Bayesian Network Theory Research Articles

Conditional Coordination Games on Cyclic Social Influence Networks

Coordinated decision making is one of the fundamental attributes of intelligent behavior, and game theory has long served as a framework within which to model interactive decision making, especially for scenarios where decision makers are motivated to compete for the distribution of scarce resources. It is not obvious, however, that the standard game-theoretic framework is appropriate for complex social scenarios where agents are influenced by the attitudes and opinions of others and, consequently, the opportunity exists to make coordinated decisions in the pursuit of coherent group behavior. Noncooperative game theory is founded on the premise that choices ought to be strategically rational—agents make best-replies to the expected actions of others. However, when modeling groups whose members are responsive to social influence, a relevant notion of behavior is for them to coordinate rather than compete. The combination of conditional game theory and Bayesian network theory provides a framework within which to formalize a theory of coordinated decision making under social influence. As originally developed, however, conditional game theory applies only to acyclical networks involving unilateral influence propagation. This paper extends the theory to account for networks with cycles, where agents are able to exert multilateral influence on each other. The Markov convergence theorem establishes conditions for convergence to steady-state coordinated decisions, and demonstrates its use with a bilateral collaboration scenario. In addition, this paper provides a mathematical analysis of several canonical network topologies.

Fault Diagnosis in the Field of Additive Manufacturing (3D Printing) Using Bayesian Networks

In this work, a new approach for fault diagnosis in the field of additive manufacturing (3d printing) using artificial intelligence will be given. This approach is based on the marriage of the Bayesian Networks theory and data acquisition techniques. Bayesian Networks are well known for their ability to infer probabilities and to give decisional support under uncertainty. In order to do so, these probability engines must be constructed and maintained by a big amount of data and information using learning algorithms. This work provides a methodology that uses sensors based data acquisition and processing to construct such networks. Some of these sensors are already available in most of the 3d printers available in the market, while other sensors were additionally embedded in a studied 3d printer in order to enrich the number of observational variables to gain a high level of fault diagnosis accuracy and support.

Online Car-Hailing System Performance Analysis Based on Bayesian Network

The concept of the sharing economy has attracted wide attention due to its huge impact on transforming traditional industries. Online car-hailing, combining the sharing economy and ICT technologies, shapes a new landscape, which can greatly shorten the traveler's waiting time and reduce the empty-run rate of cars (such as Uber and DiDi). However, a few comprehensive studies have been conducted on the sustainable development of online car-hailing considering both the user experience and the operational cost. To address this issue, this paper systematically studies the influencing factors, their relations, and their impacts on online car-hailing in an empirical way. First, an index system in four key aspects, namely service, price, safety, and traveling time, is established to evaluate the user experience. Second, the Bayesian network theory is employed to model the complexity of each factor and the extent of its influence on the online car-hailing system with expert scores. The two most important influence paths affecting the passengers' choices of online car-hailing are determined. Third, we further construct an investment allocation model with the aim of minimizing the economic cost of the online car-hailing system while maintaining the system performance, considering the limiting factors, such as the complexity and cost. Finally, we perform a simulation experiment, which generates some practical suggestions for improving the online car-hailing system.

Applications of machine learning for facies and fracture prediction using Bayesian Network Theory and Random Forest: Case studies from the Appalachian basin, USA

Identification and prediction of facies and fractures are critical to subsurface geosystems analysis and hydrocarbon exploration. However, accurate prediction of facies and fractures is hard in the absence of high-resolution advanced geophysical logs (e.g. image logs) and core samples. This study demonstrates the application of machine learning algorithms, such as Bayesian Network Theory and Random Forest to predict the presence of different facies and fractures in sedimentary rocks using common well logs. We build and train supervised machine learning models using Bayesian Network Theory and Random Forest to classify facies and fractures in unconventional shale and conventional sandstone, and carbonate reservoirs. The trained machine-learning models are cross-validated 10-fold to check their robustness. Application of Bayesian Network theory and Random Forest shows that both facies and fractures can be predicted with high accuracy using limited common well logs. In addition, the Bayesian Network shows the complex causal relationship among the input petrophysical parameters and output (facies or fracture), which is its unique feature compared to other machine learning algorithms, such as Random Forest.

La medición de la familia en las empresas familiares privadas: un modelo de red bayesiana

<p><span lang="EN-US">The objective of this analysis was to identify the causality among variables that originate the highest level of <em>familiness </em>in private family firms. The Bayesian Networks (BN) theory was applied to measure the effectiveness of resources and capabilities provided by the family members within a family business to understand causal relations among variables by using probabilistic reasoning throughout a graphic. Re­sults showed that if salary of family members was higher than salary of employees in the same position, if family members shared information among themselves, and if family firms presented family-employee bonds, there was an 83%, 70%, and 79% of probability of having a high level <em>familiness</em>, respectively. The limitation of the study is that any modification in the BN might show different outcomes. These findings expand the knowledge on family business discipline and suggest a path for family business’ leaders to increase <em>familiness</em>. If family firms want to strengthen their competitive advantage, the main variables they should focus, among all the resources and capabilities that represent <em>familiness</em>, are salaries of family members, sharing information, and family-employee bonds.</span></p>

Improved Bayesian Network-Based Risk Model and Its Application in Disaster Risk Assessment

The application of Bayesian network (BN) theory in risk assessment is an emerging trend. But in cases where data are incomplete and variables are mutually related, its application is restricted. To overcome these problems, an improved BN assessment model with parameter retrieval and decorrelation ability is proposed. First, multivariate nonlinear planning is applied to the feedback error learning of parameters. A genetic algorithm is used to learn the probability distribution of nodes that lack quantitative data. Then, based on an improved grey relational analysis that considers the correlation of variation rate, the optimal weight that characterizes the correlation is calculated and the weighted BN is improved for decorrelation. An improved risk assessment model based on the weighted BN then is built. An assessment of sea ice disaster shows that the model can be applied for risk assessment with incomplete data and variable correlation.

Bayesian network modelling for supply chain risk propagation

Supply chain risk propagation is a cascading effect of risks on global supply chain networks. The paper attempts to measure the behaviour of risks following the assessment of supply chain risk propagation. Bayesian network theory is used to analyse the multi-echelon network faced with simultaneous disruptions. The ripple effect of node disruption is evaluated using metrics like fragility, service level, inventory cost and lost sales. Developed risk exposure and resilience indices support in assessing the vulnerability and adaptability of each node in the supply chain network. The research provides a holistic measurement approach for predicting the complex behaviour of risk propagation for improved supply chain risk management.

Methods for optimising energy efficiency and renovation processes of complex public properties

The HoEff-CIM-project aims at developing automated inventory and evaluation tools, which support portfolio managers to plan and perform energy efficiency measures within large properties, considering their highly heterogeneous usages and structural specifications. The project covers models and tools, which facilitate the assessment of existing building stocks and the recognition of state of the art energetic reliance and renovation concepts. These concepts are then automatically adapted to different structural, technical and usage dependent parameters. The tools will help to analyse building physics and building services but also to empower strategic planning and energy distribution schemes. Therefore, an intuitive dataflow was needed that streamlines the process from data acquisition (building inspection), evaluation of buildings and renovation measures to the elaboration of specific building energy reports; all resulting in master plans for building complexes as well as systematic renovation strategies for larger properties. Particular emphasis was put on the determination of joint influences of building construction and building services on potential energy savings. Accompanying this quantification process, uncertainty and sensitivity analyses have been performed via Bayesian Network theory and statistical analytics. Different renovation concepts have been analysed, modelled and evaluated regarding their energy saving potential, usability as well as ecologic features. A ranking system was developed to assess specific energy demands of different building usages and to identify the most effective renovation measures. The result is a fully transparent, interactive and easy to understand decision-making methodology. The holistic approach contributes to create climate neutral building stocks and consequently to reach the energy saving and environmental goals defined by national and international governmental institutions (e.g. EED).

Contagion in Financial Systems: A Bayesian Network Approach

We develop a structural default model for interconnected financial institutions in a probabilistic framework. For all possible network structures we characterize the joint default distribution of the system using Bayesian network methodologies. Particular emphasis is given to the treatment and consequences of cyclic financial linkages. We further demonstrate how Bayesian network theory can be applied to detect contagion channels within the financial network, to measure the systemic importance of selected entities on others, and to compute conditional or unconditional probabilities of default for single or multiple institutions.

Establishment of optimal regulatory network of colorectal cancer based on p42.3 protein.

Objective: to establish regulatory network of colorectal cancer involving p42.3 protein and to provide theoretical evidence for deep functional exploration of p42.3 protein in the onset and development of colorectal cancer. Methods: with protein similarity algorithm, reference protein set of p42.3 cell apoptosis was built according to structural features of p42.3. GO and KEGG databases were used to establish regulatory network of tumor cell apoptosis involving p42.3; meanwhile, the largest possible working pathway that involves p42.3 protein was screened out based on Bayesian network theory. Besides, GO and KEGG were used to build regulatory network on early diagnosis gene markers for colorectal cancer including WWOX, K-ras, COX-2, p53, APC, DCC and PTEN, at the same time, a regulatory network of colorectal cancer cell apoptosis which involves p42.3 was established. Results: cell apoptotic regulatory network that p42.3 participates in primarily consists of Bcl-2 family genes and the largest possible pathway is p42.3 → FKBP → Bcl-2 centered as FKBP protein. Combined with colorectal cancer regulatory network that involves early diagnosis gene markers, it can be predicted that p42.3 is most likely to regulate the colorectal cancer cell apoptosis through FKBP → Bcl-2 → Bax → caspase-9 → caspase-3 pathway. Conclusion: the colorectal cancer apoptosis network based on p42.3 established in the study provides theoretical evidence for deep exploration of p42.3 regulatory mechanism and molecular targeting treatment of colorectal cancer.

A machine learning bayesian network for refrigerant charge faults of variable refrigerant flow air conditioning system

An intelligent fault diagnosis network for variable refrigerant flow air conditioning system is proposed in this study. The network is developed under the foundation of bayesian belief network theory, which comprises two main elements: the structure and parameters. The structure obtained by machine learning and experts’ experiences illustrates the relationships among faults and physical variables from the qualitative prospective, and its parameters (including prior probability distribution and conditional distribution) describe the uncertainty between them quantitatively. Once the structure and parameters are determined, the posterior probability distribution which can be used to complete fault diagnosis and isolation will be calculated by some algorithms. In comparison with other fault diagnosis approaches, the proposed approach can make full use of performance information. Moreover, it is more reasonable and precise to express the relationship between faults and variables rather than Boolean variables. Evaluation was conducted on a variable refrigerant flow air conditioning system, which demonstrated that this strategy is effective and efficient.

A methodology for assessing the reliability and security of cascade reservoir systems

The scale of cascade reservoir systems is increasing rapidly along with the development of hydroelectric projects in these systems. Related security problems are receiving increasing public attention. However, no security evaluation method for this type of system can be found in the literature. Reliability is an essential indicator to measure system security. Because of the brittleness, complexity, and polymorphism of the system, a Bayesian network and set pair analysis theory–based methodology suitable for the reliability analysis of cascade reservoir systems is suggested in this article. First, the topological structure of the Bayesian network is constructed, and the brittle effect is measured using the powerful probabilistic inference capability of the Bayesian network. Then, the probability distribution of each subsystem is obtained, and probabilities of the system in all states are integrated using set pair analysis theory. Furthermore, the composite reliability is proposed as a generic reliability representation for cascade reservoir systems. On this basis, through constructing a composite probability vector and composite probability matrix, the Bayesian network and set pair analysis theory–based reliability measure models for both subsystems and the whole system are proposed. Finally, the effectiveness and practicability of the method described above are verified by introducing an example application. The results show that this research provides a theoretical basis and practical method for assessing the reliability and security of cascade reservoir systems.

On the phylogeny graphs of degree-bounded digraphs

Hefner et al. (1991) characterized acyclic digraphs each vertex of which has indegree and outdegree at most two and whose competition graphs are interval. They called acyclic digraphs each vertex of which has indegree and outdegree at most two (2,2) digraphs. In this paper, we study the phylogeny graphs of (2,2) digraphs. Especially, we give a sufficient condition and necessary conditions for (2,2) digraphs having chordal phylogeny graphs. Phylogeny graphs are also called moral graphs in Bayesian network theory. Our work is motivated by problems related to evidence propagation in a Bayesian network for which it is useful to know which acyclic digraphs have their moral graphs being chordal.

SWARMs Ontology: A Common Information Model for the Cooperation of Underwater Robots.

In order to facilitate cooperation between underwater robots, it is a must for robots to exchange information with unambiguous meaning. However, heterogeneity, existing in information pertaining to different robots, is a major obstruction. Therefore, this paper presents a networked ontology, named the Smart and Networking Underwater Robots in Cooperation Meshes (SWARMs) ontology, to address information heterogeneity and enable robots to have the same understanding of exchanged information. The SWARMs ontology uses a core ontology to interrelate a set of domain-specific ontologies, including the mission and planning, the robotic vehicle, the communication and networking, and the environment recognition and sensing ontology. In addition, the SWARMs ontology utilizes ontology constructs defined in the PR-OWL ontology to annotate context uncertainty based on the Multi-Entity Bayesian Network (MEBN) theory. Thus, the SWARMs ontology can provide both a formal specification for information that is necessarily exchanged between robots and a command and control entity, and also support for uncertainty reasoning. A scenario on chemical pollution monitoring is described and used to showcase how the SWARMs ontology can be instantiated, be extended, represent context uncertainty, and support uncertainty reasoning.

Probabilistic analysis of natural gas pipeline network accident based on Bayesian network

Natural gas pipeline network (NGPN) accident is a kind of catastrophic disaster as the hazard of natural gas may present a large-scale extension in NGPN that can easily result in cascading accidents. In this paper, the Bayesian network (BN) was employed to probabilistically analyze natural gas pipeline network accidents. On the basis of case-studies of typical NGPN accidents, eleven BN nodes were proposed to represent the evolution process of natural gas pipeline network accidents from failure causes to consequences. The conditional probabilities of every BN node were determined by expert knowledge with weighted treatments by the Dempster-Shafer evidence theory. Through giving evidences of some BN nodes with certain state values, the probabilities of evolution stages and consequences of the natural gas pipeline network accident can be estimated. The results indicate that the combination of Bayesian network and Dempster-Shafer evidence theory is an alternative method for evaluating NGPN accident, and the proposed framework can provide a more realistic consequence analysis since it could consider the conditional dependency in the evolution process of the NGPN accident. This study could be helpful for emergency response decision-making and loss prevention.

Classification of X-Ray images of shipping containers

Smuggling has long played an important role in the inefficiency of economies. To secure the borders against this illegal act, X-Ray Inspection Systems are often deployed at the borders and customs. In this paper, we present a new method for classification of shipping containers X-Ray images, produced in the inspection lines. The aim is to improve the matching accuracy of the presented manifest, which lists the claimed contents of the shipping containers, with the real contents of the container. The proposed method is based on utilizing Scale Invariant Feature Transforms (SIFT) feature vectors, Bag of visual words (BOVW) and tree augmented naive Bayes (TAN) approach for classifying containers X-Ray images. The prior research on classification of X-Ray images of shipping containers has focused mostly on working with greedy algorithms such as sliding windows for task of classification. More recent studies introduced filter banks and visual words for extraction of features. The proposed method for the first time considers the salient points and keypoints for the task of feature extraction. In addition, this paper presents a framework using the tree augmented naive Bayes based on the theory of learning Bayesian networks, which is proved to have a significant improvements upon the prior designed systems by considering the correlations among the extracted features. For experimental evaluations, our method is compared with two recently proposed methods on containers X-Ray images categorization. The results show that the proposed method is more accurate and time-efficient in categorization of X-Ray images.

Exploration and analysis of drug modes of action through feature integration.

Identifying drug modes of action (MoA) is of paramount importance for having a good grasp of drug indications in clinical tests. Anticipating MoA can help to discover new uses for approved drugs. Here we first used a drug-set enrichment analysis method to discover significant biological activities in every mode of action category. Then, we proposed a new computational model, a probability ensemble approach based on Bayesian network theory, which integrated chemical, therapeutic, genomic and phenotypic properties of over a thousand of FDA approved drugs to assist with the prediction of MoA. 10-fold cross validation tests demonstrate that this method can achieve better performances than four other methods with the area under the receiver operating characteristic (ROC) curves. Finally, we further conducted a large-scale prediction for drug-MoA pairs. Using the Cardiovascular Agents category as an example, several predicted drug-MoA pairs were supported by literature resources.

A Bayesian Network Approach to Causation Analysis of Road Accidents Using Netica

Based on an overall consideration of factors affecting road safety evaluations, the Bayesian network theory based on probability risk analysis was applied to the causation analysis of road accidents. By taking Adelaide Central Business District (CBD) in South Australia as a case, the Bayesian network structure was established by integrating K2 algorithm with experts’ knowledge, and Expectation-Maximization algorithm that could process missing data was adopted to conduct the parameter learning in Netica, thereby establishing the Bayesian network model for the causation analysis of road accidents. Then Netica was used to carry out posterior probability reasoning, the most probable explanation, and inferential analysis. The results showed that the Bayesian network model could effectively explore the complex logical relation in road accidents and express the uncertain relation among related variables. The model not only can quantitatively predict the probability of an accident in certain road traffic condition but also can find the key reasons and the most unfavorable state combination which leads to the occurrence of an accident. The results of the study can provide theoretical support for urban road management authorities to thoroughly analyse the induction factors of road accidents and then establish basis in improving the safety performance of the urban road traffic system.

Identification and Analysis of Metro Foundation Construction Safety Risk Based on Fault Tree Analysis

Fuzzy theory and Bayesian network theory were based on the fault tree. The fault tree analysis method improved the fault analysis to obtain number of event's exact problem. Considering the characteristics of the fuzzy risk analysis at the same time, it also solved the problem of scale fault analysis modeling. The theory is applied to the cost of construction shield tunnel risk analysis, floating shield tunnel segments, dislocation of failure risk analysis and deep station foundation construction risk analysis. By security risk identification and evaluation process, it can be improved to subway construction process of deep foundation pit hazard anticipation. Before construction accidents to avoid and resolve security risks, thereby improving overall system safety performance. Based on this, the paper depth study the theory and methods of security risk identification and security risk assessment, further enrich the security risk management theory, and put forward to deal with the security risk response countermeasures.

Multi-state reliability assessment for hydraulic lifting system based on the theory of dynamic Bayesian networks

Mechanical systems and their components usually have multiple failure modes and different performance states. Most existing system reliability modelling theories are developed on the basis of binary logic, which lack sufficient ability to describe the above phenomena. In this article, dynamic Bayesian network theory is employed to evaluate the multi-state reliability of a hydraulic lifting system. First, failure mode and effect analysis and structural analysis and design technique are comprehensively applied to analyse the functionalities and failure modes of the components. Afterwards, the time factor is integrated into the model by considering the state transition of the components. In this way, the multi-state reliability model of the system is established by dynamic Bayesian network. The reliability assessment and diagnostic analysis are performed by taking advantage of the dynamic Bayesian network’s bi-directional reasoning ability, and the results are in good agreement with actual situation. It shows that the proposed approach is effective and convenient for multi-state reliability modelling and analysis for mechanical systems.