A Hybrid algorithm for estimation of the parameters of Hidden Markov Model based acoustic modeling of speech signals using constraint-based genetic algorithm and expectation maximization
- Authors: Ghosh, Ranadhir , Huda, Shamsul , Yearwood, John
- Date: 2005
- Type: Text , Conference paper
- Relation: Paper presented at the Workshop in Learning Algorithms for Pattern Recognition, in conjunction with the 18th Australian Joint Conference on Artificial Intelligence, Sydney : 5th December, 2005
- Full Text: false
- Reviewed:
- Description: E1
- Description: 2003001368
A variable initialization approach to the EM algorithm for better estimation of the parameters of hidden Markov Model based acoustic modeling of speech signals
- Authors: Huda, Shamsul , Ghosh, Ranadhir , Yearwood, John
- Date: 2006
- Type: Text , Conference paper
- Relation: Paper presented at Artificial Intelligence, Advances in Data Mining, Applications in Medicine, Web Mining, Marketing, Image and Signal Mining Conference 2006, Leipzig, Germany : 14th July, 2006 p. 416-430
- Full Text: false
- Reviewed:
- Description: The traditional method for estimation of the parameters of Hidden Markov Model (HMM) based acoustic modeling of speech uses the Expectation-Maximization (EM) algorithm. The EM algorithm is sensitive to initial values of HMM parameters and is likely to terminate at a local maximum of likelihood function resulting in non-optimized estimation for HMM and lower recognition accuracy. In this paper, to obtain better estimation for HMM and higher recognition accuracy, several candidate HMMs are created by applying EM on multiple initial models. The best HMM is chosen from the candidate HMMs which has highest value for likelihood function. Initial models are created by varying maximum frame number in the segmentation step of HMM initialization process. A binary search is applied while creating the initial models. The proposed method has been tested on TIMIT database. Experimental results show that our approach obtains improved values for likelihood function and improved recognition accuracy.
- Description: E1
- Description: 2003001542
Hybrid training approaches to Hidden Markov Model-based acoustic models for automatic speech recognition
- Authors: Huda, Shamsul
- Date: 2008
- Type: Text , Thesis , PhD
- Full Text:
- Description: Doctor of Philosophy
A constraint-based evolutionary learning approach to the expectation maximization for optimal estimation of the hidden Markov model for speech signal modeling
- Authors: Huda, Shamsul , Yearwood, John , Togneri, Roberto
- Date: 2009
- Type: Text , Journal article
- Relation: IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics Vol. 39, no. 1 (2009), p. 182-197
- Full Text:
- Reviewed:
- Description: This paper attempts to overcome the tendency of the expectation-maximization (EM) algorithm to locate a local rather than global maximum when applied to estimate the hidden Markov model (HMM) parameters in speech signal modeling. We propose a hybrid algorithm for estimation of the HMM in automatic speech recognition (ASR) using a constraint-based evolutionary algorithm (EA) and EM, the CEL-EM. The novelty of our hybrid algorithm (CEL-EM) is that it is applicable for estimation of the constraint-based models with many constraints and large numbers of parameters (which use EM) like HMM. Two constraint-based versions of the CEL-EM with different fusion strategies have been proposed using a constraint-based EA and the EM for better estimation of HMM in ASR. The first one uses a traditional constraint-handling mechanism of EA. The other version transforms a constrained optimization problem into an unconstrained problem using Lagrange multipliers. Fusion strategies for the CEL-EM use a staged-fusion approach where EM has been plugged with the EA periodically after the execution of EA for a specific period of time to maintain the global sampling capabilities of EA in the hybrid algorithm. A variable initialization approach (VIA) has been proposed using a variable segmentation to provide a better initialization for EA in the CEL-EM. Experimental results on the TIMIT speech corpus show that CEL-EM obtains higher recognition accuracies than the traditional EM algorithm as well as a top-standard EM (VIA-EM, constructed by applying the VIA to EM). © 2008 IEEE.
A stochastic version of Expectation Maximization algorithm for better estimation of Hidden Markov Model
- Authors: Huda, Shamsul , Yearwood, John , Togneri, Roberto
- Date: 2009
- Type: Text , Journal article
- Relation: Pattern Recognition Letters Vol. 30, no. 14 (2009), p. 1301-1309
- Full Text: false
- Reviewed:
- Description: This paper attempts to overcome the local convergence problem of the Expectation Maximization (EM) based training of the Hidden Markov Model (HMM) in speech recognition. We propose a hybrid algorithm, Simulated Annealing Stochastic version of EM (SASEM), combining Simulated Annealing with EM that reformulates the HMM estimation process using a stochastic step between the EM steps and the SA. The stochastic processes of SASEM inside EM can prevent EM from converging to a local maximum and find improved estimation for HMM using the global convergence properties of SA. Experiments on the TIMIT speech corpus show that SASEM obtains higher recognition accuracies than the EM. © 2009 Elsevier B.V. All rights reserved.
Constraint-based evolutionary learning approach to the non-normal process performance evaluation
- Authors: Ahmad, S. , Huda, Shamsul , Bakir, S. , Abdollahian, Mali , Zeephongsekul, P.
- Date: 2009
- Type: Text , Conference paper
- Relation: Paper presented at 3rd International Conference on Informatics and Technology 2009: ICT Opportunities in the Current Global Recession, Kuala Lumpur, Malaysia : 27th-28th October 2009 p. 26-33
- Full Text: false
- Description: Performance of industrial products is very important for an industry. Conventional methods for performance analysis consider a normality assumption and limited to low dimensional data. Different manufacturing processes very often have products with quality characteristics that do not follow normal distribution. In such cases fitting a known non-normal distribution to these quality characteristics would lead to erroneous results while assessing the performance of products. In this paper, we propose a novel method for non-normal multivariate process performance analysis. We have proposed a Constraint-based Evolutionary Algorithm (EA) approach for optimal estimation of the parameters of non-normal multivariate process. Furthermore, a geometric distance based method has been employed to reduce higher dimensionality of data to lower dimension. The efficacy of the proposed method is assessed by using the proportion of nonconformance (PNC) criterion to summarize the performance of EA approach. The experimental results from constraint-based EA have been compared to those obtained using steepest descent and simulated annealing (SA) approaches.
- Description: 2003007898
Process performance evaluation using evolutionary algorithm
- Authors: Ahmad, S. , Huda, Shamsul , Bakir, S. , Abdollahian, Mali , Zeephongsekul, P.
- Date: 2009
- Type: Text , Conference paper
- Relation: Paper presented at 2009 International Conference on Information & Knowledge Engineering, IKE 2009, Las Vegas, Nevada, U.S.A. : 13th-16th July 2009 p. 731-737
- Full Text:
- Description: Nowadays every business is using different quantitative measures and techniques to assess performance of their products / services. It is well known that different manufacturing processes very often manufacture products with quality characteristics that do not follow normal distribution. In such cases, fitting a known non-normal distribution to these quality characteristics would lead to erroneous results. Furthermore, there is always more than one characteristic Critical to Quality (CTQ) in the process outcomes and very often these quality characteristics are correlated with each other. In this paper, we assess performance of such a bivariate process data which is non-normal as well as correlated. We will use the geometric distance approach to reduce the dimension of the correlated non-normal bivariate data and then fit Burr distribution to the geometric distance variable. The optimal parameters of the fitted Burr distribution are estimated using Evolutionary Algorithm (EA). The results are compared with those using Simulated Annealing (SA) algorithm. The proportion of nonconformance (PNC) for process measurements is then obtained by using the fitted Burr distributions based on the two methods. The results based on both search algorithms are then compared with the exact proportion of nonconformance of the data. Finally, a case study using real data is presented.
- Description: 2003008140
Automatic sleep stage identification: difficulties and possible solutions
- Authors: Sukhorukova, Nadezda , Stranieri, Andrew , Ofoghi, Bahadorreza , Vamplew, Peter , Saleem, Muhammad Saad , Ma, Liping , Ugon, Adrien , Ugon, Julien , Muecke, Nial , Amiel, Hélène , Philippe, Carole , Bani-Mustafa, Ahmed , Huda, Shamsul , Bertoli, Marcello , Levy, P , Ganascia, J.G
- Date: 2010
- Type: Text , Conference proceedings
- Full Text:
- Description: The diagnosis of many sleep disorders is a labour intensive task that involves the specialised interpretation of numerous signals including brain wave, breath and heart rate captured in overnight polysomnogram sessions. The automation of diagnoses is challenging for data mining algorithms because the data sets are extremely large and noisy, the signals are complex and specialist's analyses vary. This work reports on the adaptation of approaches from four fields; neural networks, mathematical optimisation, financial forecasting and frequency domain analysis to the problem of automatically determing a patient's stage of sleep. Results, though preliminary, are promising and indicate that combined approaches may prove more fruitful than the reliance on a approach.
Cluster based rule discovery model for enhancement of government's tobacco control strategy
- Authors: Huda, Shamsul , Yearwood, John , Borland, Ron
- Date: 2010
- Type: Text , Conference proceedings
- Full Text:
- Description: Discovery of interesting rules describing the behavioural patterns of smokers' quitting intentions is an important task in the determination of an effective tobacco control strategy. In this paper, we investigate a compact and simplified rule discovery process for predicting smokers' quitting behaviour that can provide feedback to build an scientific evidence-based adaptive tobacco control policy. Standard decision tree (SDT) based rule discovery depends on decision boundaries in the feature space which are orthogonal to the axis of the feature of a particular decision node. This may limit the ability of SDT to learn intermediate concepts for high dimensional large datasets such as tobacco control. In this paper, we propose a cluster based rule discovery model (CRDM) for generation of more compact and simplified rules for the enhancement of tobacco control policy. The clusterbased approach builds conceptual groups from which a set of decision trees (a decision forest) are constructed. Experimental results on the tobacco control data set show that decision rules from the decision forest constructed by CRDM are simpler and can predict smokers' quitting intention more accurately than a single decision tree. © 2010 IEEE.
Exploring novel features and decision rules to identify cardiovascular autonomic neuropathy using a hybrid of wrapper-filter based feature selection
- Authors: Huda, Shamsul , Jelinek, Herbert , Ray, Biplob , Stranieri, Andrew , Yearwood, John
- Date: 2010
- Type: Text , Conference paper
- Relation: Paper presented at the 2010 6th International Conference on Intelligent Sensors, Sensor Networks and Information Processing, ISSNIP 2010 p. 297-302
- Full Text:
- Reviewed:
- Description: Cardiovascular autonomic neuropathy (CAN) is one of the important causes of mortality among diabetes patients. Statistics shows that more than 22% of people with type 2 diabetes mellitus suffer from CAN and which in turn leads to cardiovascular disease (heart attack, stroke). Therefore early detection of CAN could reduce the mortality. Traditional method for detection of CAN uses Ewing's algorithm where five noninvasive cardiovascular tests are used. Often for clinician, it is difficult to collect data from for the Ewing Battery patients due to onerous test conditions. In this paper, we propose a hybrid of wrapper-filter approach to find novel features from patients' ECG records and then generate decision rules for the new features for easier detection of CAN. In the proposed feature selection, a hybrid of filter (Maximum Relevance, MR) and wrapper (Artificial Neural Net Input Gain Measurement Approximation ANNIGMA) approaches (MR-ANNIGMA) would be used. The combined heuristics in the hybrid MRANNIGMA takes the advantages of the complementary properties of the both filter and wrapper heuristics and can find significant features. The selected features set are used to generate a new set of rules for detection of CAN. Experiments on real patient records shows that proposed method finds a smaller set of features for detection of CAN than traditional method which are clinically significant and could lead to an easier way to diagnose CAN. © 2010 IEEE.
Hybrid wrapper-filter approaches for input feature selection using maximum relevance and Artificial Neural Network Input Gain Measurement Approximation (ANNIGMA)
- Authors: Huda, Shamsul , Yearwood, John , Stranieri, Andrew
- Date: 2010
- Type: Text , Conference proceedings
- Full Text:
- Description: Feature selection is an important research problem in machine learning and data mining applications. This paper proposes a hybrid wrapper and filter feature selection algorithm by introducing the filter's feature ranking score in the wrapper stage to speed up the search process for wrapper and thereby finding a more compact feature subset. The approach hybridizes a Mutual Information (MI) based Maximum Relevance (MR) filter ranking heuristic with an Artificial Neural Network (ANN) based wrapper approach where Artificial Neural Network Input Gain Measurement Approximation (ANNIGMA) has been combined with MR (MR-ANNIGMA) to guide the search process in the wrapper. The novelty of our approach is that we use hybrid of wrapper and filter methods that combines filter's ranking score with the wrapper-heuristic's score to take advantages of both filter and wrapper heuristics. Performance of the proposed MRANNIGMA has been verified using bench mark data sets and compared to both independent filter and wrapper based approaches. Experimental results show that MR-ANNIGMA achieves more compact feature sets and higher accuracies than both filter and wrapper approaches alone. © 2010 IEEE.
Smokers' characteristics and cluster based quitting rule discovery model for enhancement of government's tobacco control systems
- Authors: Huda, Shamsul , Yearwood, John , Borland, Ron
- Date: 2010
- Type: Text , Conference paper
- Relation: Proceedings of the 14th Pacific Asia Conference on Information Systems (PACIS 2010)
- Full Text:
- Reviewed:
- Description: Discovery of cluster characteristics and interesting rules describing smokers' clusters and the behavioural patterns of smoker's quitting intentions is an important task in the development of an effective tobacco control systems. In this paper, we attempt to determine the characteristics smokers' cluster and simplified rule for predicting smokers' quitting behaviour that can provide feedback to build a scientific evidence-based adaptive tobacco control systems. Standard clustering algorithm groups the data based on there inherent pattern. "From abstract"
- Description: Discovery of cluster characteristics and interesting rules describing smokers' clusters and the behavioural patterns of smoker's quiiting intentios is an important task in the development of an effective tobacco control systems. In this paper, we attempt to determine the characteristics smokers' cluster and simplified rule for predicting smokers' quitting behaviour that can provide feedback to build a scientific evidence-based adaptive tobacco control systems. Standard clustering algorithm groups the data based on there inherent pattern. "From abstract"
A reinforcement learning approach with spline-fit object tracking for AIBO Robot's high level decision making
- Authors: Mukherjee, Subhasis , Huda, Shamsul , Yearwood, John
- Date: 2011
- Type: Text , Book chapter
- Relation: Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing p. 169-183
- Full Text: false
- Reviewed:
- Description: Robocup is a popular test bed for AI programs around the world. Robosoccer is one of the two major parts of Robocup, in which AIBO entertainment robots take part in the middle sized soccer event. The three key challenges that robots need to face in this event are manoeuvrability, image recognition and decision making skills. This paper focuses on the decision making problem in Robosoccer-The goal keeper problem. We investigate whether reinforcement learning (RL) as a form of semi-supervised learning can effectively contribute to the goal keeper's decision making process when penalty shot and two attacker problem are considered. Currently, the decision making process in Robosoccer is carried out using rule-base system. RL also is used for quadruped locomotion and navigation purpose in Robosoccer using AIBO. Moreover the ball distance is being calculated using IR sensors available at the nose of the robot. In this paper, we propose a reinforcement learning based approach that uses a dynamic state-action mapping using back propagation of reward and Q-learning along with spline fit (QLSF) for the final choice of high level functions in order to save the goal. The novelty of our approach is that the agent learns while playing and can take independent decision which overcomes the limitations of rule-base system due to fixed and limited predefined decision rules. The spline fit method used with the nose camera was also able to find out the location and the ball distance more accurately compare to the IR sensors. The noise source and near and far sensor dilemma problem with IR sensor was neutralized using the proposed spline fit method. Performance of the proposed method has been verified against the bench mark data set made with Upenn'03 code logic and a base line experiment with IR sensors. It was found that the efficiency of our QLSF approach in goalkeeping was better than the rule based approach in conjunction with the IR sensors. The QLSF develops a semi-supervised learning process over the rule-base system's input-output mapping process, given in the Upenn'03 code. © 2011 Springer-Verlag Berlin Heidelberg.
Multivariate control charts for surgical procedures
- Authors: Abdollahian, Malie , Ahmad, S. , Huda, Shamsul
- Date: 2011
- Type: Text , Conference proceedings
- Full Text: false
- Description: Patient's progress in the Intensive Care Unit is characterised by more than one quality characteristics. This paper employs univariate and multivariate control charts to monitor patient progress in the Intensive Care Unit. A definitive comparison is made, between the performance of univariate and multivariate control chart methods, in the monitoring of the patient recovery process. © 2011 ACM.
Reinforcement learning approach to AIBO robot's decision making process in Robosoccer's goal keeper problem
- Authors: Mukherjee, Subhasis , Yearwood, John , Vamplew, Peter , Huda, Shamsul
- Date: 2011
- Type: Text , Conference proceedings
- Full Text: false
- Description: Robocup is a popular test bed for AI programs around the world. Robosoccer is one of the two major parts of Robocup, in which AIBO entertainment robots take part in the middle sized soccer event. The three key challenges that robots need to face in this event are manoeuvrability, image recognition and decision making skills. This paper focuses on the decision making problem in Robosoccer - The goal keeper problem. We investigate whether reinforcement learning (RL) as a form of semi-supervised learning can effectively contribute to the goal keeper's decision making process when penalty shot and two attacker problem are considered. Currently, the decision making process in Robosoccer is carried out using rule-base system. RL also is used for quadruped locomotion and navigation purpose in Robosoccer using AIBO. In this paper, we propose a reinforcement learning based approach that uses a dynamic state-action mapping using back propagation of reward and space quantized Q-learning (SQQL) for the choice of high level functions in order to save the goal. The novelty of our approach is that the agent learns while playing and can take independent decision which overcomes the limitations of rule-base system due to fixed and limited predefined decision rules. Performance of the proposed method has been verified against the bench mark data set made with Upenn'03 code logic. It was found that the efficiency of our SQQL approach in goalkeeping was better than the rule based approach. The SQQL develops a semi-supervised learning process over the rule-base system's input-output mapping process, given in the Upenn'03 code. © 2011 IEEE.
Smart RFID reader protocol for malware detection
- Authors: Ray, Biplob , Huda, Shamsul , Chowdhury, Morshed
- Date: 2011
- Type: Text , Conference proceedings
- Full Text: false
- Description: Radio frequency identification (RFID) is a remote identification technique promises to revolutionize the way a specific object use to identify in our industry. However, large scale implementation of RFID sought for protection, against Malware threat, information privacy and un-traceability, for low cost RFID tag. In this paper, we propose a framework to provide privacy for tag data and to provide protection for RFID system from malware. In the proposed framework, malware infected tag is detected by analysing individual component of the RFID tag. It uses sanitization technique for analysing individual component. Here authentication based shared unique parameters is used as a method to protect privacy. This authentication protocol will be capable of handling forward and backward security and identifying rogue reader better than existing protocols. Using this framework, the RFID system will be protected from malware and the privacy of the tag will be ensured as well. © 2011 IEEE.
Investigating the relationship between neonatal mortality rate and Mother's characteristics
- Authors: Abdollahian, Mali , Ahmad, Shafiq , Huda, Shamsul , Anggraini, D
- Date: 2012
- Type: Text , Conference proceedings
- Relation: WORLDCOMP'12, USA, 16th-19th July published in Proceedings of the 2012 World Congress in Computer Science - Computer Engineering and Applied Computing pg 1-6
- Full Text: false
- Reviewed:
- Description: Neonatal mortality rate (NMR) is an increasingly important public health issues in many developing countries. Neonatal death now accounts for about two-thirds of the eight million infant deaths that occur globally each year. It is welldocumented that low birth weight (LBW) is the most significant factor influencing NMR. This paper deploys regression analysis to explore the relationship between weight of low birth weight babies and various characteristics of mother. The results indicate that there is a significant relationship between weight of low birth weight babies and mother's weight, age, gestation age and hemoglobin level.
An approach for Ewing test selection to support the clinical assessment of cardiac autonomic neuropathy
- Authors: Stranieri, Andrew , Abawajy, Jemal , Kelarev, Andrei , Huda, Shamsul , Chowdhury, Morshed , Jelinek, Herbert
- Date: 2013
- Type: Text , Journal article
- Relation: Artificial Intelligence in Medicine Vol. 58, no. 3 (2013), p. 185-193
- Full Text:
- Reviewed:
- Description: Objective: This article addresses the problem of determining optimal sequences of tests for the clinical assessment of cardiac autonomic neuropathy (CAN) We investigate the accuracy of using only one of the recommended Ewing tests to classify CAN and the additional accuracy obtained by adding the remaining tests of the Ewing battery This is important as not all five Ewing tests can always be applied in each situation in practice Methods and material: We used new and unique database of the diabetes screening research initiative project, which is more than ten times larger than the data set used by Ewing in his original investigation of CAN We utilized decision trees and the optimal decision path finder (ODPF) procedure for identifying optimal sequences of tests Results: We present experimental results on the accuracy of using each one of the recommended Ewing tests to classify CAN and the additional accuracy that can be achieved by adding the remaining tests of the Ewing battery We found the best sequences of tests for cost-function equal to the number of tests The accuracies achieved by the initial segments of the optimal sequences for 2, 3 and 4 categories of CAN are 80.80, 91.33, 93.97 and 94.14, and respectively, 79.86, 89.29, 91.16 and 91.76, and 78.90, 86.21, 88.15 and 88.93 They show significant improvement compared to the sequence considered previously in the literature and the mathematical expectations of the accuracies of a random sequence of tests The complete outcomes obtained for all subsets of the Ewing features are required for determining optimal sequences of tests for any cost-function with the use of the ODPF procedure We have also found two most significant additional features that can increase the accuracy when some of the Ewing attributes cannot be obtained Conclusions: The outcomes obtained can be used to determine the optimal sequences of tests for each individual cost-function by following the ODPF procedure The results show that the best single Ewing test for diagnosing CAN is the deep breathing heart rate variation test Optimal sequences found for the cost-function equal to the number of tests guarantee that the best accuracy is achieved after any number of tests and provide an improvement in comparison with the previous ordering of tests or a random sequence © 2013 Elsevier B.V.
- Description: 2003011130
A hybrid wrapper-filter approach to detect the source(s) of out-of-control signals in multivariate manufacturing process
- Authors: Huda, Shamsul , Abdollahian, Mali , Mammadov, Musa , Yearwood, John , Ahmed, Shafiq , Sultan, Ibrahim
- Date: 2014
- Type: Text , Journal article
- Relation: European Journal of Operational Research Vol. 237, no. 3 (2014), p. 857-870
- Full Text: false
- Reviewed:
- Description: With modern data-Acquisition equipment and on-line computers used during production, it is now common to monitor several correlated quality characteristics simultaneously in multivariate processes. Multivariate control charts (MCC) are important tools for monitoring multivariate processes. One difficulty encountered with multivariate control charts is the identification of the variable or group of variables that cause an out-of-control signal. Expert knowledge either in combination with wrapper-based supervised classifier or a pre-filter with wrapper are the standard approaches to detect the sources of out-of-control signal. However gathering expert knowledge in source identification is costly and may introduce human error. Individual univariate control charts (UCC) and decomposition of T2 statistics are also used in many cases simultaneously to identify the sources, but these either ignore the correlations between the sources or may take more time with the increase of dimensions. The aim of this paper is to develop a source identification approach that does not need any expert-knowledge and can detect out-of-control signal in less computational complexity. We propose, a hybrid wrapper-filter based source identification approach that hybridizes a Mutual Information (MI) based Maximum Relevance (MR) filter ranking heuristic with an Artificial Neural Network (ANN) based wrapper. The Artificial Neural Network Input Gain Measurement Approximation (ANNIGMA) has been combined with MR (MR-ANNIGMA) to utilize the knowledge about the intrinsic pattern of the quality characteristics computed by the filter for directing the wrapper search process. To compute optimal ANNIGMA score, we also propose a Global MR-ANNIGMA using non-functional relationship between variables which is independent of the derivative of the objective function and has a potential to overcome the local optimization problem of ANN training. The novelty of the proposed approaches is that they combine the advantages of both filter and wrapper approaches and do not require any expert knowledge about the sources of the out-of-control signals. Heuristic score based subset generation process also reduces the search space into polynomial growth which in turns reduces computational time. The proposed approaches were tested by exhaustive experiments using both simulated and real manufacturing data and compared to existing methods including independent filter, wrapper and Multivariate EWMA (MEWMA) methods. The results indicate that the proposed approaches can identify the sources of out-of-control signals more accurately than existing approaches. © 2014 Elsevier B.V. All rights reserved.
Hybrid metaheuristic approaches to the expectation maximization for estimation of the hidden markov model for signal modeling
- Authors: Huda, Shamsul , Yearwood, John , Togneri, Roberto
- Date: 2014
- Type: Text , Journal article
- Relation: IEEE Transactions on Cybernetics Vol. 44, no. 10 (2014), p. 1962-1977
- Full Text: false
- Reviewed:
- Description: The expectation maximization (EM) is the standard training algorithm for hidden Markov model (HMM). However, EM faces a local convergence problem in HMM estimation. This paper attempts to overcome this problem of EM and proposes hybrid metaheuristic approaches to EM for HMM. In our earlier research, a hybrid of a constraint-based evolutionary learning approach to EM (CEL-EM) improved HMM estimation. In this paper, we propose a hybrid simulated annealing stochastic version of EM (SASEM) that combines simulated annealing (SA) with EM. The novelty of our approach is that we develop a mathematical reformulation of HMM estimation by introducing a stochastic step between the EM steps and combine SA with EM to provide better control over the acceptance of stochastic and EM steps for better HMM estimation. We also extend our earlier work [1] and propose a second hybrid which is a combination of an EA and the proposed SASEM, (EA-SASEM). The proposed EA-SASEM uses the best constraint-based EA strategies from CEL-EM and stochastic reformulation of HMM. The complementary properties of EA and SA and stochastic reformulation of HMM of SASEM provide EA-SASEM with sufficient potential to find better estimation for HMM. To the best of our knowledge, this type of hybridization and mathematical reformulation have not been explored in the context of EM and HMM training. The proposed approaches have been evaluated through comprehensive experiments to justify their effectiveness in signal modeling using the speech corpus: TIMIT. Experimental results show that proposed approaches obtain higher recognition accuracies than the EM algorithm and CEL-EM as well. © 2014 IEEE.