Pengfei Deng


2024

pdf
ESDM: Early Sensing Depression Model in Social Media Streams
Bichen Wang | Yuzhe Zi | Yanyan Zhao | Pengfei Deng | Bing Qin
Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)

Depression impacts millions worldwide, with increasing efforts to use social media data for early detection and intervention. Traditional Risk Detection (TRD) uses a user’s complete posting history for predictions, while Early Risk Detection (ERD) seeks early detection in a user’s posting history, emphasizing the importance of prediction earliness. However, ERD remains relatively underexplored due to challenges in balancing accuracy and earliness, especially with evolving partial data. To address this, we introduce the Early Sensing Depression Model (ESDM), which comprises two modules classification with partial information module (CPI) and decision for classification moment module (DMC), alongside an early detection loss function. Experiments show ESDM outperforms benchmarks in both earliness and accuracy.

2023

pdf
C2D2 Dataset: A Resource for the Cognitive Distortion Analysis and Its Impact on Mental Health
Bichen Wang | Pengfei Deng | Yanyan Zhao | Bing Qin
Findings of the Association for Computational Linguistics: EMNLP 2023

Cognitive distortions refer to patterns of irrational thinking that can lead to distorted perceptions of reality and mental health problems in individuals. Despite previous attempts to detect cognitive distortion through language, progress has been slow due to the lack of appropriate data. In this paper, we present the C2D2 dataset, the first expert-supervised Chinese Cognitive Distortion Dataset, which contains 7,500 cognitive distortion thoughts in everyday life scenes. Additionally, we examine the presence of cognitive distortions in social media texts shared by individuals diagnosed with mental disorders, providing insights into the association between cognitive distortions and mental health conditions. We propose that incorporating information about users’ cognitive distortions can enhance the performance of existing models mental disorder detection. We contribute to a better understanding of how cognitive distortions appear in individuals’ language and their impact on mental health.