Machine learning Models for Mental Health Analysis based on Religious Impact
Date
2021-04-28
Authors
Ashraf, Waseem
Dontha, Krishna Sri
Kancheti, Tarun Kumar
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Once a taboo topic, recently we have seen a greater awareness of suicide and the factors that influence the suicide rate. There are many factors that influence the suicide rate, among these factors are religious affiliation and religious diversity. Most of the research on religion’s influence on the suicide rate has relied on the study of published articles retrieved from a plurality of databases and surveys performed on the selected population. In addition, hardly any research addresses the impact of religious diversity on the suicide rate. The present paper proposes studying the impact of religion on suicide using a quantitative approach. A data set containing the suicide rate, and religious affiliation rate of over 150 countries is constructed from 1990 to 2010. The countries that constituted a population with a single religion over a threshold percentage are identified as countries lacking religious diversity. The analysis indicates that different religions impact suicide differently. A baseline of suicide rate was generated using countries that are mostly affiliated with no religions. The preliminary research was limited to the top four religions of the world. Our research revealed that countries that are mostly affiliated with Christianity, Hinduism, and Buddhism had lower suicide rate compared to countries with no religious affiliation at all. Even the countries that are religiously diverse shield against suicide compared to countries that are not affiliated with any religion.
Description
Keywords
Machine learning, Mental health, Natural language processing