Rethinking Inconsistent Context and Imbalanced Regression in Depression Severity Prediction

被引:0
|
作者
Huang, Guanhe [1 ]
Li, Jing [2 ]
Lu, Heli [3 ]
Guo, Ming [3 ]
Chen, Shengyong [2 ]
机构
[1] Nanchang Univ, Sch Informat Engn, Nanchang 330031, Jiangxi, Peoples R China
[2] Tianjin Univ Technol, Sch Comp Sci & Engn, Tianjin 300384, Peoples R China
[3] Nanchang Univ, Affiliated Hosp 2, Dept Psychosomat Med, Nanchang 330006, Jiangxi, Peoples R China
基金
中国国家自然科学基金;
关键词
Depression severity prediction; natural language processing; topic modeling; conditional variational auto-encoder; imbalanced regression; attention mechanism; VARIATIONAL INFERENCE;
D O I
10.1109/TAFFC.2024.3405584
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
As one of the world's most prevalent mental illnesses, depression is not easy to detect since it affects different people in different ways. Recently, linguistic features extracted from transcribed texts have been widely explored in depression detection because they contain a variety of cues about psychological activities. However, the detection performance is limited due to the following two reasons: 1) the dialogue structure is ignored, which causes the Inconsistent Context problem; and 2) Imbalanced Regression occurs due to the long-tailed distribution of depression datasets. To this end, in this paper we investigate the relationship between the local topic and global context in interview transcripts, and bridge the gap between depression symptoms and depression severity. In particular, we propose a model called Conditional Variational Topic-enriched Auto-Encoder (CVTAE), which can capture the spatial features from local topics via variational inference, and the temporal features from the global context with attention mechanism. Besides, we apply the re-weighting strategies to assigning weights to the depression labels with different values. Extensive experiments on the DAIC-WOZ dataset in English and a self-constructed database NCUDID in Chinese demonstrate the effectiveness and robustness of CVTAE, while the comprehensive ablation study and case study show its interpretability.
引用
收藏
页码:2154 / 2168
页数:15
相关论文
共 50 条
  • [31] Investigating the Role of Clustering in Construction-Accident Severity Prediction Using a Heterogeneous and Imbalanced Data Set
    Salarian, Ali Akbar
    Etemadfard, Hossein
    Rahimzadegan, Ali
    Ghalehnovi, Mansour
    JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT, 2023, 149 (02)
  • [32] Establishment and validation of a logistic regression model for prediction of septic shock severity in children
    Han, Yujie
    Kang, Lili
    Liu, Xianghong
    Zhuang, Yuanhua
    Chen, Xiao
    Li, Xiaoying
    HEREDITAS, 2021, 158 (01)
  • [33] Prediction of insomnia severity based on early maladaptive schemas: a logistic regression analysis
    Vand, Hoda Doos Ali
    Hooman, Farzaneh
    Sardarzehi, Reyhaneh
    Bastami, Malek
    Jansson-Frojmark, Markus
    SLEEP AND BREATHING, 2024, 28 (02) : 919 - 927
  • [34] Establishment and validation of a logistic regression model for prediction of septic shock severity in children
    Yujie Han
    Lili Kang
    Xianghong Liu
    Yuanhua Zhuang
    Xiao Chen
    Xiaoying Li
    Hereditas, 158
  • [35] A Metamodel Structure For Regression Analysis: Application To Prediction Of Autism Spectrum Disorder Severity
    Wang, Shiyu
    Dvornek, Nicha C.
    2021 IEEE 18TH INTERNATIONAL SYMPOSIUM ON BIOMEDICAL IMAGING (ISBI), 2021, : 1338 - 1341
  • [36] Correction: Corrigendum: Rethinking Indian monsoon rainfall prediction in the context of recent global warming
    Bin Wang
    Baoqiang Xiang
    Juan Li
    Peter J. Webster
    Madhavan N. Rajeevan
    Jian Liu
    Kyung-Ja Ha
    Nature Communications, 6
  • [37] Multiattribute Based Machine Learning Models for Severity Prediction in Cross Project Context
    Sharma, Meera
    Kumari, Madhu
    Singh, R. K.
    Singh, V. B.
    COMPUTATIONAL SCIENCE AND ITS APPLICATIONS - ICCSA 2014, PT V, 2014, 8583 : 227 - +
  • [38] A Novel Feature Selection Method Based on Maximum Likelihood Logistic Regression for Imbalanced Learning in Software Defect Prediction
    Bashir, Kamal
    Li, Tianrui
    Yahaya, Mahama
    INTERNATIONAL ARAB JOURNAL OF INFORMATION TECHNOLOGY, 2020, 17 (05) : 721 - 730
  • [39] Prediction of Depression Severity Scores Based on Functional Connectivity and Complexity of the EEG Signal
    Mohammadi, Yousef
    Moradi, Mohammad Hassan
    CLINICAL EEG AND NEUROSCIENCE, 2021, 52 (01) : 52 - 60
  • [40] An Affect Prediction Approach through Depression Severity Parameter Incorporation in Neural Networks
    Gupta, Rahul
    Sahu, Saurabh
    Espy-Wilson, Carol
    Narayanan, Shrikanth
    18TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2017), VOLS 1-6: SITUATED INTERACTION, 2017, : 3122 - 3126