Abstract
Depression is a major global public health challenge and its early identification is crucial. Social media data provides a new perspective for depression detection, but existing methods face limitations such as insufficient accuracy, insufficient utilization of time series features, and high annotation costs. To this end, this study proposes the LMILAtt model, which innovatively integrates Long Short-Term Memory autoencoders and attention mechanisms: firstly, the temporal dynamic features of user tweets (such as depressive tendency evolution patterns) are extracted through unsupervised LSTM autoencoders. Secondly, the attention mechanism is used to dynamically weight key texts (such as early depression signals) and construct a multi-example learning architecture to improve the accuracy of user-level detection. Finally, the performance was verified on the WU3D dataset labeled by professional medicine. Experiments show that the model is significantly better than the baseline model in terms of accuracy, recall and F1 score. In addition, the weakly supervised learning strategy significantly reduces the cost of labeling and provides an efficient solution for large-scale social media depression screening.
Abstract (translated)
抑郁症是全球公共健康面临的主要挑战，其早期识别至关重要。社交媒体数据为抑郁检测提供了新的视角，但现有方法存在诸如准确性不足、时间序列特征利用不充分以及标注成本高等限制。为此，本研究提出了一种名为LMILAtt的模型，该模型创新性地结合了长短期记忆自动编码器（LSTM autoencoders）和注意力机制：首先，通过无监督的LSTM自动编码器提取用户推文的时间动态特征（例如抑郁倾向演变模式）。其次，利用注意力机制对关键文本进行动态加权（如早期抑郁症信号），并构建多例学习架构以提高用户级别检测的准确性。最后，在由专业医学人员标注的WU3D数据集上验证了模型性能。实验表明，该模型在准确率、召回率和F1值方面均显著优于基准模型。此外，弱监督学习策略大幅降低了标注成本，并为大规模社交媒体抑郁筛查提供了高效的解决方案。
URL
https://arxiv.org/abs/2509.26145
