Background: Sentiment analysis, a subset of natural language processing (NLP) and computer science, uses machine learning and artificial intelligence to comprehend human emotions and feelings. One critical area of focus in behavioral sciences and psychology is identifying stress levels. With the increasing prevalence of social networks, the content shared on these platforms has emerged as a valuable resource for assessing societal mental health, potentially yielding significant advancements in behavioral sciences and psychology. Consequently, employing content analysis techniques, particularly for textual data, can greatly assist mental health professionals in diagnosis and treatment processes. Specifically, automated stress detection through text analysis is a vital and impactful topic at the intersection of NLP and psychology. This research primarily investigates the identification of user stress levels based on text messages shared on social media, utilizing a combination of deep learning (DL) approaches.
Aim: The primary objective of this study is to develop a system for early detection of stress-related disorders in individuals through the application of artificial intelligence techniques. We aim to identify signs of stress from the text content shared by people in their online posts, providing valuable support to psychotherapists in diagnosing stress-related disorders and enhancing their focus on these issues. We utilized Dreaddit dataset as well as natural language processing , machine learning (ML), and deep learning algorithms for text classification. From a practical perspective, the outcome of this research is an intelligent psychotherapist assistant software system capable of analyzing and detecting stress within text. The outputs of this system will serve as a valuable resource for mental health practitioners. Furthermore, the secondary results stemming from this research, which will arise from extensive experiments and the evaluation of diverse methodo