JMIR Publications

ABSTRACT

Background:

Depression detection has recently received attention in the field of natural language processing. The task aims to detect depressed users based on their historical posts on social media. Existing works are mainly divided into using all historical posts, and selecting depression indicator posts. However, these methods fail to effectively extract deep emotional semantic features or simply concatenate emotional representation. To solve this problem, we propose a model to extract deep emotional semantic features and select indicator posts based on the emotional states.

Objective:

We developed an emotion-based reinforcement attention network for depression detection of users on social media.

Methods:

The proposed model is composed of two components: the emotion extraction network used to capture deep emotional semantic information, and the RL attention network used to select depression indicator posts based on the emotional states. Finally, we concatenate the output of these two parts and send them to the classification layer for depression detection.

Results:

Experimental results of our model on the MDDL dataset outperforms the state-of-the-art baselines. Specifically, the proposed model achieves accuracy, precision, recall, and F1-score by 90.6%, 91.2%, 89.7%, and 90.4%, respectively.

Conclusions:

The proposed model utilizes historical posts of users for depression detection that can effectively identify users’ depression tendencies. The experimental results show that the emotion extraction network and the RL selection layer based on emotional states can effectively improve the accuracy of detection. In addition, sentence-level attention layer can capture core posts.