Bayesian-based Approach Research Articles

Bayesian empirical analysis of the proportional hazards model for right-censored failure time data

Statistical analysis of right-censored failure time data has been extensively discussed in the literature as such data often occur in many fields. In this article, we propose a new Bayesian-based empirical likelihood approach for the problem under the proportional hazards model. The new method allows one to take into account the existing prior information among other advantages, and for the implementation of the method, a Metropolis–Hasting algorithm is developed. To assess the performance of the proposed approach, a simulation study is conducted and suggests that it works well. The method is applied to a set of kidney dialysis data.

Autonomous Vessels in the Yangtze River: A Study on the Maritime Accidents Using Data-Driven Bayesian Networks

The prototypes of autonomous vessels are expected to come into service within the coming years, but safety concerns remain due to complex traffic and natural conditions (e.g., Yangtze River). However, the response of autonomous vessels to potential accidents is still uncertain. The accident prevention for autonomous vessels is unconvincing due to the lack of objective studies on the causation analysis for maritime accidents. This paper constitutes an attempt to cover the aforementioned gap by studying the potential causations for maritime accidents in the Yangtze River by using a Bayesian-based network training approach. More than two hundred accidents reported between 2013 and 2019 in the Yangtze River are collected. As a result, a Bayesian network (BN) is successfully established to describe the causations among different risk influencing factors. By analysing the BN, this study reveals that the occurrence of maritime accidents (e.g., collision, grounding) can be expected to reduce with the development of autonomous vessels as the crews are removed. However, the extent of the consequences from some accidents (e.g., fire, critical weathers) could be more serious than conventional ones. Therefore, more attention and thoughts are needed to ensure the safe navigation of autonomous vessels in the Yangtze River.

A Multi-Objective Bayesian Approach with Dynamic Optimization (MOBADO). A Hybrid of Decision Theory and Machine Learning Applied to Customs Fraud Control in Spain

This paper studies the economically significant problem of the optimization of customs fraud control, which is a critical issue for many countries. The European Union (EU) alone handles 4693 tons of goods every minute (2018 figures). Even though 70% of goods are imported at zero tariff, the EU raised EUR 25.4 billions in 2018, and customs-related income transferred by member states to the EU accounts for nearly 13% of its overall budget. In this field, (a) the conflicting objectives are qualitative and cannot be reduced to a common measure (security and terrorism, health, drug market access control, taxes, etc.); (b) each submitted item has dozens of characteristics; (c) there are constraints; and (d) risk analysis systems have to make decisions in real time. Although the World Customs Organization has promoted the use of artificial intelligence to increase the precision of controls, the problem is very complex due to the data characteristics and interpretability, which is a requirement established by customs officers. In this paper, we propose a new Bayesian-based hybrid approach combining machine learning and multi-objective linear programming (MOLP), called multi-objective Bayesian with dynamic optimization (MOBADO). We demonstrate that it is possible to more than double (with a 237% increase) the precision of current inspection systems, freeing up almost 50% of human resources, and outperform past results with respect to each of the above objectives. MOBADO is an optimization technique that could be combined with any artificial intelligence approach capable of optimizing the quality of multi-objective risk analysis in real time.

A Bayesian-based approach for extracting the pion charge radius from electron-electron scattering data

In this study, we utilize a potentially versatile Bayesian parameter approach to compute the value of the pion charge radius and quantify its uncertainty from several experimental datasets for the pion vector form factor. We employ dispersion relations to model the pion vector form factor to extract the radius. Nested model selection is used to determine the order of polynomial appearing in the form factor formulation that can be supported by the data, adapting the computation of Bayes evidence and Bayesian effective complexity based on Occam's razor. Our findings indicate that five out of six used datasets favor the nine-parameter model for radius extraction, and accordingly, we average the radii from the datasets. Despite some inconsistencies with the most updated radius values, our approach may serve as a more intuitive method of addressing parameter estimations in dispersion theory.

Variational Bayesian probabilistic modeling framework for data-driven distributed process monitoring

Data-driven process monitoring has gained increasing attention because of the increasing demand in process safety and the rapid advancement of data gathering techniques. When monitoring a plant-wide multiunit process, establishing a monitor for each unit individually ignores the correlations among units, whereas establishing a global monitor for the entire process ignores the local process behavior. A variational Bayesian-based probabilistic modeling approach is proposed for efficient distributed process monitoring. A novel probabilistic latent variable model is developed to characterize the variable relationship in each local unit and among units. First, variational Bayesian-based latent variable extraction is performed in each local unit, through which variable relationship within a local unit is characterized. Second, variational Bayesian-based regression model is established between the latent variables and neighboring variables, through which the variable relationship among units is characterized. Then, modeling residuals and monitoring statistics are generated, through which the process status and the type of a detected fault are identified. The effectiveness of the proposed probabilistic modeling and monitoring method is verified by three case studies, including a numerical example, the Tennessee Eastman benchmark process, and a laboratory distillation process.

Cheminformatics-Based Identification of Potential Novel Anti-SARS-CoV-2 Natural Compounds of African Origin.

The coronavirus disease 2019 (COVID-19) pandemic caused by the severe acute respiratory syndrome virus 2 (SARS-CoV-2) has impacted negatively on public health and socioeconomic status, globally. Although, there are currently no specific drugs approved, several existing drugs are being repurposed, but their successful outcomes are not guaranteed. Therefore, the search for novel therapeutics remains a priority. We screened for inhibitors of the SARS-CoV-2 main protease and the receptor-binding domain of the spike protein from an integrated library of African natural products, compounds generated from machine learning studies and antiviral drugs using AutoDock Vina. The binding mechanisms between the compounds and the proteins were characterized using LigPlot+ and molecular dynamics simulations techniques. The biological activities of the hit compounds were also predicted using a Bayesian-based approach. Six potential bioactive molecules NANPDB2245, NANPDB2403, fusidic acid, ZINC000095486008, ZINC0000556656943 and ZINC001645993538 were identified, all of which had plausible binding mechanisms with both viral receptors. Molecular dynamics simulations, including molecular mechanics Poisson-Boltzmann surface area (MM/PBSA) computations revealed stable protein-ligand complexes with all the compounds having acceptable free binding energies <−15 kJ/mol with each receptor. NANPDB2245, NANPDB2403 and ZINC000095486008 were predicted as antivirals; ZINC000095486008 as a membrane permeability inhibitor; NANPDB2403 as a cell adhesion inhibitor and RNA-directed RNA polymerase inhibitor; and NANPDB2245 as a membrane integrity antagonist. Therefore, they have the potential to inhibit viral entry and replication. These drug-like molecules were predicted to possess attractive pharmacological profiles with negligible toxicity. Novel critical residues identified for both targets could aid in a better understanding of the binding mechanisms and design of fragment-based de novo inhibitors. The compounds are proposed as worthy of further in vitro assaying and as scaffolds for the development of novel SARS-CoV-2 therapeutic molecules.

CRAFT for NMR lipidomics: Targeting lipid metabolism in leucine-supplemented tumor-bearing mice.

Lipid profiling by 1 H-NMR has gained increasing utility in many fields because of its intrinsically quantitative, nondestructive nature and the ability to differentiate small molecules based on their spectral location. Most nuclear magnetic resonance (NMR) techniques for metabolite quantification use frequency domain analysis that involves many user-dependent steps such as phase and baseline correction and quantification by either manual integration or peak fitting. Recently, Bayesian analysis of time-domain NMR data has been shown to reduce operator bias and increase automation in NMR spectroscopy. In this study, we demonstrate the use of CRAFT (complete reduction to amplitude-frequency table), a Bayesian-based approach to automate processing in NMR-based lipidomics using lipid standards and tissue samples of healthy and tumor-bearing mice supplemented with leucine. Complex mixtures of lipid standards were prepared and examined using CRAFT to validate it against conventional Fourier transform (FT)-NMR and derive a fingerprint to be used for analyzing lipid profiles of serum and liver samples. CRAFT and FT-NMR were comparable in accuracy, with CRAFT achieving higher correlation in quantifying several lipid species. Analysis of the serum lipidome of tumor-bearing mice revealed hyperlipidemia and no signs of hepatic triglyceride accumulation compared with that of the healthy group demonstrating that the tumor-bearing mice were in a state of precachexia. Leucine-supplementation was associated with minimal changes in the lipid profile in both tissues. In conclusion, our study demonstrates that the CRAFT method can accurately identify and quantify lipids in complex lipid mixtures and murine tissue samples and, hence, will increase automation and reproducibility in NMR-based lipidomics.

Spatiotemporally Varying Coefficients (STVC) model: a Bayesian local regression to detect spatial and temporal nonstationarity in variables relationships

ABSTRACT Local regression has an advantage over global regression by allowing coefficients that qualify variables relationships being heterogeneous, where such varying regression relationships are nonstationarity. Spatiotemporally Varying Coefficients (STVC) model is the first Bayesian-based local spatiotemporal regression approach, intending to simultaneously detect spatial and temporal nonstationarity for heterogeneous response-covariate variables relationships, through separately estimating posterior local-scale coefficients over different space areas and time frames. In this paper, we first presented a general Bayesian STVC modelling paradigm as a specification guide to show its commonality in broader geospatial research. Then, we employed it to solve a real-world issue concerning spatiotemporal healthcare-socioeconomic relations, for which we derived data of county-level hospital beds number per capita, as well as data of related socioeconomic factors in northeast China during 2002–2011. Results showed that the STVC model surpassed all the other comparative regressions, in terms of both Bayesian model fitness and predictive ability. Globally, resident savings, financial institutions loans, GDP, and primary industry were identified as key socioeconomic conditions affecting healthcare resources in Northeast China. Temporally, with Time-Coefficients (TC) plots, we found that after 2011, GDP and primary industry would further help improve the overall healthcare level of northeast China. Spatially, with Space-Coefficients (SC) maps, we could directly identify the relative contribution of four socioeconomic covariates’ impacts on healthcare within each administrative county. Bayesian STVC model is an essential development and extension of the local regression family for exploring the spatiotemporal heterogeneous variables relationships, especially under Bayesian statistics, as well as GIScience and spatial statistics.

Robust delay‐dependent LPV synthesis for blood pressure control with real‐time Bayesian parameter estimation

Mean arterial blood pressure (MAP) dynamics estimation and its automated regulation could benefit the clinical and emergency resuscitation of critical patients. In order to address the variability and complexity of the MAP response of a patient to vasoactive drug infusion, a parameter-varying model with a varying time delay is considered to describe the MAP dynamics in response to drugs. The estimation of the varying parameters and the delay is performed via a Bayesian-based multiple-model square root cubature Kalman filtering approach. The estimation results validate the effectiveness of the proposed random-walk dynamics identification method using collected animal experiment data. Following the estimation algorithm, an automated drug delivery scheme to regulate the MAP response of the patient is carried out via time-delay linear parameter-varying (LPV) control techniques. In this regard, an LPV gain-scheduled output-feedback controller is designed to meet the MAP response requirements of tracking a desired reference MAP target and guarantee robustness against norm-bounded uncertainties and disturbances. In this context, parameter-dependent Lyapunov-Krasovskii functionals are used to derive sufficient conditions for the robust stabilization of a general LPV system with an arbitrarily varying time delay and the results are provided in a convex linear matrix inequality (LMI) constraint framework. Finally, to evaluate the performance of the proposed MAP regulation approach, closed-loop simulations are conducted and the results confirm the effectiveness of the proposed control method against various simulated clinical scenarios.

A Bayesian-based approach to improving acoustic Born waveform inversion of seismic data for viscoelastic media

In seismic waveform inversion, the reconstruction of the subsurface properties is usually carried out using approximative wave propagation models to ensure computational efficiency. The viscoelastic nature of the subsurface is often unaccounted for, and two popular approximations—the acoustic and linearized Born inversion—are widely used. This leads to reconstruction errors since the approximations ignore realistic (physical) aspects of seismic wave propagation in the heterogeneous Earth. In this study, we show that the Bayesian approximation error approach can be used to partially recover from errors, addressing elastic and viscous effects in acoustic Born inversion for viscoelastic media. The results of numerical examples indicate that neglecting the modeling errors induced by the approximations results in very poor recovery of the subsurface velocity fields.

Assessment of parameter uncertainty for non-point source pollution mechanism modeling: A Bayesian-based approach

Uncertainty assessment of parameters associated with non-point source pollution mechanism modeling are crucial for improving the effectiveness of pollution controlling. In this study, an approach based on Bayesian inference and integrated Markov chain Monte Carlo and multilevel factorial analysis has been developed, and it can not only apply straightforward Bayesian inference to assess parameter uncertainties, but also quantitatively investigate the main and interactive effects of multiple parameters on the model response variables by measuring the specific variations of model outputs. Its applicability and advantages are presented through the application of the Soil and Water Assessment Tool to Shitoukoumen Reservoir Catchment in northeast China. This study investigated the uncertainties of a set of sensitive parameters and their multilevel effects on model response variables, including average annual runoff (AAR), average annual sediment (AAS) and average annual total nitrogen (AAN). Results revealed that (i) soil conservation service runoff curve number for moisture condition II (CN2) had a positive effect on all response variables; (ii) available water capacity of the soil layer (SOL_AWC) had a negative effect on all response variables; (iii) the universal soil loss equation support practice (USLE_P) had a positive effect on AAS and AAN, and little effect on AAR; while the nitrate percolation coefficient (NPERCO) had a positive effect on AAN, and little effect on AAS and AAR; and (iv) the interactions amongst parameters had obvious interdependent effects on the model response variables, for example, the interaction between CN2 and SOL_AWC had a major impact on AAR. The above findings can improve the simulating and predicting capabilities of non-point source pollution mechanism model. Overall, this study highlights that the proposed approach represents a promising solution for uncertainty assessment of model parameters in non-point source pollution mechanism modeling.

Improve the Classification Efficiency of High-Frequency Phase-Tagged SSVEP by a Recursive Bayesian-Based Approach.

Among the Steady-state visual evoked potential (SSVEP)-based brain-computer interfaces (BCIs), the phase-tagged SSVEP (p-SSVEP) has been proved a reliable paradigm to extend the number of available targets, especially for high-frequency SSVEP-based BCIs. However, the recognition efficiency of the high-frequency p-SSVEP still remains relatively low. A longer data segment may achieve a higher classification accuracy, but the time consumption of computation leads to the decrease of information transfer rate. This paper presents a recursive Bayesian-based approach to improve the high-frequency p-SSVEP classification efficiency. In each signal processing period, the classification result is generated by the current scores, the condition probability and a recursive prior probability (dynamic prior probability). The experiment displays the SSVEP stimuli with 20 Hz and 30 Hz respectively, and each frequency contains six phases. This paper compared three classification approaches and the recursive Bayesian-based approach could obtain the highest classification accuracy and practical bit rate under the same data length. The mean accuracy and practical bit rate were 89.7% and 37.8 bits/min with 20Hz, and 89.0% and 36.5 bits/min with 30Hz, respectively Furthermore, the recursive Bayesian-based approach could reduce the individual differences among different subjects. Therefore, the recursive Bayesian-based approach can lead to high classification efficiency in high-frequency p-SSVEP.

Bayesian-Based Approach for Hydraulic Flow Unit Identification and Permeability Prediction: A Field Case Application in a Tight Carbonate Reservoir

Summary Flow zonation and permeability estimation is a common task in reservoir characterization. Typically, integration of openhole log data with a conventional and special core analysis solves this problem. We present a Bayesian-based method for identifying hydraulic flow units in uncored wells using the theory of hydraulic flow units (HFUs) and subsequently compute permeability using wireline log data. We use a nonlinear optimization scheme on the basis of the probability plot to determine pertinent statistical parameters of each flow unit. Next, we couple these results with the F-test and the Akaike's criteria with the purpose of establishing the optimal number of HFUs present in the core data set. Then, we allocate the core data into their respective HFUs using the Bayes’ theorem as clustering rule. Finally, we apply an inversion algorithm on the basis of Bayesian inference to predict permeability using only wireline data. We illustrate the application of the procedure with a carbonate reservoir having extensive conventional core data. The results show that the Bayesian-based clustering and inversion technique delivers permeability estimates that agree with the core data and with the results obtained from a pressure transient analysis.

Adaptive Threshold for Zero-Velocity Detector in ZUPT-Aided Pedestrian Inertial Navigation

We present a study on the adaptive threshold of the zero-velocity detector, which enables the detector to adjust to gait patterns with different speeds, from as low as walking with 80 steps per minute to as high as running with 160 steps per minute, without any tuning of design parameters during the navigation. This approach enables the zero-velocity update (ZUPT)-aided navigation algorithm to work properly with time-varying speed in a single navigation process. A Bayesian-based approach was applied to determine the adaptive threshold in the likelihood ratio test with a uniform prior information and time-varying cost function. The position error in a velocity-changing navigation scenario was demonstrated to be reduced by 12 times after applying the adaptive threshold instead of a fixed threshold.

A Bayesian-based integrated approach for identifying groundwater contamination sources

The identification of groundwater contamination sources (GCS) can provide comprehensive knowledge for the remediation and risk assessment of contaminated sites. This study develops an innovative framework that can be effectively and efficiently used for the optimal sampling well location design and GCS parameters identification. The framework is based on Bayesian theory and integrates the relative entropy (RE), a 0-1 integer programming optimization model (0-1 IPOM), Markov Chain Monte Carlo (MCMC), and a Kriging surrogate model (KSM). The expected RE is used to quantify information about unknown parameters from concentration measurements based on a Bayesian design. The optimal sampling well locations are determined through the comprehensive application of 0-1 IPOM and the expected RE. After determining the optimal sampling well locations, a Bayesian approach based on MCMC is employed to identify the GCS parameters. However, such problems are time-consuming because both determination and identification require the contaminant transport model to be run multiple times. To address this challenge, a KSM is constructed for the contaminant transport model, which greatly accelerates the determination and identification processes. The feasibility and accuracy of the proposed approach are verified by two hypothetical numerical case studies. The results show that the developed Bayesian-based integrated approach can be accurately and effectively applied for optimal sampling well location design and GCS parameter identification. Overall, this study highlights that the Bayesian-based integrated approach represents a promising solution for GCS parameter identification.

I3

The scrolling interaction is a pervasive human-computer interaction on smartphones, which can reflect intrinsic characteristics during dynamic browsings. Different from extrinsic statistical measures like frequency of visits and dwell time, intrinsic features underlying scrolling interactions reveal fine-grained implicit feedbacks about user interests. Toward this end, we explore user interest inference by extracting efficient browsing features from scrolling human-computer interactions on smartphones. In this paper, we first analyze browsing traces of 40 volunteers, and find two intrinsic browsing features underlying scrolling interactions, i.e., browsing velocity stability and browsing velocity sequence, which are tightly related to user interests. Inspired by the observation, we propose an Intelligent Interest Inference system, I3, which infers user interests through sensing scrolling interactions during browsings. Specifically, I3 first extracts the two intrinsic browsing features from users' scrolling interactions. Then, I3 applies a Naive Bayesian-based approach to construct an interest discriminator for coarse-grained user interest (i.e., like-preferred or dislike-preferred) inference. Furthermore, we develop a deep learning-based approach for I3 to train rating classifiers for fine-grained rating inference in like-preferred and dislike-preferred browsings respectively. Finally, I3 utilizes the interest discriminator and rating classifiers to infer exact user ratings about browsing contents on smartphones. Experimental results under browsing traces of 46 volunteers present that I3 achieves 92.4% overall accuracy in interest inference.

A computerized hybrid Bayesian-based approach for modelling the deterioration of concrete bridge decks

Bridges are aging and deteriorating, thus, reliable deterioration modeling is regarded as one of the vital components of Bridge management systems. This article presents an automated defect-based tool to predict the future condition of the bridge decks by calibrating the Markovian model based on a hybrid Bayesian-optimization approach. The in-state probabilities are demonstrated in the form of posterior distributions, whereas the transition from a condition state to the next lower state is a function of the severities of five types of bridge defects. In the present study, the Bayesian belief network is employed to construct the likelihood function by modeling the dependencies between the bridge defects. The maximum entropy optimization is incorporated to compute the missing conditional probabilities. The proposed approach utilizes Markov chain Monte Carlo Metropolis-Hastings algorithm to derive the posterior distributions. Finally, a stochastic optimization model is designed to build a variable transition probability matrix for each five-year zone via genetic algorithm. An automated tool is programmed using C#.net programming language to facilitate the implementation of the developed deterioration model by the users. Results show that the proposed model outperformed some commonly-utilized deterioration models as per three performance indicators which are: root-mean squared error, mean absolute error, chi-squared statistic.

ASSERT: attack synthesis and separation with entropy redistribution towards predictive cyber defense

The sophistication of cyberattacks penetrating into enterprise networks has called for predictive defense beyond intrusion detection, where different attack strategies can be analyzed and used to anticipate next malicious actions, especially the unusual ones. Unfortunately, traditional predictive analytics or machine learning techniques that require training data of known attack strategies are not practical, given the scarcity of representative data and the evolving nature of cyberattacks. This paper describes the design and evaluation of a novel automated system, ASSERT, which continuously synthesizes and separates cyberattack behavior models to enable better prediction of future actions. It takes streaming malicious event evidences as inputs, abstracts them to edge-based behavior aggregates, and associates the edges to attack models, where each represents a unique and collective attack behavior. It follows a dynamic Bayesian-based model generation approach to determine when a new attack behavior is present, and creates new attack models by maximizing a cluster validity index. ASSERT generates empirical attack models by separating evidences and use the generated models to predict unseen future incidents. It continuously evaluates the quality of the model separation and triggers a re-clustering process when needed. Through the use of 2017 National Collegiate Penetration Testing Competition data, this work demonstrates the effectiveness of ASSERT in terms of the quality of the generated empirical models and the predictability of future actions using the models.

Modeling risks in dependent systems: A Copula-Bayesian approach

This paper develops a hybrid Copula–Bayesian-based approach to model the structural health of an operational metro tunnel in a dependent system, where risk factors interact with each other. Aiming to rectify drawbacks of discrete Bayesian networks, the Gaussian copula and different marginal distributions are involved to build up complicated interactions among possible risk factors and structural health state more precisely and rationally. To achieve systematic and real-time decision support for safety insurance during the entire life cycle of risk-prone systems, model analytics is conducted to perform correlation analysis, forward reasoning, and backward reasoning. For demonstrating the effectiveness of the method, it is applied to a realistic cross-river tunnel, the Wuhan Yangtze Metro Tunnel in China, where all the concerning risk factors are water-related rooting in water leakage, poor construction quality, exterior load, and others. Due to the availability of vault displacement by sensors, it can herein represent the structural health to explore the impacts of risk factors. Larger vault displacement implies higher risk in the overall safety condition of the tunnel structure. This research contributes to (a) the state of knowledge by integrating Bayesian networks with copula, contributing to a more robust risk assessment by accurately modeling the complex dependence structure of risk factors; (b) the state of practice by providing guidelines of the whole-life-cycle safety control for complex systems under uncertainty and randomness, which not only prevent structural failure in advance but also control risk after accident occurrence.

An advanced fuzzy Bayesian-based FMEA approach for assessing maritime supply chain risks

This paper aims to develop a novel model to assess the risk factors of maritime supply chains by incorporating a fuzzy belief rule approach with Bayesian networks. The new model, compared to traditional risk analysis methods, has the capability of improving result accuracy under a high uncertainty in risk data. A real case of a world leading container shipping company is investigated, and the research results reveal that among the most significant risk factors are transportation of dangerous goods, fluctuation of fuel price, fierce competition, unattractive markets, and change of exchange rates in sequence. Such findings will provide useful insights for accident prevention.