Trusted commonsense knowledge enhanced depression detection based on three-way decision

Depression detection on social media aims to identify depressive tendencies within textual posts, providing timely intervention by the early detection of mental health issues . In predominant approaches, the Pre-trained Language Models(PLMs) are trained solely on public datasets, falling short of vertical scenarios due to insufficient domain-specific and commonsense knowledge . In addition, ambiguous commonsense knowledge could be misleading to PLMs and results in false judgments . Therefore, it poses significant challenges to select commonsense knowledge that is trusted. To address this, we propose CoKE, a model that incorporates trusted commonsense knowledge based on three-way decision theory to enhance depression detection. CoKE comprises three key modules: trusted screening, knowledge generation, and knowledge fusion. First, we utilize psychiatric clinical scales and three-way decision theory to screen out the uncertain domain from the massive user posts. Then, an adaptive framework is applied to generate and refine trusted commonsense knowledge that can explain the true semantics of posts in the uncertain domain. Finally, a dynamic integration of posts with highly trusted knowledge is achieved through a gating mechanism, resulting in embeddings enhanced by trusted commonsense knowledge that are more effective in determining depressive tendencies. We evaluate our model on two prominent datasets, eRisk2017 and eRisk2018, demonstrating its superiority over previous state-of-the-art baseline models . • We enhance social media posts with trusted commonsense knowledge to improve the accuracy of depression detection. • We employ clinical scales and three-way decision theory to screen out the parts from the vast number of user posts that are most in need of enhancement. • We utilize an adaptive framework to generate and select commonsense knowledge for the parts that most require enhancement, ensuring the trustworthiness of the knowledge. • We dynamically adjust weights using gating and attention mechanisms, allowing the model to focus on the most critical parts. • Through experiments, we demonstrate that the trustworthiness of knowledge significantly impacts the accuracy of depression detection.