Similarity matching for integrating spatial information extracted from place descriptions

摘要: Place descriptions are used in everyday communication as a common way to convey spatial information. Processing the information from place poses multiple significant challenges because these written natural language. In particular, corpora of provide plethora human knowledge beyond geographical system, even if refer same places various ways. This article focuses on resolving ambiguous or synonymous names by exploring given relationships with other features. It matches developing novel labelled graph matching process that relies solely comparison string, linguistic and similarities between identified places. uses unstructured an input, produces composite qualitative relations descriptions. The performance this exceeds current toponym resolution coping non-gazetteered