phat 22(5): e4

Research Article

Suicidal Behavior Prediction and Socioeconomic Suicide Indicators

Download211 downloads
  • @ARTICLE{10.4108/eetpht.v8i5.3175,
        author={Muhammad Nouman and Kareem Ullah and Muhammad Azam},
        title={Suicidal Behavior Prediction and Socioeconomic Suicide Indicators},
        journal={EAI Endorsed Transactions on Pervasive Health and Technology},
        volume={8},
        number={5},
        publisher={EAI},
        journal_a={PHAT},
        year={2022},
        month={12},
        keywords={Suicidal Behavior, Risk Factors, Socioeconomic Indicators},
        doi={10.4108/eetpht.v8i5.3175}
    }
    
  • Muhammad Nouman
    Kareem Ullah
    Muhammad Azam
    Year: 2022
    Suicidal Behavior Prediction and Socioeconomic Suicide Indicators
    PHAT
    EAI
    DOI: 10.4108/eetpht.v8i5.3175
Muhammad Nouman1,*, Kareem Ullah1, Muhammad Azam1
  • 1: University of Agriculture Faisalabad
*Contact email: m.nouman909@gmail.com

Abstract

INTRODUCTION: According to the WHO (World Health Organization), nearly 0.8 million people commit suicide each year, with more than 20 suicide attempts for every self-immolation. Suicidal behaviors have a profound effect on communities, societies, families, friends, and colleagues. After the recognition of public health as a priority by the WHO, various studies are being conducted to prevent it. OBJECTIVES: The investigation's goals were to improve understanding of suicide by identifying socioeconomic indicators correlated with rising suicide rates among divergent legions globally and to develop a prediction model for those who are at a higher risk of suicide by using different predictors of suicide such as tension, depression, anxiety, and so on. METHODS: We used a variety of data mining techniques to create a prediction model for suicide, including Logistic Regression, Multilayer Perceptron, Polynomial/Gaussian/Sigmoid SVM, Decision Tree, and K-Nearest Neighbors. For identifying socioeconomic suicide indicators, we used various descriptive and exploratory analysis techniques such as mean, regression, and correlation. RESULTS: Classification through the Gaussian Kernel - SVM has been shown to have the best results relative to others. Results also stated that many countries saw a decrease in suicide rates between 2006 and 2015, compared to 1996 to 2005. The highest concentrations have been reported in Europe, while the lower has been observed in South America. CONCLUSION: Things are improving, at least according to the statistics. The performance of Gaussian Kernel-SVM has been demonstrated to be superior to the other algorithms for suicide prediction. Data on suicide and suicide attempts are imprecise and difficult to gather. Suicide and suicide attempt monitoring, and surveillance must be improved for suicide prevention initiatives to be effective.