Detecting Depression in Tweets Using DistilBERT
Keywords:
Depression, Social Media, BERT, distilBERT, TweetsAbstract
Depression is a mood disorder that will affect a person's daily life. Depression can change life with all the suicidal thoughts, and the youth is straining a lot. Social media is growing tremendously day by day, and the applications like Twitter, Facebook are being used by youngsters. They share their opinions about their mood, and we can analyze a person's state of mind with those tweets written on Twitter. Our paper aims to detect that the person is depressed or not by using the Tweets using distilBERT, the distilled version of BERT. This distilBERT model is used to train the data and helps to achieve higher accuracy.
Downloads
References
David Owen, Jose Camacho-Collados, Luis Espinosa-Anke, ”Towards Preemptive Detection of Depression and Anxiety in Twitter”, December 12, 2020.
Diveesh Singh, Aileen Wang,” https://web.stanford.edu/class/archive/cs/cs224n/cs224n.1184 /reports/6879557.pdf”.
Victor Sanh,” Smaller, faster, cheaper, lighter: Introducing DistilBERT, a distilled version of BERT”, August 28, 2019. [4] Rani Horev, ”BERT Explained: State of the art language model for NLP”, Nov 10, 2018
Tulasiram, ”Detecting Depression in Social Media via Twitter usage”, Oct 30, 2020
Cristian Bucila, Rich Caruana, Alexandru Niculescu-Mizil, “Model Compression”.
Geoffrey Hinton, Oriol Vinyals, Jeff Dean, “Distilling the Knowledge in a Neural Network”.
