To see the other types of publications on this topic, follow the link: Hidden markov model training.
Dissertations / Theses on the topic 'Hidden markov model training'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 dissertations / theses for your research on the topic 'Hidden markov model training.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
1
McKee, Bill Frederick. "Optimal hidden Markov models." Thesis, University of Plymouth, 1999. http://hdl.handle.net/10026.1/1698.
Full textAbstract:
In contrast with training algorithms such as Baum-Welch, which produce solutions that are a local optimum of the objective function, this thesis describes the attempt to develop a training algorithm which delivers the global optimum Discrete ICdden Markov Model for a given training sequence. A total of four different methods of attack upon the problem are presented. First, after building the necessary analytical tools, the thesis presents a direct, calculus-based assault featuring Matrix Derivatives. Next, the dual analytic approach known as Geometric Programming is examined and then adapted to the task. After that, a hill-climbing formula is developed and applied. These first three methods reveal a number of interesting and useful insights into the problem. However, it is the fourth method which produces an algorithm that is then used for direct comparison vAth the Baum-Welch algorithm: examples of global optima are collected, examined for common features and patterns, and then a rule is induced. The resulting rule is implemented in *C' and tested against a battery of Baum-Welch based programs. In the limited range of tests carried out to date, the models produced by the new algorithm yield optima which have not been surpassed by (and are typically much better than) the Baum-Welch models. However, far more analysis and testing is required and in its current form the algorithm is not fast enough for realistic application.
2
Kapadia, Sadik. "Discriminative training of hidden Markov models." Thesis, University of Cambridge, 1998. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.624997.
Full text
3
Wilhelmsson, Anna, and Sofia Bedoire. "Driving Behavior Prediction by Training a Hidden Markov Model." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-291656.
Full textAbstract:
Introducing automated vehicles in to traffic withhuman drivers, human behavior prediction is essential to obtainoperation safety. In this study, a human behavior estimationmodel has been developed. The estimations are based on aHidden Markov Model (HMM) using observations to determinethe driving style of surrounding vehicles. The model is trainedusing two different methods: Baum Welch training and Viterbitraining to improve the performance. Both training methods areevaluated by looking at time complexity and convergence. Themodel is implemented with and without training and tested fordifferent driving styles. Results show that training is essentialfor accurate human behavior prediction. Viterbi training is fasterbut more noise sensitive compared to Baum Welch training. Also,Viterbi training produces good results if training data reflects oncurrently observed driver, which is not always the case. BaumWelch training is more robust in such situations. Lastly, BaumWelch training is recommended to obtain operation safety whenintroducing automated vehicles into traffic.N ̈ar automatiserade fordon introduceras itrafiken och beh ̈over interagera med m ̈anskliga f ̈orare ̈ar det vik-tigt att kunna f ̈orutsp ̊a m ̈anskligt beteende. Detta f ̈or att kunnaerh ̊alla en s ̈akrare trafiksituation. I denna studie har en modellsom estimerar m ̈anskligt beteende utvecklats. Estimeringarna ̈ar baserade p ̊a en Hidden Markov Model d ̈ar observationeranv ̈ands f ̈or att best ̈amma k ̈orstil hos omgivande fordon itrafiken. Modellen tr ̈anas med tv ̊a olika metoder: Baum Welchtr ̈aning och Viterbi tr ̈aning f ̈or att f ̈orb ̈attra modellens prestanda.Tr ̈aningsmetoderna utv ̈arderas sedan genom att analysera derastidskomplexitet och konvergens. Modellen ̈ar implementerad medoch utan tr ̈aning och testad f ̈or olika k ̈orstilar. Erh ̊allna resultatvisar att tr ̈aning ̈ar viktigt f ̈or att kunna f ̈orutsp ̊a m ̈anskligtbeteende korrekt. Viterbi tr ̈aning ̈ar snabbare men mer k ̈ansligf ̈or brus i j ̈amf ̈orelse med Baum Welch tr ̈aning. Viterbi tr ̈aningger ̈aven en bra estimering i de fall d ̊a observerad tr ̈aningsdataavspeglar f ̈orarens k ̈orstil, vilket inte alltid ̈ar fallet. BaumWelch tr ̈aning ̈ar mer robust i s ̊adana situationer. Slutligenrekommenderas en estimeringsmodell implementerad med BaumWelch tr ̈aning f ̈or att erh ̊alla en s ̈aker k ̈orning d ̊a automatiseradefordon introduceras i trafiken
Kandidatexjobb i elektroteknik 2020, KTH, Stockholm
4
Davis, Richard I. A. "Training Hidden Markov Models for spatio-temporal pattern recognition /." [St. Lucia, Qld.], 2004. http://www.library.uq.edu.au/pdfserve.php?image=thesisabs/absthe18500.pdf.
Full text
5
Combrink, Jan Hendrik. "Discriminative training of hidden Markov Models for gesture recognition." Master's thesis, University of Cape Town, 2018. http://hdl.handle.net/11427/29267.
Full textAbstract:
As homes and workplaces become increasingly automated, an efficient, inclusive and language-independent human-computer interaction mechanism will become more necessary. Isolated gesture recognition can be used to this end. Gesture recognition is a problem of modelling temporal data. Non-temporal models can be used for gesture recognition, but require that the signals be adapted to the models. For example, the requirement of fixed-length inputs for support-vector machine classification. Hidden Markov models are probabilistic graphical models that were designed to operate on time-series data, and are sequence length invariant. However, in traditional hidden Markov modelling, models are trained via the maximum likelihood criterion and cannot perform as well as a discriminative classifier. This study employs minimum classification error training to produce a discriminative HMM classifier. The classifier is then applied to an isolated gesture recognition problem, using skeletal features. The Montalbano gesture dataset is used to evaluate the system on the skeletal modality alone. This positions the problem as one of fine-grained dynamic gesture recognition, as the hand pose information contained in other modalities are ignored. The method achieves a highest accuracy of 87.3%, comparable to other results reported on the Montalbano dataset using discriminative non-temporal methods. The research will show that discriminative hidden Markov models can be used successfully as a solution to the problem of isolated gesture recognition
6
Majewsky, Stefan. "Training of Hidden Markov models as an instance of the expectation maximization algorithm." Bachelor's thesis, Saechsische Landesbibliothek- Staats- und Universitaetsbibliothek Dresden, 2017. http://nbn-resolving.de/urn:nbn:de:bsz:14-qucosa-226903.
Full textAbstract:
In Natural Language Processing (NLP), speech and text are parsed and generated with language models and parser models, and translated with translation models. Each model contains a set of numerical parameters which are found by applying a suitable training algorithm to a set of training data.
Many such training algorithms are instances of the Expectation-Maximization (EM) algorithm. In [BSV15], a generic EM algorithm for NLP is described. This work presents a particular speech model, the Hidden Markov model, and its standard training algorithm, the Baum-Welch algorithm. It is then shown that the Baum-Welch algorithm is an instance of the generic EM algorithm introduced by [BSV15], from which follows that all statements about the generic EM algorithm also apply to the Baum-Welch algorithm, especially its correctness and convergence properties.
7
Fang, Eric. "Investigation of training algorithms for hidden Markov models applied to automatic speech recognition." Connect to this title online, 2009. http://etd.lib.clemson.edu/documents/1249065572/.
Full text
8
Lam, Tin Yin. "HMM converter a tool box for hidden Markov models with two novel, memory efficient parameter training algorithms." Thesis, University of British Columbia, 2008. http://hdl.handle.net/2429/5786.
Full textAbstract:
Hidden Markov models (HMMs) are powerful statistical tools for biological sequence analysis.
Many recently developed Bioinformatics applications employ variants of HMMs to analyze
diverse types of biological data. It is typically fairly easy to design the states and the topological
structure of an HMM. However, it can be difficult to estimate parameter values which yield a
good prediction performance. As many HMM-based applications employ similar algorithms
for generating predictions, it is also time-consuming and error-prone to have to re-implement
these algorithms whenever a new HMM-based application is to be designed.
This thesis addresses these challenges by introducing a tool-box, called HMMC0NvERTER,
which only requires an XML-input file to define an HMM and to use it for sequence decoding
and parameter training. The package not only allows for rapid proto-typing of HMM-based
applications, but also incorporates several algorithms for sequence decoding and parameter
training, including two new, linear memory algorithms for parameter training. Using this
software package, even users without programming knowledge can quickly set up sophisticated
HMMs and pair-HMMs and use them with efficient algorithms for parameter training and
sequence analyses. We use HMMCONVERTER to construct a new comparative gene prediction
program, called ANNOTAID, which can predict pairs of orthologous genes by integrating prior
information about each input sequence probabilistically into the gene prediction process and
into parameter training. ANNOTAID can thus be readily adapted to predict orthologous gene
pairs in newly sequenced genomes.
9
Varga, Tamás. "Off-line cursive handwriting recognition using synthetic training data." Berlin Aka, 2006. http://deposit.d-nb.de/cgi-bin/dokserv?id=2838183&prov=M&dok_var=1&dok_ext=htm.
Full text
10
Do, Trinh-Minh-Tri. "Regularized bundle methods for large-scale learning problems with an application to large margin training of hidden Markov models." Paris 6, 2010. http://www.theses.fr/2010PA066163.
Full textAbstract:
L'apprentissage automatique est souvent posé sous la forme d'un problème d'optimisation où l'on cherche le meilleur modèle parmi une famille de modèle paramétré par optimisation d'une fonction réelle de l'ensemble des paramètres. Le meilleur modèle est défini comme celui qui correspond à une l'ensemble de paramètres qui minimise cette fonction objective appelée critère. Les progrès rapide de l'apprentissage automatique ces dernières années sont allés de pair avec le développement de méthodes d'optimisation efficaces et adaptées aux particularités des fonctionnelles à minimiser, notamment pour permettre le traitement de grands jeux de données ou pour réaliser des taches d’apprentissage complexes. Dans cette thèse, nous travaillons sur les techniques d'optimisation nouvelles, génériques et efficaces afin de faciliter l'apprentissage de modèles complexes pour des applications de grande taille, et pour des critères quelconques. En particulier, nous nous sommes focalisés sur des problèmes d'optimisation non contraints dans lesquels la fonction objective peut être non-convexe et non partout différentiable. Etre capable de traiter ce genre de situation permet de pouvoir aborder des problèmes réels avec des modèles complexes et des critères d'apprentissage performants. Les contributions de cette thèse sont présentées en deux parties. La première partie présente nos travaux sur l'optimisation non contrainte. La seconde partie décrit les systèmes que nous avons développés pour l'apprentissage discriminant de modèles graphiques pour l'étiquetage de signaux et séquences, en nous appuyant lorsque nécessaire sur les algorithmes décrits dans la première partie.
11
Li, Jinyu. "Soft margin estimation for automatic speech recognition." Diss., Atlanta, Ga. : Georgia Institute of Technology, 2008. http://hdl.handle.net/1853/26613.
Full textAbstract:
Thesis (Ph.D)--Electrical and Computer Engineering, Georgia Institute of Technology, 2009.Committee Chair: Dr. Chin-Hui Lee; Committee Member: Dr. Anthony Joseph Yezzi; Committee Member: Dr. Biing-Hwang (Fred) Juang; Committee Member: Dr. Mark Clements; Committee Member: Dr. Ming Yuan. Part of the SMARTech Electronic Thesis and Dissertation Collection.
12
Koenig, Lionel. "Masquage de pertes de paquets en voix sur IP." Thesis, Toulouse, INPT, 2011. http://www.theses.fr/2011INPT0010/document.
Full textAbstract:
Les communications téléphoniques en voix sur IP souffrent de la perte de paquets causée par les problèmes d'acheminement dus aux nœuds du réseau. La perte d'un paquet de voix induit la perte d'un segment de signal de parole (généralement 10ms par paquet perdu). Face à la grande diversité des codeurs de parole, nous nous sommes intéressés dans le cadre de cette thèse à proposer une méthode de masquage de pertes de paquets générique, indépendante du codeur de parole utilisé. Ainsi, le masquage de pertes de paquets est appliqué au niveau du signal de parole reconstruit, après décodage, s'affranchissant ainsi du codeur de parole. Le système proposé repose sur une modélisation classique de type « modèles de Markov cachés » afin de suivre l'évolution acoustique de la parole. À notre connaissance, une seule étude a proposé l'utilisation des modèles de Markov cachés dans ce cadre [4]. Toutefois, Rødbro a utilisé l'utilisation de deux modèles, l'un pour la parole voisée, l'autre pour les parties non voisées, posant ainsi le problème de la distinction voisée/non voisée. Dans notre approche, un seul modèle de Markov caché est mis en œuvre. Aux paramètres classiques (10 coefficients de prédiction linéaire dans le domaine cepstral (LPCC) et dérivées premières) nous avons adjoint un nouvel indicateur continu de voisement [1, 2]. La recherche du meilleur chemin avec observations manquantes conduit à une version modifiée de l'algorithme de Viterbi pour l'estimation de ces observations. Les différentes contributions (indice de voisement, décodage acoutico-phonétique et restitution du signal) de cette thèse sont évaluées [3] en terme de taux de sur et sous segmentation, taux de reconnaissance et distances entre l'observation attendue et l'observation estimée. Nous donnons une indication de la qualité de la parole au travers d'une mesure perceptuelle : le PESQPacket loss due to misrouted or delayed packets in voice over IP leads to huge voice quality degradation. Packet loss concealment algorithms try to enhance the perceptive quality of the speech. The huge variety of vocoders leads us to propose a generic framework working directly on the speech signal available after decoding. The proposed system relies on one single "hidden Markov model" to model time evolution of acoustic features. An original indicator of continuous voicing is added to conventional parameters (Linear Predictive Cepstral Coefficients) in order to handle voiced/unvoiced sound. Finding the best path with missing observations leads to one major contribution: a modified version of the Viterbi algorithm tailored for estimating missing observations. All contributions are assessed using both perceptual criteria and objective metrics
13
Ter-Hovhannisyan, Vardges. "Unsupervised and semi-supervised training methods for eukaryotic gene prediction." Diss., Atlanta, Ga. : Georgia Institute of Technology, 2008. http://hdl.handle.net/1853/26645.
Full textAbstract:
Thesis (Ph.D)--Biology, Georgia Institute of Technology, 2009.Committee Chair: Mark Borodovky; Committee Member: Jung H. Choi; Committee Member: King Jordan; Committee Member: Leonid Bunimovich; Committee Member: Yury Chernoff. Part of the SMARTech Electronic Thesis and Dissertation Collection.
14
Tang, Shiyuyun. "Improving algorithms of gene prediction in prokaryotic genomes, metagenomes, and eukaryotic transcriptomes." Diss., Georgia Institute of Technology, 2016. http://hdl.handle.net/1853/54998.
Full textAbstract:
Next-generation sequencing has generated enormous amount of DNA and RNA sequences that potentially carry volumes of genetic information, e.g. protein-coding genes. The thesis is divided into three main parts describing i) GeneMarkS-2, ii) GeneMarkS-T, and iii) MetaGeneTack.
In prokaryotic genomes, ab initio gene finders can predict genes with high accuracy. However, the error rate is not negligible and largely species-specific. Most errors in gene prediction are made in genes located in genomic regions with atypical GC composition, e.g. genes in pathogenicity islands. We describe a new algorithm GeneMarkS-2 that uses local GC-specific heuristic models for scoring individual ORFs in the first step of analysis. Predicted atypical genes are retained and serve as ‘external’ evidence in subsequent runs of self-training. GeneMarkS-2 also controls the quality of training process by effectively selecting optimal orders of the Markov chain models as well as duration parameters in the hidden semi-Markov model. GeneMarkS-2 has shown significantly improved accuracy compared with other state-of-the-art gene prediction tools.
Massive parallel sequencing of RNA transcripts by the next generation technology (RNA-Seq) provides large amount of RNA reads that can be assembled to full transcriptome. We have developed a new tool, GeneMarkS-T, for ab initio identification of protein-coding regions in RNA transcripts. Unsupervised estimation of parameters of the algorithm makes unnecessary several steps in the conventional gene prediction protocols, most importantly the manually curated preparation of training sets. We have demonstrated that the GeneMarkS-T self-training is robust with respect to the presence of errors in assembled transcripts and the accuracy of GeneMarkS-T in identifying protein-coding regions and, particularly, in predicting gene starts compares favorably to other existing methods.
Frameshift prediction (FS) is important for analysis and biological interpretation of metagenomic sequences. Reads in metagenomic samples are prone to sequencing errors. Insertion and deletion errors that change the coding frame impair the accurate identification of protein coding genes. Accurate frameshift prediction requires sufficient amount of data to estimate parameters of species-specific statistical models of protein-coding and non-coding regions. However, this data is not available; all we have is metagenomic sequences of unknown origin. The challenge of ab initio FS detection is, therefore, twofold: (i) to find a way to infer necessary model parameters and (ii) to identify positions of frameshifts (if any). We describe a new tool, MetaGeneTack, which uses a heuristic method to estimate parameters of sequence models used in the FS detection algorithm. It was shown on several test sets that the performance of MetaGeneTack FS detection is comparable or better than the one of earlier developed program FragGeneScan.
15
Chong, Fong Ho. "Frequency-stream-tying hidden Markov model /." View Abstract or Full-Text, 2003. http://library.ust.hk/cgi/db/thesis.pl?ELEC%202003%20CHONG.
Full textAbstract:
Thesis (M. Phil.)--Hong Kong University of Science and Technology, 2003.Includes bibliographical references (leaves 119-123). Also available in electronic version. Access restricted to campus users.
16
Schimert, James. "A high order hidden Markov model /." Thesis, Connect to this title online; UW restricted, 1992. http://hdl.handle.net/1773/8939.
Full text
17
Kotsalis, Georgios. "Model reduction for Hidden Markov models." Thesis, Massachusetts Institute of Technology, 2006. http://hdl.handle.net/1721.1/38255.
Full textAbstract:
Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2006.Includes bibliographical references (leaves 57-60).
The contribution of this thesis is the development of tractable computational methods for reducing the complexity of two classes of dynamical systems, finite alphabet Hidden Markov Models and Jump Linear Systems with finite parameter space. The reduction algorithms employ convex optimization and numerical linear algebra tools and do not pose any structural requirements on the systems at hand. In the Jump Linear Systems case, a distance metric based on randomization of the parametric input is introduced. The main point of the reduction algorithm lies in the formulation of two dissipation inequalities, which in conjunction with a suitably defined storage function enable the derivation of low complexity models, whose fidelity is controlled by a guaranteed upper bound on the stochastic L2 gain of the approximation error. The developed reduction procedure can be interpreted as an extension of the balanced truncation method to the broader class of Jump Linear Systems. In the Hidden Markov Model case, Hidden Markov Models are identified with appropriate Jump Linear Systems that satisfy certain constraints on the coefficients of the linear transformation. This correspondence enables the development of a two step reduction procedure.
(cont.) In the first step, the image of the high dimensional Hidden Markov Model in the space of Jump Linear Systems is simplified by means of the aforementioned balanced truncation method. Subsequently, in the second step, the constraints that reflect the Hidden Markov Model structure are imposed by solving a low dimensional non convex optimization problem. Numerical simulation results provide evidence that the proposed algorithm computes accurate reduced order Hidden Markov Models, while achieving a compression of the state space by orders of magnitude.
by Georgios Kotsalis.
Ph.D.
18
Kato, Akihiro. "Hidden Markov model-based speech enhancement." Thesis, University of East Anglia, 2017. https://ueaeprints.uea.ac.uk/63950/.
Full textAbstract:
This work proposes a method of model-based speech enhancement that uses a network of HMMs to first decode noisy speech and to then synthesise a set of features that enables a speech production model to reconstruct clean speech. The motivation is to remove the distortion and residual and musical noises that are associated with conventional filteringbased methods of speech enhancement. STRAIGHT forms the speech production model for speech reconstruction and requires a time-frequency spectral surface, aperiodicity and a fundamental frequency contour. The technique of HMM-based synthesis is used to create the estimate of the timefrequency surface, and aperiodicity after the model and state sequence is obtained from HMM decoding of the input noisy speech. Fundamental frequency were found to be best estimated using the PEFAC method rather than synthesis from the HMMs. For the robust HMM decoding in noisy conditions it is necessary for the HMMs to model noisy speech and consequently noise adaptation is investigated to achieve this and its resulting effect on the reconstructed speech measured. Even with such noise adaptation to match the HMMs to the noisy conditions, decoding errors arise, both in terms of incorrect decoding and time alignment errors. Confidence measures are developed to identify such errors and then compensation methods developed to conceal these errors in the enhanced speech signal. Speech quality and intelligibility analysis is first applied in terms of PESQ and NCM showing the superiority of the proposed method against conventional methods at low SNRs. Three way subjective MOS listening test then discovers the performance of the proposed method overwhelmingly surpass the conventional methods over all noise conditions and then a subjective word recognition test shows an advantage of the proposed method over speech intelligibility to the conventional methods at low SNRs.
19
Lott, Paul Christian. "StochHMM| A Flexible Hidden Markov Model Framework." Thesis, University of California, Davis, 2014. http://pqdtopen.proquest.com/#viewpdf?dispub=3602142.
Full textAbstract:
In the era of genomics, data analysis models and algorithms that provide the means to reduce large complex sets into meaningful information are integral to further our understanding of complex biological systems. Hidden Markov models comprise one such data analysis technique that has become the basis of many bioinformatics tools. Its relative success is primarily due to its conceptually simplicity and robust statistical foundation. Despite being one of the most popular data analysis modeling techniques for classification of linear sequences of data, researchers have few available software options to rapidly implement the necessary modeling framework and algorithms. Most tools are still hand-coded because current implementation solutions do not provide the required ease or flexibility that allows researchers to implement models in non-traditional ways. I have developed a free hidden Markov model C++ library and application, called StochHMM, that provides researchers with the flexibility to apply hidden Markov models to unique sequence analysis problems. It provides researchers the ability to rapidly implement a model using a simple text file and at the same time provide the flexibility to adapt the model in non-traditional ways. In addition, it provides many features that are not available in any current HMM implementation tools, such as stochastic sampling algorithms, ability to link user-defined functions into the HMM framework, and multiple ways to integrate additional data sources together to make better predictions. Using StochHMM, we have been able to rapidly implement models for R-loop prediction and classification of methylation domains. The R-loop predictions uncovered the epigenetic regulatory role of R-loops at CpG promoters and protein coding genes 3' transcription termination. Classification of methylation domains in multiple pluripotent tissues identified epigenetics gene tracks that will help inform our understanding of epigenetic diseases.
20
Yi, Kwan 1963. "Text classification using a hidden Markov model." Thesis, McGill University, 2005. http://digitool.Library.McGill.CA:80/R/?func=dbin-jump-full&object_id=85214.
Full textAbstract:
Text categorization (TC) is the task of automatically categorizing textual digital documents into pre-set categories by analyzing their contents. The purpose of this study is to develop an effective TC model to resolve the difficulty of automatic classification. In this study, two primary goals are intended. First, a Hidden Markov Model (HAM is proposed as a relatively new method for text categorization. HMM has been applied to a wide range of applications in text processing such as text segmentation and event tracking, information retrieval, and information extraction. Few, however, have applied HMM to TC. Second, the Library of Congress Classification (LCC) is adopted as a classification scheme for the HMM-based TC model for categorizing digital documents. LCC has been used only in a handful of experiments for the purpose of automatic classification. In the proposed framework, a general prototype for an HMM-based TC model is designed, and an experimental model based on the prototype is implemented so as to categorize digitalized documents into LCC. A sample of abstracts from the ProQuest Digital Dissertations database is used for the test-base. Dissertation abstracts, which are pre-classified by professional librarians, form an ideal test-base for evaluating the proposed model of automatic TC. For comparative purposes, a Naive Bayesian model, which has been extensively used in TC applications, is also implemented. Our experimental results show that the performance of our model surpasses that of the Naive Bayesian model as measured by comparing the automatic classification of abstracts to the manual classification performed by professionals.
21
Stanke, Mario. "Gene prediction with a Hidden Markov model." Doctoral thesis, [S.l.] : [s.n.], 2004. http://deposit.ddb.de/cgi-bin/dokserv?idn=970841310.
Full text
22
Beattie, Valerie L. "Hidden Markov Model state-based noise compensation." Thesis, University of Cambridge, 1992. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.259519.
Full text
23
Schwardt, Ludwig. "Efficient Mixed-Order Hidden Markov Model Inference." Thesis, Link to the online version, 2007. http://hdl.handle.net/10019/709.
Full text
24
Seward, D. C. (DeWitt Clinton). "Graphical analysis of hidden Markov model experiments." Thesis, Massachusetts Institute of Technology, 1994. http://hdl.handle.net/1721.1/36469.
Full textAbstract:
Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 1994.Includes bibliographical references (leaves 60-61).
by DeWitt C. Seward IV.
Ph.D.
25
Kadhem, Safaa K. "Model fit diagnostics for hidden Markov models." Thesis, University of Plymouth, 2017. http://hdl.handle.net/10026.1/9966.
Full textAbstract:
Hidden Markov models (HMMs) are an efficient tool to describe and model the underlying behaviour of many phenomena. HMMs assume that the observed data are generated independently from a parametric distribution, conditional on an unobserved process that satisfies the Markov property. The model selection or determining the number of hidden states for these models is an important issue which represents the main interest of this thesis. Applying likelihood-based criteria for HMMs is a challenging task as the likelihood function of these models is not available in a closed form. Using the data augmentation approach, we derive two forms of the likelihood function of a HMM in closed form, namely the observed and the conditional likelihoods. Subsequently, we develop several modified versions of the Akaike information criterion (AIC) and Bayesian information criterion (BIC) approximated under the Bayesian principle. We also develop several versions for the deviance information criterion (DIC). These proposed versions are based on the type of likelihood, i.e. conditional or observed likelihood, and also on whether the hidden states are dealt with as missing data or additional parameters in the model. This latter point is referred to as the concept of focus. Finally, we consider model selection from a predictive viewpoint. To this end, we develop the so-called widely applicable information criterion (WAIC). We assess the performance of these various proposed criteria via simulation studies and real-data applications. In this thesis, we apply Poisson HMMs to model the spatial dependence analysis in count data via an application to traffic safety crashes for three highways in the UK. The ultimate interest is in identifying highway segments which have distinctly higher crash rates. Selecting an optimal number of states is an important part of the interpretation. For this purpose, we employ model selection criteria to determine the optimal number of states. We also use several goodness-of-fit checks to assess the model fitted to the data. We implement an MCMC algorithm and check its convergence. We examine the sensitivity of the results to the prior specification, a potential problem given small sample sizes. The Poisson HMMs adopted can provide a different model for analysing spatial dependence on networks. It is possible to identify segments with a higher posterior probability of classification in a high risk state, a task that could prioritise management action.
26
Bulla, Jan. "Application of Hidden Markov and Hidden Semi-Markov Models to Financial Time Series." Doctoral thesis, [S.l. : s.n.], 2006. http://swbplus.bsz-bw.de/bsz260867136inh.pdf.
Full text
27
Farges, Eric P. "An analysis-synthesis hidden Markov model of speech." Diss., Georgia Institute of Technology, 1987. http://hdl.handle.net/1853/14775.
Full text
28
Le, Riche Pierre (Pierre Jacques). "Handwritten signature verification : a hidden Markov model approach." Thesis, Stellenbosch : Stellenbosch University, 2000. http://hdl.handle.net/10019.1/51784.
Full textAbstract:
Thesis (MEng)--University of Stellenbosch, 2000.ENGLISH ABSTRACT: Handwritten signature verification (HSV) is the process through which handwritten signatures are analysed in an attempt to determine whether the person who made the signature is who he claims to be. Banks and other financial institutions lose billions of rands annually to cheque fraud and other crimes that are preventable with the aid of good signature verification techniques. Unfortunately, the volume of cheques that are processed precludes a thorough HSV process done in the traditional manner by human operators. It is the aim of this research to investigate new methods to compare signatures automatically, to eventually speed up the HSV process and improve on the accuracy of existing systems. The new technology that is investigated is the use of the so-called hidden Markov models (HMMs). It is only quite recently that the computing power has become commonly available to make the real-time use of HMMs in pattern recognition a possibility. Two demonstration programs, SigGrab and Securitlheque, have been developed that make use of this technology, and show excellent improvements over other techniques and competing products. HSV accuracies in excess of99% can be attained.
AFRIKAANSE OPSOMMING: Handgeskrewe handtekening verifikasie (HHV) is die proses waardeur handgeskrewe handtekeninge ondersoek word in 'n poging om te bevestig of die persoon wat die handtekening gemaak het werklik is wie hy voorgee om te wees. Banke en ander finansiele instansies verloor jaarliks biljoene rande aan tjekbedrog en ander misdrywe wat voorkom sou kon word indien goeie metodes van handtekening verifikasie daargestel kon word. Ongelukkig is die volume van tjeks wat hanteer word so groot, dat tradisionele HHV deur menslike operateurs 'n onbegonne taak is. Dit is die doel van hierdie navorsmg om nuwe metodes te ondersoek om handtekeninge outomaties te kan vergelyk en so die HHV proses te bespoedig en ook te verbeter op die akkuraatheid van bestaande stelsels. Die nuwe tegnologie wat ondersoek is is die gebruik van die sogenaamde verskuilde Markov modelle (VMMs). Dit is eers redelik onlangs dat die rekenaar verwerkingskrag algemeen beskikbaar geraak het om die intydse gebruik van VMMs in patroonherkenning prakties moontlik te maak. Twee demonstrasieprogramme, SigGrab en SecuriCheque, is ontwikkel wat gebruik maak van hierdie tegnologie en toon uitstekende verbeterings teenoor ander tegnieke en kompeterende produkte. 'n Akkuraatheid van 99% of hoer word tipies verkry.
29
Dey, Arkajit. "Hidden Markov model analysis of subcellular particle trajectories." Thesis, Massachusetts Institute of Technology, 2011. http://hdl.handle.net/1721.1/66307.
Full textAbstract:
Thesis (M. Eng.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2011.This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections.
Cataloged from student submitted PDF version of thesis.
Includes bibliographical references (p. 71-73).
How do proteins, vesicles, or other particles within a cell move? Do they diffuse randomly or ow in a particular direction? Understanding how subcellular particles move in a cell will reveal fundamental principles of cell biology and biochemistry, and is a necessary prerequisite to synthetically engineering such processes. We investigate the application of several variants of hidden Markov models (HMMs) to analyzing the trajectories of such particles. And we compare the performance of our proposed algorithms with traditional approaches that involve fitting a mean square displacement (MSD) curve calculated from the particle trajectories. Our HMM algorithms are shown to be more accurate than existing MSD algorithms for heterogeneous trajectories which switch between multiple phases of motion.
by Arkajit Dey.
M.Eng.
30
Alneberg, Johannes. "Movement of a prawn: a Hidden Markov Model approach." Thesis, Uppsala universitet, Analys och tillämpad matematik, 2011. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-155994.
Full text
31
Wynne-Jones, Michael. "Model building in neural networks with hidden Markov models." Thesis, University of Edinburgh, 1994. http://hdl.handle.net/1842/284.
Full textAbstract:
This thesis concerns the automatic generation of architectures for neural networks and other pattern recognition models comprising many elements of the same type. The requirement for such models, with automatically determined topology and connectivity, arises from two needs. The first is the need to develop commercial applications of the technology without resorting to laborious trial and error with different network sizes; the second is the need, in large and complex pattern processing applications such as speech recognition, to optimise the allocation of computing resources for problem solving. The state of the art in adaptive architectures is reviewed, and a mechanism is proposed for adding new processing elements to models. The scheme is developed in the context of multi-layer perceptron networks, and is linked to the best network-pruning mechanism available using a numerical criterion with construction required at one extreme and pruning at the other. The construction mechanism does not work in the multi-layer perceptron for which it was developed, owing to the long-range effects occurring in many applications of these networks. It works demonstrably well in density estimation models based on Gaussian mixtures, which are of the same family as the increasingly popular radial basis function networks. The construction mechanism is applied to the initialization of the density estimators embedded in the states of a hidden Markov model for speaker-independent speech recognition, where it offers a considerable increase in recogniser performance, provided cross-validation is used to prevent over-training.
32
Chan, Kin Wah. "Pruning of hidden Markov model with optimal brain surgeon /." View Abstract or Full-Text, 2003. http://library.ust.hk/cgi/db/thesis.pl?ELEC%202003%20CHAN.
Full textAbstract:
Thesis (M. Phil.)--Hong Kong University of Science and Technology, 2003.Includes bibliographical references (leaves 72-76). Also available in electronic version. Access restricted to campus users.
33
TALARICO, ERICK COSTA E. SILVA. "SEISMIC TO FACIES INVERSION USING CONVOLVED HIDDEN MARKOV MODEL." PONTIFÍCIA UNIVERSIDADE CATÓLICA DO RIO DE JANEIRO, 2018. http://www.maxwell.vrac.puc-rio.br/Busca_etds.php?strSecao=resultado&nrSeq=36004@1.
Full textAbstract:
PONTIFÍCIA UNIVERSIDADE CATÓLICA DO RIO DE JANEIROA indústria de óleo e gás utiliza a sísmica para investigar a distribuição de tipos de rocha (facies) em subsuperfície. Por outro lado, apesar de seu corriqueiro uso em geociências, medidas sísmicas costumam ser ruidosas, e a inversão do dado sísmico para a distribuição de facies é um problema mal posto. Por esta razão, diversos autores estudam esta inversão sob o ponto de vista probabilístico, para ao menos estimar as incertezas da solução do problema inverso. O objetivo da presente dissertação é desenvolver método quantitativo para estimar a probabilidade de reservatório com hidrocarboneto, dado um traço sísmico de reflexão, integrando modelagem sísmica direta, e conhecimento geológico a priori. Utiliza-se, um dos métodos mais recentes para resolver o problema inverso: Modelo de Markov Oculto com Efeito Convolucional (mais especificamente, a Aproximação por Projeção de (1)). É demonstrado que o método pode ser reformulado em termos do Modelo de Markov Oculto (MMO) ordinário. A teoria de sísmica de AVA é apresentada, e usada conjuntamente com MMO com Efeito Convolucional para resolver a inversão de sísmica para facies. A técnica de inversão é avaliada usando-se medidas difundidas em Aprendizado de Máquina, em um conjunto de experimentos variados e realistas. Apresenta-se uma técnica para medir a capacidade do algoritmo em estimar valores confiáveis de probabilidade. Pelos testes realizados a aproximação por projeção apresenta distorções de probabilidade inferiores a 5 por cento, tornando-a uma técnica útil para a indústria de óleo e gás.
Oil and Gas Industry uses seismic data in order to unravel the distribution of rock types (facies) in the subsurface. But, despite its widespread use, seismic data is noisy and the inversion from seismic data to the underlying rock distribution is an ill-posed problem. For this reason, many authors have studied the topic in a probabilistic formulation, in order to provide uncertainty estimations about the solution of the inversion problem. The objective of the present thesis is to develop a quantitative method to estimate the probability of hydrocarbon bearing reservoir, given a seismic reflection profile, and, to integrate geological prior knowledge with geophysical forward modelling. One of the newest methods for facies inversion is used: Convolved Hidden Markov Model (more specifically the Projection Approximation from (1)). It is demonstrated how Convolved HMM can be reformulated as an ordinary Hidden Markov Model problem (which models geological prior knowledge). Seismic AVA theory is introduced, and used with Convolved HMM theory to solve the seismic to facies problem. The performance of the inversion technique is measured with common machine learning scores, in a broad set of realistic experiments. The technique capability of estimating reliable probabilities is quantified, and it is shown to present distortions smaller than 5 percent. As a conclusion, the studied Projection Approximation is applicable for risk management in Oil and Gas applications, which integrates geological and geophysical knowledge.
34
Jiang, Zuliang. "Hidden Markov Model with Binned Duration and Its Application." ScholarWorks@UNO, 2010. http://scholarworks.uno.edu/td/1108.
Full textAbstract:
Hidden Markov models (HMM) have been widely used in various applications such as speech processing and bioinformatics. However, the standard hidden Markov model requires state occupancy durations to be geometrically distributed, which can be inappropriate in some real-world applications where the distributions on state intervals deviate signi cantly from the geometric distribution, such as multi-modal distributions and heavy-tailed distributions. The hidden Markov model with duration (HMMD) avoids this limitation by explicitly incor- porating the appropriate state duration distribution, at the price of signi cant computational expense. As a result, the applications of HMMD are still quited limited. In this work, we present a new algorithm - Hidden Markov Model with Binned Duration (HMMBD), whose result shows no loss of accuracy compared to the HMMD decoding performance and a com- putational expense that only diers from the much simpler and faster HMM decoding by a constant factor. More precisely, we further improve the computational complexity of HMMD from (TNN +TND) to (TNN +TND ), where TNN stands for the computational com- plexity of the HMM, D is the max duration value allowed and can be very large and D generally could be a small constant value.
35
Dawson, Colin Reimer, and Colin Reimer Dawson. "HaMMLeT: An Infinite Hidden Markov Model with Local Transitions." Diss., The University of Arizona, 2017. http://hdl.handle.net/10150/626170.
Full textAbstract:
In classical mixture modeling, each data point is modeled as arising i.i.d. (typically) from a weighted sum of probability distributions. When data arises from different sources that may not give rise to the same mixture distribution, a hierarchical model can allow the source contexts (e.g., documents, sub-populations) to share components while assigning different weights across them (while perhaps coupling the weights to "borrow strength" across contexts). The Dirichlet Process (DP) Mixture Model (e.g., Rasmussen (2000)) is a Bayesian approach to mixture modeling which models the data as arising from a countably infinite number of components: the Dirichlet Process provides a prior on the mixture weights that guards against overfitting. The Hierarchical Dirichlet Process (HDP) Mixture Model (Teh et al., 2006) employs a separate DP Mixture Model for each context, but couples the weights across contexts. This coupling is critical to ensure that mixture components are reused across contexts.
An important application of HDPs is to time series models, in particular Hidden Markov Models (HMMs), where the HDP can be used as a prior on a doubly infinite transition matrix for the latent Markov chain, giving rise to the HDP-HMM (first developed, as the "Infinite HMM", by Beal et al. (2001), and subsequently shown to be a case of an HDP by Teh et al. (2006)). There, the hierarchy is over rows of the transition matrix, and the distributions across rows are coupled through a top-level Dirichlet Process.
In the first part of the dissertation, I present a formal overview of Mixture Models and Hidden Markov Models. I then turn to a discussion of Dirichlet Processes and their various representations, as well as associated schemes for tackling the problem of doing approximate inference over an infinitely flexible model with finite computa- tional resources. I will then turn to the Hierarchical Dirichlet Process (HDP) and its application to an infinite state Hidden Markov Model, the HDP-HMM.
These models have been widely adopted in Bayesian statistics and machine learning. However, a limitation of the vanilla HDP is that it offers no mechanism to model correlations between mixture components across contexts. This is limiting in many applications, including topic modeling, where we expect certain components to occur or not occur together. In the HMM setting, we might expect certain states to exhibit similar incoming and outgoing transition probabilities; that is, for certain rows and columns of the transition matrix to be correlated. In particular, we might expect pairs of states that are "similar" in some way to transition frequently to each other. The HDP-HMM offers no mechanism to model this similarity structure.
The central contribution of the dissertation is a novel generalization of the HDP- HMM which I call the Hierarchical Dirichlet Process Hidden Markov Model With Local Transitions (HDP-HMM-LT, or HaMMLeT for short), which allows for correlations between rows and columns of the transition matrix by assigning each state a location in a latent similarity space and promoting transitions between states that are near each other. I present a Gibbs sampling scheme for inference in this model, employing auxiliary variables to simplify the relevant conditional distributions, which have a natural interpretation after re-casting the discrete time Markov chain as a continuous time Markov Jump Process where holding times are integrated out, and where some jump attempts "fail". I refer to this novel representation as the Markov Process With Failed Jumps. I test this model on several synthetic and real data sets, showing that for data where transitions between similar states are more common, the HaMMLeT model more effectively finds the latent time series structure underlying the observations.
36
Lindberg, David Seaman III. "Enhancing Individualized Instruction through Hidden Markov Models." The Ohio State University, 2014. http://rave.ohiolink.edu/etdc/view?acc_num=osu1405350981.
Full text
37
Lancaster, Joseph Paul Jr. "Toward autism recognition using hidden Markov models." Thesis, Manhattan, Kan. : Kansas State University, 2008. http://hdl.handle.net/2097/777.
Full text
38
Cavalin, Paulo Rodrigo. "Adaptive systems for hidden Markov model-based pattern recognition systems." Mémoire, École de technologie supérieure, 2011. http://espace.etsmtl.ca/976/1/CAVALIN_Paulo_Rodrigo.pdf.
Full textAbstract:
Cette thèse porte sur l’étude des systèmes adaptatifs pour la reconnaissance de formes. Habituellement
les systèmes de reconnaissance reposent sur une connaissance statique du problème
à résoudre et cela pour la durée de vie du système. Cependant il y a des circonstances où la
connaissance du problème est partielle lors de l’apprentissage initial à l’étape de la conception.
Pour cette raison, les systèmes de classification adaptatifs de nouvelle génération permettent
au système de base de s’adapter à la fois en apprenant sur les nouvelles données et sont également
capables de s’adapter à l’environnement lors de la généralisation. Cette thèse propose
une nouvelle définition d’un système de reconnaissance adaptatif où les MMCs (Modèles de
Markov Cachés) sont considérés comme étude de cas.
La première partie de la thèse présente une évaluation des principaux algorithmes d’apprentissage
incrémental utilisés pour l’estimation des paramètres des MMCs. L’objectif de cette étude
est de dégager les stratégies d’apprentissage incrémental dont la performance en généralisation
se rapproche de cette obtenue avec un apprentissage hors-ligne (batch). Les résultats obtenus
sur le problème de la reconnaissance de chiffres et de lettres manuscrits montrent la supériorité
des approches basées sur les ensembles de modèles. De plus, nous avons montré l’importance
de conserver dans une mémoire à court terme des exemples utilisés en validation, ce qui permet
d’obtenir un niveau de performance qui peut même dépasser celui obtenu en mode batch.
La deuxième partie de cette thèse est consacrée à la formulation d’une nouvelle approche pour
la sélection dynamique des ensembles de classifieurs. Inspiré du concept de fusion appelé
« organisation multi-niveau » (multistage organizations), nous avons formulé une variante de
ce concept appelé DMO (dynamic multistage organization - DMO) qui permet d’adapter la
fonction de fusion dynamiquement pour chaque exemple de test à classer. De plus, le concept
DMO a été intégré à la méthode DSA proposée par Dos Santos et al pour la sélection
dynamique d’ensembles de classifieurs. Ainsi, deux nouvelles variantes, DSAm et DSAc, ont
été proposées et évaluées. Dans le premier cas (DSAm), plusieurs fonctions de sélection permettent
une généralisation de la structure DMO. Pour ce qui est de la variante DSAc, nous
utilisons l’information contextuelle (représentée par les profils de décisions des classifieurs de
base) acquise par le système et qui est associée à la base de validation conservée dans une mémoire
à court terme. L’évaluation des deux approches sur des bases de données de petite et de
grande échelle ont montré que la méthode DSAc domine DSAm sur la plupart des cas étudiés.
Ce résultat montre que l’utilisation d’informations contextuelles permet une meilleure performance
en généralisation comparées aux méthodes non informées. Une propriété importante de
l’approche DSAc est qu’elle peut également servir pour apprendre de nouvelles données dans
le temps, une propriété très importante pour la conception de systèmes de reconnaissance adaptatifs dans les environnements dynamiques caractérisés par un niveau important d’incertitude
sur le problème à résoudre.
Finalement, un nouveau framework appelé LoGID (Local and Global Incremental Learning for
Dynamic Selection) est proposé pour la conception d’un système de reconnaissance adaptatif
basé sur les MMC, et capable de s’adapter dans le temps durant les phases d’apprentissage de
généralisation. Le système est composé d’un pool de classifieurs de base et l’adaptation durant
la phase de généralisation est effectuée par la sélection dynamique des membres du pool les
plus compétents pour classer chaque exemple de test. Le mécanisme de sélection dynamique
est basé sur l’algorithme des K plus proches vecteurs de décision, tandis que l’adaptation durant
la phase d’apprentissage consiste à la mise à jour et à l’ajout de classifieurs de base dans
le système. Durant la phase d’apprentissage, deux stratégies sont proposées pour apprendre
incrémentalement sur des nouvelles données: l’apprentissage local et l’apprentissage global.
L’apprentissage incrémentale local implique la mise à jour du pool de classifieurs de base en
ajoutant des nouveaux membres à cet ensemble. Les nouveaux membres sont générés avec
l’algorithme Learn++. L’apprentissage incrémental global consiste à la mise à jour de la base
de connaissances composée des vecteurs de décisions qui seront utilisés en généralisation pour
la sélection dynamique des membres les plus compétents.
Le système LoGID a été validé sur plusieurs bases de données et les résultats comparés à
ceux publiés dans la littérature. En général, la méthode proposée domine les autres méthodes
incluant les méthodes d’apprentissage hors-ligne. Enfin, le système LoGID évalué en mode
adaptatif montre qu’il est en mesure d’apprendre de nouvelles connaissances dans le temps au
moment où les nouvelles données sont disponibles. Cette faculté d’adaptation est très importante
également lorsque les données disponibles pour l’apprentissage sont peu nombreuses.
39
Salfner, Felix. "Event-based failure prediction an extended hidden Markov model approach." Berlin dissertation.de, 2008. http://d-nb.info/990430626/04.
Full text
40
Rooney, Thomas J. A. "On improving the forecast accuracy of the hidden Markov model." Master's thesis, University of Cape Town, 2016. http://hdl.handle.net/11427/22977.
Full textAbstract:
The forecast accuracy of a hidden Markov model (HMM) may be low due first, to the measure of forecast accuracy being ignored in the parameterestimation method and, second, to overfitting caused by the large number of parameters that must be estimated. A general approach to forecasting is described which aims to resolve these two problems and so improve the forecast accuracy of the HMM. First, the application of extremum estimators to the HMM is proposed. Extremum estimators aim to improve the forecast accuracy of the HMM by minimising an estimate of the forecast error on the observed data. The forecast accuracy is measured by a score function and the use of some general classes of score functions is proposed. This approach contrasts with the standard use of a minus log-likelihood score function. Second, penalised estimation for the HMM is described. The aim of penalised estimation is to reduce overfitting and so increase the forecast accuracy of the HMM. Penalties on both the state-dependent distribution parameters and transition probability matrix are proposed. In addition, a number of cross-validation approaches for tuning the penalty function are investigated. Empirical assessment of the proposed approach on both simulated and real data demonstrated that, in terms of forecast accuracy, penalised HMMs fitted using extremum estimators generally outperformed unpenalised HMMs fitted using maximum likelihood.
41
lin, yea-chau, and 林業超. "Training Profile Hidden Markov Model with a Combinatorial Method." Thesis, 2005. http://ndltd.ncl.edu.tw/handle/77674125649641383823.
Full textAbstract:
碩士逢甲大學
資訊工程所
93
Hidden Markov Model have many application in signal processing, pattern recognition. And in computational biology profile Hidden Mar kov Model is famous for fold recognition, multiple sequence alignment, and DNA prediction. But these models involved data sets (DNA sequence or RNA sequence), usually contain noise signal like intro etc. it is difficult to analyze the multiple observation training problem without certain assumptions. In tradition method assumption all sequence is are independent of each other. This paper presents a multiple sequence training use combinatorial method. This combinatorial method defines different dependence-independence assumptions for training sequence. And show the different dependence-independence profile HMM in this result.
42
Purnell, Darryl William. "Discriminative and Bayesian techniques for hidden Markov model speech recognition systems." Thesis, 2001. http://hdl.handle.net/2263/29158.
Full textAbstract:
The collection of large speech databases is not a trivial task (if done properly). It is not always possible to collect, segment and annotate large databases for every task or language. It is also often the case that there are imbalances in the databases, as a result of little data being available for a specific subset of individuals. An example of one such imbalance is the fact that there are often more male speakers than female speakers (or vice-versa). If there are, for example, far fewer female speakers than male speakers, then the recognizers will tend to work poorly for female speakers (as compared to performance for male speakers). This thesis focuses on using Bayesian and discriminative training algorithms to improve continuous speech recognition systems in scenarios where there is a limited amount of training data available. The research reported in this thesis can be divided into three categories: • Overspecialization is characterized by good recognition performance for the data used during training, but poor recognition performance for independent testing data. This is a problem when too little data is available for training purposes. Methods of reducing overspecialization in the minimum classification error algo¬rithm are therefore investigated. • Development of new Bayesian and discriminative adaptation/training techniques that can be used in situations where there is a small amount of data available. One example here is the situation where an imbalance in terms of numbers of male and female speakers exists and these techniques can be used to improve recognition performance for female speakers, while not decreasing recognition performance for the male speakers. • Bayesian learning, where Bayesian training is used to improve recognition perfor¬mance in situations where one can only use the limited training data available. These methods are extremely computationally expensive, but are justified by the improved recognition rates for certain tasks. This is, to the author's knowledge, the first time that Bayesian learning using Markov chain Monte Carlo methods have been used in hidden Markov model speech recognition. The algorithms proposed and reviewed are tested using three different datasets (TIMIT, TIDIGITS and SUNSpeech), with the tasks being connected digit recognition and con¬tinuous speech recognition. Results indicate that the proposed algorithms improve recognition performance significantly for situations where little training data is avail¬able.Thesis (PhD (Electronic Engineering))--University of Pretoria, 2006.
Electrical, Electronic and Computer Engineering
unrestricted
43
Yi-RuChen and 陳奕儒. "An Automatic Assessment Framework for Exercise Training System using Hidden Markov Model and K-Means Clustering." Thesis, 2017. http://ndltd.ncl.edu.tw/handle/9a7nc5.
Full textAbstract:
碩士國立成功大學
工程科學系
105
Since the launch of RGB-D sensors, these sensors are applied to exercise training systems. They are used to record users’ exercise processes and extract human skeletal data. By monitoring/reviewing users’ exercise video or skeletal data, users or a computer program could check if the poses are correct, especially for key poses. This assessment of a key pose does not appropriately present the relationship between a user’s posture and time. This research proposes a software framework (1) for professionals to build standard reference key poses of some exercise, (2) users would perform the same exercise, and the system automatically performs assessment. This framework transforms the professionals’ demonstration into sequences of continuous movements through preprocessing, feature extracting and a clustering algorithm. These sequences of continuous movements become training data sources of Hidden Markov Models that correspond to each movement primitive. A user records his/her training process by RGB-D sensors, and through the same way above to generate sequences of the entire training process. These sequences are segmented into movement primitives, and compared to each trained HMMs. Thereby automatically assess if the training process is close to the professional’s demonstration. After viewing the feedback of training process and practicing repeatedly to reach the goal of training, the user is expected to gain improvements in the exercise.
44
Yin, Yan. "A study of convex optimization for discriminative training of hidden Markov models in automatic speech recognition /." 2008. http://gateway.proquest.com/openurl?url_ver=Z39.88-2004&res_dat=xri:pqdiss&rft_val_fmt=info:ofi/fmt:kev:mtx:dissertation&rft_dat=xri:pqdiss:MR45978.
Full textAbstract:
Thesis (M.Sc.)--York University, 2008. Graduate Programme in Computer Science.Typescript. Includes bibliographical references (leaves 101-109). Also available on the Internet. MODE OF ACCESS via web browser by entering the following URL: http://gateway.proquest.com/openurl?url_ver=Z39.88-2004&res_dat=xri:pqdiss&rft_val_fmt=info:ofi/fmt:kev:mtx:dissertation&rft_dat=xri:pqdiss:MR45978
45
Nair, Nishanth Ulhas. "Joint Evaluation Of Multiple Speech Patterns For Speech Recognition And Training." Thesis, 2009. http://hdl.handle.net/2005/630.
Full textAbstract:
Improving speech recognition performance in the presence of noise and interference continues to be a challenging problem. Automatic Speech Recognition (ASR) systems work well when the test and training conditions match. In real world environments there is often a mismatch between testing and training conditions. Various factors like additive noise, acoustic echo, and speaker accent, affect the speech recognition performance. Since ASR is a statistical pattern recognition problem, if the test patterns are unlike anything used to train the models, errors are bound to occur, due to feature vector mismatch. Various approaches to robustness have been proposed in the ASR literature contributing to mainly two topics: (i) reducing the variability in the feature vectors or (ii) modify the statistical model parameters to suit the noisy condition. While some of those techniques are quite effective, we would like to examine robustness from a different perspective. Considering the analogy of human communication over telephones, it is quite common to ask the person speaking to us, to repeat certain portions of their speech, because we don't understand it. This happens more often in the presence of background noise where the intelligibility of speech is affected significantly. Although exact nature of how humans decode multiple repetitions of speech is not known, it is quite possible that we use the combined knowledge of the multiple utterances and decode the unclear part of speech. Majority of ASR algorithms do not address this issue, except in very specific issues such as pronunciation modeling. We recognize that under very high noise conditions or bursty error channels, such as in packet communication where packets get dropped, it would be beneficial to take the approach of repeated utterances for robust ASR. In this thesis, we have formulated a set of algorithms for both joint evaluation/decoding for recognizing noisy test utterances as well as utilize the same formulation for selective training of Hidden Markov Models (HMMs), again for robust performance.
We first address joint recognition of multiple speech patterns given that they belong to the same class. We formulated this problem considering the patterns as isolated words. If there are K test patterns (K ≥ 2) of a word by a speaker, we show that it is possible to improve the speech recognition accuracy over independent single pattern evaluation of test speech, for the case of both clean and noisy speech. We also find the state sequence which best represents the K patterns. This formulation can be extended to connected word recognition or continuous speech recognition also.
Next, we consider the benefits of joint multi-pattern likelihood for HMM training. In the usual HMM training, all the training data is utilized to arrive at a best possible parametric model. But, it is possible that the training data is not all genuine and therefore may have labeling errors, noise corruptions, or plain outlier exemplars. Such outliers will result in poorer models and affect speech recognition performance. So it is important to selectively train them so that the outliers get a lesser weightage. Giving lesser weight to an entire outlier pattern has been addressed before in speech recognition literature. However, it is possible that only some portions of a training pattern are corrupted. So it is important that only the corrupted portions of speech are given a lesser weight during HMM training and not the entire pattern. Since in HMM training, multiple patterns of speech from each class are used, we show that it is possible to use joint evaluation methods to selectively train HMMs such that only the corrupted portions of speech are given a lesser weight and not the entire speech pattern.
Thus, we have addressed all the three main tasks of a HMM, to jointly utilize the availability of multiple patterns belonging to the same class. We experimented the new algorithms for Isolated Word Recognition in the case of both clean speech and noisy speech. Significant improvement in speech recognition performance is obtained, especially for speech affected by transient/burst noise.
46
Liu, Yang. "A study of Hidden Markov Model." 2004. http://etd.utk.edu/2004/LiuYang.pdf.
Full textAbstract:
Thesis (M.S.)--University of Tennessee, Knoxville, 2004.Title from title page screen (viewed Sept. 21, 2004). Thesis advisor: Jan Rosinski. Document formatted into pages (vi, 86 p. : ill.). Vita. Includes bibliographical references (p. 54-56).
47
Tsai, Tsung-Yu, and 蔡宗祐. "Underground Stratification Using Hidden Markov Model." Thesis, 2018. http://ndltd.ncl.edu.tw/handle/w6ma7f.
Full textAbstract:
碩士國立臺灣大學
土木工程學研究所
106
In the work of site investigation, investigating the underground soil layers is a quite important part. Once we know the distribution of the soil layers, we could understand the strength of stratum bearing capacity which leads the design of subsequent engineering project. The main contents of investigating underground soil layers are to obtain the thickness of the layers, the elevations of the interfaces and the types of the soils. The conventional way is to perform a standard penetration test (SPT) and take out the soil samples to identify the interfaces and the types of the soil layers. Although the cone penetration test (CPT) can not sample the soils, soil stratification still can be performed based on it. And CPT is more simple and convenient than SPT. Soil behavior type index (Ic), which has been proved to be able to distinguish soil types effectively according to its value, can be calculated from CPT data. Therefore, many scholars have devoted themselves to the development of soil stratification methods based in CPT data. This study developed a method of soil stratification using the hidden Markov model (HMM) and Gibbs sampling, which is called HMM soil stratification method. The approach is to regard the soil types as the hidden states of the hidden Markov model, and to regard Ic as the output sequence of the model. Using Ic as the analytical data, describe the spatial variability of Ic with one-dimensional stationary Gaussian random field. Then based on Bayes '' theorem, the mean (μ) and the standard deviation (σ) of Ic are estimated by Gibbs sampling. According to Ic and its mean and standard deviation, use forward-backward recursions (FB recursions) to find the most likely soil type at each point. The above steps are performed for iterative calculations to obtain convergent results, and the types and interfaces of the soil layers can be found by this method. Finally, the likelihood of each number of cluster is calculated by the likelihood recursions to find the optimal number of clusters. In terms of case studied, this study used the in-situ CPT data from Hollywood, South Carolina, to verify the results of HMM soil stratification method. And another stable 1D soil stratificaiton method—the wavelet transform modulus maxima method (WTMM method, Ching et al., 2015) was performed for comparison and discussion with HMM method. The conclusion is as following: the advantages of HMM soil stratification method is that the number of clusters can be changed from 1 to 10, and HMM can analyze the change of Ic and automatically classified similar soil layers into one layer. However, it was also found that the irrational thin layer problem and the cluster scores problem need to be addressed by subsequent studies. The second part of this study is trying to combine the WTMM method with the generalized coupled Markov chain (GCMC) model developed by Park (2010). We conducted the case studies of the predictions of 2D and 3D soil stratification profiles. And we explored the feasibility of using CPT data to build a multidimensional soil stratification model and analyzed two cases in Hollywood and South Parklands in Adelaide, South Australia, respectively. Both cases have received reasonable results.
48
Chiang, Chang-Hsuan, and 江長軒. "Mixture Markov Model and Hidden Markov Model for Blood Donation Sequence Data." Thesis, 2018. http://ndltd.ncl.edu.tw/handle/896u26.
Full textAbstract:
碩士國立臺灣科技大學
工業管理系
106
Blood transfusion is essential for certain medical treatments. In recent years, considerable concern has arisen over the issue of how to maintain stable supply of blood components. While a blood center can either hold blood drive campaigns to recruit new donors or encourage regular donors to return to ensure sufficient supply of blood, having a donor donate blood regularly seems to be more valuable than recruiting a new donor. In this study, sequence data which contains blood donation history of donors from 2010 to 2014 were analyzed. In particular, the donors who donate first time in the first half of 2010 were followed up for five years and model-based clustering methods, including mixture Markov model and mixture hidden Markov model, were used to identify the clusters of the donors. After obtaining and interpreting clusters, logistic regression models and random forest models were adopted to investigate how demographic characteristics and the short-term behavior affect a donor's long-term return behavior. Results show that the short-term donation behavior is the most important indicator for predicting a donor's long-term donation behavior. Furthermore, “age” is also significantly associated with a donor's behavior, and those who are older than 40 years old are more likely to return regularly.
49
Huang, Gen-Kai, and 黃俊凱. "Applying Hidden Markov Model and Observable Markov Model for Audio Content Identification." Thesis, 2007. http://ndltd.ncl.edu.tw/handle/f5jbsf.
Full textAbstract:
碩士國立臺北科技大學
電機工程系研究所
95
In this thesis, both Hidden Markov Model and Observable Markov Model (OMM) are developed as the audio fingerprints for each audio signal. Each state of both Markov Models is classified by a set of gaussian mixture probabilities and the features Mel-Frequency Cepstral Coefficients (MFCC) are taken into consideration in the experiments. The framework consists of two phases, one is the database training phase and the other is the identification phase. The audio database used in the experiments is divided into 12 categories, including 9 kinds of musical instruments , symphony and males and females singing. Three classifiers that consist of Gaussian Mixture Model, Hidden Markov Model, and Observable Markov Model are investigated. The experimental results show that the OMM(MFCC) scheme can execute faster than the HMM(MFCC) and performs graceful degradation even when suffering various distortion, such as clipping , MP3 compression, AAC compression, amplitude modification, and time-scale modification, etc.
50
Romberg, Justin Keith. "A universal hidden Markov tree image model." Thesis, 1999. http://hdl.handle.net/1911/17293.
Full textAbstract:
Wavelet-domain hidden Markov models have proven to be useful tools for statistical signal and image processing. The hidden Markov tree (HMT) model captures the key features of the joint density of the wavelet coefficients of real-world data. One potential drawback to the HMT framework is the need for computationally expensive iterative training. We propose two reduced-parameter HMT models that capture the general structure of a broad class of real-world images. In the image HMT model, we use the fact that for real-world images the structure of the HMT is self-similar across scale, allowing us to reduce the complexity of the model to just nine parameters. In the universal HMT we fix these nine parameters, eliminating training while retaining nearly all of the key structure modeled by the full HMT. Finally, we propose a fast shift-invariant HMT estimation algorithm that outperforms all other wavelet-based estimators in the current literature.