Evaluating the High Risk Groups for Suicide: A Comparison of Logistic Regression, Support Vector Machine, Decision Tree and Artificial Neural Network
Abstract
Background: We aimed to assess the high-risk group for suicide using different classification methods includinglogistic regression (LR), decision tree (DT), artificial neural network (ANN), and support vector machine (SVM).
Methods: We used the dataset of a study conducted to predict risk factors of completed suicide in Hamadan Province, the west of Iran, in 2010. To evaluate the high-risk groups for suicide, LR, SVM, DT and ANNwere performed. The applied methods were compared using sensitivity, specificity, positive predicted value, negative predicted value, accuracy and the area under curve. Cochran-Q test was implied to check differences in proportion among methods. To assess the association between the observed and predicted values, Ø coefficient, contingency coefficient, and Kendall tau-b were calculated.
Results: Gender, age, and job were the most important risk factors for fatal suicide attempts in common for four methods. SVM method showed the highest accuracy 0.68 and 0.67 for training and testing sample, respectively. However, this method resulted in the highest specificity (0.67 for training and 0.68 for testing sample) and the highest sensitivity for training sample (0.85), but the lowest sensitivity for the testing sample (0.53). Cochran-Q test resulted in differences between proportions in different methods (P<0.001). The association of SVM predictions and observed values, Ø coefficient, contingency coefficient, and Kendall tau-b were 0.239, 0.232 and 0.239, respectively.
Conclusion: SVM had the best performance to classify fatal suicide attempts comparing to DT, LR and ANN.
Simon M, Chang E, Zeng P, Dong X (2013). Prevalence of suicidal ideation, attempts, and completed suicide rate in Chinese aging populations: a systematic review. Arch Gerontol Geriatr, 57(3):250-6.
Organization WHO (2009). Preventing suicide: a resource for police, firefighters, and other first line responders.
www.who.int/mental_health/resources/preventingsuicide/en/
Reddy M (2010). Suicide incidence and epidemiology. Indian J Psychol Med, 32(2):77.
Bertolote JM, Fleischmann A (2009). Oxford Textbook of Suicidology and Suicide Prevention.In: A global perspective on the magnitude of suicide mortality. Oxford University Press, UK. pp.: 91-8.
Organization WHO (2014) . Preventing suicide: a global imperative. Available from: www.who.int/mental_health/suicide-prevention/world_report_2014/en/
Naghavi M, Jafari N (2007). Mortality of 29 provinces-2004. Tehran: Ministry of Health and Medical Education:(Persian).
Poorolajal J, Esmailnasab N, Ahmadzadeh J, Motlagh TA (2012). The Burden of Premature Mortality in Hamadan Province in 2006 and 2010 Using Standard Expected Years of Potential Life Lost: A Population-based Study. Epidemiol Health, 34:e2012005.
Memari AM, Ramim T, Amirmoradi F, Khosravi K, Godarzi Z (2006). Causes of suicide in married women. Hayat. [Research], 12(1):47-53.
Mohammadkhani P (2004). Epidemiology of Suicide thoughts and attempts in young girls from highrisk regions in Iran. Social Wellfare, 13(4):(persian).
Hajebi A, Ahmadzad Asl M, Zaman M, Naserbakht M, Mohammadi N, Davoudi F, et al (2011). Designing a Registration System for Suicide in Iran. Ira J Psychiatr Clin Psychol, 17(2):106-9.
Han J, Kamber M (2006). Data Mining, Southeast Asia Edition: Concepts and Techniques. Morgan kaufmann.
Pang-Ning T, Steinbach M, Kumar V (2006). Introduction to data mining. Pearson.
Kantardzic M (2011). Data mining: concepts, models, methods, and algorithms. John Wiley & Sons.
Witten IH, Frank E (2005). Data Mining: Practical machine learning tools and techniques. Morgan Kaufmann.
Jenhani I, Amor NB, Elouedi Z (2008). Decision trees as possibilistic classifiers. Int J Approx Reason, 48(3):784-807.
Moreno JJM, Pol AP, Gracia PM (2011). Artificial neural networks applied to forecasting time series. Psicothema, 23(2):322-9.
Delen D, Oztekin A, Tomak L (2012). An analytic approach to better understanding and management of coronary surgeries. Decis Support Syst, 52(3):698-705.
Amiri B, Pourreza A, Rahimi Foroushani A, Hosseini SM, Poorolajal J (2012). Suicide and associated risk factors in Hamadan Province, West of Iran, in 2008 and 2009. J Res Health Sci, 12(2):88-92.
Lin TH (2010). A comparison of multiple imputation with EM algorithm and MCMC method for quality of life missing data. Qual Quant, 44(2):277-87.
Ayat S, Farahani HA, Aghamohamadi M, Alian M, Aghamohamadi S, Kazemi Z (2013). A comparison of artificial neural
| Files | ||
| Issue | Vol 45 No 9 (2016) | |
| Section | Original Article(s) | |
| Keywords | ||
| Suicide Support vector machine Neuralnetworks Logistic regression Decision tree Classification | ||
| Rights and permissions | |
|
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
