| Yazarlar (3) |
Dr. Öğr. Üyesi İbrahim ŞANLIALP
Kırşehir Ahi Evran Üniversitesi, Türkiye |
Elif Şanlıalp
Türkiye |
|
Tuncay Yiğit
Süleyman Demirel Üniversitesi, Türkiye |
| Özet |
| Predicting suicide risk is a critical issue for the future of public health. Failure to accurately predict suicide risk limits solutions to this major public health problem. The aim of the presented study is to predict the risk of suicide by using the machine learning approach in Turkey. This study uses the Turkish Statistical Institute’s public database for the prediction of suicide risk. The dataset consists of 30,811 patients committing suicide. Subject data includes all cities (81 cities) of Turkey and covers a 10-year period (2009–2018). Population information grouped by attributes in the data set is also taken from the Turkish Statistical Institute’s public database (for all cities in Turkey). The structured patient’s feature includes city, age-group, gender, and specific mortality rate. Multiple linear regression model is implemented and results indicate that age-group, gender, and city variables are promising success predictors of specific mortality rate in predicting future risk of suicide. (i.e., MAE: 0.0386959, RMSE: 0.0621640, R2: 0.5648034). The findings are expected to help suicide prevention rehabilitation programs and to assist developers in Machine learning-based suicide risk assessment tools. |
| Anahtar Kelimeler |
| Linear regression | Machine learning | Prevention | Suicide risk |
| Bildiri Türü | Tebliğ/Bildiri |
| Bildiri Alt Türü | Tam Metin Olarak Yayımlanan Tebliğ (Uluslararası Kongre/Sempozyum) |
| Bildiri Niteliği | Alanında Hakemli Uluslararası Kongre/Sempozyum |
| Bildiri Dili | İngilizce |
| Kongre Adı | ICAIAME'20 |
| Kongre Tarihi | 24-10-2020 / |
| Basıldığı Ülke | Türkiye |
| Basıldığı Şehir | Antalya |
| Atıf Sayıları |
