Hybrid Deep Learning Models for Efficient Detection of Depressive Disorders from Twitter Data

Social media occupy an important part of people’s lives through which they share their experiences, feelings and interests. This leads to the generation of huge volumes of data, especially in textual form. Several methods for natural language processing tasks have been used so as to extract valuable information from social media. In particular, various deep learning approaches have been proposed to study depression detection through twitter data analysis. In this respect, this study proposes a well-defined methodology for building efficient models that allow predicting depression through Tweets analysis. The idea consists in devising an efficient process for combining different types of neural networks, in order to improve the general performances. Once the training phase performed, we proceed to a deployment phase in which the well-performing variants are used to analyse the sentiments of Twitter users in UK and US during COVID-19 period toward depressive disorders behaviours. The proposal is validated through several experiments and comparisons according to some evaluation metrics. Overall, the obtained results are satisfactory and encouraging.