Real Time Detection of Depression in Social Media Using Regex

Abstract

Depression is a severe mental condition that impacts people everywhere, regardless of age, gender, caste, or religion. Since social media sites make it easier for people to express their opinions, many spend nearly their whole day there. With the use of user posts uploaded on a social networking website, this study aims to investigate a data model for identifying sorrow. On this project, we offered a data model based on user datasets in the social media platform of all social media websites. The dataset's social media postings must be used to estimate the user's depression levels. A technique that comprises data validation, data preparation, and training the model using user test data to predict depression levels is used to identify individual depression. With the use of data models created from people's tweets, we will categorise persons with clinical depression and symptoms associated with it. In this project, we'll create a machine learning system to assess the severity of depression using user information from social networking sites. The Support Vector Machine (SVM) and Nave Bayes algorithms were used with Natural Language Processing (NLP) to diagnose depression in the simplest and most effective manner.