Grounding linked open data in wordnet: the case of the OSM semantic network

  • Authors:

  • Andrea Ballatore;Michela Bertolotto;David C. Wilson

  • Affiliations:

  • School of Computer Science and Informatics, University College Dublin, Ireland;School of Computer Science and Informatics, University College Dublin, Ireland;Department of Software and Information Systems, University of North Carolina, Charlotte, NC

  • Venue:
  • W2GIS'13 Proceedings of the 12th international conference on Web and Wireless Geographical Information Systems
  • Year:
  • 2013

Quantified Score

Hi-index 0.00

Visualization

Abstract

In recent years, the open data (LOD) paradigm has emerged as a promising approach to structuring, publishing, and sharing data online, using Semantic Web standards. From a geospatial perspective, one of the key challenges consists of bridging the gap between the vast amount of crowdsourced, semi-structured or unstructured geo-information and the Semantic Web. Notably, OpenStreetMap (OSM) has gathered billions of objects from its contributors in a spatial folksonomy. The contribution of this paper is twofold. First, we add a piece to the LOD jigsaw, the OSM Semantic Network, structuring it as a W3C Simple Knowledge Organization System (SKOS) vocabulary, and discussing its role in the constellation of geo-knowledge bases. Second, we devise Voc2WordNet, a mapping approach between a given vocabulary and WordNet, a pivotal component in the LOD cloud. Our approach is evaluated on the OSM Semantic Network against a human-generated alignment, obtaining high precision and recall.