest Approach

Mental health is considered one of the major challenges for the generations. Generalized anxiety disorder (GAD) is one of many mental health complications. However, individuals with the disorder experience hyperbolic concerns and tensions regarding daily events. Furthermore, it is reported that approximately 5% of the population of developed countries suffer from GAD. Additionally, women are affected by this disease twice as often as men, and it is an increasing disorder among women, particularly female students. This paper aims to predict generalized anxiety disorder among female students using the random decision forest algorithm. The data mining method was utilized for prediction. Female students of Shiraz Azad University developed the research community. Therefore, 150 female students were selected by simple random method and tested with a DSM-IV questionnaire. Accordingly, a random forest algorithm is proposed to generate a prediction model. Moreover, NetBeans IDE was applied for operationalization. Java was the programming language to code the prototype, and the WEKA library was involved in the operation. However, the results showed that the prediction accuracy with the random forest algorithm exceeds 0.9, which indicates that the algorithm is likely to predict GAD accurately. The random decision forest algorithm consistently predicts an individual not suffering from GAD. The results are relatively consistent compared to the baseline employed in the R. However, the random decision forest algorithm produces high predictive performance and may display significant relationships between the proposed and dependent parameters.