Weight of Evidence Approach to Maritime Accident Risk Assessment Based on Bayesian Network Classifier

Probabilistic maritime accident models based on Bayesian Networks are typically built upon the data available in accident records and the data obtained from human experts knowledge on accident. The drawback of such models is that they do not take explicitly into the account the knowledge on non-accidents as would be required in the probabilistic modelling of rare events. Consequently, these models have difficulties with delivering interpretation of influence of risk factors and providing sufficient confidence in the risk assessment scores. In this work, modelling and risk score interpretation, as two aspects of the probabilistic approach to complex maritime system risk assessment, are addressed. First, the maritime accident modelling is posed as a classification problem and the Bayesian network classifier that discriminates between accident and non-accident is developed which assesses state spaces of influence factors as the input features of the classifier. Maritime accident risk are identified as adversely influencing factors that contribute to the accident. Next, the weight of evidence approach to reasoning with Bayesian network classifier is developed for an objective quantitative estimation of the strength of factor influence, and a weighted strength of evidence is introduced. Qualitative interpretation of strength of evidence for individual accident influencing factor, inspired by Bayes factor, is defined. The efficiency of the developed approach is demonstrated within the context of collision of small passenger vessels and the results of collision risk assessments are given for the environmental settings typical in Croatian nautical tourism. According to the results obtained, recommendations for navigation safety during high density traffic have been distilled.

The Expressive Triad: Structure, Color, and Texture Similarity of Emotion Expressions Predict Impressions of Neutral Faces

Previous research has demonstrated how emotion resembling cues in the face help shape impression formation (i. e., emotion overgeneralization). Perhaps most notable in the literature to date, has been work suggesting that gender-related appearance cues are visually confounded with certain stereotypic expressive cues (see Adams et al., 2015 for review). Only a couple studies to date have used computer vision to directly map out and test facial structural resemblance to emotion expressions using facial landmark coordinates to estimate face shape. In one study using a Bayesian network classifier trained to detect emotional expressions structural resemblance to a specific expression on a non-expressive (i.e., neutral) face was found to influence trait impressions of others (Said et al., 2009). In another study, a connectionist model trained to detect emotional expressions found different emotion-resembling cues in male vs. female faces (Zebrowitz et al., 2010). Despite this seminal work, direct evidence confirming the theoretical assertion that humans likewise utilize these emotion-resembling cues when forming impressions has been lacking. Across four studies, we replicate and extend these prior findings using new advances in computer vision to examine gender-related, emotion-resembling structure, color, and texture (as well as their weighted combination) and their impact on gender-stereotypic impression formation. We show that all three (plus their combination) are meaningfully related to human impressions of emotionally neutral faces. Further when applying the computer vision algorithms to experimentally manipulate faces, we show that humans derive similar impressions from them as did the computer.

An efficient dynamic Bayesian network classifier structure learning algorithm: application to sport epidemiology

Exposing an athlete to intense physical exertion when their organism is not ready for the mobilization of such resources can lead to musculoskeletal injury. In turn, sport practitioners regularly monitor athlete readiness in hopes of mitigating these tragic events. Rapid developments in athlete monitoring technologies has thus resulted in sport practitioners aspiring to siphon meaningful insight from high-throughput datasets. However, revealing the temporal sequence of biological adaptation while yielding accurate probabilistic predictions of an event, demands computationally efficient and accurate algorithms. The purpose of the present study is to create a model in the form of the intuitively appealing dynamic Bayesian network (DBN). Existing DBN approaches can be split into two varieties: either computationally burdensome and thus unscalable, or place structural constraints to increase scalability. This article introduces a novel algorithm ‘rapid incremental search for time-varying associations’ $(Rista)$, to be time-efficient without imposing structural constraints. Furthermore, it offers such flexibility and computational efficiency without compromising prediction performance. The present algorithm displays comparable results to contemporary algorithms in classification accuracy while maintaining superior speed.

Intelligent Predictive Analytics for Sustainable Business Investment in Renewable Energy Sources

Willingness to invest in renewable energy sources (RES) is predictable under data mining classification methods. Data was collected from the area of Evia in Greece via a questionnaire survey by using a sample of 360 respondents. The questions focused on the respondents’ perceptions and offered benefits for wind energy, solar photovoltaics (PVs), small hydro parks and biomass investments. The classification algorithms of Bayesian Network classifier, Logistic Regression, Support Vector Machine (SVM), C4.5, k-Nearest Neighbors (k-NN) and Long Short Term Memory (LSTM) were used. The Bayesian Network classifier was the best method, with a prediction accuracy of 0.7942. The most important variables for the prediction of willingness to invest were the level of information, the level of acceptance and the contribution to sustainable development. Future studies should include data on state incentives and their impact on willingness to invest.

A Timeliness-Enhanced Traffic Identification Method in Airborne Network

High dynamic topology and limited bandwidth of the airborne network make it difficult to provide reliable information interaction services for diverse combat mission of aviation swarm operations. Therefore, it is necessary to identify the elephant flows in the network in real time to optimize the process of traffic control and improve the performance of airborne network. Aiming at this problem, a timeliness-enhanced traffic identification method based on machine learning Bayesian network model is proposed. Firstly, the data flow training subset is obtained by preprocessing the original traffic dataset, and the sub-classifier is constructed based on Bayesian network model. Then, the multi-window dynamic Bayesian network classifier model is designed to enable the early identification of elephant flow. The simulation results show that compared with the existing elephant flow identification method, the proposed method can effectively improve the timeliness of identification under the condition of ensuring the accuracy of identification.