Sökning: id:"swepub:oai:DiVA.org:su-95262" >
Bayesian Block-Diag...
-
Corander, Jukka
(författare)
Bayesian Block-Diagonal Predictive Classifier for Gaussian Data
- Artikel/kapitelEngelska2013
Förlag, utgivningsår, omfång ...
-
Berlin, Heidelberg :Springer Berlin/Heidelberg,2013
-
printrdacarrier
Nummerbeteckningar
-
LIBRIS-ID:oai:DiVA.org:su-95262
-
https://urn.kb.se/resolve?urn=urn:nbn:se:su:diva-95262URI
-
https://doi.org/10.1007/978-3-642-33042-1_58DOI
-
https://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-117792URI
Kompletterande språkuppgifter
-
Språk:engelska
-
Sammanfattning på:engelska
Ingår i deldatabas
Klassifikation
-
Ämneskategori:ref swepub-contenttype
-
Ämneskategori:kap swepub-publicationtype
Anmärkningar
-
QC 20130205
-
The paper presents a method for constructing Bayesian predictive classifier in a high-dimensional setting. Given that classes are represented by Gaussian distributions with block-structured covariance matrix, a closed form expression for the posterior predictive distribution of the data is established. Due to factorization of this distribution, the resulting Bayesian predictive and marginal classifier provides an efficient solution to the high-dimensional problem by splitting it into smaller tractable problems. In a simulation study we show that the suggested classifier outperforms several alternative algorithms such as linear discriminant analysis based on block-wise inverse covariance estimators and the shrunken centroids regularized discriminant analysis.
Ämnesord och genrebeteckningar
Biuppslag (personer, institutioner, konferenser, titlar ...)
-
Koski, TimoKTH,Matematisk statistik(Swepub:kth)u1m4flfv
(författare)
-
Pavlenko, TatjanaKTH,Matematisk statistik(Swepub:kth)u13cu9mc
(författare)
-
Tillander, Annika,1971-Karolinska Institutet, Sweden(Swepub:su)anti4790
(författare)
-
KTHMatematisk statistik
(creator_code:org_t)
Sammanhörande titlar
-
Ingår i:Synergies of Soft Computing and Statistics for Intelligent Data AnalysisBerlin, Heidelberg : Springer Berlin/Heidelberg, s. 543-55197836423304149783642330421
Internetlänk
Hitta via bibliotek
Till lärosätets databas