Depression Analysis of Social Media Activists Using the Gated Architecture Bi-LSTM

Twitter is one of the social media platforms that has evolved into an incredible environment for users to communicate with friends and other users to trade thoughts, videos, and photographs that reflect their present mood. Using social media allows academics to investigate people’ online data, revealing their moods and habits, and analyze their mental states. Since social media is becoming a part of our life, this study implies users’ moods by what they post in their tweets. This study will analyze Twitter data for depression sentiment using a verified public internet portal. Social information is presented as an important predictor of depression. This study suggests distinct deep learning classification strategies which can perform as a competent and adaptable strategy and have already been effectively utilized in Natural Language Processing jobs to explore the impact of depression diagnosis on Twitter. LSTM and Convolutional Neural Networks were used to analyze Twitter data for depression analysis. Bi-LSTM proven the highest accuracy of 95% for real time depression analysis. We also developed a depression analysis tool based on this study.