Högskolan i Skövde

his.sePublikasjoner
Endre søk
RefereraExporteraLink to record
Permanent link

Direct link
Referera
Referensformat
  • apa
  • apa-cv
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annet format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annet språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf
Inconsistency: Friend or Foe
The School of Business and Informatics, University of Borås, Sweden.
The School of Business and Informatics, University of Borås, Sweden.
Högskolan i Skövde, Institutionen för kommunikation och information. Högskolan i Skövde, Forskningscentrum för Informationsteknologi. (Skövde Cognition and Artificial Intelligence Lab (SCAI))
2007 (engelsk)Inngår i: The 2007 International Joint Conferenceon Neural Networks: IJCNN 2007 Conference Proceedings, IEEE, 2007, s. 1383-1388Konferansepaper, Publicerat paper (Fagfellevurdert)
Abstract [en]

One way of obtaining accurate yet comprehensible models is to extract rules from opaque predictive models. When evaluating rule extraction algorithms, one frequently used criterion is consistency; i.e. the algorithm must produce similar rules every time it is applied to the same problem. Rule extraction algorithms based on evolutionary algorithms are, however, inherently inconsistent, something that is regarded as their main drawback. In this paper, we argue that consistency is an overvalued criterion, and that inconsistency can even be beneficial in some situations. The study contains two experiments, both using publicly available data sets, where rules are extracted from neural network ensembles. In the first experiment, it is shown that it is normally possible to extract several different rule sets from an opaque model, all having high and similar accuracy. The implication is that consistency in that perspective is useless; why should one specific rule set be considered superior? Clearly, it should instead be regarded as an advantage to obtain several accurate and comprehensible descriptions of the relationship. In the second experiment, rule extraction is used for probability estimation. More specifically, an ensemble of extracted trees is used in order to obtain probability estimates. Here, it is exactly the inconsistency of the rule extraction algorithm that makes the suggested approach possible.

sted, utgiver, år, opplag, sider
IEEE, 2007. s. 1383-1388
Serie
Proceedings of the International Joint Conference on Neural Networks, ISSN 2161-4393, E-ISSN 2161-4407
HSV kategori
Forskningsprogram
Teknik
Identifikatorer
URN: urn:nbn:se:his:diva-2104DOI: 10.1109/IJCNN.2007.4371160ISI: 000254291101059Scopus ID: 2-s2.0-51749099818ISBN: 978-1-4244-1380-5 (digital)ISBN: 1-4244-1380-X (tryckt)ISBN: 978-1-4244-1379-9 (tryckt)OAI: oai:DiVA.org:his-2104DiVA, id: diva2:32380
Konferanse
The 2007 International Joint Conference on Neural Networks, IJCNN 2007, August 12-17, 2007, Renaissance Orlando Resort, Florida, USA
Tilgjengelig fra: 2008-05-30 Laget: 2008-05-30 Sist oppdatert: 2021-04-22bibliografisk kontrollert

Open Access i DiVA

Fulltekst mangler i DiVA

Andre lenker

Forlagets fulltekstScopus

Person

Niklasson, Lars

Søk i DiVA

Av forfatter/redaktør
Niklasson, Lars
Av organisasjonen

Søk utenfor DiVA

GoogleGoogle Scholar

doi
isbn
urn-nbn

Altmetric

doi
isbn
urn-nbn
Totalt: 528 treff
RefereraExporteraLink to record
Permanent link

Direct link
Referera
Referensformat
  • apa
  • apa-cv
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annet format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annet språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf