Part-of-Speech (POS) tagging for code-mixed text is notably challenging due to frequent language switching, non-standard orthography, transliteration issues, and the prevalence of informal syntactic structures in user-generated content. This study presents a Hidden Markov Model (HMM)-based approach tailored to English–Punjabi code-mixed text, specifically addressing bilingual interactions in which Punjabi is written in Romanized script. A code-mixed corpus was strategically compiled from platforms such as YouTube, Facebook, and WhatsApp, and meticulously annotated at the token level. The resulting dataset comprises 900 sentences totalling 10,117 words, showcasing diverse mixing patterns and typical social media artefacts, including abbreviations, emojis, and irregular punctuation. The proposed framework conceptualizes POS tagging as a sequence labelling problem. It estimates emission and transition probabilities through the annotated corpus and employs the Viterbi algorithm to decode the most probable tag sequences. Experimental evaluations yield an overall tagging accuracy of 71.52%, establishing a probabilistic baseline for English–Punjabi code-mixed POS tagging. This work lays the groundwork for future research to integrate richer feature sets and leverage neural architectures to enhance performance.

ABSTRACT In this article, we introduce a regime-switching matrix factor model in which the mean, loading matrix, and noise processes change with different regimes, regulated by a hidden Markov chain. We devise an iterative algorithm to estimate parameters of the matrix factor model, starting with an initial estimation of Markov states to estimate the factor loadings and dimensions of latent factors based on eigenanalysis, and subsequently utilizing the Viterbi algorithm to estimate the Markov states, and repeat the procedures until the convergence criterion is satisfied. Furthermore, we establish the asymptotic properties of the parameter estimates and demonstrate that the convergence rate can be improved by the presence of strong factors in stages with weak factors, as supported by numerical simulations. Empirical studies on nine indicators of 10 component stocks in Energy industry of S&P 500 show that the regime-switching matrix factor model is reasonable since it successfully distinguishes between two scenarios of market states which are stable and volatile downward with high and low frequency, respectively, and also works better than the matrix factor model according to RSS/TSS.

Accurate parental haplotype information is crucial for noninvasive prenatal diagnosis (NIPD) of recessive single-gene disorders (SGD) (NIPD-SGD). However, conventional approaches rely on complex experimental techniques or trio-based SNP linkage analysis requiring family members, which limit clinical application. Here, we present a novel direct haplotyping based NIPD approach termed DiHNIPD, utilizing single-tube long fragment read (stLFR) sequencing. First, parental genome-wide haplotypes were reconstructed by stLFR-based whole genome sequencing (WGS). Second, SNPs within and surrounding the target gene in maternal plasma were identified by WGS. Finally, fetal haplotypes were determined by implementing a parental haplotype-assisted hidden Markov model combined with the Viterbi algorithm. The DiHNIPD results were further confirmed by invasive prenatal diagnosis. This study recruited 23 couples at risk of having a fetus with SGD. DiHNIPD directly phased all parental haplotypes of the target gene and accurately deduced 23 fetal genotypes, achieving 100% concordance with diagnostic results. These results demonstrate that DiHNIPD is a sensitive, user-friendly and inexpensive strategy for NIPD-SGD. It eliminates the need for expensive devices and complex procedures and does not rely on family members. This method shows significant promise for clinical use in high-risk pregnancies without prior offspring.

Accurate segmentation of heart sound signal stages is critical in cardiovascular disease analysis. This study proposed the integration of a duration hidden Markov model (DHMM) with a temporal convolutional network (TCN) and an adaptive calibration mechanism (based on electrocardiogram signals) to enable the precise segmentation of complex heart sound signals. Multiple features of heart sound signals are extracted and utilized as model inputs, constructed a segmentation model architecture improved by TCN-based observation probability estimation and an attention mechanism integrated into the Viterbi algorithm. The experimental results demonstrated that the average accuracy of this method is 94.71 ± 2.64% at a segmentation error of 50ms. The enhanced Viterbi algorithm elevated performance by approximately 9 percentage points. Furthermore, the adaptive calibration mechanism yielded an additional average accuracy increase of 1.41 percentage points and reduced the standard deviation by 1.21 percentage points.Conclusion: Compared to traditional methods employing Gaussian distribution-based observation probability estimation, the utilization of a TCN substantially enhanced state discrimination accuracy, achieving an improvement of approximately 3 percentage points.The refined Viterbi algorithm demonstrated superior performance relative to prior methodologies. This method enables effective segmentation of complex heart sound data, delivering a high-precision solution for the automated analysis of heart sounds. Our code can be found in https://github.com/KC-Y-bjut/Heart-sound-segmentation.

Periapical radiographs remain the first-line imaging modality in endodontics due to accessibility and low radiation dose, whereas cone-beam computed tomography (CBCT) is reserved for inconclusive cases or suspected anatomical complexity. We propose a physics- and geometry-aware preprocessing pipeline coupled with sliding-window Viterbi tracking to enhance canal visibility and recover plausible root canal trajectories directly from routine periapical images. The pipeline standardizes row-wise brightness, compensates for the cone-like tooth density profile (Tukey window), and suppresses noise prior to dynamic-programming inference, requiring only minimal operator input (two-point orientation and region of interest). In a retrospective evaluation against micro-computed tomography (micro-CT)/CBCT reference anatomy, the approach accurately localized canals on periapicals under study conditions, suggesting potential as a rapid, chairside aid when 3D imaging is unavailable or deferred.

Uncovering molecular trajectories in follicular lymphoma using hidden markov models of ctdna dynamics

Error correction in the results of atrial fibrillation (AF) detected with a convolutional neural network (CNN) classifier was investigated using the Viterbi algorithm. State transition probabilities were calculated using annotations of Holter ECG recordings obtained from 60 subjects. Errors in the detection results were corrected based on the Viterbi algorithm using the pre‐calculated state transition probabilities and the emission probabilities output by the CNN classifier. As a result, the detection results were appropriately corrected, and the detection performance was improved, demonstrating the effectiveness of the proposed method. © 2025 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.

The paper presents a calculation method and analytical relationships for determining the efficiency of using error-correcting channel coding in systems with a binary block code of Bose – Chaudhuri – Hocquenghem (BCH), M -position modulation and decoding using the Viterbi algorithm with a soft decision (SOVA). A generalized decoding efficiency indicator was used, which took into account the code parameters, modulation format and decoder operating mode. In comparison with the known method, the proposed method does not require knowledge of the weight components of the code spectrum and the use of computer modeling procedures in calculations. The obtained analytical relationships are presented in a compact form and general for research. They are used to calculate the error immunity, energy gain from coding, the correcting ability of the decoder and the information efficiency of communication channels with variations in code parameters and error probabilities at the decoder output.

A new nano-based architectural design of multiple-stream convolutional homeomorphic error-control coding will be conducted, and a corresponding hierarchical implementation of important class of the homeomorphic Viterbi algorithm within convolutional homeomorphic error-control coding using lattice networks via nano carbon-based field emission controlled-switching will be achieved. Error-correction coding is highly important in modern wireless networking and digital data transaction since channel coding is required to counteract data errors that are encountered in wireless data networking due to the corresponding existence of unavoidable channel noise. Further, the new lattice nano-based implementation will be useful for enhancing the error-control system performance such as the corresponding enhancements within error-correcting capability, regular synthesis, speed improvement and the minimization of power consumption. Logic homeomorphism describes properties-preserving logic mapping that is intrinsically bijective. Homeomorphic property in error-control coding is important since it is shown that the homeomorphism relationship between multiple-streams of data can be used for further correction of errors that are uncorrectable using the implemented decoding algorithm such as in the case of triple-errors that are uncorrectable using the classical Viterbi algorithm. Applications of the new regular nano-based homeomorphic architecture include low-power design of circuits and systems for enhanced and more reliable wireless data networking and transaction.

Photonics-aided terahertz (THz) transmission leverages the abundant spectral resources of the THz band and enables seamless integration with optical fiber networks, representing a key pathway toward 6G communication. Meanwhile, spectrally efficient frequency division multiplexing (SEFDM) can improve spectral efficiency for optical and wireless communications, thereby further increasing the capacity of photonics-aided THz transmission systems, especially when the bandwidth of existing THz devices is limited due to cost constraints. In this work, we experimentally investigate and compare the performance of conventional SEFDM and the proposed non-orthogonal discrete Fourier transform spread (NO-DFT-S) SEFDM in a photonics-aided THz transmission system. In order to eliminate the inter-carrier interference (ICI) intentionally induced by bandwidth compression, three optional signal detection algorithms, including Volterra nonlinear equalization (VNLE), I/Q separation iterative detection (ID), and maximum a-posteriori (MAP)-Viterbi algorithm, are individually employed for performance comparison. Moreover, due to bandwidth compression relative to orthogonal frequency division multiplexing (OFDM) and a reduction in peak-to-average power ratio (PAPR) compared to conventional SEFDM, the capacity of NO-DFT-S SEFDM is increased by 31.3% over OFDM and 23.5% over conventional SEFDM under a BER threshold of 2.4e-2 @20% soft-decision forward error correction (SD-FEC). Specifically, by employing high-gain THz modules such as THz lenses and low-noise amplifiers (LNAs), a maximum line rate of 168 Gbit/s and a net rate of 129.2 Gbit/s are successfully achieved over a 100-m THz wireless transmission at 300 GHz using the NO-DFT-S SEFDM scheme.

The development of the Internet of Things (IoT) opens up new opportunities for creating intelligent services that enhance user interaction with surrounding devices. Modern IoT systems primarily use touchscreens and mobile applications for control; however, gesture-based methods can significantly expand their functionality. This work proposes a gesture recognition system applied to the control of IoT devices. The core of the system is the classification of finger movement trajectories using a Hidden Markov Model (HMM). The system consists of three main stages: initial hand segmentation using colour and depth information, fingertip detection based on hand contours, and the use of clustering in polar coordinates to extract dynamic features. The Baum-Welch and Viterbi algorithms are applied for training and gesture recognition, respectively. Experimental results show that the developed system is capable of classifying gestures with consideration of spatiotemporal variability with high accuracy. In particular, the average recognition rate reached 98.61% for the training set and 93.06% for the test data. The proposed approach demonstrates effectiveness under challenging conditions, including changes in lighting and partial occlusion of objects in the scene.

Hidden Markov Models (HMMs) are widely used in many sequential decision-making problems due to their ability to model time-related dependencies. The standard decoding methods in these models, such as the Viterbi algorithm, are limited by their dependence on past observations only. Thus, this leads to unpredictability when future information is available. In this work, we propose a decoding strategy called Beam-Limited k-Step Lookahead that looks k-step ahead, drawing parallels to k-step discrete control synthesis, to make use of future information. The proposed method achieves a balance between decoding accuracy and computational complexity by constraining the search space to the top M most promising paths. Experimental results on synthetic HMM data show that our new decoding strategy significantly improves decoding accuracy over classical Viterbi decoding. The findings highlight the potential of this new strategy to improve the performance of sequential decoding systems.

ABSTRACT An effective tactical asset management program, incorporating condition and degradation assessments, is essential for asset‐intensive organizations to make informed decisions about maintenance and renewal. Since asset degradation is an inherently stochastic phenomenon, models based on stochastic processes are the most suitable approach for accurately predicting it. Asset condition assessment and event data are collected and recorded during inspection and maintenance activities to model equipment degradation. A degradation model that integrates these data associated with assets is highly desirable for predicting the effective and reliable remaining useful life (RUL). This study develops a novel Degradation Hidden Semi‐Markov Model that uses both failure and condition data to predict RUL of critical pumps in the wastewater network of a regional town in South Australia. To evaluate the model's performance and outcomes, a portion of the condition data was reserved as the test dataset, while the remaining historical condition data was used for degradation and RUL modeling. In addition, the predicted RUL was validated using both forward‐backward and Viterbi algorithms. Results showed not only the expected asset health state closely followed of the actual, but also the absolute prediction error of the estimated RUL using these algorithms was minimal.

In recent years, the photoconductive antenna (PCA) has been widely utilized for terahertz (THz) signal sampling and analysis due to its broadband response and ultrafast temporal resolution. When illuminated by ultrashort optical pulses from the mode-locked laser (MLL), PCAs exhibit transient photoconductivity, enabling high-speed THz signal sampling and analysis in the optical domain, which has been successfully applied to THz frequency measurement and THz communication signal analysis. However, the impact of noise interference, particularly time jitter and phase noise, on PCA-based THz signal sampling remains insufficiently explored. In this paper, a theoretical analysis of the time jitter and phase noise interference and effects on the sampling of on-off keying (OOK) and quadrature phase-shift keying (QPSK) signals is presented, followed by numerical simulations. To eliminate sampling errors caused by time jitter, the MLL repetition frequency is locked to an external reference source using a locking module while phase noise interference is compensated through the Viterbi and Viterbi algorithm. The effectiveness of the methods is validated through numerical simulations and experimental verification. The findings of this study provide valuable insights into optimizing PCA-based THz communication sampling systems, contributing to the advancement of broadband THz signal analysis.

This paper details the development of a morphosyntactic label set for the Adamawa dialect of the Fulani language (Fulfulde), addressing the critical lack of digital resources and automatic processing tools for this significant African language. The primary objective is to facilitate the creation of a training corpus for morphosyntactic tagging, there by aiding linguists and advancing Natural Language Processing (NLP) applications for Fulani. The proposed label set is meticulously constructed based on a dual methodological approach: it draws heavily from the well-established EAGLES (Expert Advisory Group on Language Engineering Standards) recommendations to ensure corpus reuse and cross-linguistic comparability, while simultaneously incorporating an in-depth analysis of Fulani grammatical specificities. This adaptation is crucial given the morphological richness and complex grammatical structure of Fulani, including its elaborate system of approximately 25 noun classes, unique adjective derivations, and intricate verbal conjugations. The resulting tagset comprises 15 mandatory labels and 54 recommended labels. While some EAGLES categories like "article" and "residual" are not supported, new categories such as "participle," "ideophone," "determiner," and "particle" are introduced to capture the nuances of Fulani grammar. The recommended tags further detail the mandatory categories, subdividing nouns into proper, common singular, and common plural; verbs based on voice and conjugation (infinitive active, middle, passive; conjugated active affirmative/negative, middle affirmative/negative, passive affirmative/negative); and adjectives and pronouns into more specific types based on demonstrative, possessive, subject, object, relative, emphatic, interrogative, and indefinite functions. Participles are divided into singular and plural, adverbs into time, place, manner, and negation, numbers into singular and plural, and determiners into singular and plural. Particles are further broken down into dicto-modal, abdominal, interrogative, emphatic, postposed, and postposed negative. The categories of preposition, conjunction, interjection, unique, punctuation, and ideophone remain indivisible. This meticulously defined tag set was utilized to manually annotate 5,186 words from Dominique Noye’s Fulfulde-French dictionary, creating a valuable, publicly accessible resource for linguistic research and NLP development. Furthermore, the paper outlines a robust workflow for automatic morphosyntactic tagging of Fulfulde sentences, leveraging a Hidden Markov Model (HMM) in conjunction with the Viterbi algorithm. This approach, which extracts transition and emission probabilities from the annotated corpus, enables the disambiguation of morphosyntactic categories within context, considering the specific syntactic and lexical patterns of the Adamawa dialect. Ultimately, this work significantly contributes to the digitization and standardization of the Fulani language, enhancing the performance of linguistic tools and fostering its integration into digital technologies and multilingual systems.

This study explores the identification and assessment of regime shifts in Bitcoin markets through the application of advanced statistical models, namely HMMs, MSMs, and Threshold Models. The analysis utilizes key financial indicators including market capitalization, volatility, trading volume, and historical Bitcoin price data, along with statistical measures such as mean, minimum, and maximum values to enhance the detection of market patterns. Distinctions are made between bullish (sustained price increases exceeding 20%), bearish (sustained price declines exceeding 20%), and neutral (periods of low volatility and sideways movement) market regimes. HMMs provide predictive insights into market transitions, MSMs are employed to capture structural regime changes, and the Threshold Model identifies significant price behaviors. The findings indicate that HMMs outperform the other models in forecasting regime shifts, particularly in detecting transitions among bullish, bearish, and neutral phases information crucial for strategic trading decisions. Unlike traditional models, HMMs effectively accommodate the non-stationary characteristics of Cryptocurrency markets by incorporating variables such as market sentiment, regulatory developments, technological advancements, and macroeconomic conditions. The study presents a comprehensive HMM-based framework for regime detection and market forecasting that supports traders in optimizing entry, exit, and holding strategies to maximize profits while managing risk. Furthermore, the Viterbi algorithm is employed to evaluate the accuracy and reliability of HMM-based forecasts, confirming the robustness of HMMs in modeling and predicting complex market regimes.

To address the challenge of assessing carbon emissions in the iron ore sintering process, characterized by dynamic complexity, strong temporal correlations, and multi-stage coupling, this study innovatively introduces the Hidden Markov Model (HMM) into the field of carbon emission analysis. We propose a method for process stage identification and carbon emission modeling based on a Gaussian Hidden Markov Model (Gaussian HMM). The model defines the four stages of the sintering process as hidden states and uses six-dimensional flue gas monitoring data (temperature, SO2 concentration, NO concentration, NOx concentration, O2 content, CO concentration) as the observation sequence, with Gaussian distributions describing the emission characteristics of each stage. The research employs the Random Forest algorithm to impute missing values and correct outliers in the raw data, followed by standardization to eliminate scale differences. Model parameters are initialized using Maximum Likelihood Estimation (MLE) and iteratively optimized via the Forward-Backward and Baum-Welch algorithms to enhance the model's fitting capability for complex temporal data. The Viterbi algorithm dynamically decodes the hidden state sequence, enabling an online "predict-until-cooling" monitoring strategy. This strategy accurately determines the optimal cooling timing to balance combustion efficiency with the reaction endpoint. This approach prevents incomplete iron ore combustion and low raw material utilization, while simultaneously reducing emissions of harmful gases and greenhouse gases, thereby achieving the goal of lowering carbon emissions. It provides technical support for the refined management of carbon emissions.

Aims: This study presents a comprehensive exploration of Hidden Markov Models (HMMs) for modeling and decoding hidden structures within sequential agricultural data, specifically the oilseed area from 1992 to 2022. Study Design: HMMs with varying numbers of hidden states, ranging from two to eight, were constructed to analyze the underlying patterns in the time series. Place and Duration of Study: The study was based on historical agricultural data collected from India, covering a period of 30 years, with model development and analysis conducted across different model state configurations ranging from two to eight. Methodology: For each model configuration, key parameters, including the Transition Probability Matrix (TPM), Emission Probability Matrix (EPM), and initial state distribution (π), were estimated. Model performance was evaluated using standard selection criteria such as the Akaike Information Criterion (AIC) and the Bayesian Information Criterion (BIC) to determine the optimal number of states. The Viterbi algorithm was then employed to decode the most probable sequence of hidden states corresponding to the observed data. Results: Results indicate that the model with two hidden states provides the best fit, effectively capturing the temporal dynamics of the oilseed area. Conclusion: This work highlights the potential of Hidden Markov Models in uncovering latent structures in agricultural datasets and supports data-driven decision-making in crop planning and agricultural policy design.

Part-of-speech (POS) tagging plays a pivotal role in Natural Language Processing, providing valuable metadata for various downstream tasks such as syntactic parsing, information retrieval, sentiment analysis, machine translation, named entity recognition, and speech recognition. While significant progress has been made in POS tagging for English, the landscape for Hindi remains relatively underexplored. Limited availability of tagged datasets poses a unique challenge in training accurate POS taggers for Hindi. In this study, we aim to explore this gap by training a Hidden Markov Model (HMM) and evaluating its performance using the Viterbi algorithm. We specifically focus on assessing the model’s ability to handle unseen or unknown words, crucial for real-world applications, and evaluate the impact of smoothing techniques too. Our results show that the non-smoothed model achieves higher overall accuracy and significantly better performance on unknown words compared to the smoothed version. Interestingly, precision and recall across POS tags remain consistent between both models, suggesting comparable effectiveness in tagging individual categories despite differences in overall performance.

With the development of the marine economy and the increase in marine activities, deep saturation diving has gained significant attention. Helium speech communication is indispensable for saturation diving operations and is a critical technology for deep saturation diving, serving as the sole communication method to ensure the smooth execution of such operations. This study introduces deep learning into helium speech recognition and proposes a spectrogram-based dual-model helium speech recognition method. First, we extract the spectrogram features from the helium speech. Then, we combine a deep fully convolutional neural network with connectionist temporal classification (CTC) to form an acoustic model, in which the spectrogram features of helium speech are used as an input to convert speech signals into phonetic sequences. Finally, a maximum entropy hidden Markov model (MEMM) is employed as the language model to convert the phonetic sequences to word outputs, which is regarded as a dynamic programming problem. We use a Viterbi algorithm to find the optimal path to decode the phonetic sequences to word sequences. The simulation results show that the method can effectively recognize helium speech with a recognition rate of 97.89% for isolated words and 95.99% for continuous helium speech.