Search or ask a question

How to solve hidden Markov model?

Hidden Markov model

Maximum-entropy Markov model

Answers from top 7 papers

PDF

Open Access

More filters

Papers (7)	Insight
Proceedings Article•DOI Internet banking fraud detection using HMM Sunil Mhamane, L. M. R. J. Lobo - Show less +1 more 26 Jul 2012 16 Citations	We propose a model to overcome all these difficulties using Hidden Markov Model.
Journal Article•DOI Speech recognition using hidden Markov models: a CMU perspective Kai-Fu Lee, Hsiao-Wuen Hon, Mei-Yuh Hwang, Xuedong Huang - Show less +3 more 01 Dec 1990-Speech Communication 96 Citations	Publisher Summary This chapter discusses that Hidden Markov models (HMMs) have become the predominant approach for speech recognition systems.
Journal Article•DOI Use of HMM for evaluation of maintenance activities Pascal Vrignat, Manuel Avila, Florent Duculty, Frédéric Kratz - Show less +3 more 24 Aug 2010-International Journal of Adaptive and Innovative Systems 14 Citations	We also show that hidden Markov models can be used according to the right choice of parameters.
Open access•Proceedings Article•DOI Modeling interleaved hidden processes Niels Landwehr 05 Jul 2008 34 Citations	It is shown to be more accurate than a standard hidden Markov model in this domain.
Proceedings Article•DOI Probabilistic face authentication using Hidden Markov Models Manuele Bicego, Enrico Grosso, Massimo Tistarelli - Show less +2 more 28 Mar 2005 9 Citations	The results obtained are really promising, showing the wide applicability of the Hidden Markov Models methodology.
Journal Article•DOI Asymptotic behavior of Bayes estimators for hidden Markov models with application to ion channels M. C. M. De Gunst, O. Shcherbakova - Show less +1 more 01 Dec 2008-Mathematical Methods of Statistics 20 Citations	We show that our theorem is applicable to a wide class of hidden Markov models.
Journal Article•DOI Simulation of hidden Markov models with EXCEL W. H. Laverty, M. J. Miket, I. W. Kelly - Show less +2 more 01 Mar 2002-The Statistician 15 Citations	This can be a very valuable aid in the understanding of hidden Markov models.

My columns

Related Questions

How hidden markov models and concatentative helped in education?5 answersHidden Markov Models (HMMs) have been used in education to teach and learn about various topics. HMMs are particularly useful in temporal pattern recognition tasks such as speech, handwriting, and gesture recognition, as well as robot localization. They provide a framework for modeling problems where external sequential evidence is used to derive underlying state information, such as intended words and gestures. In the field of data-driven learning and inference, HMMs are employed to infer unknown variables and quantities, along with other techniques like Bayesian inference, Monte Carlo Markov Chain methods, and maximum-likelihood estimation. Additionally, HMMs have been used in educational web servers like HMMTeacher, which helps users learn and create their first HMM without programming skills. Overall, HMMs and concatenative methods have contributed to improving the intelligibility and distinguishability of synthesized voices in education.

How effective are Hidden Markov Models in predicting market trends for algorithmic trading?4 answersHidden Markov Models (HMMs) have shown effectiveness in predicting market trends for algorithmic trading. HMMs have been applied in various fields, including quantitative investment in the financial market. Studies have used HMMs to forecast stock prices and have achieved accurate predictions for different stock features such as open, close, high, and low prices. HMMs have also been used to understand finance variables in the stock market, exploring relationships between changing share values and influencing indicators. Additionally, HMMs have been utilized to predict stock exchange indices, resulting in improved accuracy and less prediction errors compared to other Markov family models. HMMs combined with other models like ARIMA have been used to find financial market trends, aiding decision-making in stock trading. Overall, HMMs have demonstrated their effectiveness in predicting market trends for algorithmic trading, providing valuable insights for investment strategies.

How can hidden Markov models be used for portfolio optimization? Code using python?5 answersHidden Markov models (HMMs) can be used for portfolio optimization by incorporating regime-switching behavior and capturing the time-varying nature of financial markets. HMMs allow for modeling the assets of a portfolio through a hidden state process, where the drift and volatility can switch between different states. This enables the portfolio to react to changes in market conditions and avoid left tail events. By utilizing HMMs in portfolio optimization, researchers have found that their strategies often outperform naive investment strategies, such as equal weights. Additionally, model predictive control can be used to dynamically optimize the portfolio based on forecasts from the HMM. Python provides an open-source implementation of HMMs, called PyHHMM, which includes features like different initialization algorithms, missing data inference, and model order selection criteria.

How is a markov chain used in a hidden markov model?3 answersA hidden Markov model (HMM) uses a Markov chain to model signals. Markov chains represent data as a sequence of states in time, where each new state depends only on the previous state. HMMs are a type of Markov chain that incorporate hidden states to solve the problem of incomplete observations. They allow a sequence of observations to determine the real states of the system. HMMs are useful in various applications, such as modeling multivariate processes and non-stationary processes with a life-cycle component. They minimize the information loss between the HMM and the true data-generating process by providing an optimal grid and transition probability matrix. HMMs have been used in diverse fields, including asset-pricing models and life-cycle consumption-savings models. Additionally, HMMs have been applied to determine the behavior of animals, such as sharks, based on their trajectories.

How can the Hidden Markov Model be used in computer vision?5 answersHidden Markov Models (HMMs) are widely used in computer vision for various applications. One application is in recognizing human motion, such as artistic performances like Peking opera, where the poses are more complex. In this case, a Gaussian Mixture based HMM (GMM-HMM) is used for motion recognition, along with a method to filter abnormal and missing data. HMMs are also used in proximity capacitive sensors to detect user gestures, where the models are trained using probability concepts to obtain correct input signal probability feedback. In visual inspection, HMMs are applied to analyze fixations' sequences during the inspection of front panels, allowing for the differentiation between expert and novice operators. Additionally, HMMs are used in hand gesture recognition, where they are combined with trajectory code words and other techniques to achieve high accuracy and real-time performance.

Where is hidden Markov model is used?8 answers

See what other people are reading

Purposeful and purposive sampling the same?

"Purposeful sampling" and "purposive sampling" are often used interchangeably in research literature. Both terms refer to the intentional selection of samples based on specific criteria or purposes relevant to the study. Purposeful sampling involves selecting information-rich cases related to the phenomenon of interest, commonly used in qualitative research. It is emphasized that good sampling techniques are crucial for research design and interpretation, highlighting the importance of purposeful sampling strategies. Contrary to traditional beliefs that random sampling may introduce bias, recent studies suggest that both random and purposive sampling methods can generate consistent and unbiased estimators of population parameters. Therefore, while the terms may have slight nuances in specific contexts, they generally refer to the deliberate selection of samples based on defined criteria for research purposes.Is there a protein timing window?

Nutrient timing, particularly around exercise sessions, has been a topic of interest in the context of protein and carbohydrate consumption for optimizing body composition and exercise performance. The post-exercise period is often highlighted as crucial for nutrient timing, with claims of an anabolic "window of opportunity" to enhance muscular adaptations. However, recent evidence has challenged the classical view of this post-exercise window's significance in relation to anabolism. Studies have shown that the timing of protein synthesis inhibition in memory formation processes can vary, with some indicating a single period of sensitivity lasting between 10 minutes and 1 hour after training. In the realm of protein conformational dynamics, innovative methods have been proposed to enhance the robustness and power of Markov state models without disrupting their Markovian properties, particularly beneficial for slow dynamic modes in complex biological processes.In what ways does algorithmic decision-making impact the eligibility criteria, terms, and consumer targeting strategies of BNPL providers?

Algorithmic decision-making significantly influences the eligibility criteria, terms, and consumer targeting strategies of Buy Now Pay Later (BNPL) providers. BNPL schemes have evolved to allow for the purchase of smaller items over time, attracting merchants in Islamic countries to explore BNPL options. Algorithms play a crucial role in determining eligibility for targeted social policies, including BNPL, affecting who benefits from these schemes. Moreover, algorithmic decision autonomy impacts consumer purchase decisions, with self-efficacy mediating this relationship and power distance moderating it, influencing consumer targeting strategies for BNPL providers. The use of AI methods in poverty-based targeting, like BNPL, can enhance accuracy and extend coverage to more individuals without increasing costs, showcasing the impact of algorithmic rules on eligibility criteria and consumer targeting strategies.What are the key advantages of using machine learning for speech-to-text conversion compared to traditional methods?

Machine learning offers significant advantages over traditional methods for speech-to-text conversion. By utilizing machine learning techniques, such as Hidden Markov Models (HMM), state-of-the-art models can accurately transcribe audio inputs into text. These models have been tested on various Indian languages, showcasing their effectiveness in converting audio to text with high precision. Additionally, machine learning enables the development of integrated systems that can convert not only audio but also text and images into text, expanding the scope of applications for speech-to-text conversion. Furthermore, machine learning-based systems, like the proposed text-to-speech converter, enhance accessibility for visually impaired individuals by converting text from images into audio format. Overall, machine learning empowers more accurate, versatile, and inclusive speech-to-text conversion systems compared to traditional methods.How does the post-hoc analysis help researchers determine which groups have significantly different means?

Post-hoc analysis aids researchers in identifying groups with significantly different means after an omnibus test indicates an overall difference across multiple groups. When the omnibus test is significant, post-hoc tests are crucial for pinpointing specific group differences, ensuring a comprehensive understanding of the data. These tests, such as the Fisher LSD, Tukey HSD, Bonferroni, Newman-Keuls, Dunnett, and Scheffe test, help determine which group or groups contribute to the observed differences in means, enhancing the interpretability of research findings. By conducting post-hoc pairwise group comparisons following a significant omnibus test, researchers can unveil the nuances of group differences that might not be apparent from the initial analysis, leading to more informed conclusions in biomedical and psychosocial research studies.How hidden Markov chains are used for diagnosis and prognosis of faults in technical processes ?

Hidden Markov Models (HMMs) and their variants play a crucial role in the diagnosis and prognosis of faults in technical processes by leveraging their ability to model stochastic processes and capture the dynamics of systems with hidden states. For diagnosis, HMMs are employed to detect anomalies and identify the cause of faults in industrial processes. For instance, a novel fault detection scheme based on HMM is presented for multimode processes with transitions, effectively separating measurement data of stable and transitional modes for accurate fault detection. Similarly, a diagnoser based on HMM is built to identify the cause of alarm signals in industrial settings, demonstrating high accuracy in determining the probable cause of generated alarms. This approach is further enhanced by incorporating partially observed stochastic Petri nets for characterizing detected faults by estimating their occurrence date, thereby improving diagnosis. For prognosis, HMMs and their extensions are utilized to predict the remaining useful life (RUL) of equipment. A hidden semi-Markov model (HSMM) methodology introduces explicit state durational distribution parameters, overcoming limitations of traditional HMMs and improving diagnostic and prognostic accuracy for fatigue cracks in helicopter main gearbox planet carriers. Additionally, a first-order uncertain hidden semi-Markov process (1-UHSMP) is defined to build a degradation model from scarce data and derive adaptively the RUL with associated uncertainty interval, showcasing its effectiveness in practical engineering applications where data may be limited. Moreover, a hybrid system combining HMM with Bayesian Networks (BN) is employed for predicting and isolating faults in the Tennessee Eastman process, demonstrating the system's capability to predict and accurately isolate identified faults. This multifaceted approach, leveraging the strengths of HMMs and their variants, underscores the versatility and effectiveness of these models in both diagnosing the current state of a system and prognosticating its future state, thereby ensuring timely maintenance and preventing potential failures in technical processes.What is an experimental design with two independent combinations?

An experimental design with two independent combinations refers to a methodology that allows researchers to simultaneously test the effects of multiple independent variables. Factorial experimental designs, as highlighted in, are commonly used for this purpose, enabling the exploration of mediator and moderator effects in social science research. These designs are valuable for theory development and testing. Additionally, optimal experimental designs, as discussed in, play a crucial role in models with simultaneous equations, particularly when considering the impact of exogenous and endogenous variables. By utilizing factorial designs and optimal experimental designs, researchers can efficiently investigate the influence of different variables on outcomes, leading to a comprehensive understanding of complex systems and phenomena.Is there classification error with self-report survey?

Self-report surveys often lead to classification errors due to factors like underestimation or overestimation. These errors can impact various fields, such as healthcare, where accurately identifying conditions like dementia is crucial. The presence of measurement errors in self-reported data is a common issue, leading to discrepancies between reported and true values. In the case of surveys with multiple recall windows, methods have been developed to address misclassification issues caused by recall errors. Over time, the accuracy of self-reported data in classifying obesity has shown a trend of increasing underreporting, particularly in the overweight and obese categories. This highlights the need for caution when relying solely on self-reported data for classification purposes.What are the advantages and disadvantages of modeling of a nonlinear machine?

Modeling a nonlinear machine presents various advantages and disadvantages. Advantages include the ability to integrate magnetic saturation, spatial harmonics, and cross coupling in a mathematical expression for improved accuracy and real-time application. Additionally, combining a Takagi-Sugeno fuzzy model with long-short term memory cells can enhance interpretability and approximation abilities. On the other hand, nonlinear models may face challenges such as high sensitivity to non-linear distortions like intermodulation distortion, requiring complex mathematical modeling for evaluation. Furthermore, issues related to nonlinear modeling can arise when dealing with duration, count data, or nonparametric estimation in the presence of individual or time-specific effects. Despite these challenges, advancements in modeling techniques aim to address these limitations and improve the overall performance of nonlinear machine models.What is the definition for active sampling in analytical chmistry?

Active sampling in analytical chemistry refers to a strategic approach that involves iteratively selecting optimal subsamples for data collection based on machine learning predictions on unseen data. This method aims to overcome computational and economic constraints by improving the efficiency of parameter estimation in processes with shared and private parameters. It entails dynamic decision-making on which processes to sample, determining stopping criteria for data collection, and employing estimators for all parameters involved. By utilizing data-driven sampling decisions and conditional estimation cost functions, active sampling enables the sequential estimation of parameters with fewer samples while ensuring reliable estimates. This approach has shown significant performance enhancements compared to traditional sampling methods, particularly in scenarios like safety assessment of advanced driver assistance systems.How is the triangle distribution used in statistical analysis, and what are some common applications of this distribution?

The triangle distribution is utilized in statistical analysis for network description, model building, and estimation tasks. It is particularly relevant in the context of network analysis, where triangle counts per vertex and edge play a crucial role. Applications of the triangle distribution include generating random graphs with specified properties, such as degree sequences and triangle counts. Additionally, the triangle distribution is employed in classification frameworks that involve mean/location and covariance matrix statistics, leveraging Riemannian geometry for symmetric positive definite matrices. Furthermore, the triangle distribution is instrumental in demonstrating quantum nonlocality in network scenarios, allowing for the characterization of underlying quantum strategies and certifiable randomness.