Search

Remote control for the position and status of a machine or an equipment can often be teleoperated by gestures in an intelligent manufacturing environment. In order to solve the problems that gestures with two directions such as left and right cannot be detected by single ultrasonic frequency, double different ultrasonic frequencies are used to detect gestures by the Doppler shift, and an algorithm of the recognition gesture based on the DAG‐SVMs mixed Hidden Markov Model (HMM) is proposed to identify and classify the extracted feature sequences. Thus, four more types of gestures are expanded other than that of reading display screen information, and the comparative experiments to classify and recognise gestures of teleoperation are made with DAG‐SVMs, the HMM, the DAG‐SVMs mixed HMM, and other improved HMM algorithms. The test results have shown that the mean rate of gesture recognition for the algorithm based on the DAG‐SVMs mixed HMM is 94.917%, which is 9.497% higher than that of the unimproved HMM, and its recognition accuracy of complex teleoperation gestures is improved by 2.3% compared with other improved HMM algorithms. The experimental results show that the DAG‐SVMs mixed HMM algorithm has a good effect on recognition for the gestures of teleoperation and it can perform gesture recognition accurately.

The wireless power terminals are deployed in harsh public places and lack strict control, facing security problems. Thus, they are faced with security problems such as illegal and counterfeit terminal access, unlawful control of connected terminals, etc. The intrusion detection system based on machine learning and artificial intelligence significantly improve the terminal side’s abnormal detection capacity. In this article, we aim at identifying the abnormal behavior of wireless power terminals based on a double Hidden Markov Model (HMM), which solves the computational complexity problem caused by high dimensions in intrusion detection systems using a single HMM. The lower-layer HMM is used to identify the discrete single network abnormal behavior. Simultaneously, the upper-layer can obtain more extended period attack behavior in multiple independent abnormal events identified by the low-level. The experiment results indicate that the intrusion detection system using proposed double HMM can effectively detect the terminal’s abnormal behavior and identify the network attack behavior for an extended period.

This paper investigates the problem of sliding mode controller design based on output feedback for discrete-time Markov jump systems. Considering the typical asynchronous phenomenon for jump systems, here we adopt the Hidden Markov Model (HMM) to quantify the asynchronous degree. Based on this, a static output feedback sliding surface is constructed. Sufficient conditions in terms of bilinear matrix inequalities are proposed ensuring the sliding motion dynamic to be stochastically stable and its $\mathcal {H}_\infty $ performance for the considered discrete time Markov jump systems. Moreover, the reachability of the sliding mode surface is ensured by a predesigned control law. The whole design scheme is presented by using an iterative algorithm. Finally, simulation results are presented to illustrate the effectiveness of the design methodology.

This paper proposes a novel phonography-based method for Fetal Breathing Movement (FBM) detection by its excitation sounds. It requires significantly less effort than the current procedures, and it allows long-term measurement, even at home. More than 50 pregnancies in the third trimester were examined, for a minimum of 20 minutes, taking synchronous long-term measurements using a commercial phonocardiographic fetal monitor and a 3D ultrasound machine. To analyze the gained chaotic signal, the frequency band was split into single test-frequencies in the 15-35 Hz frequency band, and their signal-free (silent) zones were regarded as the starting point (SP) of the next motions. The analysis made other disturbing signals, such as fetal hiccups, trunk rotation and limb movements, or maternal heart beats, distinguishable. The dominant test-frequencies of the analysis were predicted by a Hidden Markov Model (HMM). The SPs of the motion units (episodes) were determined by some features of the FBM, applying weighting factors. The recorded material lasted for 16 hours altogether (with nearly 3.5 hours of FBM). Based on the results of HMM method, nearly 7500 FBM episodes were identified in the phonogram signal with an average length of 0.96±0.13 seconds. The procedure for phonography-based breathing movement detection can be combined with a fetal heart activity measurement, and thus allows very intensive, long-term monitoring of the fetus.

The historical training or exam data of an athlete produced in the past sport exercise or test activities have provided a promising way to objectively and accurately evaluate the real-time sport performance of the athlete. However, the continuous generation of sport training or exam data has placed a heavy transmission and processing burden on the traditional centralized data processing paradigm (e.g., cloud platform). Considering this drawback, a decentralized blockchain-based athlete sport data transmission and utilization solution is proposed in this research work. Moreover, the available athlete sport data produced in past sport exercise or test activities is often sparse and time-related, which call for a robust and time-aware data fusion and processing solution. In this situation, HMM model is employed in this article to cope with the data sparsity and dynamics and further make accurate sports performance prediction for athletes accordingly. Finally, we design a set of experiments on a real-world dataset to validate the feasibility of our proposal in terms of effectiveness and efficiency.

Hill musculoskeletal model (HMM) is commonly used to estimate human motion intentions. HMM utilizes electromyography (EMG) signals as the nonlinear model input to obtain muscle forces or torques. However, due to the fact that it contains many physiological parameters that are difficult to measure, HMM is generally applied in simple continuous intention estimation of a single joint. In this work, we aimed at recognizing shoulder and elbow joints angles and their angular velocities continuously in real time. Firstly, we used MYO armband as the EMG sensor. Then, a reasonable prediction model was deduced based on HMM and human dynamics to realize online continuous recognition of the four angles and angular velocities of shoulder and elbow joints. Nonlinear autoregressive with external input neural network (NARX) replaced the prediction equation. In addition, the framework of state space model was completed by constructing an observation equation. Thus, the closed-loop characteristic was realized to eliminate the influence of cumulative error and ensure good estimation performance. Experimental results verified the feasibility and accuracy of the algorithm. For predefined trajectory and random trajectory separately, the RMSE were 0.955 and 1.15 (degree) for angles estimation and 2.8, 3.40 for angular velocities (degree/s). Compared with the normally used back-propagation neural network (BPNN), the method proposed in this paper obviously got more accurate and smooth results.

Climbing stairs is a fundamental part of daily life, adding additional demands on the postural control system compared to level walking. Although real-world gait analysis studies likely contain stair ambulation sequences, algorithms dedicated to the analysis of such activities are still missing. Therefore, we propose a new gait analysis pipeline for foot-worn inertial sensors, which can segment, parametrize, and classify strides from continuous gait sequences that include level walking, stair ascending, and stair descending. For segmentation, an existing approach based on the hidden Markov model and a feature-based gait event detection were extended, reaching an average segmentation F1 score of 98.5% and gait event timing errors below ±10msfor all conditions. Stride types were classified with an accuracy of 98.2% using spatial features derived from a Kalman filter-based trajectory reconstruction. The evaluation was performed on a dataset of 20 healthy participants walking on three different staircases at different speeds. The entire pipeline was additionally validated end-to-end on an independent dataset of 13 Parkinson’s disease patients. The presented work aims to extend real-world gait analysis by including stair ambulation parameters in order to gain new insights into mobility impairments that can be linked to clinically relevant conditions such as a patient’s fall risk and disease state or progression. This work was supported by the Mobilise-D project that has received funding from the Innovative Medicines Initiative 2 Joint Undertaking under grant agreement No. 820820. This Joint Undertaking receives support from the European Union's Horizon 2020 research and innovation program and the European Federation of Pharmaceutical Industries and Associations (EFPIA). Content in this publication reflects the authors’ view and neither IMI nor the European Union, EFPIA, or any Associated Partners are responsible for any use that may be made of the information contained herein.

Antimicrobial peptides (AMPs) are distributed across all kingdoms of life and are an indispensable component of host defenses. They consist of predominantly short cationic peptides with a wide variety of structures and targets. Given the ever-emerging resistance of various pathogens to existing antimicrobial therapies, AMPs have recently attracted extensive interest as potential therapeutic agents. As the discovery of new AMPs has increased, many databases specializing in AMPs have been developed to collect both fundamental and pharmacological information. In this review, we summarize the sources, structures, modes of action, and classifications of AMPs. Additionally, we examine current AMP databases, compare valuable computational tools used to predict antimicrobial activity and mechanisms of action, and highlight new machine learning approaches that can be employed to improve AMP activity to combat global antimicrobial resistance.

We are developing a system for long term Semi-Automated Rehabilitation At the Home (SARAH) that relies on low-cost and unobtrusive video-based sensing. We present a cyber-human methodology used by the SARAH system for automated assessment of upper extremity stroke rehabilitation at the home. We propose a hierarchical model for automatically segmenting stroke survivor's movements and generating training task performance assessment scores during rehabilitation. The hierarchical model fuses expert therapist knowledge-based approaches with data-driven techniques. The expert knowledge is more observable in the higher layers of the hierarchy (task and segment) and therefore more accessible to algorithms incorporating high level constraints relating to activity structure (i.e., type and order of segments per task). We utilize an HMM and a Decision Tree model to connect these high level priors to data driven analysis. The lower layers (RGB images and raw kinematics) need to be addressed primarily through data driven techniques. We use a transformer based architecture operating on low-level action features (tracking of individual body joints and objects) and a Multi-Stage Temporal Convolutional Network(MS-TCN) operating on raw RGB images. We develop a sequence combining these complimentary algorithms effectively, thus encoding the information from different layers of the movement hierarchy. Through this combination, we produce a robust segmentation and task assessment results on noisy, variable and limited data, which is characteristic of low cost video capture of rehabilitation at the home. Our proposed approach achieves 85% accuracy in per-frame labeling, 99% accuracy in segment classification and 93% accuracy in task completion assessment. Although the methodology proposed in this paper applies to upper extremity rehabilitation using the SARAH system, it can potentially be used, with minor alterations, to assist automation in many other movement rehabilitation contexts (i.e., lower extremity training for neurological accidents). National Science FoundationNational Science Foundation (NSF) [2014499]; National Institute on Disability, Independent Living, and Rehabilitation ResearchUnited States Department of Health & Human Services [90REGE0010] Published version This material was based upon work supported by the National Science Foundation under Grant No. (2014499) and the National Institute on Disability, Independent Living, and Rehabilitation Research under Award No. 90REGE0010.

Service-oriented architecture provides the ability to combine several web services in order to fulfil a user-specific requirement. In dynamic environments, the appearance of several unforeseen events can destabilize the composite web service (CWS) and affect its quality. To deal with these issues, the composite web service must be dynamically reconfigured. Dynamic reconfiguration may be enhanced by avoiding the invocation of degraded web services by predicting QoS for the candidate web service. In this paper, we propose a dynamic reconfiguration method based on HMM (Hidden Markov Model) states to predict the imminent degradation in QoS and prevent the invocation of partner web services with degraded QoS values. PSO (Particle Swarm Optimization) and SFLA (Shuffled Frog Leaping Algorithm) are used to improve the prediction efficiency of HMM. Through extensive experiments on a real-world dataset, WS-Dream, the results demonstrate that the proposed approach can achieve better prediction accuracy. Moreover, we carried out a case study where we revealed that the proposed approach outperforms several state-of-the-art methods in terms of execution time.

As part of our contributions to researches on the ongoing COVID-19 pandemic worldwide, we have studied the cough changes to the infected people based on the Hidden Markov Model (HMM) speech recognition classification, formants frequency and pitch analysis. In this paper, An HMM-based cough recognition system was implemented with 5 HMM states, 8 Gaussian Mixture Distributions (GMMs) and 13 dimensions of the basic Mel-Frequency Cepstral Coefficients (MFCC) with 39 dimensions of the overall feature vector. A comparison between formants frequency and pitch extracted values is realized based on the cough of COVID-19 infected people and healthy ones to confirm our cough recognition system results. The experimental results present that the difference between the recognition rates of infected and non-infected people is 6.7%. Whereas, the formant analysis variation based on the cough of infected and non-infected people is clearly observed with F1, F3, and F4 and lower for F0 and F2.

Background The rapid development of the (meta-)omics fields has produced an unprecedented amount of high-resolution and high-fidelity data. Through the use of these datasets we can infer the role of previously functionally unannotated proteins from single organisms and consortia. In this context, protein function annotation can be described as the identification of regions of interest (i.e., domains) in protein sequences and the assignment of biological functions. Despite the existence of numerous tools, challenges remain in terms of speed, flexibility, and reproducibility. In the big data era, it is also increasingly important to cease limiting our findings to a single reference, coalescing knowledge from different data sources, and thus overcoming some limitations in overly relying on computationally generated data from single sources. Results We implemented a protein annotation tool, Mantis, which uses database identifiers intersection and text mining to integrate knowledge from multiple reference data sources into a single consensus-driven output. Mantis is flexible, allowing for the customization of reference data and execution parameters, and is reproducible across different research goals and user environments. We implemented a depth-first search algorithm for domain-specific annotation, which significantly improved annotation performance compared to sequence-wide annotation. The parallelized implementation of Mantis results in short runtimes while also outputting high coverage and high-quality protein function annotations. Conclusions Mantis is a protein function annotation tool that produces high-quality consensus-driven protein annotations. It is easy to set up, customize, and use, scaling from single genomes to large metagenomes. Mantis is available under the MIT license at https://github.com/PedroMTQ/mantis.

In this paper, various methodologies of acoustic and language models, as well as labeling methods for automatic speech recognition for spoken dialogues in emergency call centers were investigated and comparatively analyzed. Because of the fact that dialogue speech in call centers has specific context and noisy, emotional environments, available speech recognition systems show poor performance. Therefore, in order to accurately recognize dialogue speeches, the main modules of speech recognition systems—language models and acoustic training methodologies—as well as symmetric data labeling approaches have been investigated and analyzed. To find an effective acoustic model for dialogue data, different types of Gaussian Mixture Model/Hidden Markov Model (GMM/HMM) and Deep Neural Network/Hidden Markov Model (DNN/HMM) methodologies were trained and compared. Additionally, effective language models for dialogue systems were defined based on extrinsic and intrinsic methods. Lastly, our suggested data labeling approaches with spelling correction are compared with common labeling methods resulting in outperforming the other methods with a notable percentage. Based on the results of the experiments, we determined that DNN/HMM for an acoustic model, trigram with Kneser–Ney discounting for a language model and using spelling correction before training data for a labeling method are effective configurations for dialogue speech recognition in emergency call centers. It should be noted that this research was conducted with two different types of datasets collected from emergency calls: the Dialogue dataset (27 h), which encapsulates call agents’ speech, and the Summary dataset (53 h), which contains voiced summaries of those dialogues describing emergency cases. Even though the speech taken from the emergency call center is in the Azerbaijani language, which belongs to the Turkic group of languages, our approaches are not tightly connected to specific language features. Hence, it is anticipated that suggested approaches can be applied to the other languages of the same group.

BACKGROUND: An important part of the rehabilitation process using exoskeleton robots has been the creation of a friendly Human Robot Interaction (HRI) system. OBJECTIVE: In order to combine SEMG signal into the HRI system, a SEMG-angle model based on Hidden Markov Model (HMM) was put forward in this paper. METHODS: Feature extraction as a critical issue of signal preprocessing was handled by Principal Component Analysis (PCA) which realized signal data dimension reduction and solved the common problem of redundant features. A comparison study was given to show the different performance of various EMG-angle model separately based on HMM, Back Propagation (BP) neural network and Radial Basis Function (RBF) neural network. RESULTS: The HMM modeling method which with lower calculation complexity can achieve a better modeling performance (average accuracy 93.063%) compared with BP neural network (average accuracy 88.180%) and RBF neural network (average accuracy 88.752%). CONCLUSIONS: SEMG signals have some characteristic properties which is similar to a quasi-stationary filtered white noise stochastic process, the structure of HMMs makes it ideally suited for classification and modeling SEMG signals, and the results of this study show that it can achieve a better performance than the commonly used methods (BP and RBF).

Datasets of the same geographic space at different scales and temporalities are increasingly abundant, paving the way for new scientific research. These datasets require data integration, which implies linking homologous entities in a process called data matching that remains a challenging task, despite a quite substantial literature, because of data imperfections and heterogeneities. In this paper, we present an approach for matching spatial networks based on a hidden Markov model (HMM) that takes full benefit of the underlying topology of networks. The approach is assessed using four heterogeneous datasets (streets, roads, railway, and hydrographic networks), showing that the HMM algorithm is robust in regards to data heterogeneities and imperfections (geometric discrepancies and differences in level of details) and adaptable to match any type of spatial networks. It also has the advantage of requiring no mandatory parameters, as proven by a sensitivity exploration, except a distance threshold that filters potential matching candidates in order to speed-up the process. Finally, a comparison with a commonly cited approach highlights good matching accuracy and completeness.

This paper addresses the development of an Automatic Speech Recognition (ASR) system for the Moroccan Dialect. Dialectal Arabic (DA) refers to the day-to-day vernaculars spoken in the Arab world. In fact, Moroccan Dialect is very different from the Modern Standard Arabic (MSA) because it is highly influenced by the French Language. It is observed throughout all Arab countries that standard Arabic widely written and used for official speech, news papers, public administration and school but not used in everyday conversation and dialect is widely spoken in everyday life but almost never written. we propose to use the Mel Frequency Cepstral Coefficient (MFCC) features to specify the best speaker identification system. The extracted speech features are quantized to a number of centroids using vector quantization algorithm. These centroids constitute the codebook of that speaker. MFCC’s are calculated in training phase and again in testing phase. Speakers uttered same words once in a training session and once in a testing session later. The Euclidean distance between the MFCC’s of each speaker in training phase to the centroids of individual speaker in testing phase is measured and the speaker is identified according to the minimum Euclidean distance. The code is developed in the MATLAB environment and performs the identification satisfactorily.

Conventional large vocabulary automatic speech recognition (ASR) systems require a mapping from words into sub-word units to generalize over the words that were absent in the training data and to enable the robust estimation of acoustic model parameters. This paper surveys the research done during the last 15 years on the topic of word to sub-word mappings for Lithuanian ASR systems. It also compares various phoneme and grapheme based mappings across a broad range of acoustic modelling techniques including monophone and triphone based Hidden Markov models (HMM), speaker adaptively trained HMMs, subspace gaussian mixture models (SGMM), feed-forward time delay neural network (TDNN), and state-of-the-art low frame rate bidirectional long short term memory (LFR BLSTM) recurrent deep neural network. Experimental comparisons are based on a 50-hour speech corpus. This paper shows that the best phone-based mapping significantly outperforms a grapheme-based mapping. It also shows that the lowest phone error rate of an ASR system is achieved by the phoneme-based lexicon that explicitly models syllable stress and represents diphthongs as single phonetic units.

This paper deals with a project of Automatic Bird Species Recognition Based on Bird Vocalization. Eighteen bird species of 6 different families were analyzed. At first, human factor cepstral coefficients representing the given signal were calculated from particular recordings. In the next phase, using the voice activity detection system, segments of bird vocalizations were detected from which a likelihood rate, with which the given code value corresponds to the given model, was calculated using individual hidden Markov models. For each bird species, just one respective hidden Markov model was trained. The interspecific success of 81.2% has been reached. For classification into families, the success has reached 90.45%.

Occupant presence and behaviour in buildings is considered a key element towards building intelligent and pervasive environments. Yet, practical applications of energy intelligent buildings typically suffer from high sensor unreliability. In this work, we propose a layered probabilistic framework for occupancy-based control in intelligent buildings. We adopt a cascade of layers, where each layer addresses different aspects of the occupancy detection problem in a probabilistic manner rather than in a hard rule engine. We show that propagating uncertainty through each layer instead of standard hard decision outcomes improves the overall system performance. This finding suggests that smart building interfaces and communication data formats may need to input and output probabilistic data rather than simple discrete classification outputs. System performance and user comfort were evaluated with real life radar sensor data, based on an algorithm that allows real-time (casual) processing. Energy savings of up to 30% were obtained, compared to baseline measurements, while maintaining user comfort.

An automatic speech recognition (ASR) system translates spoken words or utterances (isolated, connected, continuous, and spontaneous) into text format. State-of-the-art ASR systems mainly use Mel frequency (MF) cepstral coefficient (MFCC), perceptual linear prediction (PLP), and Gammatone frequency (GF) cepstral coefficient (GFCC) for extracting features in the training phase of the ASR system. Initially, the paper proposes a sequential combination of all three feature extraction methods, taking two at a time. Six combinations, MF-PLP, PLP-MFCC, MF-GFCC, GF-MFCC, GF-PLP, and PLP-GFCC, are used, and the accuracy of the proposed system using all these combinations was tested. The results show that the GF-MFCC and MF-GFCC integrations outperform all other proposed integrations. Further, these two feature vector integrations are optimized using three different optimization methods, particle swarm optimization (PSO), PSO with crossover, and PSO with quadratic crossover (Q-PSO). The results demonstrate that the Q-PSO-optimized GF-MFCC integration show significant improvement over all other optimized combinations.

International audience; Time series have met high interest in various fields for measuring the evolution of a quantity over time. Several machine learning techniques are proposed to extract knowledge from this type of data and make them more meaningful. Clustering is one such prominent technique, for detecting homogeneous subgroups from a data set when there is no prior knowledge about classes. It has been called upon various fields to discover hidden models that arise from the data. In this paper, we propose a new framework called Generative time series Clustering with Bagging (GCBag). It combines the power of several techniques designed for time series.Most existing works use DTW to provide a starting point for HMMs to estimate their parameters. As a result, the estimation becomes dependent on this single provided initialization, which can be biased. The originality of GCBag lies in the use of bagging during the clustering where it significantly raises the stability of models. Consequently, we have succeeded in improving the quality of clustering while preserving the descriptive aspect. Several experiments are conducted to demonstrate the effectiveness of GCBag over the existing models for time series clustering, on both synthetic and real time series data.

We present ARCHes, a fast and accurate haplotype-based approach for inferring an individual’s ancestry composition. Our approach works by modeling haplotype diversity from a large, admixed cohort of hundreds of thousands, then annotating those models with population information from reference panels of known ancestry. The running time of ARCHes does not depend on the size of a reference panel because training and testing are separate processes, and the inferred population-annotated haplotype models can be written to disk and reused to label large test sets in parallel (in our experiments, it averages less than one minute to assign ancestry from 32 populations to 1,001 sections of a genotype using 10 CPU). We test ARCHes on public data from the 1,000 Genomes Project and HGDP as well as simulated examples of known admixture. Our results demonstrate that ARCHes outperforms RFMix at correctly assigning both global and local ancestry at finer population scales regardless of the amount of population admixture.Author SummaryHuman DNA is inherited from ancestors that come from different populations across the globe and across time. Being able to identify which of those populations make up an individual’s DNA, how much they contribute, and on which chromosomes, is currently an important open research problem with many applications in the study of human diversity and history. As DNA sequencing and genotyping technology has developed, we have greater and greater amounts of data, which allows for the development of new sophisticated machine learning methods to approach this problem, and presents a need to process large amounts of data efficiently. These methods learn from examples of DNA data from known populations, and must be robust to differences in size and diversity among those reference populations. We present a new approach to this problem called ARCHes (Ancestry inference usingReference labeledClusters ofHaplotypes), that models the global diversity of small segments of human DNA sequence (“haplotypes”), and the extent to which these haplotypes are associated with each of a set of population reference panels. It then computes the most likely population assignments and the points along the genome where the populations change. Our experiments show that ARCHes has superior accuracy compared to a state-of-the-art method in identifying source populations and their locations on the genome, regardless of the number of different populations present in the genome, how closely related those populations are. ARCHes is also able to model populations despite having a small amount of population reference DNA data.

Background Speech recognition is a technology that enables machines to understand human language. Objective In this study, speech recognition of isolated words from a small vocabulary was applied to the field of mental health counseling. Methods A software platform was used to establish a human-machine chat for psychological counselling. The software uses voice recognition technology to decode the user's voice information. The software system analyzes and processes the user's voice information according to many internal related databases, and then gives the user accurate feedback. For users who need psychological treatment, the system provides them with psychological education. Results The speech recognition system included features such as speech extraction, endpoint detection, feature value extraction, training data, and speech recognition. Conclusions The Hidden Markov Model was adopted, based on multithread programming under a VC2005 compilation environment, to realize the parallel operation of the algorithm and improve the efficiency of speech recognition. After the design was completed, simulation debugging was performed in the laboratory. The experimental results showed that the designed program met the basic requirements of a speech recognition system.

With the widespread development of location-based services, the demand for accurate indoor positioning is getting more and more urgent. Floor positioning, as a prerequisite for indoor positioning in multi-story buildings, is particularly important. Though lots of work has been done on floor positioning, the existing studies on floor positioning in complex multi-story buildings with large hollow areas through multiple floors still cannot meet the application requirements because of low accuracy and robustness. To obtain accurate and robust floor estimation in complex multi-story buildings, we propose a novel floor positioning method, which combines the Wi-Fi based floor positioning (BWFP), the barometric pressure-based floor positioning (BPFP) with HMM and the XGBoost based user motion detection. Extensive experiments show that using our proposed method can achieve 99.2% accuracy, which outperforms other state-of-the-art floor estimation methods.

Point of interest (POI) recommendation can benefit users and merchants. It is a very important and popular service in modern life. In this paper, we aim to study the next new POI recommendation problem with the consideration of privacy preserving in edge computing. The challenge lies in capturing the transition patterns between POIs precisely and meanwhile protecting users’ location. In this paper, first, we propose to model users’ check-in sequences with their latent states based on HMM, and EM algorithm is used to estimate the parameters of the model. Second, we propose to protect users’ location information by a weighted noise injection method. Third, we predict users’ next movement according to his current location based on Forward algorithm. Experimental results on two large-scale LBSNs datasets show that our proposed model without noise injection can achieve better recommendation accuracy than several state-of-the-art techniques, and the proposed weighted noise injection approach can achieve better performance on privacy preserving than traditional one with a little cost on accuracy.

Článek se zabývá syntézou expresivní řeči v dialogu. Pro popis expresivity jsou použity dialogové akty - diskrétní expresivní kategorie. Cílem této práce je popsat postup vývoje expresivní syntézy řeči pro dialogový systém v omezené oblasti. Zde je tato oblast omezena na rozhovory mezi člověkem a počítačem na témata týkající se osobních fotografií. Aby bylo možné začlenit do syntézy řeči expresivitu, je potřeba modifikovat stávající algoritmy používané pro syntézu neutrální řeči. Byl nahrán expresivní řečový korpus, data byla anotována předdefinovanou množinou dialogových aktů, a byla provedena akustická analýza tohoto korpusu. Pro syntézu expresivní řeči byly využity metody výběru jednotek a HMM. Jako výsledky jsou v článku uvedené výstupy z poslechových testů. Posluchači v nich hodnotili dva aspekty expresivní syntézy řeči pro izolované promluvy: kvalitu řeči a vnímání expresivity. Vyhodnocení je provedeno také pro promluvy v rámci dialogu pro ověření vhodnosti syntetické expresivní řeči. Závěrem je, že syntetická expresivní řeč je hodnocena pozitivně i když je o něco horší kvality než syntetická řeč neutrální. Syntetická expresivní řeč však skutečně umožňuje přenášet expresivitu na posluchače a tím zlepšuje přirozenost syntetické řeči. This paper deals with expressive speech synthesis in a dialogue. Dialogue acts - discrete expressive categories - are used for expressivity description. The aim of the work is to create a procedure for development of expressive speech synthesis for a dialogue system in a limited domain. The domain is here limited to dialogues between a human and a computer on a given topic of reminiscing about personal photographs. To incorporate expressivity into synthetic speech, modifications of current algorithms used for neutral speech synthesis are made. An expressive speech corpus is recorded, annotated using a predefined set of dialogue acts, and its acoustic analysis is performed. Unit selection and HMM-based methods are used to synthesize expressive speech, and an evaluation using listening tests is presented. The listeners asses two basic aspects of synthetic expressive speech for isolated utterances: speech quality and expressivity perception. The evaluation is also performed for utterances in a dialogue to asses appropriateness of synthetic expressive speech. It can be concluded that synthetic expressive speech is rated positively even though it is of worse quality when comparing with the neutral speech synthesis. However, synthetic expressive speech is able to transmit expressivity to listeners and to improve the naturalness of the synthetic speech.

Reliable and efficient sensing and tracking of multiple weak or time-varying frequency line components in underwater acoustic signals is the topic of this paper. We propose a method for automatic detection and tracking of multiple frequency lines in lofargram based on hidden Markov model (HMM). Instead of being directly subjected to frequency line tracking, the whole lofargram is first segmented into several sub-lofargrams. Then, the sub-lofargrams suspected to contain frequency lines are screened. In these sub-lofargrams, the HMM-based method is used for detection of multiple frequency lines. Using image stitching and statistical model method, the frequency lines with overlapping parts detected by different sub-lofargrams are merged to obtain the final detection results. The method can effectively detect multiple time-varying frequency lines of underwater acoustic signals while ensuring the performance under the condition of low signal-to-noise ratio (SNR). It can be concluded that the proposed algorithm can provide better multiple frequency lines sensing ability while greatly reducing the amount of calculations and providing potential techniques for feature sensing and tracking processing of unattended equipment such as sonar buoys and submerged buoys.

Financial time-series are well known for their non-linearity and non-stationarity nature. The application of conventional econometric models in prediction can incur significant errors. The fast advancement of soft computing techniques provides an alternative approach for estimating and forecasting volatile stock prices. Soft computing approaches exploit tolerance for imprecision, uncertainty, and partial truth to progressively and adaptively solve practical problems. In this study, a comprehensive review of latest soft computing tools is given. Then, examples incorporating a series of machine learning models, including both single and hybrid models, to predict prices of two representative indexes and one stock in Hong Kong’s market are undertaken. The prediction performances of different models are evaluated and compared. The effects of the training sample size and stock patterns (viz. momentum and mean reversion) on model prediction are also investigated. Results indicate that artificial neural network (ANN)-based models yield the highest prediction accuracy. It was also found that the determination of optimal training sample size should take the pattern and volatility of stocks into consideration. Large prediction errors could be incurred when stocks exhibit a transition between mean reversion and momentum trend.

In supervised Activities of Daily Living (ADL) recognition systems, annotating collected sensor readings is an essential, yet exhaustive, task. Readings are collected from activity-monitoring sensors in a 24/7 manner. The size of the produced dataset is so huge that it is almost impossible for a human annotator to give a certain label to every single instance in the dataset. This results in annotation gaps in the input data to the adopting learning system. The performance of the recognition system is negatively affected by these gaps. In this work, we propose and investigate three different paradigms to handle these gaps. In the first paradigm, the gaps are taken out by dropping all unlabeled readings. A single &ldquo, Unknown&rdquo, or &ldquo, Do-Nothing&rdquo, label is given to the unlabeled readings within the operation of the second paradigm. The last paradigm handles these gaps by giving every set of them a unique label identifying the encapsulating certain labels. Also, we propose a semantic preprocessing method of annotation gaps by constructing a hybrid combination of some of these paradigms for further performance improvement. The performance of the proposed three paradigms and their hybrid combination is evaluated using an ADL benchmark dataset containing more than 2.5 &times, 10 6 sensor readings that had been collected over more than nine months. The evaluation results emphasize the performance contrast under the operation of each paradigm and support a specific gap handling approach for better performance.

The context analysis of customer requests in a natural language call routing problem is investigated in the paper. One of the most significant problems in natural language call routing is a comprehension of client request. With the aim of finding a solution to this issue, the Hybrid HMM and ANFIS models become a subject to an examination. Combining different types of models (ANFIS and HMM) can prevent misunderstanding by the system for identification of user intention in dialogue system. Based on these models, the hybrid system may be employed in various language and call routing domains due to nonusage of lexical or syntactic analysis in classification process.

In this paper, we propose ultra-wideband (UWB) radar target detection approach based on Hidden Markov Models (HMMs). HMMs are used as a classifier to identify signal with the presence of target in a background clutter and the pure clutter response signal. Time-frequency features are extracted and features have less correlation to each other are selected based on the feature covariance matrix and fed into HMMs. The detection experiments are conducted in two different scenarios: sense-through-foliage target detection and sense-through-wall human detection. The sense-through-foliage data set contains poor quality UWB radar return echoes using low amplitude transmitting pulses. Data collected from different radar locations are tested and detection results are presented. Sense-through-wall data are collected using different UWB radar and the target is human standing behind different types of walls. HMMs parameters are also investigated to optimally model UWB radar signals for target detection.

Named entity (NE) recognition (NER) is a process to identify and classify atomic elements such as person name, organization name, place/location name, quantities, temporal expressions, and monetary expressions in running text. In this paper, the Hindi NER task has been mapped into a multiclass learning problem, where the classes are NE tags. This paper presents a solution to this Hindi NER problem using a memory-based learning method. A set of simple and composite features, which includes binary, nominal, and string features, has been defined and incorporated into the proposed model. A relatively small Hindi Gazetteer list has also been employed to enhance the system performance. A comparative study on the experimental results obtained by the memory-based NER system proposed in this paper and a hidden Markov model (HMM)-based NER system shows that the performance of the proposed memory-based NER system is comparable to the HMM-based NER system.

In this paper the research related to the development of a deep neural network based speech synthesizer for the Serbian language, trained on recorded utterances of a single female voice talent, is described. Two separate networks are used for prediction of acoustic features and phonetic segment durations. Through a set of experiments the optimal values of the hyper-parameters of the neural networks are established, and then the influence of the amount of training data on the quality of synthesized speech is examined. The quality is evaluated through objective measures as well as appropriate listening tests. It has been confirmed that 4-layer deep neural networks with 512 units per hidden layer, trained on 3 hours of data, produce speech of very good quality. The results also suggest that a further increase in the amount of training data may contribute to further improvement in quality.

Spamdexing is the art of black hat SEO. Features which are more influential for high rank and visibility are manipulated for the SEO task. The motivation behind the work is utilizing the state of the art Website optimization features to enhance the performance of spamdexing detection. Features which play a focal role in current SEO strategies show a significant deviation for spam and non-spam samples. This paper proposes 44 features named as NLSDF (New Link Spamdexing Detection Features). Social media creates an impact in search engine results ranking. Features pertaining to the social media were incorporated with the NLSDF features to boost the recital of the spamdexing classification. The NLSDF features with 44 attributes along with 5 social media features boost the classification performance of the WEBSPAM-UK 2007 dataset. The one tailed paired t-test with 95% confidence, performed on the AUC values of the learning models shows significance of the NLSDF.

International audience; This paper describes a weakly-supervised approach to Automatic Chord Estimation (ACE) task that aims to estimate a sequence of chords from a given music audio signal at the frame level, under a realistic condition that only non-aligned chord annotations are available. In conventional studies assuming the availability of time-aligned chord annotations, Deep Neural Networks (DNNs) that learn frame-wise mappings from acoustic features to chords have attained excellent performance. The major drawback of such frame-wise models is that they cannot be trained without the time alignment information. Inspired by a common approach in automatic speech recognition based on non-aligned speech transcriptions, we propose a two-step method that trains a Hidden Markov Model (HMM) for the forced alignment between chord annotations and music signals, and then trains a powerful frame-wise DNN model for ACE. Experimental results show that although the frame-level accuracy of the forced alignment was just under 90%, the performance of the proposed method was degraded only slightly from that of the DNN model trained by using the ground-truth alignment data. Furthermore, using a sufficient amount of easily collected non-aligned data, the proposed method is able to reach or even outperform the conventional methods based on ground-truth time-aligned annotations.

Background Accurate and reliable identification of sequence variants, including single nucleotide polymorphisms (SNPs) and insertion-deletion polymorphisms (INDELs), plays a fundamental role in next-generation sequencing (NGS) applications. Existing methods for calling these variants often make simplified assumptions of positional independence and fail to leverage the dependence between genotypes at nearby loci that is caused by linkage disequilibrium (LD). Results and conclusion We propose vi-HMM, a hidden Markov model (HMM)-based method for calling SNPs and INDELs in mapped short-read data. This method allows transitions between hidden states (defined as “SNP,” “Ins,” “Del,” and “Match”) of adjacent genomic bases and determines an optimal hidden state path by using the Viterbi algorithm. The inferred hidden state path provides a direct solution to the identification of SNPs and INDELs. Simulation studies show that, under various sequencing depths, vi-HMM outperforms commonly used variant calling methods in terms of sensitivity and F1 score. When applied to the real data, vi-HMM demonstrates higher accuracy in calling SNPs and INDELs. Electronic supplementary material The online version of this article (10.1186/s40246-019-0194-6) contains supplementary material, which is available to authorized users.

Computational visual attention is a hot topic in computer vision. However, most efforts are devoted to model saliency, whilst the actual eye guidance problem, which brings into play the sequence of gaze shifts characterising overt attention, is overlooked. Further, in those cases where the generation of gaze behaviour is considered, stimuli of interest are by and large static (still images) rather than dynamic ones (videos). Under such circumstances, the work described in this note has a twofold aim: (i) addressing the problem of estimating and generating visual scan paths, that is the sequences of gaze shifts over videos; (ii) investigating the effectiveness in scan path generation offered by features dynamically learned on the base of human observers attention dynamics as opposed to bottom-up derived features. To such end a probabilistic model is proposed. By using a publicly available dataset, our approach is compared against a model of scan path simulation that does not rely on a learning step.

Background Full Bayesian inference for detecting copy number variants (CNV) from whole-genome sequencing (WGS) data is still largely infeasible due to computational demands. A recently introduced approach to perform Forward–Backward Gibbs sampling using dynamic Haar wavelet compression has alleviated issues of convergence and, to some extent, speed. Yet, the problem remains challenging in practice. Results In this paper, we propose an improved algorithmic framework for this approach. We provide new space-efficient data structures to query sufficient statistics in logarithmic time, based on a linear-time, in-place transform of the data, which also improves on the compression ratio. We also propose a new approach to efficiently store and update marginal state counts obtained from the Gibbs sampler. Conclusions Using this approach, we discover several CNV candidates in two rat populations divergently selected for tame and aggressive behavior, consistent with earlier results concerning the domestication syndrome as well as experimental observations. Computationally, we observe a 29.5-fold decrease in memory, an average 5.8-fold speedup, as well as a 191-fold decrease in minor page faults. We also observe that metrics varied greatly in the old implementation, but not the new one. We conjecture that this is due to the better compression scheme. The fully Bayesian segmentation of the entire WGS data set required 3.5 min and 1.24 GB of memory, and can hence be performed on a commodity laptop.

Prediction plays a key role in recent computational models of the brain and it has been suggested that the brain constantly makes multisensory spatiotemporal predictions. Inspired by these findings we tackle the problem of audiovisual fusion from a new perspective based on prediction. We train predictive models which model the spatiotemporal relationship between audio and visual features by learning the audio-to-visual and visual-to-audio feature mapping for each class. Similarly, we train predictive models which model the time evolution of audio and visual features by learning the past-to-future feature mapping for each class. In classification, all the class-specific regression models produce a prediction of the expected audio / visual features and their prediction errors are combined for each class. The set of class-specific regressors which best describes the audiovisual feature relationship, i.e., results in the lowest prediction error, is chosen to label the input frame. We perform cross-database experiments, using the AMI, SAL and MAHNOB databases, in order to classify laughter and speech and subject-independent experiments on the AVIC database in order to classify laughter, hesitation and consent. In virtually all cases prediction-based audiovisual fusion consistently outperforms the two most commonly used fusion approaches, decision-level and feature-level fusion.

We propose a novel variance reduction strategy based on control variables for simulating the averaged equation of a stochastic slow-fast system. In this system, we assume that the fast equation is ergodic, implying the existence of an invariant measure, for every fixed value of the slow variable. The right hand side of the averaged equation contains an integral with respect to this unknown invariant measure, which is approximated by the heterogeneous multiscale method (HMM). The HMM method corresponds to a Markov chain Monte Carlo method in which samples are generated by simulating the fast equation. As a consequence, the variance of the HMM estimator decays slowly. Therefore, we introduce a variance-reduced HMM estimator based on control variables: from the current time HMM estimation, we subtract a second HMM estimator at the previous time step using the exact same seed as the current time HMM estimator. To avoid introducing a bias, we add the previously calculated variance-reduced estimator. We analyze convergence of the proposed estimator and apply it to a linear and nonlinear model problem. ispartof: pages:1255-1266 ispartof: Procedia Computer Science vol:80 pages:1255-1266 ispartof: International Conference on Computational Science 2016, ICCS 2016 location:San Diego date:6 Jun - 8 Jun 2016 status: published

The hand is an important part of the body used to express information through gestures, and its movements can be used in dynamic gesture recognition systems based on computer vision with practical applications, such as medical, games and sign language. Although depth sensors have led to great progress in gesture recognition, hand gesture recognition still is an open problem because of its complexity, which is due to the large number of small articulations in a hand. This paper proposes a novel approach for hand gesture recognition with depth maps generated by the Microsoft Kinect Sensor (Microsoft, Redmond, WA, USA) using a variation of the CIPBR (convex invariant position based on RANSAC) algorithm and a hybrid classifier composed of dynamic time warping (DTW) and Hidden Markov models (HMM), called the hybrid approach for gesture recognition with depth maps (HAGR-D). The experiments show that the proposed model overcomes other algorithms presented in the literature in hand gesture recognition tasks, achieving a classification rate of 97.49% in the MSRGesture3D dataset and 98.43% in the RPPDI dynamic gesture dataset.

© 2015 The Authors. Published by Elsevier B.V.In modern computer systems, the intermittent behaviour of infrequent, additional loads affects performance. Often, representative traces of storage disks or remote servers can be scarce and obtaining real data is sometimes expensive. Therefore, stochastic models, through simulation and profiling, provide cheaper, effective solutions, where input model parameters are obtained. A typical example is the Markov-modulated Poisson process (MMPP), which can have its time index discretised to form a hidden Markov model (HMM). These models have been successful in capturing bursty behaviour and cyclic patterns of I/O operations and Internet traffic, using underlying properties of the discrete (or continuous) Markov chain. However, learning on such models can be cumbersome in terms of complexity through re-training on data sets. Thus, we provide an online learning HMM (OnlineHMM), which is composed of two existing variations of HMMs: first, a sliding HMM using a moving average technique to update its parameters on-the-fly and, secondly, a multi-input HMM capable of training on multiple discrete traces simultaneously. The OnlineHMM reduces data processing times significantly and thence synthetic workloads become computationally more cost effective. We measure the accuracy of reproducing representative traces through comparisons of moments and autocorrelation on original data points and HMM-generated synthetic traces. We present, analytically, the training steps saved through the OnlineHMMs adapted Baum-Welch algorithm and obtain, through simulation, mean waiting times of a queueing model. Finally, we conclude our work and offer model extensions for the future.

Productivity is a very important part of any organisation in general and software industry in particular. Now a day’s Software Effort estimation is a challenging task. Both Effort and Productivity are inter-related to each other. This can be achieved from the employee’s of the organization. Every organisation requires emotionally stable employees in their firm for seamless and progressive working. Of course, in other industries this may be achieved without man power. But, software project development is labour intensive activity. Each line of code should be delivered from software engineer. Tools and techniques may helpful and act as aid or supplementary. Whatever be the reason software industry has been suffering with success rate. Software industry is facing lot of problems in delivering the project on time and within the estimated budget limit. If we want to estimate the required effort of the project it is significant to know the emotional state of the team member. The responsibility of ensuring emotional contentment falls on the human resource department and the department can deploy a series of systems to carry out its survey. This analysis can be done using a variety of tools, one such, is through study of emotion recognition. The data needed for this is readily available and collectable and can be an excellent source for the feedback systems. The challenge of recognition of emotion in speech is convoluted primarily due to the noisy recording condition, the variations in sentiment in sample space and exhibition of multiple emotions in a single sentence. The ambiguity in the labels of training set also increases the complexity of problem addressed. The existing models using probabilistic models have dominated the study but present a flaw in scalability due to statistical inefficiency. The problem of sentiment prediction in spontaneous speech can thus be addressed using a hybrid system comprising of a Convolution Neural Network and Hidden Markov Model.

Online signature verification is a prominent behavioral biometric trait. It offers many dynamic features along with static two dimensional signature image. In this paper, the Hybrid Wavelet Transform (HWT) was generated using Kronecker product of two orthogonal transform such as DCT, DHT, Haar, Hadamard and Kekre. HWT has the ability to analyze the signal at global as well as local level like wavelet transform. HWT-1 and -2 was applied on the first 128 samples of the pressure parameter and first 16 samples of the output were used as feature vector for signature verification. This feature vector is given to Left to Right HMM classifier to identify the genuine and forged signature. For HWT-1, DCT HAAR offers best FAR and FRR. . For HWT-2, KEKRE 128 offers best FAR and FRR. HWT-1 offers better performance than HWT- 2 in terms of FAR and FRR. As the number of states increase, the performance of the system improves. For HWT - 1, KEKRE 128 offers best performance at 275 symbols whereas for HWT - 2, best performance is at 475 symbols by KEKRE 128.

A combination of spectral kurtosis (SK), based on Choi–Williams distribution (CWD) and hidden Markov models (HMM), accurately identifies initial gearbox failures and diagnoses fault types of gearboxes. First, using the LMD algorithm, five types of gearbox vibration signals are collected and decomposed into several product function (PF) components and the multicomponent signals are decomposed into single-component signals. Then, the kurtosis value of each component is calculated, and the component with the largest kurtosis value is selected for the CWD-SK analysis. According to the calculated CWD-SK value, the characteristics of the initial failure of the gearbox are extracted. This method not only avoids the difficulty of selecting the window function, but also provides original eigenvalues for fault feature classification. In the end, from the CWD-SK characteristic parameters at each characteristic frequency, the characteristic sequence based on CWD-SK is obtained with HMM training and diagnosis. The experimental results show that this method can effectively identify the initial fault characteristics of the gearbox, and also accurately classify the fault characteristics of different degrees.

This work addresses the problem of presence detection as Distributed Sensing of a Hidden Markov Model (DS-HMM). We propose an efficient communication strategy to limit sensor communication. For optimum detection under constrained communication, it does not suffice to just send only the latest sensor observations. It requires tracing back in the HMM algorithm to the latest moment that all data from all sensors was available. Alternatively, we investigate a more pragmatic approach, that is, to send a few recent observations in every message (packet) instead of the full history. The effect of unreliable communication, subject to channel erasure, is also investigated.

The majority of handwritten word recognition strategies are constructed on learning-based generative frameworks from letter or word training samples. Theoretically, constructing recognition models through discriminative learning should be the more effective alternative. The primary goal of this research is to compare the performances of discriminative and generative recognition strategies, which are described by generatively-trained hidden Markov modeling (HMM), discriminatively-trained conditional random fields (CRF) and discriminatively-trained hidden-state CRF (HCRF). With learning samples obtained from two dissimilar databases, we initially trained and applied an HMM classification scheme. To enable HMM classifiers to effectively reject incorrect and out-of-vocabulary segmentation, we enhance the models with adaptive threshold schemes. Aside from proposing such schemes for HMM classifiers, this research introduces CRF and HCRF classifiers in the recognition of offline Arabic handwritten words. Furthermore, the efficiencies of all three strategies are fully assessed using two dissimilar databases. Recognition outcomes for both words and letters are presented, with the pros and cons of each strategy emphasized.