Högskolan i Skövde

his.sePublikationer
Ändra sökning
RefereraExporteraLänk till posten
Permanent länk

Direktlänk
Referera
Referensformat
  • apa
  • apa-cv
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annat format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annat språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf
Utilizing Information on Uncertainty for In Silico Modeling using Random Forests
Högskolan i Skövde, Institutionen för kommunikation och information. Högskolan i Skövde, Forskningscentrum för Informationsteknologi. Stockholm University, Sweden. (Skövde Artificial Intelligence Lab (SAIL))ORCID-id: 0000-0001-8382-0300
AstraZeneca R&D, Södertälje, Sweden.
2009 (Engelska)Ingår i: Proceedings of the 3rd Skövde Workshop on Information Fusion Topics (SWIFT 2009), Skövde: University of Skövde , 2009, s. 59-62Konferensbidrag, Publicerat paper (Refereegranskat)
Abstract [en]

Information on uncertainty of measurements or estimates of molecular properties are rarely utilized by in silico predictive models. In this study, different approaches to handling uncertain numerical features are explored when using the stateof- the-art random forest algorithm for generating predictive models. Two main approaches are considered: i) sampling from probability distributions prior to tree generation, which does not require any change to the underlying tree learning algorithm, and ii) adjusting the algorithm to allow for handling probability distributions, similar to how missing values typically are handled, i.e., partitions may include fractions of examples. An experiment with six datasets concerning the prediction of various chemical properties is presented, where 95% confidence intervals are included for one of the 92 numerical features. In total, five approaches to handling uncertain numeric features are compared: ignoring the uncertainty, sampling from distributions that are assumed to be uniform and normal respectively, and adjusting tree learning to handle probability distributions that are assumed to be uniform and normal respectively. The experimental results show that all approaches that utilize information on uncertainty indeed outperform the single approach ignoring this, both with respect to accuracy and area under ROC curve. A decomposition of the squared error of the constituent classification trees shows that the highest variance is obtained by ignoring the information on uncertainty, but that this also results in the highest mean squared error of the constituent trees.

Ort, förlag, år, upplaga, sidor
Skövde: University of Skövde , 2009. s. 59-62
Serie
Skövde University Studies in Informatics, ISSN 1653-2325 ; 2009:3
Nationell ämneskategori
Data- och informationsvetenskap
Forskningsämne
Teknik; Skövde Artificial Intelligence Lab (SAIL)
Identifikatorer
URN: urn:nbn:se:his:diva-3542ISBN: 978-91-978513-2-9 (digital)OAI: oai:DiVA.org:his-3542DiVA, id: diva2:284576
Konferens
The 3rd Annual Skövde Workshop on Information Fusion Topics (SWIFT 2009), 12-13 Oct 2009, Skövde, Sweden
Anmärkning

[CD-ROM]

Tillgänglig från: 2010-01-07 Skapad: 2010-01-07 Senast uppdaterad: 2020-11-30Bibliografiskt granskad

Open Access i DiVA

Fulltext saknas i DiVA

Person

Boström, Henrik

Sök vidare i DiVA

Av författaren/redaktören
Boström, Henrik
Av organisationen
Institutionen för kommunikation och informationForskningscentrum för Informationsteknologi
Data- och informationsvetenskap

Sök vidare utanför DiVA

GoogleGoogle Scholar

isbn
urn-nbn

Altmetricpoäng

isbn
urn-nbn
Totalt: 536 träffar
RefereraExporteraLänk till posten
Permanent länk

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