Year: 2012
Type: conference-paper
Abstract: Extraction of spatial information from natural language is a challenging problem in applications such as robotics, geographical information systems and human-machine interaction. The task of spatial role labeling (SpRL) is a newly introduced task (Kordjamshidi et al., 2011) that formalizes the spatial concepts and relations in the language to be mapped to formal spatial calculus using machine learning. The first level of this task is to extract spatial roles called spatial indicators(ind), trajectories(tr) and landmarks(lm) and their link. Spatial indicators indicate the existence of spatial information in a sentence. Trajector is an entity whose location is described and landmark is a reference object for describing the location of a trajector. For example in the sentence: The book on AI in on the table behind you., the second on is a spatial indicator. book is a trajector and its landmark is table. A spatial relation is a triplet of spatial roles. Hence, there are two spatial relations in this sentence: and .
Source: Lirias
Authors: Parisa Kordjamshidi, Marie‐Francine Moens
Institution: KU Leuven
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2520082489