by Gilles Bisson
Abstract:
Currently, in the Objects-Based Representation Systems, both classification and categorization are based on subsumption criterion. In practice, this criterion seems too strong as soon as we need to deal with incomplete or incoherent knowledge. The notion of similarity has been successfully used in many domains such as Data Analysis, Pattern Recognition, or Machine Learning to compare and to structure noisy knowledge. In this paper we present a preliminary work aiming at demonstrating the advantage of using a similarity measure instead of a subsumption criterion in the object-based representations.
Reference:
Why and How to Define a Similarity Measure for Object Based Representation Systems (Gilles Bisson), Chapter in Towards Very Large Knowledge Bases (N. J. I. Mars, ed.), IOS Press, 1995.
Bibtex Entry:
@incollection{Bisson1995,
abstract = {Currently, in the Objects-Based Representation Systems, both classification and categorization are based on subsumption criterion. In practice, this criterion seems too strong as soon as we need to deal with incomplete or incoherent knowledge. The notion of similarity has been successfully used in many domains such as Data Analysis, Pattern Recognition, or Machine Learning to compare and to structure noisy knowledge. In this paper we present a preliminary work aiming at demonstrating the advantage of using a similarity measure instead of a subsumption criterion in the object-based representations.},
author = {Bisson, Gilles},
booktitle = {Towards Very Large Knowledge Bases},
edition = {1st},
editor = {Mars, N. J. I.},
isbn = {9051992173},
keywords = {SML-LIB-BIBLIO,lang:ENG},
mendeley-tags = {SML-LIB-BIBLIO,lang:ENG},
pages = {236--246},
publisher = {IOS Press},
title = {{Why and How to Define a Similarity Measure for Object Based Representation Systems}},
url = {http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.49.3998},
year = {1995}
}