Probabilistic Prediction of Significant Wave Height Using Dynamic Bayesian Network and Information Flow

Mathematical Problems in Engineering

Yang

2021

Self Cite

Piracy is a major threat to maritime safety. Assessing piracy risk is crucial to ship safety, travel security, and emergency plan preparation. In the absence of a thorough understanding of the factors and mechanisms that influence piracy, no perfect mathematical equation can be set up for such risk assessment. Therefore, the major factors that influence piracy were identified to construct an indicator system for assessment. These factors were analyzed, keeping in view the overall hazard, vulnerability, and antirisk properties, and then the Bayesian network was introduced into the risk assessment model to fuse multiresource information. For some indicators, which have only qualitative information or fragmentary statistical data, the cloud model theory was adopted to realize prior probability settings of the Bayesian network and thus made up for the deficiency in parameter settings. Finally, the inherent hazard of the South China Sea was assessed, as an example for the model, and two real piracy cases were studied to validate the proposed model. The assessment model constructed here can be applied to all cases, similar to the ones studied here.

Section: E Construction Of Bayesian Network Bayesian Network (Bn)mentioning

confidence: 99%

Application of Cloud Model and Bayesian Network to Piracy Risk Assessment

Mathematical Problems in Engineering

Yang

2021

Self Cite

“…Input variable selection is of vital importance in ML. Irrelevant variables, used as inputs of a regression machine, can unnecessarily increase the time consuming of a prediction system, as well as degrade the generalization ability and interpretability (Li and Liu, 2020). In previous studies about time series prediction with ML algorithms, methods based on correlation coefficient and univariate regression are still the most common data-based tools to select predictors by analyzing associations (Li and Liu, 2019).…”

Section: Causal Analysismentioning

confidence: 99%

“…Wang et al (2017) applied the convolutional neural network (CNN) to estimate SIC in the Gulf of Saint Lawrence from synthetic-aperture-radar (SAR) imagery, showing the superiority of CNN model in SIC estimation. Compared with numerical models and classic statistical models, ML algorithms simplify the tedious and intricate calculations, and they are good at expressing non-linear relationships of variables (Li and Liu, 2020). However, the previous studies using ML techniques still focused on the long-term prediction of SIC.…”

Section: Introductionmentioning

confidence: 99%

Machine Learning Incorporated With Causal Analysis for Short-Term Prediction of Sea Ice

Zhang

2021

Front. Mar. Sci.

Self Cite

Accurate and fast prediction of sea ice conditions is the foundation of safety guarantee for Arctic navigation. Aiming at the imperious demand of short-term prediction for sea ice, we develop a new data-driven prediction technique for the sea ice concentration (SIC) combined with causal analysis. Through the causal analysis based on kernel Granger causality (KGC) test, key environmental factors affecting SIC are selected. Then multiple popular machine learning (ML) algorithms, namely self-adaptive differential extreme learning machine (SaD-ELM), classification and regression tree (CART), random forest (RF) and support vector regression (SVR), are employed to predict daily SIC, respectively. The experimental results in the Barents-Kara (B-K) sea show: (1) compared with correlation analysis, the input variables of ML models screened out by causal analysis achieve better prediction; (2) when lead time is short (<3 d), the four ML algorithms are all suitable for short-term prediction of daily SIC, while RF and SaD-ELM have better prediction performance with long lead time (>3 d); (3) RF has the best prediction accuracy and generalization ability but hugely time consuming, while SaD-ELM achieves more favorable performance when taking computational complexity into consideration. In summary, ML is applicable to short-term prediction of daily SIC, which develops a new way of sea ice prediction and provides technical support for Arctic navigation.

“…Structural learning is the basis and prerequisite of parameter learning, which mines causal relationships from data and expresses them in the form of a network. It can be described as a process where, based on an observed dataset D of the node set X, the network structure G that best matches D can be found through intelligent learning algorithms [26,27]. SS algorithms are widely used for structural learning and achieving satisfactory results.…”

Section: Assuming a Set Of Variablesmentioning

confidence: 99%

A New Ensemble Learning Algorithm Combined with Causal Analysis for Bayesian Network Structural Learning

Zhang

2020

Symmetry

Self Cite

The Bayesian Network (BN) has been widely applied to causal reasoning in artificial intelligence, and the Search-Score (SS) method has become a mainstream approach to mine causal relationships for establishing BN structure. Aiming at the problems of local optimum and low generalization in existing SS algorithms, we introduce the Ensemble Learning (EL) and causal analysis to propose a new BN structural learning algorithm named C-EL. Combined with the Bagging method and causal Information Flow theory, the EL mechanism for BN structural learning is established. Base learners of EL are trained by using various SS algorithms. Then, a new causality-based weighted ensemble way is proposed to achieve the fusion of different BN structures. To verify the validity and feasibility of C-EL, we compare it with six different SS algorithms. The experiment results show that C-EL has high accuracy and a strong generalization ability. More importantly, it is capable of learning more accurate structures under the small training sample condition.