QQ客服 官方微博 反馈留言

	An Evidence-based Approach for Toponym Disambiguation
	Wang, Xingguang ; Zhang, Yi ; Chen, Min ; Lin, Xing ; Yu, Hao ; Liu, Yu
	2010
关键词	Toponym Disambiguation Semantic relations Evidence D-S theory
英文摘要	Toponym Disambiguation (TD) in Geographic Information Retrieval (GIR) systems is a crucial technique, which makes a direct impact on the quality of subsequent assignment of geographic focus to a document and that of establishment of spatial index as well as the effectiveness of the entire retrieval model as a whole. We explore the mechanism for human beings' dealing with the problem of TD. Human's way of TD follows some linguistic rules since that documents are organized according to these rules. From a linguistic perspective, involved rules fall into three categories: semantic rules, syntactic rules and pragmatic rules. In this paper we set our focus on semantic rules. Moreover, co-occurring toponyms without any ambiguity in the same context of document can be regarded as evidences or clues to support the process of TD. The candidate referent of a toponym that has the closest semantic relationship with the referent of co-occurring toponym as evidence in context is chosen as the intended one. To achieve this goal, the strength of semantic relationships among co-occurring toponyms is required to be computed properly. In response to this requirement, a model is proposed in this paper for the needed calculation. Combining the evidential effects of toponyms as evidences is essential when there are multiple of them in context. In this paper, we propose an evidence-based approach to fulfill the task of TD in two main steps: 1) calculate the semantic relationships between any two geographical referents based on a computing model of semantic relations among them; 2) use Dempster-Shafer (D-S) theory of evidence to combine multiple co-occurring toponyms as evidences. Finally, case studies are presented to evaluate the effectiveness of our approach.; Computer Science, Information Systems; Engineering, Electrical & Electronic; EI; CPCI-S(ISTP); 0
语种	英语
DOI标识	10.1109/GEOINFORMATICS.2010.5567805
内容类型	其他
源URL	[http://ir.pku.edu.cn/handle/20.500.11897/311940]
专题	地球与空间科学学院
推荐引用方式 GB/T 7714	Wang, Xingguang,Zhang, Yi,Chen, Min,et al. An Evidence-based Approach for Toponym Disambiguation. 2010-01-01.

个性服务

查看访问统计

相关权益政策

暂无数据

收藏/分享

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。