Bayesian Network Model Research Articles

Quantitative analysis of aeolian dust incidence and diagnosis using a dynamic Bayesian network model: A case study of estuary area in central Taiwan

The presence of high concentrations of PM10 in aeolian dust within the estuary area of central Taiwan has been a focal point for air quality protection and remediation efforts over the past 15 years. This study explores the quantitative analysis of aeolian dust occurrences concerning various levels of PM10 concentration, aiming to offer early warning information to local communities and government agencies. Employing Dynamic Bayesian network (DBN) models, the research interrelates conditional probability dependencies among air quality and meteorological factors monitored in the downstream area of the Jhuoshuei River, predicting the probability distribution of aeolian dust events. The constructed DBNs demonstrate the inference of probability distributions of meteorological factors' states based on PM10 concentrations, enabling diagnostic analysis of meteorological conditions during high PM10 concentration dust events. The DBN models exhibit robust predictive ability, achieving an overall accuracy of 85.72% in validation and 85.8% in testing, with high discriminatory capacity that the area under the receiver operating characteristic curve (AUC) values of 0.924 and 0.915 during validation and testing, respectively. Sensitivity analysis highlights wind direction, wind speed, relative humidity as significant factors affecting PM10 concentrations. The probability of low PM10 concentrations was notably higher compared to higher concentrations, indicating effective aeolian dust remediation efforts. However, attention is warranted toward the 10% incidence of aeolian dust events deemed unhealthy or hazardous to human health. Diagnosis of meteorological conditions during high PM10 concentration events suggests low temperature, high wind speeds, and dry conditions, particularly with northeastern or southwestern winds, as conducive factors. This study underscores the physical relationship between meteorological factors and PM10 concentrations, employing DBNs to consider temporal relationships and uncertainties in data, aligning with human health impact standards proposed by the Ministry of Environment. Future research should incorporate soil and land surface conditions to provide a more comprehensive understanding of aeolian dust dynamics and mitigation strategies.

Socio-Demographic Factors Driving the Choice of Alternative Safe Water Sources and Their Implications for Public Health: Lessons from Goalmari, Bangladesh

This study examined how socio-economic driving forces influence households’ choice of water, ranging from a piped water supply provided by Veolia to untreated sources contaminated with high levels of arsenic and pathogens. Households fall into three cluster groups based on variations in socio-economic status and physical, infrastructure, and institutional elements. About 64% of the variations are related to differences in awareness, willingness, and ability to pay for safe water sources. Families with higher monthly income showed interest in accepting Veolia’s house connection options, resulting in the shutdown of six community tap points and ultimately affecting the low-income households’ accessibility to Veolia water. A causal loop diagram showed five feedback loops influencing the choice of drinking contaminated water. Bayesian Network models were sensitive to the ability, accessibility, and willingness to pay for safe water, deep tube well distributions, installation and maintenance costs, ownership of tube wells, household income level, and the level of awareness. Results suggest that the risks of drinking contaminated water can be minimized by raising awareness; accepting arsenic removal techniques; sharing expenses; training for deep tube well installations and maintenance; increasing Veolia pipeline coverage; and redesigning the tap point distributions. These results help identify the relative importance of such interventions to improve water security in safe water-poor areas.

Bayesian approach to energy dependence of fission product yields of 235U by data augmentation

ABSTRACT We utilized a two hidden-layer Bayesian neural network (BNN) model along with data augmentation (DA) to predict the energy dependence of fission product yields (FPY). In the BNN model, the JENDL-5 FPY data are separated into 80 % for training and 20 % for validation. Additionally, the training data are combined with experimental cumulative fission yields and calculated values by five Gaussian model. The number of units in each layer and activation function were selected carefully to reproduce the global and fine structure of the FPY data. Through comparing the results with and without DA, we found that DA is particularly valuable for specific nuclides. The evaluation results with DA demonstrate reliable and accurate predictions of the energy dependence of fission product yields of 235U.

Development of an improved Bayesian network method for maritime accident safety assessment based on multiscale scenario analysis theory

The strait is infrastructure for connecting ports, ensuring the national economy and human life. Assessing maritime safety is crucial for strait operation and maintenance. Current methods over-rely on statistical analysis of historical data, which lacks scientific algorithms and cannot accurately assess safety timeliness. Therefore, this study proposes an improved Bayesian network (BN) model based on the multiscale scenario analysis theory to process to the maritime safety assessment process. Firstly, the maritime safety assessment model is constructed based on the knowledge element theory (KET) and scenario analysis theory (SAT). Secondly, a heuristic learning algorithm is used to construct the BN model of maritime accidents to define scenario elements and key parameters. Thirdly, a scenario simulation is conducted with the Bohai Strait as a case to assessment its emergency capability in catastrophic accidents. This paper successfully developed a maritime accident safety assessment framework using an improved BN algorithm within a scenario analysis perspective, significantly enhancing the accuracy and reliability of safety assessments. This methodology used in this study can be used a tool suitable for the safety assessment of major infrastructure accidents at the national level and the results obtained can serve as a reference for dynamically revising emergency plans.

An international relations quantitative evaluation model (IRQEM) based on an integrated method.

International relations (IR) have great uncertainty and instability. Bad IR or conflicts will bring about heavy economic losses and widespread social unrest domestically and internationally. The accurate prediction for bilateral relations can support decision making for timely responses, which will be used to find ways to maintain development in the complex international situation. An international relations quantitative evaluation model (IRQEM) is proposed by integrating a variety of research models and methods like the interpretative structural modeling method (ISM), Bayesian network (BN) model, the Bayesian search (BS), and the expectation-maximization (EM) algorithm, which is novel for IR research. Factors from several different fields are identified as BN nodes. Each node is assigned different state values. The hierarchical structure of these BN nodes is obtained by ISM. The data collection of 192 cases is used to construct the BN model by GeNIe 4.0. The IRQEM can be used to evaluate the influence of emergencies on IR. The critical factors of IR also can be explored through our proposed model. Results show that the prediction of bilateral relations under emergencies can be realized by updating the indicator set when emergencies occur. The capability to anticipate threats of IR changes is advanced by optimizing the reporting information of IR forecasting through a combination of qualitative and quantitative methods, charts, and texts. Relevant analysis results can provide support for national security decision making.

Bridging the Gap Between Human Toxicology and Ecotoxicology Under One Health Perspective by a Cross-Species Adverse Outcome Pathway Network for Reproductive Toxicity.

Although ecotoxicological and toxicological risk assessments are performed separately from each other, recent efforts have been made in both disciplines to reduce animal testing and develop predictive approaches instead, for example, via conserved molecular markers, and in vitro and in silico approaches. Among them, adverse outcome pathways (AOPs) have been proposed to facilitate the prediction of molecular toxic effects at larger biological scales. Thus, more toxicological data are used to inform on ecotoxicological risks and vice versa. An AOP has been previously developed to predict reproductive toxicity of silver nanoparticles via oxidative stress on the nematode Caenorhabditis elegans (AOPwiki ID 207). Following this previous study, our present study aims to extend the biologically plausible taxonomic domain of applicability (tDOA) of AOP 207. Various types of data, including in vitro human cells, in vivo, and molecular to individual, from previous studies have been collected and structured into a cross-species AOP network that can inform both human toxicology and ecotoxicology risk assessments. The first step was the collection and analysis of literature data to fit the AOP criteria and build a first AOP network. Then, key event relationships were assessed using a Bayesian network modeling approach, which gave more confidence in our overall AOP network. Finally, the biologically plausible tDOA was extended using in silico approaches (Genes-to-Pathways Species Conservation Analysis and Sequence Alignment to Predict Across Species Susceptibility), which led to the extrapolation of our AOP network across over 100 taxonomic groups. Our approach shows that various types of data can be integrated into an AOP framework, and thus facilitates access to knowledge and prediction of toxic mechanisms without the need for further animal testing. Environ Toxicol Chem 2024;00:1-14. © 2024 The Authors. Environmental Toxicology and Chemistry published by Wiley Periodicals LLC on behalf of SETAC.

Prediction of residential building occupancy using Machine learning with integrated sensor and survey Data: Insights from a living lab in Morocco

Building occupancy information is essential for effective energy management in buildings through the adoption of energy conservation and occupant-centric control strategies. These strategies endeavor to contribute to optimizing energy consumption while ensuring occupant comfort. This study focuses on advanced occupancy modeling techniques to enhance energy efficiency in residential buildings, utilizing various data-driven techniques. Various Machine Learning models, such as Random Forest, Bayesian Network, Decision Trees, Support Vector Machines, K-Nearest Neighbors, eXtreme Gradient Boosting, and Regularized Greedy Forest, have been evaluated for predicting and classifying building occupancy. Employing a living lab approach within a residential setting, the research evaluates model performance using two ground truth datasets: IoT sensor and survey data. Experimental tests use Accuracy and F1_score as evaluation criteria, demonstrating accuracy rates ranging from 70% to 95.96%. The results highlight the performance of these models in predicting residential occupancy, offering valuable insight into two approaches to occupancy modeling. The Random Forest model performs exceptionally well in capturing occupancy trends, while the Bayesian Network model, combined with expert knowledge, provides detailed predictions of occupancy types and zones. The research also addresses challenges related to data collection, including privacy concerns. It presents effective occupancy modeling strategies for residential energy management.

Identifying influencing factors of metabolic syndrome in patients with major depressive disorder: A real-world study with Bayesian network modeling

BackgroundThe bidirectional relationships between metabolic syndrome (MetS) and major depressive disorder (MDD) were discovered, but the influencing factors of the comorbidity were barely investigated. We aimed to fully explore the factors and their associations with MetS in MDD patients. MethodsThe data were retrieved from the electronic medical records of a tertiary psychiatric hospital in Beijing from 2016 to 2021. The influencing factors were firstly explored by univariate analysis and multivariate logistic regressions. The propensity score matching was used to reduce the selection bias of participants. Then, the Bayesian networks (BNs) with hill-climbing algorithm and maximum likelihood estimation were preformed to explore the relationships between influencing factors with MetS in MDD patients. ResultsTotally, 4126 eligible subjects were included in the data analysis. The proportion rate of MetS was 32.6 % (95 % CI: 31.2 %–34.1 %). The multivariate logistic regression suggested that recurrent depression, uric acid, duration of depression, marriage, education, number of hospitalizations were significantly associated with MetS. In the BNs, number of hospitalizations and uric acid were directly connected with MetS. Recurrent depression and family history psychiatric diseases were indirectly connected with MetS. The conditional probability of MetS in MDD patients with family history of psychiatric diseases, recurrent depression and two or more times of hospitalizations was 37.6 %. ConclusionUsing the BNs, we found that number of hospitalizations, recurrent depression and family history of psychiatric diseases contributed to the probability of MetS, which could help to make health strategies for specific MDD patients.

Uncovering hidden and complex relations of pandemic dynamics using an AI driven system

The COVID-19 pandemic continues to challenge healthcare systems globally, necessitating advanced tools for clinical decision support. Amidst the complexity of COVID-19 symptomatology and disease severity prediction, there is a critical need for robust decision support systems to aid healthcare professionals in timely and informed decision-making. In response to this pressing demand, we introduce BayesCovid, a novel decision support system integrating Bayesian network models and deep learning techniques. BayesCovid automates data preprocessing and leverages advanced computational methods to unravel intricate patterns in COVID-19 symptom dynamics. By combining Bayesian networks and Bayesian deep learning models, BayesCovid offers a comprehensive solution for uncovering hidden relationships between symptoms and predicting disease severity. Experimental validation demonstrates BayesCovid ’s high prediction accuracy (83.52–98.97%). Our work represents a significant stride in addressing the urgent need for clinical decision support systems tailored to the complexities of managing COVID-19 cases. By providing healthcare professionals with actionable insights derived from sophisticated computational analysis, BayesCovid aims to enhance clinical decision-making, optimise resource allocation, and improve patient outcomes in the ongoing battle against the COVID-19 pandemic.

Analyzing ship collision accidents in China: A framework based on the N-K model and Bayesian networks

Maritime accidents are frequent in China, with collisions constituting a significant portion, leading to substantial casualties, financial losses, and environmental harm. Analyzing collision risk is crucial for safe shipping operations, as such accidents arise when multiple risk factors interact beyond certain thresholds. Thus, understanding the coupled relationships among these factors is essential for accident prevention. This study presents a quantitative coupling risk assessment approach for collisions using the N–K model and Bayesian networks (BNs). Through an analysis of 304 accident reports from the China Maritime Safety Administration (CMSA), four primary risk factors (human, ship, environment, and management) are identified. The N–K model quantifies the coupling degree between these factors, while a BN model further elucidates and quantifies the collision risk based on the calculations of N–K model. Validation of the BN model is followed by sensitivity analysis to assess the impact of risk factors on accident occurrence. Subsequently, the method is applied to a case study on collision risk assessment in Chinese waters. Findings indicate that multifactor coupling has a higher occurrence probability than two-factor coupling, with human-ship-environment-management coupling presenting the highest likelihood. Human and management factors emerge as pivotal in collision incidents, while the impact of ship and environmental factors on coupling becomes more pronounced as their probabilities vary. Leveraging the capability of BNs to handle uncertainties and logical relationships, combined with the data-driven N–K model, mitigates subjectivity. The results propose the formulation of risk management strategies and measures for practitioners to enhance maritime safety.

Integrating Multiple Bacterial Phenotypes and Bayesian Network for Analyzing Health Risks of Pathogens in Plastisphere.

Plastic pollution represents a critical threat to soil ecosystems and even humans, as plastics can serve as a habitat for breeding and refuging pathogenic microorganisms against stresses. However, evaluating the health risk of plastispheres is difficult due to the lack of risk factors and quantification model. Here, DNA sequencing, single-cell Raman-D2O labeling, and transformation assay were used to quantify key risk factors of plastisphere, including pathogen abundance, phenotypic resistance to various stresses (antibiotic and pesticide), and ability to acquire antibiotic resistance genes. A Bayesian network model was newly introduced to integrate these three factors and infer their causal relationships. Using this model, the risk of pathogen in the plastisphere is found to be nearly 3 magnitudes higher than that in free-living state. Furthermore, this model exhibits robustness for risk prediction, even in the absence of one factor. Our framework offers a novel and practical approach to assessing the health risk of plastispheres, contributing to the management of plastic-related threats to human health.

Navigating uncertainty: A dynamic Bayesian network-based risk assessment framework for maritime trade routes

Maritime safety is crucial for international seaborne trade and the global economy. Acknowledging the inevitable and multifaceted risks present in maritime navigation, we introduce a novel approach to evaluate the time-dependent risk performance of various routes, with the objective to mitigate vulnerabilities and fortify resilience. To achieve this goal, we have developed an effective framework using a Dynamic Bayesian network (DBN) model to capture the interactive relationships among the influential factors in probabilistic terms. This framework employs a combination of the synthetic minority over-sampling technique and edited nearest neighbor approach to address the issue of imbalanced maritime accident reports. Furthermore, the gradient descent algorithm is utilized to calculate the conditional probabilities among the influential factors, while the Markov chain model is applied to develop the DBN model. The framework enables dynamical consideration of the risk performance of the maritime trade routes and identifies the key influential factors during different time intervals. Our research scientifically identifies the devastating impact of terrorism on shipping routes, followed by other risk factors such as piracy, military conflicts, and various induced risks, ranked in descending order of severity. The case study on the Indian Ocean shipping route, which applies the proposed methodology, reveals an initial phase of fluctuating increases, succeeded by a prevailing downward trend over the span of 2009–2019. The methodology and results provide valuable support for maritime stakeholders in making informed decisions.

Improving the accuracy of genomic prediction in dairy cattle using the biologically annotated neural networks framework

BackgroundBiologically annotated neural networks (BANNs) are feedforward Bayesian neural network models that utilize partially connected architectures based on SNP-set annotations. As an interpretable neural network, BANNs model SNP and SNP-set effects in their input and hidden layers, respectively. Furthermore, the weights and connections of the network are regarded as random variables with prior distributions reflecting the manifestation of genetic effects at various genomic scales. However, its application in genomic prediction has yet to be explored.ResultsThis study extended the BANNs framework to the area of genomic selection and explored the optimal SNP-set partitioning strategies by using dairy cattle datasets. The SNP-sets were partitioned based on two strategies–gene annotations and 100 kb windows, denoted as BANN_gene and BANN_100kb, respectively. The BANNs model was compared with GBLUP, random forest (RF), BayesB and BayesCπ through five replicates of five-fold cross-validation using genotypic and phenotypic data on milk production traits, type traits, and one health trait of 6,558, 6,210 and 5,962 Chinese Holsteins, respectively. Results showed that the BANNs framework achieves higher genomic prediction accuracy compared to GBLUP, RF and Bayesian methods. Specifically, the BANN_100kb demonstrated superior accuracy and the BANN_gene exhibited generally suboptimal accuracy compared to GBLUP, RF, BayesB and BayesCπ across all traits. The average accuracy improvements of BANN_100kb over GBLUP, RF, BayesB and BayesCπ were 4.86%, 3.95%, 3.84% and 1.92%, and the accuracy of BANN_gene was improved by 3.75%, 2.86%, 2.73% and 0.85% compared to GBLUP, RF, BayesB and BayesCπ, respectively across all seven traits. Meanwhile, both BANN_100kb and BANN_gene yielded lower overall mean square error values than GBLUP, RF and Bayesian methods.ConclusionOur findings demonstrated that the BANNs framework performed better than traditional genomic prediction methods in our tested scenarios, and might serve as a promising alternative approach for genomic prediction in dairy cattle.

Remote Coastal Weed Infestation Management Using Bayesian Networks

The increasing prevalence of species that are detrimental to biodiversity is a major concern, particularly for managers of national parks. To develop effective programmes for controlling weeds, it is essential to have a thorough understanding of the extent and severity of infestations, as well as the contributing factors such as temperature, rainfall, and disturbance. Predicting these factors on a regional scale requires models that can incorporate a wide range of variables in a quantifiable manner, while also assisting with on-ground operations. In this study, we present two Bayesian Network models specifically designed for six significant weed species found along the southern coast of Australia. Our models are based on empirical data collected during a coastal weed survey conducted in 2015 and repeated in 2016. We applied these models to the coastal national parks in the isolated and pristine East Gippsland region. Importantly, the prediction models were developed at two different spatial scales that directly corresponded to the scale of the observations. Our findings indicate that coastal habitats, with their vulnerable environments and prevalence of open dune systems, are particularly susceptible to weed infestations. Moreover, adjacent regions also have the potential for colonization if these infestations are not effectively controlled. Climate-related factors play a role in moderating the potential for colonization, which is a significant concern for weed control efforts in the context of global climate change.

Product resilience evaluation: A Bayesian network modeling based method

Future products will have a higher degree of intelligence, and more complex and changing use environments, so resilience has been introduced into the design and operation of products as a concept that helps them cope with high-impact shocks and the damage they cause. With the work on product resilience, there is a need to find methods that can objectively reflect product resilience. Existing methods that can be used to evaluate product resilience are usually based on performance curves, but it is difficult to reflect the multifactorial and stochastic character of product resilience process. Therefore, this paper firstly reviews the resilience related research, and analyzes the factors affecting product resilience and their interrelationships layer by layer to construct the index system of product resilience. Then, a Bayesian network model is established based on the results of the above analysis, and the corresponding calculation method is proposed. Finally, the proposed method is illustrated with a case study of a complex terrain drilling rig and its improvement program. After discussion, the proposed method can be applied to the quantitative evaluation of product resilience, and the possible design direction of resilient products can be suggested using this method.

Can Bayesian Networks Improve Ground-Strike Point Classification?

Studying cloud-to-ground lightning strokes and ground-strike points provides an alternative method of lightning mapping for lightning risk assessment. Various k-means algorithms have been used to verify the ground-strike points from lightning locating systems, producing results with room for improvement. This paper proposes using Bayesian networks (BNs), a model not previously used for this purpose, to classify lightning ground-strike points. A Bayesian network is a probabilistic graphical model that uses Bayes’ theorem to represent the conditional dependencies of variables. The networks created for this research were trained from the data using a score-based structure-learning procedure and the Bayesian information criterion score function. The models were evaluated using confusion matrices and kappa indices and produced accuracy values ranging from 86% to 94% and kappa indices of up to 0.76. While BN models do not outperform k-means algorithms, they offer an alternative by not requiring predetermined distances. However, the easy implementation of the k-means approach means that no significant gain is made by implementing the more complex Bayesian network approach.

Power generation forecasting for solar plants based on Dynamic Bayesian networks by fusing multi-source information

A Dynamic Bayesian network (DBN) model for solar power generation forecasting in photovoltaic (PV) solar plants is proposed in this paper. The key idea is to fuse sensor data, operational indicators, meteorological data, lagged output power information, and model errors for more accurate short-term (e.g., hours) and mid-term (e.g., days to weeks) power generation forecasting. The proposed DBN augments automated data-driven structure learning with expert knowledge encoding using continuous and categorical data given constraints to represent causal relationships within a solar inverter system. In addition, an error compensation mechanism that uses an error node to capture the temporal fluctuation caused by system degradation or failures is proposed to capture temporal fluctuation. The effectiveness of the DBN on solar power generation forecasting was evaluated by rolling window analysis with one-year testing data collected from a local solar plant. The proposed DBN is compared with four state-of-art methods including support-vector regression (SVR), k-nearest neighbors (kNN), artificial neural network (ANN), and long short-term memory (LSTM) models. The results show that the proposed DBN achieves better accuracy in general, and it is not as data-hungry as some neural network-based models. The proposed DBN is also shown to have robust and consistent forecasting power with different forecasting horizons. The accuracy is 92 %–95 % from 1 h to one week ahead forecasting.

Changes in expression of VGF, SPECC1L, HLA-DRA and RANBP3L act with APOE E4 to alter risk for late onset Alzheimer’s disease

While there are currently over 40 replicated genes with mapped risk alleles for Late Onset Alzheimer’s disease (LOAD), the Apolipoprotein E locus E4 haplotype is still the biggest driver of risk, with odds ratios for neuropathologically confirmed E44 carriers exceeding 30 (95% confidence interval 16.59–58.75). We sought to address whether the APOE E4 haplotype modifies expression globally through networks of expression to increase LOAD risk. We have used the Human Brainome data to build expression networks comparing APOE E4 carriers to non-carriers using scalable mixed-datatypes Bayesian network (BN) modeling. We have found that VGF had the greatest explanatory weight. High expression of VGF is a protective signal, even on the background of APOE E4 alleles. LOAD risk signals, considering an APOE background, include high levels of SPECC1L, HLA-DRA and RANBP3L. Our findings nominate several new transcripts, taking a combined approach to network building including known LOAD risk loci.

Statistical machine learning models for prediction of China's maritime emergency patients in dynamic: ARIMA model, SARIMA model, and dynamic Bayesian network model.

Rescuing individuals at sea is a pressing global public health issue, garnering substantial attention from emergency medicine researchers with a focus on improving prevention and control strategies. This study aims to develop a Dynamic Bayesian Networks (DBN) model utilizing maritime emergency incident data and compare its forecasting accuracy to Auto-regressive Integrated Moving Average (ARIMA) and Seasonal Auto-regressive Integrated Moving Average (SARIMA) models. In this research, we analyzed the count of cases managed by five hospitals in Hainan Province from January 2016 to December 2020 in the context of maritime emergency care. We employed diverse approaches to construct and calibrate ARIMA, SARIMA, and DBN models. These models were subsequently utilized to forecast the number of emergency responders from January 2021 to December 2021. The study indicated that the ARIMA, SARIMA, and DBN models effectively modeled and forecasted Maritime Emergency Medical Service (EMS) patient data, accounting for seasonal variations. The predictive accuracy was evaluated using Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and Coefficient of Determination (R 2) as performance metrics. In this study, the ARIMA, SARIMA, and DBN models reported RMSE of 5.75, 4.43, and 5.45; MAE of 4.13, 2.81, and 3.85; and R 2 values of 0.21, 0.54, and 0.44, respectively. MAE and RMSE assess the level of difference between the actual and predicted values. A smaller value indicates a more accurate model prediction. R 2 can compare the performance of models across different aspects, with a range of values from 0 to 1. A value closer to 1 signifies better model quality. As errors increase, R 2 moves further from the maximum value. The SARIMA model outperformed the others, demonstrating the lowest RMSE and MAE, alongside the highest R 2, during both modeling and forecasting. Analysis of predicted values and fitting plots reveals that, in most instances, SARIMA's predictions closely align with the actual number of rescues. Thus, SARIMA is superior in both fitting and forecasting, followed by the DBN model, with ARIMA showing the least accurate predictions. While the DBN model adeptly captures variable correlations, the SARIMA model excels in forecasting maritime emergency cases. By comparing these models, we glean valuable insights into maritime emergency trends, facilitating the development of effective prevention and control strategies.

Risk Analysis of Pirate Attacks on Southeast Asian Ships Based on Bayesian Networks

As a bridge for international trade, maritime transportation security is crucial to the global economy. Southeast Asian waters have become a high-incidence area of global piracy attacks due to geographic location and complex security situations, posing a great threat to the development of the Maritime Silk Road. In this study, the factors affecting the risk of pirate attacks are analyzed in depth by using the Global Ship Piracy Attacks Report from the IMO Global Integrated Shipping Information System (GISIS) database (i.e., 2013–2022) in conjunction with a Bayesian Network (BN) model, and the Expectation Maximization algorithm is used to train the model parameters. The results show that piracy behaviors and the ship’s risk are the key factors affecting the risk of pirate attacks, and suggestions are made to reduce the risk of pirate attacks. This study develops a theoretical basis for preventing and controlling the risk of pirate attacks on ships, which helps maintain the safety of ship operations.