Abstract
Obsessive-compulsive disorder (OCD) is extremely common, but early detection is difficult because symptoms do not appear until puberty. Therefore, it is crucial to identify the causes of this mental illness. Making an early and accurate diagnosis of OCD in children and adolescents is essential to preventing the long-term problems. Several studies have looked at ways to recognise OCD in children, but their accuracy was not very high and they only included a few features and participants. Therefore, the purpose of this study was to examine the detection of OCD utilising machine learning algorithms and 667 features from Young Minds Matter (YMM), Australia’s nationally representative mental health survey of children and adolescents aged 4 to 17 years. According to the internal CV score of the Tree-based Pipeline Optimization Tool (TPOTClassifier), the performance of the suggested technique has been evaluated on the YMM dataset using three of the most optimal algorithms, including Random Forest (RF), Decision Tree (DT), and Gaussian Naïve Bayes (GaussianNB). GaussianNB outperformed all other methods in classifying OCD with 91% accuracy, 76% precision, and 96% specificity, despite significant variation in model performance.