• Mihir Parte
• Devansh Choudhury
• Varun Sovani
• Priyanka Shahane

Nowadays, the use of social media for self-expression has increased widely with growing social media platforms. It could share facts and opinions, record events, update daily living, and more. Social media and virtual communication are gradually replacing in-person conversations. Teenagers’ active social media usage is one facet of these shifting dynamics. Today’s youth regularly use the microblogging platform Twitter. The enormous amount of tweets and the different variables related to tweeting, retweeting, and commenting make behavioural research and human sentiment analysis possible. Research on Twitter user behaviour and mood swings is desperately needed in this large and largely uncharted topic. Among these is the field of depression detection. Treatment for depression necessitates early detection to take preventative action to lessen or mitigate problems—especially when lives are at stake. Despite this, many proposals only consider a reactive solution, ignoring the temporal problem. This research represents the initial attempts to develop a model to identify depression. In our proposed system, we have used the Twitter dataset for depression detection. We have selected the Bidirectional Encoder Representations from Transformers (BERT) & Convolutional Neural Networks (CNN) as these two algorithms have not been tested before for the problem of depression detection. We have incorporated advanced pre-processing techniques such as Stemming, Lemmatization, PoS tagging, & Stop word removal to achieve higher accuracy.

Depression Detection, Deep Learning, Machine Learning, Natural Language Processing, Bidirectional Encoder Representations from Transformers (BERT) & Convolutional Neural Network (CNN).

"Depression Detection from Social Media using Bidirectional Encoder Representations from Transformers & Convolutional Neural Network", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 5, page no.o71-o79, May 2024, Available :http://www.jetir.org/papers/JETIR2405F07.pdf

"Depression Detection from Social Media using Bidirectional Encoder Representations from Transformers & Convolutional Neural Network", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 5, page no. ppo71-o79, May 2024, Available at : http://www.jetir.org/papers/JETIR2405F07.pdf