Extracting Spatial Knowledge from the Web
SAINT(2003)
摘要
The content of the world-wide web is pervaded by informationof a geographical or spatial nature, particularlysuch location information as addresses, postal codes, andtelephone numbers. We present a system for extracting spatialknowledge from collections of web pages gathered byweb-crawling programs. For each page determined to containlocation information, we apply geocoding techniques tocompute geographic coordinates, such as latitude-longitudepairs. Next, we augment the location information with keyworddescriptors extracted from the web page contents. Wethen apply spatial data mining techniques on the augmentedlocation information to derive spatial knowledge.
更多查看译文
关键词
andtelephone number,web data mining,spatial nature,geoparsing,keyword extraction,crawl,geocoding,geographic information system gis,extracting spatial knowledge,particularlysuch location information,dimension reduction,labeling,location information,web page,spatial data mining technique,spatial knowledge,world-wide web,web page content,augmentedlocation information,clustering,information extraction,data mining,web pages,geographic information system,search engines,web crawling,world wide web,information retrieval
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络