Mining Geospatial Path Data from Natural Language Descriptions

Nate Blaylock, Bradley Swain, James F. Allen · 2009

In this paper, we describe the TEGUS system for mining geospatial path data from natural language descriptions. TEGUS uses natural language processing, GIS entity databases, and graph-based path finding to predict lat/lon paths based only on natural language text input. We also report on preliminary results from experiments on a corpus of path descriptions.

Read the paper · More papers on PaperTik