Real Time Depression and Anxiety Detection Using Machine Learning

Abstract

Depression affects a staggering 264 million individuals worldwide, constituting a significant cause of disability. The detrimental impact of a negative workplace environment extends beyond productivity loss to encompass physical and medical ailments. Unfortunately, individuals often refrain from seeking help due to the pervasive stigma surrounding mental health issues. Leveraging the potential of machine learning, we have embarked on a journey to predict depression using diverse algorithms. Our study draws upon routine survey data, delving into factors such as home and workplace environments, family history of mental illness, among others. Recognizing and understanding the mental state of individuals, be it stress, anxiety, or depression, holds paramount importance in averting untoward incidents. Recent events, such as economic downturns, pandemic-induced fears, and social isolation, have contributed to a surge in depression and anxiety cases. Furthermore, there is compelling evidence indicating heightened social media usage among individuals with mental health disorders. Thus, we delve into the potential of online personas on social media platforms. Our work presents a comprehensive review of various methodologies employed in the literature for detecting depression, thereby shedding light on emerging trends and challenges. By identifying gaps in existing research, we aim to provide fresh insights and directions for researchers committed to advancing the field of depression detection. Key Words: Depression Detection, Machine Learning, Support Vector Machine (SVM), Accuracies, Human Being.