作者: Cungen Cao , Shi Wang , Lin Jiang
DOI: 10.1007/978-3-319-12096-6_19
关键词:
摘要: Geographical information extraction is a special case of extraction. In this paper, we present practical method extracting both names geographical entities and their relations from the Web. The composed three major phases. First, manually designed list 493 Chinese lexico-syntactical patterns for matching Web page excerpts which contain relations; second, developed knowledge extractor those to generate graph whose nodes are entities, edges represent entities; third, several methods handling problems or errors in generated graph. Experimental results show that OMKast-Googling system has satisfactory performance entity name relation