Bayesian Network Inference Research Articles

Sensitivity of Bayesian Networks to Noise in Their Parameters

There is a widely spread belief in the Bayesian network (BN) community that the overall accuracy of results of BN inference is not too sensitive to the precision of their parameters. We present the results of several experiments in which we put this belief to a test in the context of medical diagnostic models. We study the deterioration of accuracy under random symmetric noise but also biased noise that represents overconfidence and underconfidence of human experts.Our results demonstrate consistently, across all models studied, that while noise leads to deterioration of accuracy, small amounts of noise have minimal effect on the diagnostic accuracy of BN models. Overconfidence, common among human experts, appears to be safer than symmetric noise and much safer than underconfidence in terms of the resulting accuracy. Noise in medical laboratory results and disease nodes as well as in nodes forming the Markov blanket of the disease nodes has the largest effect on accuracy. In light of these results, knowledge engineers should moderately worry about the overall quality of the numerical parameters of BNs and direct their effort where it is most needed, as indicated by sensitivity analysis.

Approaches for Behavior Intensity Estimation in Groups of Heterogeneous Individuals: Precision and Applicability for Data with Uncertainty

In socially oriented areas, there arises the problem of assessing the cumulative characteristics of behavior, such as intensity, that are realized in groups of individuals. All individuals vary in their behavior and the available data is limited and may be associated with significant uncertainty: only a few episodes may be known and only a few individualsi the group may be observed. Mathematical models of behavior are used for estimation of key characteristics of the behavior. One of them is based on the gamma–Poisson point process, that reflects the heterogeneity of individuals in a form of a mixing distribution. This general model allows to formulate several methods of frequency estimation: the Cox regression, estimation of the copula parameter, and a posteriori inference in Bayesian belief networks. The aim of the paper is to assess their determine the precision of these methods based on the Kantorovich–Rubinstein distance between estimates and the true distribution of the desired parameter. The analysis of assumptions of those methods allows to formulate rules, that allow to chose the appropriate method in various sutuations of data availability. It has been shown that the copula-based approach provides the most accurate estimates and has the mild assumptions for the number of observed objects, but it cannot take into account external factors that may influence the behavior. Among methods that can take into account process covariants, estimates based on a posteriori inference in hybrid Bayesian belief networks have the highest precision. The paper considers a method for quantification of a hybrid BBNs with the approximation of mixtures of truncated exponents, that is data-demanding at the stage of calculating a priori estimates. However, it is noted that there are other approaches to setting hybrid BSDs in which a priori estimates can be set completely expertly.

Approximate inference on optimized quantum Bayesian networks

In recent years, there has been a significant upsurge in the interest surrounding Quantum machine learning, with researchers actively developing methods to leverage the power of quantum technology for solving highly complex problems across various domains. However, implementing gate-based quantum algorithms on noisy intermediate quantum devices (NISQ) presents notable challenges due to limited quantum resources and inherent noise. In this paper, we propose an innovative approach for representing Bayesian networks on quantum circuits, specifically designed to address these challenges and highlight the potential of combining optimized circuits with quantum hybrid algorithms for Bayesian network inference. Our aim is to minimize the required quantum resource needed to implement a Quantum Bayesian network (QBN) and implement quantum approximate inference algorithm on a quantum computer. Through simulations and experiments on IBM Quantum computers, we show that our circuit representation significantly reduces the resource requirements without decreasing the performance of the model. These findings underscore how our approach can better enable practical applications of QBN on currently available quantum hardware.

A Bayesian Network Inference Approach for Dynamic Risk Assessment Using Multisource-Based Information Fusion in an Interval Type-2 Fuzzy Set Environment

A Bayesian Network Inference Approach for Dynamic Risk Assessment Using Multisource-Based Information Fusion in an Interval Type-2 Fuzzy Set Environment

Superior probabilistic computing using operationally stable probabilistic-bit constructed by manganite nanowire

Abstract Probabilistic computing has emerged as a viable approach to treat optimization problems. To achieve superior computing performance, the key aspect during computation is massive sampling and tuning on the probability states of each probabilistic bit (p-bit), demanding its high stability under extensive operations. Here, we demonstrate a p-bit constructed by manganite nanowire that shows exceptionally high stability. The p-bit contains an electronic domain that fluctuates between metallic (low resistance) and insulating (high resistance) states near its transition temperature. The probability for the two states can be directly controlled by nano-ampere electrical current. Under extensive operations, the standard error of its probability values is less than 1.3%. Simulations show that our operationally stable p-bit plays the key role to achieve correct inference in Bayesian network by strongly suppressing the relative error, displaying the potential for superior computing performance. Our p-bit also serves as high quality random number generator without extra data-processing, beneficial for cryptographic applications.

Network analyses on photographic surveys reveal that invertebrate predators do not structure epibenthos in the deep (~2000m) rocky Powell Basin, Weddell Sea, Antarctica

Predator-prey interactions in marine ecosystems control population sizes, maintain species richness, and provide intermediate disturbance. Such ecosystem structuring interactions may be rare in Antarctic epibenthic communities, which are unique among marine ecosystems worldwide for their dominance of soft bodied fauna (sponges, soft and hard corals, and echinoderms) and a simultaneous paucity of shell crushing predators (sharks, rays and durophagous decapods). In the shallow benthos, instead of durophagy, important Antarctic predators such as starfish, pycnogonids (sea spiders), nemertean worms, and nudibranchs employ grazing, scavenging, or sucking strategies. Far less is known about deep sea (>1000 m) Antarctic benthic communities due to the challenging nature of polar data collection, so that photographic surveys provide one of the only means of making in situ observations of these deep sea communities. We used seabed photographs of the deep (~2000m) slope of the Powell Basin, northwest Weddell Sea, taken by the Ocean Floor Observation and Bathymetry System on board the RV Polarstern (PS118, April 2019) to investigate the epibenthic community composition, and Bayesian Network Inference (BNI) to determine the ecological network, namely the ecological associations, including potential invertebrate predator-prey relationships between taxa. Photographs show that the rocky substrates of the basin slope support between 10-22 morphotaxa per photo, and highly abundant communities (density between 106 to 553 individuals/m2). BNI results reveal a network of associations between the sessile and mobile suspension and filter feeding organisms and their physical environment. However, associations between invertebrate predators like starfish, and other organisms, were not detected in the network. This lack of inclusion within the network suggests that, despite the presence of these normally important mobile predators, invertebrate predator-prey interactions on the rocky Powell Basin slope do not have the same ecosystem-regulating impact that they do on shallow Antarctic epibenthic communities.

A modeling framework for detecting and leveraging node-level information in Bayesian network inference.

Bayesian graphical models are powerful tools to infer complex relationships in high dimension, yet are often fraught with computational and statistical challenges. If exploited in a principled way, the increasing information collected alongside the data of primary interest constitutes an opportunity to mitigate these difficulties by guiding the detection of dependence structures. For instance, gene network inference may be informed by the use of publicly available summary statistics on the regulation of genes by genetic variants. Here we present a novel Gaussian graphical modeling framework to identify and leverage information on the centrality of nodes in conditional independence graphs. Specifically, we consider a fully joint hierarchical model to simultaneously infer (i) sparse precision matrices and (ii) the relevance of node-level information for uncovering the sought-after network structure. We encode such information as candidate auxiliary variables using a spike-and-slab submodel on the propensity of nodes to be hubs, which allows hypothesis-free selection and interpretation of a sparse subset of relevant variables. As efficient exploration of large posterior spaces is needed for real-world applications, we develop a variational expectation conditional maximization algorithm that scales inference to hundreds of samples, nodes and auxiliary variables. We illustrate and exploit the advantages of our approach in simulations and in a gene network study which identifies hub genes involved in biological pathways relevant to immune-mediated diseases.

A New Fuzzy Bayesian Inference Approach for Risk Assessments

Bayesian network (BN) inference is an important statistical tool with additive symmetry. However, BN inference cannot deal with the uncertain, fuzzy, random, and conflicting information from experts’ knowledge in the process of conducting a risk assessment. To tackle this issue, a new fuzzy BN inference approach for risk assessments was proposed based on cloud model (CM), interval type-2 fuzzy set (IT2 FS), interval type-2 fuzzy logic system (IT2 FLS), modified Dempster–Shafer (D-S) evidence theory (ET), and Latin hypercube sampling (LHS) methods along the following lines. Firstly, CM was integrated into IT2 FS, and CM-based IT2 FS (CM-IT2 FS) was defined in the IT2 FLS. Secondly, modified D-S ET was utilized to determine the CM-IT2 FS-based a priori probabilities, and the CM-IT2 FS-based BN model was established. Thirdly, the CM-IT2 FS-based a priori probabilities were reduced to the CM-IT1 FS-based ones using a type reducer in the IT2 FLS, LHS was applied to propose a new fuzzy BN inference algorithm, and then, the new algorithm was used in a typical case to perform the fuzzy BN positive inference for risk prediction and the fuzzy BN reverse inference for risk sensitivity analysis. Finally, the BN inference results were analyzed using the proposed algorithm and the two common BN inference algorithms, and the effectiveness of the proposed approach was validated. It can be concluded that the proposed approach was both accurate and promising.

Identification of User Requirements and their Influencing Factors Based on Online Reviews and Operational Data

It is crucial for enterprises to clearly identify user needs during the process of formulating product design improvement plans. Therefore, it is essential to comprehensively and accurately identify user needs, explore the reasons behind the emergence of these needs, and incorporate user opinions into the process of product design improvement. A method is proposed to comprehensively and accurately capture user requirements and address the challenge of identifying the underlying causes of user requirements. This method utilizes online comments and operational data to identify user requirements and their influencing factors. First, text sentiment analysis techniques are employed to quantify the importance and performance values of product feature topics. Second, we construct a quadrant model to identify product features requiring improvement, and the original negative comments related to these features are traced. However, the quadrant model alone is insufficient to reflect specific product issues that users are concerned about. Therefore, a functional structure model based on product issues is designed to filter and identify factors that influence user requirements using operational data. Finally, a Bayesian network inference approach is utilized to identify the key influencing factors on user requirements, enabling analysis of the causes behind user requirements and the proposal of product design improvement strategies. The feasibility and effectiveness of the proposed method are validated through experiments conducted on heavy-duty truck data. By analyzing the original negative comments related to the power characteristics, specific user demands regarding the insufficient power of the product were identified, such as “obviously insufficient power when climbing slopes” and other issues. Based on the vehicle power system functional structure model, combined with expert knowledge and operational data, factors related to the state of parts and user behavior that may affect “insufficient vehicle power” were identified. Based on the analysis results, suggestions were made to improve the engine intake air temperature control strategy and to enhance vehicle performance by promoting correct user behavior through informational campaigns.

Assessing vessel transportation delays affected by tropical cyclones using AIS data and a bayesian network: A case study of veronica in northwestern Australia

Tropical cyclones frequently disrupt maritime transportation systems, impacting the normal operation of vessels and causing transportation delays. Analyzing the occurrence and degree of vessel transportation delays under disaster conditions is one of the critical aspects of maritime transportation management. This article, based on Automatic Identification System (AIS) data, examined the real behavioral trajectories of vessels and delay characteristics during tropical cyclones. Taking the example of Tropical Cyclone Veronica, which occurred in the waters off northwestern Australia in 2019, we conducted a comprehensive analysis. This involved numerical simulations of the entire disaster process and the cleaning and matching of trajectory data for all affected vessels in the area. We delved deeply into the relationship between vessel behavior characteristics and tropical cyclones, identified factors contributing to delays, and, based on our findings, constructed a Bayesian network inference model for vessel transportation delays under disaster conditions. This model uses information such as tropical cyclone intensity, vessel basic attributes, behavior choices, and port disaster avoidance measures as its primary nodes. The research results indicate that vessel rerouting, vessel loss of control, and waiting for port entry are the three major sources of vessel delays during disasters. Key influencing factors contributing to these consequences include port closure measures, the duration of vessels being adversely affected by strong winds, vessel tendencies toward loss of control, and risk preferences. To control the extent of vessel delays more effectively, it is crucial to prioritize these sensitive factors and strengthen monitoring and management efforts. This model is driven by real data, providing an objective reflection of real-world scenarios, and its effectiveness has been validated through sensitivity analysis. The research perspective and algorithmic framework presented in this paper offer a novel paradigm and fresh perspective for the study of maritime transportation disasters. The research findings have significant implications for bolstering the resilience of maritime transportation networks, enhancing our comprehension and anticipation of delays, and ensuring the continuity, security, and efficiency of the global supply chain for goods.

Deciphering the role of zinc homeostasis in the tumor microenvironment and prognosis of prostate cancer

BackgroundDysregulation of zinc homeostasis is widely recognized as a hallmark feature of prostate cancer (PCa) based on the compelling clinical and experimental evidence. Nevertheless, the implications of zinc dyshomeostasis in PCa remains largely unexplored.MethodsIn this research, the zinc homeostasis pattern subtype (ZHPS) was constructed according to the profile of zinc homeostasis genes. The identified subtypes were assessed for their immune functions, mutational landscapes, biological peculiarities and drug susceptibility. Subsequently, we developed the optimal signature, known as the zinc homeostasis-related risk score (ZHRRS), using the approach won out in multifariously machine learning algorithms. Eventually, clinical specimens, Bayesian network inference and single-cell sequencing were used to excavate the underlying mechanisms of MT1A in PCa.ResultsThe zinc dyshomeostasis subgroup, ZHPS2, possessed a markedly worse prognosis than ZHPS1. Moreover, ZHPS2 demonstrated a more conspicuous genomic instability and better therapeutic responses to docetaxel and olaparib than ZHPS1. Compared with traditional clinicopathological characteristics and 35 published signatures, ZHRRS displayed a significantly improved accuracy in prognosis prediction. The diagnostic value of MT1A in PCa was substantiated through analysis of clinical samples. Additionally, we inferred and established the regulatory network of MT1A to elucidate its biological mechanisms.ConclusionsThe ZHPS classifier and ZHRRS model hold great potential as clinical applications for improving outcomes of PCa patients.

Variational Bayesian Neural Networks via Resolution of Singularities

In this work, we advocate for the importance of singular learning theory (SLT) as it pertains to the theory and practice of variational inference in Bayesian neural networks (BNNs). To begin, we lay to rest some of the confusion surrounding discrepancies between downstream predictive performance measured via the test log predictive density and the variational objective. Next, we use the SLT-corrected asymptotic form for singular posterior distributions to inform the design of the variational family itself. Specifically, we build upon the idealized variational family introduced in Bhattacharya, Pati, and Plummer which is theoretically appealing but practically intractable. Our proposal takes shape as a normalizing flow where the base distribution is a carefully-initialized generalized gamma. We conduct experiments comparing this to the canonical Gaussian base distribution and show improvements in terms of variational free energy and variational generalization error. Supplemental appendices and code for the article are available online.

The privacy preserving auction mechanisms in IoT-based trading market: A survey

Nowadays, the Internet of Things (IoT) technology has greatly promoted the transaction behavior. It has led to a shift from the traditional face-to-face, bartering business to today’s online virtual goods business, such as energy trading, spectrum, and so on. However, there exist two critical challenges in IoT-based trading market, which include “efficiency” and “safety”. “Efficiency” indicates a fair-trading decision-making rule, and “safety” indicates that the bidders’ privacy must be protected. To address the above challenges, numerous researchers are working on privacy-preserving auction mechanisms for IoT-based transactional markets, drawing insights from electronic auction theory and information security theory. However, a predominant focus on specific scenarios characterizes most existing works, lacking a comprehensive synthesis of the collective contributions to date. Therefore, this paper systematically elucidates privacy-preserving auction mechanisms for IoT-based transactional markets. Firstly, we expound the foundational knowledge of IoT-based auction markets, delving into auction theory and privacy theory through concise categorization. Secondly, we address both theoretical considerations (from the integration perspective of various privacy protection methods and auction mechanisms) and practical aspects (evaluating various real-world application scenarios). Each aspect undergoes meticulous scrutiny, providing practical assessments and prospects. Lastly, we propose future research directions. Key challenges include the absence of inference attack models for differential privacy, a dearth of algorithmic designs with privacy-preserving capabilities at the auction mechanism level, and the intricate balance between privacy and efficiency. Proposed solutions for future research directions include leveraging Bayesian inference and neural networks for effective attacks, designing autonomous privacy protection mechanisms, and addressing the privacy-efficiency trade-off through the application of Markov decision processes.

A Multisource Uncertainty Fusion Reliability Evaluation Method for the Control Rod Drive Mechanism of Nuclear Power Plants

The reliability of a pressurized water reactor power plant’s control rod drive mechanism (CRDM) is affected by many factors, such as operation states, unit performance, and dynamic environments. Multiple sources of uncertainties, including random, interval, and fuzzy, exist when analyzing the reliability of CRDMs. Modeling reliability without considering the fusion of multisource uncertainties may result in model distortion and may not provide realistic assessment results. This paper proposes a multisource uncertainty fusion method powered by the margin-based variable transformation and the Bayesian network propagation. The interval and fuzzy variables are transformed into random variables to obtain the corresponding generalized probability density function of the CRDM’s feature parameters. After that, the CRDM’s reliability can be evaluated via probability quantization of the Bayesian network inference result through sampling algorithms. A magnetic-jack type CRDM case is presented to verify the proposed method, and the results show that this method can fuse three types of uncertainty variable in a unified way and effectively obtain reliability evaluation results.

The landscape of immune checkpoint-related long non-coding RNAs core regulatory circuitry reveals implications for immunoregulation and immunotherapy responses

Long non-coding RNAs (lncRNAs) could modulate expression of immune checkpoints (ICPs) by cooperating with immunity genes in tumor immunization. However, precise functions in immunity and potential for predicting ICP inhibitors (ICI) response have been described for only a few lncRNAs. Here we present an integrated framework that leverages network-based analyses and Bayesian network inference to identify the regulated relationships including lncRNA, ICP and immunity genes as ICP-related LncRNAs mediated Core Regulatory Circuitry Triplets (ICP-LncCRCTs) that can make robust predictions. Hub ICP-related lncRNAs such as MIR155HG and ADAMTS9-AS2 were highlighted to play central roles in immune regulation. Specific ICP-related lncRNAs could distinguish cancer subtypes. Moreover, the ICP-related lncRNAs are likely to significantly correlated with immune cell infiltration, MHC, CYT. Some ICP-LncCRCTs such as CXCL10-MIR155HG-ICOS could better predict one-, three- and five-year prognosis compared to single molecule in melanoma. We also validated that some ICP-LncCRCTs could effectively predict ICI-response using three kinds of machine learning algorithms follow five independent datasets. Specially, combining ICP-LncCRCTs with the tumor mutation burden (TMB) improves the prediction of ICI-treated melanoma patients. Altogether, this study will improve our grasp of lncRNA functions and accelerating discovery of lncRNA-based biomarkers in ICI treatment.

Causal association study of the dynamic development of the metabolic syndrome based on longitudinal data

The dynamic progression of metabolic syndrome (MetS) includes developmental deterioration and reverse recovery; however, the key factors in this bidirectional progression have not been identified. Our study aimed to use the data obtained from the China Health and Retirement Longitudinal Study (CHARLS) and construct a Bayesian network to explore the causal relationship between influential factor and the development and recovery of MetS. Followed up at 4 years, forward progression of MetS occurred in 1543 and reverse recovery of MetS occurred in 1319 of 5581 subjects. Bayesian Networks showed that hyperuricemia and body mass index (BMI) levels directly influenced progression of MetS, and gender, exercise and age play an indirect role through hyperuricemia and BMI levels; high hemoglobin A1c (HbA1c) and BMI levels directly influenced recovery of MetS, and gender and exercise play an indirect role through BMI levels. Bayesian Network inference found that the rate of progression of MetS in subjects with hyperuricemia increases from 36 to 60%, the rate of progression of MetS in subjects with overweight or obese increases from 36 to 41% and the rate of reverse recovery rate of MetS in subjects with high HbA1c decreased from 33 to 20%. Therefore, attention to individuals at high risk of hyperuricemia, high HbA1c levels, and overweight/obesity should be enhanced, with early detection and following healthy behavioral interventions to prevent, control and delay the progression of MetS and its components.

Scalable and rapid building damage detection after hurricane Ian using causal Bayesian networks and InSAR imagery

Timely and accurate assessment of hurricane-induced building damage is crucial for effective post-hurricane response and recovery efforts. Recently, remote sensing technologies provide large-scale optical or Interferometric Synthetic Aperture Radar (InSAR) imagery data immediately after a disastrous event, which can be readily used to conduct rapid building damage assessment. Compared to optical satellite imageries, the Synthetic Aperture Radar can penetrate cloud cover and provide more complete spatial coverage of damaged zone in various weather conditions. However, these InSAR imageries often contain highly noisy and mixed signals induced by co-occurring or co-located building damage, flood, flood/wind-induced vegetation changes, as well as anthropogenic activities, making it challenging to extract accurate building damage information. In this paper, we introduced a causality-informed Bayesian network inference approach for rapid post-hurricane building damage detection from InSAR imagery. This approach encoded complex causal dependencies among wind, flood, building damage, and InSAR imagery using a holistic causal Bayesian network. Based on the causal Bayesian network, we further jointly inferred the large-scale unobserved building damage by fusing the information from InSAR imagery with prior physical models of flood and wind, without the need for ground truth labels. Furthermore, we validated our estimation results in a real-world devastating hurricane—the 2022 Hurricane Ian. We gathered and annotated building damage ground truth data in Lee County, Florida, and compared the introduced method’s estimation results with the ground truth and also benchmarked it against state-of-the-art models to assess the effectiveness of our proposed method. The results show that our method advances building damage assessment after hurricanes by accurately reflecting the complex dynamics between wind and flood impacts. Notably, it achieves this without the need for a ground truth label, which is a substantial step forward from traditional methods. Our model registers a 22.6% increase in the Area Under the Curve (AUC) and a 46.29% enhancement in the True Positive Rate (TPR). Moreover, it expedites the detection of building damage, cutting down processing times by up to 83.8%. These improvements mark a considerable leap in efficiency, demonstrating our method’s ability to streamline the assessment process markedly over conventional methods.

Gravity Investigation to Characterize Enceladus's Ocean and Interior

A key objective for the future exploration of the icy moon Enceladus is the characterization of the habitable conditions in its internal ocean. Radio science instrumentation on board a spacecraft in a low-altitude orbit about Enceladus would enable gravity measurements that are fundamental to providing constraints on its internal structure. We present here the concept of operations and expected results of the gravity investigation for a New Frontiers–class mission. Numerical simulations are carried out to determine the gravity field in spherical harmonics to degree and order 30 and the Love number k 2 and its phase. By combining Enceladus’s shape measured by Cassini and the geophysical constraints obtained through the processing of the simulated radio science data, a Bayesian inference network is used for the interior model inversion. Our results indicate that the gravity investigation would enable tight constraints on core radius and density, ocean depth and density, and ice shell rigidity. By assuming a high core rigidity and a preliminary modeling of dissipation in the ice shell, our interior model inversion also yields information on the ice shell viscosity. Further data on the hydrosphere properties might be gathered through optical navigation data by accurately measuring Enceladus’s orientation model.

Enhancing Residential Electricity Safety and Management: A Novel Non-Intrusive Load Monitoring-Based Methodology for Accurate Appliance Operational State Identification

Non-intrusive load monitoring (NILM) technology, crucial for intelligent electricity management, has gained considerable attention in residential electricity usage studies. NILM enables monitoring of total electrical current and voltage in homes, offering insights vital for enhancing safety and preventing domestic electrical accidents. Despite its importance, accurately discerning the operational status of appliances using non-intrusive methods remains a challenging area within this field. This paper presents a novel methodology that integrates an advanced clustering algorithm with a Bayesian network for the identification of appliance operational states. The approach involves capturing the electrical current signals during appliance operation via NILM, followed by their decomposition into odd harmonics. An enhanced clustering algorithm is then employed to ascertain the central coordinates of the signal clusters. Building upon this, a three-layer Bayesian network inference model, incorporating leak nodes, is developed. Within this model, harmonic signals are used as conditions for node activation. The operational states of the appliances are subsequently determined through probabilistic reasoning. The proposed method’s effectiveness is validated through a series of simulation experiments conducted in a laboratory environment. The results of these experiments (low mode 89.1%, medium mode 94.4%, high mode 92.0%, and 98.4% for combination) provide strong evidence of the method’s accuracy in inferring the operational status of household electrical appliances based on NILM technology.

Carbon trading price forecasting in digitalization social change era using an explainable machine learning approach: The case of China as emerging country evidence

Carbon trading prices are considered important reference indicators for policy formulation and enterprise decision-making, and play an important role in low-carbon development. However, predictive approaches that balance the accuracy and interpretability of carbon trading prices remain limited. This study proposes a hybrid approach that integrates group variable selection regularization and uncertainty inference in Bayesian neural networks (BNN) with inherent interpretability to predict carbon trading prices. We also identify effective predictive indicators in an emerging market context. Based on the carbon data of emerging markets, a dataset of price information was constructed for model training. The results indicate that under the same regularized dataset, all BNN results surpass other artificial neural networks (ANNs). The group smoothly clipped absolute deviation-BNN performs best in all models and has interpretability. The model effectively identified new influencing factors for predicting carbon trading prices in emerging countries, including coal prices and blockchain-related information. The proposed new prediction approach provides a new basis for the prediction and evaluation of carbon trading prices and provides the necessary references for policy formulation in the digitalization era of social change.