Emotional Conversation Generation Orientated Syntactically Constrained Bidirectional-Asynchronous Framework

doi:10.1109/TAFFC.2019.2923619

CORC > 自动化研究所 > 中国科学院自动化研究所 > 模式识别国家重点实验室 > 智能交互

	Emotional Conversation Generation Orientated Syntactically Constrained Bidirectional-Asynchronous Framework
	Sun, Xiao 1; Li, Jingyuan 1; Tao, Jianhua 2
刊名	IEEE TRANSACTIONS ON AFFECTIVE COMPUTING
	2022
卷号	13 期号:1 页码:187-198
关键词	Emotional conversation generation bidirectional-asynchronous processing syntactic constraint compound information affective computing
ISSN号	1949-3045
DOI	10.1109/TAFFC.2019.2923619
通讯作者	Sun, Xiao(sunx@hfut.edu.cn)
英文摘要	The field of open-domain conversation generation using deep neural networks has attracted increasing attention from researchers for several years. However, traditional neural language models tend to generate safe, generic reply with poor logic and no emotion. In this paper, an emotional conversation generation orientated syntactically constrained bidirectional-asynchronous framework called E-SCBA is proposed to generate meaningful (logical and emotional) reply. In E-SCBA, pre-generated emotion keyword and topic keyword are asynchronously introduced into the reply during the generation, and the process of decoding is much different from the most existing methods that generates reply from the first word to the end. A newly designed bidirectional-asynchronous decoder with the multi-stage strategy is proposed to support this idea, which ensures the fluency and grammaticality of reply by making full use of syntactic constraint. Through the experiments, the results show that our framework not only improves the diversity of replies, but gains a boost on both logic and emotion compared with baselines as well.
资助项目	State Key Program of the National Natural Science of China[61432004] ; State Key Program of the National Natural Science of China[61461045] ; China Postdoctoral Science Foundation[2017T100447] ; National Laboratory of Pattern Recognition (NLPR)
WOS关键词	RECOGNITION ; FEATURES
WOS研究方向	Computer Science
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:000766268600015
资助机构	State Key Program of the National Natural Science of China ; China Postdoctoral Science Foundation ; National Laboratory of Pattern Recognition (NLPR)
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/47960]
专题	模式识别国家重点实验室_智能交互
通讯作者	Sun, Xiao
作者单位	1.Hefei Univ Technol, Sch Comp & Informat, Hefei 230009, Peoples R China 2.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Sun, Xiao,Li, Jingyuan,Tao, Jianhua. Emotional Conversation Generation Orientated Syntactically Constrained Bidirectional-Asynchronous Framework[J]. IEEE TRANSACTIONS ON AFFECTIVE COMPUTING,2022,13(1):187-198.
APA	Sun, Xiao,Li, Jingyuan,&Tao, Jianhua.(2022).Emotional Conversation Generation Orientated Syntactically Constrained Bidirectional-Asynchronous Framework.IEEE TRANSACTIONS ON AFFECTIVE COMPUTING,13(1),187-198.
MLA	Sun, Xiao,et al."Emotional Conversation Generation Orientated Syntactically Constrained Bidirectional-Asynchronous Framework".IEEE TRANSACTIONS ON AFFECTIVE COMPUTING 13.1(2022):187-198.