This is the authors' abstract. We don't add key points for this paper.
Depression is a major public health concern and a leading cause of disability. Globally, more than 332 million people of all ages suffer from depression. Several studies in the literature show that people language usage is indicative of their psychological states. That is why, there is growing interest in the application of natural language processing techniques for predicting depression. In this current work, we present a comparative study of different machine learning methods and alternative ways of representing documents in order to automatically detect users who have reported to have been diagnosed with depression. The obtained results have demonstrated that a Deep Learning approach had the superior classification performance, when combined with a Synthetic Minority Oversampling Technique to deal with the problem of class imbalances in the dataset used in our experiments. The F1 score achieved was 82.93% with an accuracy of more than 94%.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Dondena et al. (2017) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: