Associated near sets of distance functions in pattern analysis

Authors:
James F. Peters
Affiliations:
Computational Intelligence Laboratory, Department of Electrical & Computer Engineering, Univ. of Manitoba, Winnipeg, MB, Canada
Venue:
MIWAI'11 Proceedings of the 5th international conference on Multi-Disciplinary Trends in Artificial Intelligence
Year:
2011

Citing 4
Cited 0

Near Sets. Special Theory about Nearness of Objects

Fundamenta Informaticae - New Frontiers in Scientific Discovery - Commemorating the Life and Work of Zdzislaw Pawlak
Foundations of near sets

Information Sciences: an International Journal
Grill Determined L-Approach Merotopological Spaces

Fundamenta Informaticae
ε-near collections

RSKT'11 Proceedings of the 6th international conference on Rough sets and knowledge technology

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper introduces description-based associated sets of distance functions, where members are topological structures helpful in pattern analysis and machine intelligence. An associated set of a function is a collection containing members with one or more common properties. This study has important implications in discerning patterns shared by members of an associated set. The focus in this paper is on defining and characterising distance functions relative to structures that are collections of sufficiently near (far) neighbourhoods, filters, grills and clusters. Naimpally-Peters-Tiwari distance functions themselves define approach spaces that generalise the traditional notion of a metric space. An important side-effect of this work is the discovery of various patterns that arise from the descriptions (perceptions) of associated set members. An application of the proposed approach is given in the context of camouflaged objects.