A Multi-Modal Hierarchical Recurrent Neural Network for Depression Detection
Open Access
- 15 October 2019
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 65-71
Abstract
We propose a multi-modal method with a hierarchical recurrent neural structure to integrate vision, audio and text features for depression detection. Such a method contains two hierarchies of bidirectional long short term memories to fuse multi-modal features and predict the severity of depression. An adaptive sample weighting mechanism is introduced to adapt to the diversity of training samples. Experiments on the testing set of a depression detection challenge demonstrate the effectiveness of the proposed method. CCS CONCEPTS • Computing methodologies → Supervised learning by regression.Keywords
Funding Information
- National Science Foundation of China (917418129)
- National Key R&D Program of China (2018YFB1307102)
This publication has 28 references indexed in Scilit:
- Multi task sequence learning for depression scale prediction from video2015
- Automated Depression Diagnosis Based on Facial Dynamic Analysis and Sparse CodingIEEE Transactions on Information Forensics and Security, 2015
- CROWDSOURCING A WORD–EMOTION ASSOCIATION LEXICONComputational Intelligence, 2012
- A blog emotion corpus for emotional expression analysis in ChineseComputer Speech & Language, 2010
- Opinion Mining and Sentiment AnalysisFoundations and Trends® in Information Retrieval, 2008
- Critical Analysis of the Impact of Glottal Features in the Classification of Clinical Depression in SpeechIEEE Transactions on Biomedical Engineering, 2007
- Framewise phoneme classification with bidirectional LSTM and other neural network architecturesNeural Networks, 2005
- Measuring praise and criticismACM Transactions on Information Systems, 2003
- Long Short-Term MemoryNeural Computation, 1997
- A Decision-Theoretic Generalization of On-Line Learning and an Application to BoostingJournal of Computer and System Sciences, 1997