부산대학교 도서관

India is one of the countries in the world with the highest rates of mental illness. As a consequence of the COVID-19 pandemic, 20 percent of the population is likely to be affected by some kind of mental health disorder. Around 56 million Indians struggle with depression, and 38 million have an anxiety illness of some kind. Although there is a clear need of assistance with such issues, there are only 3 psychiatrists and psychologists for every 100000 people in India. In the age of Artificial Intelligence, attempts are being made to develop smart systems that can aid medical professionals to cater to the increasing number of patients requiring assistance with mental health issues. The automatic extraction of positive or negative attitude expressions from text, known as sentiment analysis, has drawn a lot of attention from researchers in the last ten years. This technique is currently being used by social networking companies like Twitter, Instagram, Facebook, etc. to identify how their users are feeling based on their comments and posts in order to present them with targeted content in the form of advertisements and promotional posts. The same techniques can be extended to predict moods of individuals on a daily basis and use the records to depict the alterations over different periods of time. This can be of great use to individuals and healthcare professionals to keep track of behavioral changes and maintain a routine of self check-ins. In this paper, we have made use of the Long Short Term Memory model, which is a deep learning approach, to carry out sentiment analysis and have integrated it into a web application which is an artificially intelligent digital diary called’ Amigo: My Virtual Friend.’