Sadeque, Farig YousufNoman, Mirza Abdullah AlKhan, Waleed Bin HabibChowdhuruy, Pratick Roy2023-12-052023-12-0520232023-05ID 19301244ID 19301178ID 19301065http://hdl.handle.net/10361/21915Cataloged from PDF version of thesis.Includes bibliographical references (pages 43-48).This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science, 2023.In our world, 3.8% of the total population is suffering from depression and it is the fourth major cause of death in 15-29 year olds. It is estimated that more than 75% of people suffering from it in low and middle income countries receive no treatment. Also, in these countries, so many people live with such conditions without even recognizing it because of the lack of proper diagnosis and mental health facilities. However, a huge chunk of the population is connected and active on different social media platforms. Detection of depression from social media activities can help in recognizing the problems in an individual level and in a public health level to know its prevalence in different demographics. The early prediction of such can help us to work on the problem before the onset. In our work, we propose to use state of the art machine learning and deep learning models to provide an efficient early detection model for diagnosis of such. We hope that it would help individuals and relevant authorities to find out the illness and its severity for the betterment of global and regional mental health.48 pagesenBrac University theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission.Natural language processingNeural networkDepressionMajor depressive disorderEarly risk detectionTransformerPsycholinguisticsNatural language processing (Computer science)Neural networks (Computer science)Thesis