Automated ontology generation using spatial reasoning

  • Authors:
  • Alton Coalter;Jennifer L. Leopold

  • Affiliations:
  • Missouri University of Science and Technology, Department of Computer Science, Rolla, Missouri;Missouri University of Science and Technology, Department of Computer Science, Rolla, Missouri

  • Venue:
  • KSEM'10 Proceedings of the 4th international conference on Knowledge science, engineering and management
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

Recently there has been much interest in using ontologies to facilitate knowledge representation, integration, and reasoning. Correspondingly, the extent of the information embodied by an ontology is increasing beyond the conventional is_a and part_of relationships. To address these requirements, a vast amount of digitally available information may need to be considered when building ontologies, prompting a desire for software tools to automate at least part of the process. The main efforts in this direction have involved textual information retrieval and extraction methods. For some domains extension of the basic relationships could be enhanced further by the analysis of 2D and/or 3D images. For this type of media, image processing algorithms are more appropriate than textual analysis methods. Herein we present an algorithm that, given a collection of 3D image files, utilizes Qualitative Spatial Reasoning (QSR) to automate the creation of an ontology for the objects represented by the images, relating the objects in terms of is_a and part_of relationships and also through unambiguous Relational Connection Calculus (RCC) relations.