INDU: An Interval and Duration Network

  • Authors:
  • Arun K. Pujari;G. Vijaya Kumari;Abdul Sattar

  • Affiliations:
  • -;-;-

  • Venue:
  • AI '99 Proceedings of the 12th Australian Joint Conference on Artificial Intelligence: Advanced Topics in Artificial Intelligence
  • Year:
  • 1999

Quantified Score

Hi-index 0.00

Visualization

Abstract

The significance of representing duration information along with the qualitative information of the time intervals is well argued in the literature. A new framework INVU (INterval and DUration) network consisting of 25 basic relations, is proposed here. INDU cam handle qualitative information of time interval and duration in one single structure. It inherits many interesting properties of Allen's Interval Algebra (of 13 basic relations) but it also exhibits severed interesting additional features. We present several representations of INDU (ORD-clause, Geometric and Lattice) and chatracterise its tractable subclasses such as the Convex and Pre-convex classes. The important contribution of the current study is to show that for the tractable subclasses (Convex as well as Pre-convex) 4-consistency is necessary to guarantee global consistency of INDU-network.