In recent times, significant research is carried out in the fields of psychology and computer science regarding the detection and modeling of stress. The foremost step in identifying early signs of stress - related issues involves assessing and monitoring sleep quality, understanding the impact of stress on physical health. Accurately detecting stress during sleep can provide valuable insights into an individual’s mental and physical health. Machine learning techniques are used in analyzing the sleep patterns and early diagnosis of stress. The goal of this study is to develop an AI model for predicting the detection of stress and to potentially substitute the supervised machine learning classification models. The performance of various supervised algorithms like Decision Tree, Adaboost, SVM and MLP are compared and the results are evaluated in terms of accuracy.