Translating verbally communicated local geographic knowledge using semantic technologies: A Balinese example

Abstract

Using cognitive linguistic strategies, people can verbally encode and convey their spatial realities with little effort (i.e. “my house is right across the street from the grocery store”). However, to date there are a limited number of ways to transform such spatial information into forms that are useful for computational analysis in a geographic information system (GIS), and for sharing across research communities. This paper uses a case study in the Balinese language to investigate the spatial and linguistic information necessary to compute such transformations. That is, to transform verbally communicated spatial scenes into GIS-suitable data. We propose an ontology which captures reference frames used in certain Balinese locative expressions together with the parameters (ground, direction and template) required for transformation. The approach allows for the sharing of translation methods and the reuse of contextual information on the Web. Based on this model, we identify open research questions on the way to supporting approximate transformations of locative expressions.