- El-Cezeri
- Vol: 9 Issue: 4 Özel Sayı
- Detection of Post Traumatic Stress Disorder with Deep Learning Methods
Detection of Post Traumatic Stress Disorder with Deep Learning Methods
Authors : Engin Seven, Cansın Turguner, Muhammed Ali Aydin
Pages : 1274-1281
Doi:10.31202/ecjse.1133463
View : 11 | Download : 4
Publication Date : 2022-12-31
Article Type : Research
Abstract :Post-traumatic stress disorder (PTSD) is a psychiatric problem that negatively affects a person\'s mental and physical life after a traumatic event. If the disease is not recognized and treated at an early stage, negative consequences such as bipolar disorder, anxiety or suicidality can occur. An artificial intelligence-based model has been developed for the early detection of PTSD. In the study, K-Nearest Neighbor algorithm, Support Vector Machines, Decision Trees, Gaus Naive Bayes and Artificial Neural Networks were used and tests were carried out on the dataset collected from medical students during the Covid-19 pandemic. In the study; accuracy, precision, recall and f1 score values were examined comparatively. Artificial neural networks achieved the best result with an accuracy rate of 0,987. In addition, artificial neural networks found the best PTSD prediction with an f1 score of 0,966.Keywords : Travma sonrası stres bozukluğu, Derin Öğrenme, Akıllı teşhis, Özellik Seçimi