A Concept-Based Knowledge Representation Model for Semantic Entailment Inference

CASIA OpenIR > 综合信息系统研究中心

	A Concept-Based Knowledge Representation Model for Semantic Entailment Inference
	Zhao MJ(赵美静); Ni WC(倪晚成); Zhang HD(张海东); Yang YP(杨一平); Ni WC(倪晚成)
	2014
会议名称	Proceedings of 33rd Control Conference (CCC),Chinese, 2014
会议录名称	Proceedings of 33rd Control Conference (CCC),Chinese, 2014
页码	522 - 527
会议日期	2014,0728-0730
会议地点	南京
摘要	Semantic entailment is a fundamental problem in natural language understanding that has a large number of applica-tions. Knowledge acquisition and knowledge representation are usually the crucial parts in semantic inference strategy. This paper presents a principled approach to semantic entailment problem that builds on a concept-based knowledge representation model (CKR). This model formally defines the concept as a triple (attribute, relation and behavior) and the knowledge of a concept can be illustrated with the triple. We propose a semantic inference strategy that against identify text segments which dissimilar on their surface form but share a common meaning. The inference strategy avoids syntactic analysis steps. A prelim-inary evaluation on the PASCAL text collection is presented. Experimental results show that our concept-based inference strat-egy is effective and has strong potential space.
关键词	Semantic Inference Knowledge Representation Concept Ckr Semantic Entailment
收录类别	EI
语种	英语
文献类型	会议论文
条目标识符	http://ir.ia.ac.cn/handle/173211/11215
专题	综合信息系统研究中心
通讯作者	Ni WC(倪晚成)
作者单位	中国科学院自动化研究所
第一作者单位	中国科学院自动化研究所
推荐引用方式 GB/T 7714	Zhao MJ,Ni WC,Zhang HD,et al. A Concept-Based Knowledge Representation Model for Semantic Entailment Inference[C],2014:522 - 527.

条目包含的文件		下载所有文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
A Concept-Based Know（433KB）	会议论文		开放获取	CC BY-NC-SA	浏览下载