Analogies Between Binary Images

Application to Chinese Characters

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

The purpose of this chapter is to show how it is possible to efficiently extract the structure of a set of objects by use of the notion of proportional analogy.As a proportional analogy involves four objects, the very naïve approach to the problem, has basically a complexity of O(n4) for a given set of n objects. We show, under some conditions on proportional analogy, how to reduce this complexity to O(n2) by considering an equivalent problem, that of enumerating analogical clusters that are informative and not redundant. We further show how some improvements make the task tractable. We illustrate our technique with a task related with natural language processing, that of clustering Chinese characters. In this way, we re-discover the graphical structure of these characters.

Original languageEnglish
Pages (from-to)25-57
Number of pages33
JournalStudies in Computational Intelligence
Volume548
DOIs
Publication statusPublished - 2014

Fingerprint

Binary images
Processing

ASJC Scopus subject areas

  • Artificial Intelligence

Cite this

Analogies Between Binary Images : Application to Chinese Characters. / Lepage, Yves.

In: Studies in Computational Intelligence, Vol. 548, 2014, p. 25-57.

Research output: Contribution to journalArticle

@article{325b1dd8f0a94b6ea880a24987d212f1,
title = "Analogies Between Binary Images: Application to Chinese Characters",
abstract = "The purpose of this chapter is to show how it is possible to efficiently extract the structure of a set of objects by use of the notion of proportional analogy.As a proportional analogy involves four objects, the very na{\"i}ve approach to the problem, has basically a complexity of O(n4) for a given set of n objects. We show, under some conditions on proportional analogy, how to reduce this complexity to O(n2) by considering an equivalent problem, that of enumerating analogical clusters that are informative and not redundant. We further show how some improvements make the task tractable. We illustrate our technique with a task related with natural language processing, that of clustering Chinese characters. In this way, we re-discover the graphical structure of these characters.",
author = "Yves Lepage",
year = "2014",
doi = "10.1007/978-3-642-54516-0_2",
language = "English",
volume = "548",
pages = "25--57",
journal = "Studies in Computational Intelligence",
issn = "1860-949X",
publisher = "Springer Verlag",

}

TY - JOUR

T1 - Analogies Between Binary Images

T2 - Application to Chinese Characters

AU - Lepage, Yves

PY - 2014

Y1 - 2014

N2 - The purpose of this chapter is to show how it is possible to efficiently extract the structure of a set of objects by use of the notion of proportional analogy.As a proportional analogy involves four objects, the very naïve approach to the problem, has basically a complexity of O(n4) for a given set of n objects. We show, under some conditions on proportional analogy, how to reduce this complexity to O(n2) by considering an equivalent problem, that of enumerating analogical clusters that are informative and not redundant. We further show how some improvements make the task tractable. We illustrate our technique with a task related with natural language processing, that of clustering Chinese characters. In this way, we re-discover the graphical structure of these characters.

AB - The purpose of this chapter is to show how it is possible to efficiently extract the structure of a set of objects by use of the notion of proportional analogy.As a proportional analogy involves four objects, the very naïve approach to the problem, has basically a complexity of O(n4) for a given set of n objects. We show, under some conditions on proportional analogy, how to reduce this complexity to O(n2) by considering an equivalent problem, that of enumerating analogical clusters that are informative and not redundant. We further show how some improvements make the task tractable. We illustrate our technique with a task related with natural language processing, that of clustering Chinese characters. In this way, we re-discover the graphical structure of these characters.

UR - http://www.scopus.com/inward/record.url?scp=84926662794&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84926662794&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-54516-0_2

DO - 10.1007/978-3-642-54516-0_2

M3 - Article

VL - 548

SP - 25

EP - 57

JO - Studies in Computational Intelligence

JF - Studies in Computational Intelligence

SN - 1860-949X

ER -