Prediction of depression using Machine Learning and NLP approach

: Today, for Internet users, micro-blogging has become a popular networking forum. Millions of people exchange views on different aspects of their lives. Thus, microblogging websites are a rich source of opinion mining data or Sentiment Analysis (SA) information. Because of the recent advent of microblogging, there are a few research papers dedicated to this subject. In our paper, we concentrate on Reddit.com, one of the leading microblogging sites, to explore the opinion of the public. We will demonstrate how to collect real-time Reddit data and use algorithms such as Support Vector Machine, KNN, and Multinomial Naive Bayes (MNB) for sentiment analysis or opinion mining purposes. We are able to assess positive and negative feelings using the algorithms selected above for the real-time Reddit depression info. The following experimental evaluations show that the algorithms used are accurate and can be used as an application for diagnosing the depression of individuals. After deployment User can write the content as input and after submitting the input text model API will be called from where result will come as User is going through depression or suicide. We worked with English in this post, but it can be used with any other language. English in this document, but it can be used for any other language this will be in future scope.