X-A-BiLSTM: a Deep Learning Approach for Depression Detection in Imbalanced Data

Abstract

An increasing number of people suffering from mental health conditions resort to online resources (specialized websites, social media, etc.) to share their feelings. Early depression detection using social media data through deep learning models can help to change life trajectories and save lives. But the accuracy of these models was not satisfying due to the real-world imbalanced data distributions. To tackle this problem, we propose a deep learning model (X-A-BiLSTM) for depression detection in imbalanced social media data. The X-A-BiLSTM model consists of two essential components: the first one is XGBoost, which is used to reduce data imbalance; and the second one is an Attention-BiLSTM neural network, which enhances classification capacity. The Reddit Self-reported Depression Diagnosis (RSDD) dataset was chosen, which included approximately 9,000 users who claimed to have been diagnosed with depression (”diagnosed users and approximately 107,000 matched control users. Results demonstrate that our approach significantly outperforms the previous state-of-the-art models on the RSDD dataset.