Littérature scientifique sur le sujet « Continuous Time Bayesian Network Classifier »

AbstractMost of the sampled data in complex industrial processes are sequential in time. Therefore, the traditional BN learning mechanisms have limitations on the value of probability and cannot be applied to the time series. The model established in Chap. 10.1007/978-981-16-8044-1_13 is a graphical model similar to a Bayesian network, but its parameter learning method can only handle the discrete variables. This chapter aims at the probabilistic graphical model directly for the continuous process variables, which avoids the assumption of discrete or Gaussian distributions.

Bayesian Network classifiers (BNCs) are Bayesian Network (BN) models specifically tailored for classification tasks. There is a wide range of existing models that vary in complexity and efficiency. All of them have in common the ability to deal with uncertainty in a very natural way, at the same time providing a descriptive environment. In this chapter, the authors focus on the family of semi-naïve Bayesian classifiers (naïve Bayes, AODE, TAN, kDB, etc.), motivated by the good trade-off between efficiency and performance they provide. The domain of the BNs is generally of discrete nature, but since the presence of continuous variables is very common, the chapter discusses more classical and novel approaches to handling numeric data. In this chapter the authors also discuss more recent techniques such as multi-dimensional and dynamic models. Last but not least, they focus on applications and recent developments, including some of the BNCs approaches to the multi-class problem together with other traditionally successful and cutting edge cases regarding real-world applications.

Chronic wound (CW) treatment by large is a burden for the government and society due to its high cost and time consuming treatment. It becomes more serious for the old age patient with the lack of moving flexibility. Proper wound recovery management is needed to resolve this problem. Careful and accurate documentation is required for identifying the patient's improvement and or deterioration timely for early diagnostic purposes. This paper discusses the comprehensive wound diagnostic method using three important modules, viz. Wounds Data Acquisition (WDA) module, Tele-Wound Technology Network (TWTN) module and Wound Screening and Diagnostic (WSD) module. Here the wound image characterization and diagnosis tool has been proposed under telemedicine to classify the percentage wise wound tissue based on the color variation over regular intervals for providing a prognostic treatment with better degree of accuracy. The Bayesian classifier based wound characterization (BWC) technique is proposed that able to identify wounded tissue and correctly predict the wound status with a good degree of accuracy. Results show that BWC technique provides very good accuracy, i.e. 87.40%, whereas the individual tissue wise accuracy for granulation tissue is 89.44%, slough tissue is 81.87% and for necrotic tissue is 90.91%.

This paper presents a tele-wound framework for monitoring chronic wound status based on color variation over a period of time. This will facilitate patients at remote locations to connect to medical experts through mobile devices. Further this will help medical professionals to monitor and manage the wounds in more timely, accurate and precise manner using the proposed framework. Tele-medical agent (TMA) collects the chronic wound data using smart phone and send it to the Tele-medical hub (TMH). In TMH, the wound image has been segmented using Fuzzy C-Means which gives highest segmented accuracy i.e. 92.60%, then the wound tissue is classified using proposed Bayesian classifier. The smart phone supported prototype system has been demonstrated with snapshots using very compatible and easy to integrate Hypertext preprocessor (PHP) and MySqL. The proposed system may facilitate better wound management and treatment by providing percentage of wound tissues.

A basic task of machine learning and data mining is to automatically uncover <b>patterns</b> that reflect regularities in a data set. When dealing with a large database, especially when domain knowledge is not available or very weak, this can be a challenging task. The purpose of <b>pattern discovery</b> is to find non-random relations among events from data sets. For example, the “exclusive OR” (XOR) problem concerns 3 binary variables, A, B and C=A<img src="http://resources.igi-global.com/Marketing/Preface_Figures/x_symbol.png">B, i.e. C is true when either A or B, but not both, is true. Suppose not knowing that it is the XOR problem, we would like to check whether or not the occurrence of the compound event [A=T, B=T, C=F] is just a random happening. If we could estimate its frequency of occurrences under the random assumption, then we know that it is not random if the observed frequency deviates significantly from that assumption. We refer to such a compound event as an event association pattern, or simply a <b>pattern</b>, if its frequency of occurrences significantly deviates from the default random assumption in the statistical sense. For instance, suppose that an XOR database contains 1000 samples and each primary event (e.g. [A=T]) occurs 500 times. The expected frequency of occurrences of the compound event [A=T, B=T, C=F] under the independence assumption is 0.5×0.5×0.5×1000 = 125. Suppose that its observed frequency is 250, we would like to see whether or not the difference between the observed and expected frequencies (i.e. 250 – 125) is significant enough to indicate that the compound event is not a random happening.<div><br></div><div>In statistics, to test the correlation between random variables, <b>contingency table</b> with chi-squared statistic (Mills, 1955) is widely used. Instead of investigating variable correlations, pattern discovery shifts the traditional correlation analysis in statistics at the variable level to association analysis at the event level, offering an effective method to detect statistical association among events.</div><div><br></div><div>In the early 90’s, this approach was established for second order event associations (Chan & Wong, 1990). A higher order <b>pattern discovery</b> algorithm was devised in the mid 90’s for discrete-valued data sets (Wong & Yang, 1997). In our methods, patterns inherent in data are defined as statistically significant associations of two or more primary events of different attributes if they pass a statistical test for deviation significance based on <b>residual analysis</b>. The discovered high order patterns can then be used for classification (Wang & Wong, 2003). With continuous data, events are defined as Borel sets and the pattern discovery process is formulated as an optimization problem which recursively partitions the sample space for the best set of significant events (patterns) in the form of high dimension intervals from which probability density can be estimated by Gaussian kernel fit (Chau & Wong, 1999). Classification can then be achieved using Bayesian classifiers. For data with a mixture of discrete and continuous data (Wong & Yang, 2003), the latter is categorized based on a global optimization discretization algorithm (Liu, Wong & Yang, 2004). As demonstrated in numerous real-world and commercial applications (Yang, 2002), pattern discovery is an ideal tool to uncover subtle and useful patterns in a database. </div><div><br></div><div>In pattern discovery, three open problems are addressed. The first concerns learning where noise and uncertainty are present. In our method, noise is taken as inconsistent samples against statistically significant patterns. Missing attribute values are also considered as noise. Using a standard statistical <b>hypothesis testing</b> to confirm statistical patterns from the candidates, this method is a less ad hoc approach to discover patterns than most of its contemporaries. The second problem concerns the detection of polythetic patterns without relying on exhaustive search. Efficient systems for detecting monothetic patterns between two attributes exist (e.g. Chan & Wong, 1990). However, for detecting polythetic patterns, an exhaustive search is required (Han, 2001). In many problem domains, polythetic assessments of feature combinations (or higher order relationship detection) are imperative for robust learning. Our method resolves this problem by directly constructing polythetic concepts while screening out non-informative pattern candidates, using statisticsbased heuristics in the discovery process. The third problem concerns the representation of the detected patterns. Traditionally, if-then rules and graphs, including networks and trees, are the most popular ones. However, they have shortcomings when dealing with multilevel and multiple order patterns due to the non-exhaustive and unpredictable hierarchical nature of the inherent patterns. We adopt <b>attributed hypergraph</b> (AHG) (Wang & Wong, 1996) as the representation of the detected patterns. It is a data structure general enough to encode information at many levels of abstraction, yet simple enough to quantify the information content of its organized structure. It is able to encode both the qualitative and the quantitative characteristics and relations inherent in the data set.<br></div>

Non-stationary continuous time Bayesian networks are introduced. They allow the parents set of each node in a continuous time Bayesian network to change over time. Structural learning of nonstationary continuous time Bayesian networks is developed under different knowledge settings. A macroeconomic dataset is used to assess the effectiveness of learning non-stationary continuous time Bayesian networks from real-world data.

Recent research in the field of composite materials has shown that it is theoretically possible to produce composite materials with macroscopic mechanical stiffness and loss properties that surpass those of conventional composites. This research explores the possibility of designing and fabricating these composite materials by embedding small volume fractions of negative stiffness inclusions in a continuous host material. Achieving high stiffness and loss from these materials by design, however, is a nontrivial task. This paper presents a hierarchical multiscale material model for these materials, coupled with a set-based, multilevel design approach based on Bayesian network classifiers. Bayesian network classifiers are used to map promising regions of the design space at each hierarchical modeling level, and then the maps are intersected to identify sets of multilevel or multiscale solutions that are likely to provide desirable system performance. Length scales range from the behavior of the structured microscale negative stiffness inclusions to the effective properties of mesoscale composite materials to the performance of an illustrative macroscale component — a vibrating beam coated with the high stiffness, high loss composite material.

Exploration of a design space is the first step in identifying sets of high-performing solutions to complex engineering problems. For this purpose, Bayesian network classifiers (BNCs) have been shown to be effective for mapping regions of interest in the design space, even when those regions of interest exhibit complex topologies. However, identifying sets of desirable solutions can be difficult with a BNC when attempting to map a space where high-performance designs are spread sparsely among a disproportionately large number of low-performance designs, resulting in an imbalanced classifier. In this paper, a method is presented that utilizes probabilities of class membership for known training points, combined with interpolation between those points, to generate synthetic high-performance points in a design space. By adding synthetic design points into the BNC training set, a designer can rebalance an imbalanced classifier and improve classification accuracy throughout the space. For demonstration, this approach is applied to an acoustics metamaterial design problem with a sparse design space characterized by a combination of discrete and continuous design variables.

In many applications, one can define a large set of features to support the classification task at hand. At test time, however, these become prohibitively expensive to evaluate, and only a small subset of features is used, often selected for their information-theoretic value. For threshold-based, Naive Bayes classifiers, recent work has suggested selecting features that maximize the expected robustness of the classifier, that is, the expected probability it maintains its decision after seeing more features. We propose the first algorithm to compute this expected same-decision probability for general Bayesian network classifiers, based on compiling the network into a tractable circuit representation. Moreover, we develop a search algorithm for optimal feature selection that utilizes efficient incremental circuit modifications. Experiments on Naive Bayes, as well as more general networks, show the efficacy and distinct behavior of this decision-making approach.

Abstract History matching is critical in subsurface flow modeling. It is to align the reservoir model with the measured data. However, it remains challenging since the solution is not unique and the implementation is expensive. The traditional approach relies on trial and error, which are exhaustive and labor-intensive. In this study, we propose a new workflow utilizing Bayesian Markov Chain Monte Carlo (MCMC) to automatically and accurately perform history matching. We deliver four novelties within the workflow: 1) the use of multi-resolution low-fidelity models to guarantee high-quality matching, 2) updating the ranges of priors to assure convergence, 3) the use of Long-Short Term Memory (LSTM) network as a low-fidelity model to produce continuous time-response, and 4) the use of Bayesian optimization to obtain the optimum low-fidelity model for Bayesian MCMC runs. We utilize the first SPE comparative model as the physical and high-fidelity model. It is a gas injection into an oil reservoir case, which is the gravity-dominated process. The coarse low-fidelity model manages to provide updated priors that increase the precision of Bayesian MCMC. The Bayesian-optimized LSTM has successfully captured the physics in the high-fidelity model. The Bayesian-LSTM MCMC produces an accurate prediction with narrow uncertainties. The posterior prediction through the high-fidelity model ensures the robustness and precision of the workflow. This approach provides an efficient and high-quality history matching for subsurface flow modeling.

Modern manufacturing processes are increasing becoming cyber-physical in nature, where a computational system monitors the system performance, provides real-time process control by analyzing sensor data collected regarding process and product characteristics, in order to increase the quality of the manufactured product. Such real-time process monitoring and control techniques are useful in precision and ultra-precision machining processes. However, the output product quality is affected by several uncertainty sources in various stages of the manufacturing process such as the sensor uncertainty, computational system uncertainty, control input uncertainty, and the variability in the manufacturing process. The computational system may be a single computing node or a distributed computing network; the latter scenario introduces additional uncertainty due to the communication between several computing nodes. Due to the continuous monitoring process, these uncertainty sources aggregate and compound over time, resulting in variations of product quality. Therefore, characterization of the various uncertainty sources and their impact on the product quality are necessary to increase the efficiency and productivity of the overall manufacturing process. To this end, this paper develops a two-level dynamic Bayesian network methodology, where the higher level captures the uncertainty in the sensors, control inputs, and the manufacturing process while the lower level captures the uncertainty in the communication between several computing nodes. In addition, we illustrate the use of a variance-based global sensitivity analysis approach for dimension reduction in a high-dimensional manufacturing process, in order to enable real-time analysis for process control. The proposed methodologies of process control under uncertainty and dimension reduction are illustrated for a cyber-physical turning process.

Gearbox fault diagnosis is one of the core research areas in the field of condition monitoring of rotating machines. The aim of this paper is to present an intelligent method for fault diagnosis of a kind of automotive gearbox in run-up condition based on vibration signals. The vibration signals are obtained from an acceleration sensor and sampled at constant time increment by AdvantechTM PCI-1712 card. Automotive gearbox test setup has been designed and constructed in Acoustics Research Laboratory in Amirkabir University of Technology. To process the non-stationary vibration signals, the re-sampling technique at constant angle increment is combined with the continuous wavelet transform (CWT) and the wavelet coefficients of the signals are obtained. The statistical parameters of the wavelet coefficients are extracted, and then the principle component analysis (PCA) is introduced to enhance the pattern recognition and reduce the dimensionality of the original feature space. Gearbox is considered in healthy, chipped tooth and worn teeth gears conditions. Finally, a feedforward multilayer perceptron (MLP) neural network is used for classification. The experimental results show that the adoption of PCA diagnosis method leads to higher accuracy and less training time for fault detection of the gear chip and wear.

Abstract Despite the outstanding improvements achieved by artificial intelligence in the Structural Health Monitoring (SHM) field, some challenges need to be coped with. Among them, the necessity to reduce the complexity of the models and the data-to-user latency time which are still affecting state-of-the-art solutions. This is due to the continuous forwarding of a huge amount of data to centralized servers, where the inference process is usually executed in a bulky manner. Conversely, the emerging field of Tiny Machine Learning (TinyML), promoted by the recent advancements by the electronic and information engineering community, made sensor-near data inference a tangible, low-cost and computationally efficient alternative. In line with this observation, this work explored the embodiment of the One Class Classifier Neural Network, i.e., a neural network architecture solving binary classification problems for vibration-based SHM scenarios, into a resource-constrained device. To this end, OCCNN has been ported on the Arduino Nano 33 BLE Sense platform and validated with experimental data from the Z24 bridge use case, reaching an average accuracy and precision of 95% and 94%, respectively.

The paper presents a method for analyzing data from sensors and developing the predictive models based on learning methods. There are some methods, described on scientific literature, such as statistical methods (linear regression, logistic regression, and Bayesian models), advanced methods based on machine learning and data mining (decision trees and artificial neural networks) and survival models. All of these methods are intended to discover the correlation and covariance between biomedical parameters. This paper presents the decision tree method for predictive health modeling based on machine learning and data mining. Based on this method used can be developed a decision support system for healthcare. Machine learning is used in healthcare predictive modeling for learning to recognize complex patterns within big data received from biomedical sensors. The sensors data fusion refers to the usage of the sensors wireless network and data fusion on the same level (for similar sensors - e. g. temperature sensors) and on different levels (different sensors category - pulse, breath, temperature, moisture sensors) for developing the decision systems. Big data concept is familiar for medical sciences (genomics, biomedical research) and also for physical sciences (meteorology, physics and chemistry), financial institutions (banking and capital markets) and government (defense). For predictive models in clinical analysis is important to establish the time steps discretization of occurrence of a particular event (critical state required continuous monitoring) for observe the impact of the correlated values for biomedical parameters. These aspects presented are useful for healthcare learning about correlation between diseases and biomedical parameters.