Investigation of the Unemployment Based on Poor Mental Conditions

Abstract

Psychological wellness is essential for everyone in society since it fosters personal growth and success. In a variety of ways, people's work status is linked to their psychological well-being. Individuals who become unemployed are more likely than those who stay employed to suffer from poor psychological health. Unemployed people who have poor psychological conditions are less inclined to look for jobs. As a result, while work has a noteworthy impression on one's mental security, one's mental health determines whether or not one finds work. To minimize this burning issue, it’s necessary to find out the practical solutions through developed techniques. This study focus to address this issue through secondary data. This data was collected from 334 individuals by Michael Corley in 2019. The logistic regression model is used as a baseline model. In contrast, the Support Vector Machine, Decision Tree, and Random Forest models were chosen to check their performance in the study field. The study focuses on five mental health explanatory factors: namely, sadness, tiredness, obsessive thinking, panic attacks, and anxiety. The Support Vector Machine model outperforms logistic regression, Decision Tree, and Random Forest when predicting unemployment based on poor mental health. The Support Vector Machine produced a 71% predictive accuracy, and anxiety significantly influenced unemployment prediction. Future research should focus on detecting destructive mental health issues early on is critical to avoid undesirable actions. These conditions can be lessened with technology, and we can prevent individuals from adopting drastic measures that further safeguard their future.