A Genetic Algorithm-aided Hidden Markov Model Topology Estimation for Phoneme Recognition of Thai Continuous Speech

Bhuriyakorn, P.; Punyabukkana, Proadpran; Suchato, Atiwong

doi:10.1109/snpd.2008.73

Cited by 15 publications

(7 citation statements)

References 10 publications

(10 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The following research works are focused on the speech recognition using HMMs, as in Oudelha et al [26] combining the Baum-Welch algorithm (BW); in Cheshomi et al [27] also uses the BW algorithm. Bhuriyakorn et al [28] present approaches for HMMs topologies generation, in Yang et al [29] address the optimization problem combining a Tabu search and BW algorithm; Yang et al [30] uses Particle Swarm Optimization (PSO) and GA on recognition performance, and in Ogawa et al [31] determine the structure of a Partly HMM with GA. -There is a work of Won et al [2] that presents the use of GA for evolving HMMs, used on information prediction of secondary structure for protein sequences.…”

Section: Parameters Optimization Of An Hmmmentioning

confidence: 99%

Network anomaly detection by continuous hidden markov models: An evolutionary programming approach

Flores

Calderón

Antolino

et al. 2015

IDA

View full text Add to dashboard Cite

Information security is an important and growing need. The most common schemes used for detection systems include pattern-or signature-based and anomaly-based. Anomaly-based schemes use a set of metrics, which outline the normal system behavior and any significant deviation from the established profile will be treated as an anomaly. This paper contributes with an anomaly-based scheme that monitors the bandwidth consumption of a subnetwork, at the Universidad Michoacana, in Mexico. A normal behavior model is based on bandwidth consumption of the subnetwork. The presence of an anomaly indicates that something is misusing the network (viruses, worms, denial of service, or any other kind of attack). This work also presents a scheme for an automatic architecture design and parameters optimization of Hidden Markov Models (HMMs), based on Evolutionary Programming (EP). The variables to be used by the HMMs are: the bandwidth consumption of network (IN and OUT), and the associated time where the network activity occurs. The system was tested with univariate and bivariate observation sequences to analyze and detect anomaly behavior. The HMMs, designed and trained by EP, were compared against semi-random HMMs trained by the Baum-Welch algorithm. On a second experiment, the HMMs, designed and trained by EP, were compared against HMMs created by an expert user. The HMMs outperformed the other methods in all cases. Finally, we made the HMMs time-aware, by including time as another variable. This inclusion made the HMMs capable of detecting activity patterns that are normal during a period of time but anomalous at other times. For instance, a heavy load on the network may be completely normal during working times, but anomalous at nights or weekends.speech and gesture recognition, as well as applications to anomaly detection. In an HMM, the estimation of good model parameters affects the performance of the recognition or detection processes [3]. The HMM parameters can be determined during an iterative process called "training". The Baum-Welch algorithm [4] is one method applied in setting an HMM's parameters, but this method has a drawback: being a gradient-based method, it may converge to a local optimum.Global search techniques can be used to optimize an HMM's parameters. Genetic Algorithms (GA) and Evolutionary Programming (EP) are global optimization techniques that can be used to optimize an HMM's parameters [3]. Studies of using GA to train HMMs are relatively rare, particularly in comparison with the large literature on applying GA to neural network training [2].In the study presented in this paper we want to know (given certain events or behaviors of the study objects), what is expected or what can be anticipated as output. That is, we are interested in studying the behavior of a set of random variables over time, working with stochastic processes. In other words, given the behavior of a computer network, the problem is to automatically develop an HMM capable of discerning between normal and anomalous behavior. This tas...

show abstract

Section: Parameters Optimization Of An Hmmmentioning

confidence: 99%

Network anomaly detection by continuous hidden markov models: An evolutionary programming approach

Flores

Calderón

Antolino

et al. 2015

IDA

View full text Add to dashboard Cite

show abstract

“…GA will be used to get transition probability matrix, unlike [16] and [14] Baum Welch algorithm will not be used at all to adjust the transition and emission probabilities. However the rest of the steps will be used as per [1] which is using the Baum-Welch algorithm to adjust the probabilities of the HMM and then using Viterbi algorithm to determine the most likely path that a user is going to take then assigning reach buzz rank formula to finally calculate influence of the user.…”

Section: A Buzz Approachmentioning

confidence: 99%

“…Baum-Welch [13] algorithm is considered a hill climbing algorithm [14] and different initial values of HMM provide different results. [15] and [16] both have proposed use of Genetic Algorithms (GAs) to evolve structure of HMM. The structure of HMM are the number of hidden states and the topology of hidden states, evolutionary algorithm was used to identify the best hidden states that are true representative of the problem.…”

Section: Introductionmentioning

confidence: 99%

Evolving HMM for ranking Twitter influence

Thawerani

Ghani

2015

2015 International Conference on Information and Communication Technologies (ICICT)

View full text Add to dashboard Cite

Identifying influence of users in a twitter network has been researched from quite some time. Many researchers have proposed different models for calculating influence of a particular user in a twitter network. The motivation has been to target such users for digital marketing or to solicit users who might be performing terrorist activities. The static influence of user has been captured through topology based methods and temporal influence is captured through HMM model. In this research an evolutionary based HMM model for capturing the temporal influence of a twitter user has beenproposed. The reason is Baum Welch algorithm normally used to determine the emission and transition probabilities may converge to a local optimum point. Evolutionary algorithms search random portions of entire solution space and the probability of finding global optima increases.

show abstract

“…Lastly, the training of HMM is computationally intensive and there is no known method that can guarantee to obtain the most optimal model. Kwong et al [11] and Bhuriyakorn et al [12] have applied Genetic Algorithms to evolve the structure of the HMM for speech recognition. In similar work, Kyoung-Jae et al [8] applied this technique to DNA recognition.…”

Section: Introductionmentioning

confidence: 99%

An evolutionary approach for determining Hidden Markov Model for medical image analysis

Goh

Tang

Saleh

2012

2012 IEEE Congress on Evolutionary Computation

View full text Add to dashboard Cite

Abstract-Hidden Markov Model (HMM) is a technique highly capable of modelling the structure of an observation sequence. In this paper, HMM is used to provide the contextual information for detecting clinical signs present in diabetic retinopathy screen images. However, there is a need to determine a feature set that best represents the complexity of the data as well as determine an optimal HMM. This paper addresses these problems by automatically selecting the best feature set while evolving the structure and obtaining the parameters of a Hidden Markov Model. This novel algorithm not only selects the best feature set, but also identifies the topology of the HMM, the optimal number of states, as well as the initial transition probabilities.

show abstract

A Genetic Algorithm-aided Hidden Markov Model Topology Estimation for Phoneme Recognition of Thai Continuous Speech

Cited by 15 publications

References 10 publications

Network anomaly detection by continuous hidden markov models: An evolutionary programming approach

Network anomaly detection by continuous hidden markov models: An evolutionary programming approach

Evolving HMM for ranking Twitter influence

An evolutionary approach for determining Hidden Markov Model for medical image analysis

Contact Info

Product

Resources

About