Show simple item record

2006-11-21Buch DOI: 10.18452/2470
Modeling Event-driven Time Series with Generalized Hidden Semi-Markov Models
dc.contributor.authorSalfner, Felix
dc.date.accessioned2017-06-15T17:11:39Z
dc.date.available2017-06-15T17:11:39Z
dc.date.created2006-12-07
dc.date.issued2006-11-21
dc.identifier.issn0863-095X
dc.identifier.urihttp://edoc.hu-berlin.de/18452/3122
dc.description.abstractThis report introduces a new model for event-driven temporal sequence processing: Generalized Hidden Semi-Markov Models (GHSMMs). GHSMMs are an extension of hidden Markov models to continuous time that builds on turning the stochastic process of hidden state traversals into a semi-Markov process. A large variety of probability distributions can be used to specify transition durations. It is shown how GHSMMs can be used to address the principle problems of temporal sequence processing: sequence generation, sequence recognition and sequence prediction. Additionally, an algorithm is described how the parameters of GHSMMs can be determined from a set of training data: The Baum-Welch algorithm is extended by an embedded expectation-maximization algorithm. Under some conditions the procedure can be simplified to the estimation of distribution moments. A proof of convergence and a complexity assessment are provided.eng
dc.language.isoeng
dc.publisherHumboldt-Universität zu Berlin, Mathematisch-Naturwissenschaftliche Fakultät II, Institut für Informatik
dc.subject.ddc004 Informatik
dc.titleModeling Event-driven Time Series with Generalized Hidden Semi-Markov Models
dc.typebook
dc.identifier.urnurn:nbn:de:kobv:11-10071480
dc.identifier.doihttp://dx.doi.org/10.18452/2470
dc.subject.dnb28 Informatik, Datenverarbeitung
local.edoc.container-titleInformatik-Berichte
local.edoc.pages53
local.edoc.type-nameBuch
local.edoc.container-typeseries
local.edoc.container-type-nameSchriftenreihe
local.edoc.container-volume2006
local.edoc.container-issue208
local.edoc.container-year2006
local.edoc.container-erstkatid2942054-4

Show simple item record