Application of Graph Entropy for Knowledge Discovery and Data Mining

André Calero Valdez, Matthias Dehmer, Andreas Holzinger

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

Entropy, originating from statistical physics, is an interesting and challenging concept with many diverse definitions and various applications. Considering all the diverse meanings, entropy can be used as a measure of disorder in the range between total order (structured) and total disorder (unstructured) as long as by “order” we understand that objects are segregated by their properties or parameter values. States of lower entropy occur when objects become organized, and ideally when everything is in complete order, the entropy value is 0. These observations generated a colloquial meaning of entropy. In this chapter we investigate the state of the art in graph-theoretical approaches and how they are connected to text mining. This prepares us to understand how graph entropy could be used in data-mining processes
Next, we show how different graphs can be constructed from bibliometric data and what research problems can be addressed by each of those. We then focus on coauthorship graphs to identify collaboration styles using graph entropy. For this purpose, we selected a subgroup of the DBLP database and prepared it for our analysis. The results show how two entropy measures
describe our data set. From these results, we conclude our discussion of the
results and consider different extensions on how to improve our approach.
LanguageEnglish
Title of host publicationMathematical Foundations and Applications of Graph Entropy
EditorsMatthias Dehmer, Frank Emmert-Streib, Zengqiang Chen, Xueliang Li, Yongtang Shi
PublisherJohn Wiley & Sons, Inc
Pages259-276
ISBN (Electronic)978-3-527-69322-1
ISBN (Print)978-3-527-33909-9
StatusPublished - 24 Sep 2016

Publication series

NameQuantitative and Network Biology Series
PublisherWiley-VCH

Fingerprint

Data mining
Entropy
Physics

Keywords

  • Knowledge Discovery
  • Machine Learning
  • entropy
  • Graph entropy

ASJC Scopus subject areas

  • Computer Science Applications

Fields of Expertise

  • Information, Communication & Computing

Treatment code (Nähere Zuordnung)

  • Basic - Fundamental (Grundlagenforschung)
  • Experimental

Cite this

Calero Valdez, A., Dehmer, M., & Holzinger, A. (2016). Application of Graph Entropy for Knowledge Discovery and Data Mining. In M. Dehmer, F. Emmert-Streib, . Z. Chen, . X. Li, & Y. Shi (Eds.), Mathematical Foundations and Applications of Graph Entropy (pp. 259-276). (Quantitative and Network Biology Series). John Wiley & Sons, Inc.

Application of Graph Entropy for Knowledge Discovery and Data Mining. / Calero Valdez, André; Dehmer, Matthias; Holzinger, Andreas.

Mathematical Foundations and Applications of Graph Entropy. ed. / Matthias Dehmer; Frank Emmert-Streib; Zengqiang Chen; Xueliang Li; Yongtang Shi. John Wiley & Sons, Inc, 2016. p. 259-276 (Quantitative and Network Biology Series).

Research output: Chapter in Book/Report/Conference proceedingChapter

Calero Valdez, A, Dehmer, M & Holzinger, A 2016, Application of Graph Entropy for Knowledge Discovery and Data Mining. in M Dehmer, F Emmert-Streib, Z Chen, X Li & Y Shi (eds), Mathematical Foundations and Applications of Graph Entropy. Quantitative and Network Biology Series, John Wiley & Sons, Inc, pp. 259-276.
Calero Valdez A, Dehmer M, Holzinger A. Application of Graph Entropy for Knowledge Discovery and Data Mining. In Dehmer M, Emmert-Streib F, Chen Z, Li X, Shi Y, editors, Mathematical Foundations and Applications of Graph Entropy. John Wiley & Sons, Inc. 2016. p. 259-276. (Quantitative and Network Biology Series).
Calero Valdez, André ; Dehmer, Matthias ; Holzinger, Andreas. / Application of Graph Entropy for Knowledge Discovery and Data Mining. Mathematical Foundations and Applications of Graph Entropy. editor / Matthias Dehmer ; Frank Emmert-Streib ; Zengqiang Chen ; Xueliang Li ; Yongtang Shi. John Wiley & Sons, Inc, 2016. pp. 259-276 (Quantitative and Network Biology Series).
@inbook{68bd29467829494e80f1a287e26c43f3,
title = "Application of Graph Entropy for Knowledge Discovery and Data Mining",
abstract = "Entropy, originating from statistical physics, is an interesting and challenging concept with many diverse definitions and various applications. Considering all the diverse meanings, entropy can be used as a measure of disorder in the range between total order (structured) and total disorder (unstructured) as long as by “order” we understand that objects are segregated by their properties or parameter values. States of lower entropy occur when objects become organized, and ideally when everything is in complete order, the entropy value is 0. These observations generated a colloquial meaning of entropy. In this chapter we investigate the state of the art in graph-theoretical approaches and how they are connected to text mining. This prepares us to understand how graph entropy could be used in data-mining processesNext, we show how different graphs can be constructed from bibliometric data and what research problems can be addressed by each of those. We then focus on coauthorship graphs to identify collaboration styles using graph entropy. For this purpose, we selected a subgroup of the DBLP database and prepared it for our analysis. The results show how two entropy measuresdescribe our data set. From these results, we conclude our discussion of theresults and consider different extensions on how to improve our approach.",
keywords = "Knowledge Discovery, Machine Learning, entropy, Graph entropy",
author = "{Calero Valdez}, Andr{\'e} and Matthias Dehmer and Andreas Holzinger",
year = "2016",
month = "9",
day = "24",
language = "English",
isbn = "978-3-527-33909-9",
series = "Quantitative and Network Biology Series",
publisher = "John Wiley & Sons, Inc",
pages = "259--276",
editor = "Matthias Dehmer and Frank Emmert-Streib and Chen, { Zengqiang} and Li, { Xueliang} and Yongtang Shi",
booktitle = "Mathematical Foundations and Applications of Graph Entropy",
address = "United States",

}

TY - CHAP

T1 - Application of Graph Entropy for Knowledge Discovery and Data Mining

AU - Calero Valdez,André

AU - Dehmer,Matthias

AU - Holzinger,Andreas

PY - 2016/9/24

Y1 - 2016/9/24

N2 - Entropy, originating from statistical physics, is an interesting and challenging concept with many diverse definitions and various applications. Considering all the diverse meanings, entropy can be used as a measure of disorder in the range between total order (structured) and total disorder (unstructured) as long as by “order” we understand that objects are segregated by their properties or parameter values. States of lower entropy occur when objects become organized, and ideally when everything is in complete order, the entropy value is 0. These observations generated a colloquial meaning of entropy. In this chapter we investigate the state of the art in graph-theoretical approaches and how they are connected to text mining. This prepares us to understand how graph entropy could be used in data-mining processesNext, we show how different graphs can be constructed from bibliometric data and what research problems can be addressed by each of those. We then focus on coauthorship graphs to identify collaboration styles using graph entropy. For this purpose, we selected a subgroup of the DBLP database and prepared it for our analysis. The results show how two entropy measuresdescribe our data set. From these results, we conclude our discussion of theresults and consider different extensions on how to improve our approach.

AB - Entropy, originating from statistical physics, is an interesting and challenging concept with many diverse definitions and various applications. Considering all the diverse meanings, entropy can be used as a measure of disorder in the range between total order (structured) and total disorder (unstructured) as long as by “order” we understand that objects are segregated by their properties or parameter values. States of lower entropy occur when objects become organized, and ideally when everything is in complete order, the entropy value is 0. These observations generated a colloquial meaning of entropy. In this chapter we investigate the state of the art in graph-theoretical approaches and how they are connected to text mining. This prepares us to understand how graph entropy could be used in data-mining processesNext, we show how different graphs can be constructed from bibliometric data and what research problems can be addressed by each of those. We then focus on coauthorship graphs to identify collaboration styles using graph entropy. For this purpose, we selected a subgroup of the DBLP database and prepared it for our analysis. The results show how two entropy measuresdescribe our data set. From these results, we conclude our discussion of theresults and consider different extensions on how to improve our approach.

KW - Knowledge Discovery

KW - Machine Learning

KW - entropy

KW - Graph entropy

UR - http://eu.wiley.com/WileyCDA/WileyTitle/productCd-3527339094.html

M3 - Chapter

SN - 978-3-527-33909-9

T3 - Quantitative and Network Biology Series

SP - 259

EP - 276

BT - Mathematical Foundations and Applications of Graph Entropy

PB - John Wiley & Sons, Inc

ER -