Change search
ReferencesLink to record
Permanent link

Direct link
Optimal Viterbi Bayesian predictive classification for data from finite alphabets
KTH, School of Engineering Sciences (SCI), Mathematics (Dept.), Mathematical Statistics.ORCID iD: 0000-0003-1489-8512
2013 (English)In: Journal of Statistical Planning and Inference, ISSN 0378-3758, Vol. 143, no 2, 261-275 p.Article in journal (Refereed) Published
Abstract [en]

A family of Viterbi Bayesian predictive classifiers has been recently popularized for speech recognition applications with continuous acoustic signals modeled by finite mixture densities embedded in a hidden Markov framework. Here we generalize such classifiers to sequentially observed data from multiple finite alphabets and derive the optimal predictive classifier under exchangeability of the emitted symbols. We demonstrate that the optimal predictive classifier which learns from unlabelled test items improves considerably upon marginal maximum a posteriori rule in the presence of sparse training data. It is shown that the learning process saturates when the amount of test data tends to infinity, such that no further gain in classification accuracy is possible upon arrival of new test items in the long run.

Place, publisher, year, edition, pages
2013. Vol. 143, no 2, 261-275 p.
Keyword [en]
Bayesian learning, Hidden Markov models, Predictive classification
National Category
URN: urn:nbn:se:kth:diva-107602DOI: 10.1016/j.jspi.2012.07.013ISI: 000310942200004ScopusID: 2-s2.0-84867736475OAI: diva2:577043

QC 20121214

Available from: 2012-12-14 Created: 2012-12-14 Last updated: 2012-12-14Bibliographically approved

Open Access in DiVA

No full text

Other links

Publisher's full textScopus

Search in DiVA

By author/editor
Koski, Timo
By organisation
Mathematical Statistics
In the same journal
Journal of Statistical Planning and Inference

Search outside of DiVA

GoogleGoogle Scholar
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Altmetric score

Total: 47 hits
ReferencesLink to record
Permanent link

Direct link