Measuring driver cognitive distraction through lips and eyebrows

<span>Cognitive distraction is one of the several contributory factors in road accidents. A number of cognitive distraction detection methods have been developed. One of the most popular methods is based on physiological measurement. Head orientation, gaze rotation, blinking and pupil diameter are among popular physiological parameters that are measured for driver cognitive distraction. In this paper, lips and eyebrows are studied. These new features on human facial expression are obvious and can be easily measured when a person is in cognitive distraction. There are several types of movement on lips and eyebrows that can be captured to indicate cognitive distraction. Correlation and classification techniques are used in this paper for performance measurement and comparison. Real time driving experiment was setup and faceAPI was installed in the car to capture driver’s facial expression. Linear regression, support vector machine (SVM), static Bayesian network (SBN) and logistic regression (LR) are used in this study. Results showed that lips and eyebrows are strongly correlated and have a significant role in improving cognitive distraction detection. Dynamic Bayesian network (DBN) with different confidence of levels was also used in this study to classify whether a driver is distracted or not.</span>

A decomposition structure learning algorithm in Bayesian network based on a two-stage combination method

Decomposition hybrid algorithms with the recursive framework which recursively decompose the structural task into structural subtasks to reduce computational complexity are employed to learn Bayesian network (BN) structure. Merging rules are commonly adopted as the combination method in the combination step. The direction determination rule of merging rules has problems in using the idea of keeping v-structures unchanged before and after combination to determine directions of edges in the whole structure. It breaks down in one case due to appearances of wrong v-structures, and is hard to operate in practice. Therefore, we adopt a novel approach for direction determination and propose a two-stage combination method. In the first-stage combination method, we determine nodes, links of edges by merging rules and adopt the idea of permutation and combination to determine directions of contradictory edges. In the second-stage combination method, we restrict edges between nodes that do not satisfy the decomposition property and their parent nodes by determining the target domain according to the decomposition property. Simulation experiments on four networks show that the proposed algorithm can obtain BN structure with higher accuracy compared with other algorithms. Finally, the proposed algorithm is applied to the thickening process of gold hydrometallurgy to solve the practical problem.

Application of Bayesian network in risk assessment for website deployment scenarios

Abstract—The rapid development of web-based systems in the digital transformation era has led to a dramatic increase in the number and the severity of cyber-attacks. Current attack prevention solutions such as system monitoring, security testing and assessment are installed after the system has been deployed, thus requiring more cost and manpower. In that context, the need to assess cyber security risks before the deployment of web-based systems becomes increasingly urgent. This paper introduces a cyber security risk assessment mechanism for web-based systems before deployment. We use the Bayesian network to analyze and quantify the cyber security risks posed by threats to the deployment components of a website. First, the deployment components of potential website deployment scenarios are considered assets, so that their properties are mapped to specific vulnerabilities or threats. Next, the vulnerabilities or threats of each deployment component will be assessed according to the considered risk criteria in specific steps of a deployment process. The risk assessment results for deployment components are aggregated into the risk assessment results for their composed deployment scenario. Based on these results, administrators can compare and choose the least risky deployment scenario. Tóm tắt—Sự phát triển mạnh mẽ của các hệ thống trên nền tảng web trong công cuộc chuyển đổi số kéo theo sự gia tăng nhanh chóng về số lượng và mức độ nguy hiểm của các cuộc tấn công mạng. Các giải pháp phòng chống tấn công hiện nay như theo dõi hoạt động hệ thống, kiểm tra và đánh giá an toàn thông tin mạng được thực hiện khi hệ thống đã được triển khai, do đó đòi hỏi chi phí và nhân lực thực hiện lớn. Trong bối cảnh đó, nhu cầu đánh giá rủi ro an toàn thông tin mạng cho các hệ thống website trước khi triển khai thực tế trở nên cấp thiết. Bài báo này giới thiệu một cơ chế đánh giá rủi ro an toàn thông tin mạng cho các hệ thống website trước khi triển khai thực tế. Chúng tôi sử dụng mạng Bayes để phân tích và định lượng rủi ro về an toàn thông tin do các nguồn đe dọa khác nhau gây ra trên các thành phần triển khai của một website. Đầu tiên, các thành phần triển khai của các kịch bản triển khai website tiềm năng được mô hình hoá dưới dạng các tài sản, sao cho các thuộc tính của chúng đều được ánh xạ với các điểm yếu hoặc nguy cơ cụ thể. Tiếp đó, các điểm yếu, nguy cơ của từng thành phần triển khai sẽ được đánh giá theo các tiêu chí rủi ro đang xét tại mỗi thời điểm cụ thể trong quy trình triển khai. Kết quả đánh giá của các thành phần triển khai được tập hợp lại thành kết quả đánh giá hệ thống trong một kịch bản cụ thể. Căn cứ vào kết quả đánh giá rủi ro, người quản trị có thể so sánh các kịch bản triển khai tiềm năng với nhau để lựa chọn kịch bản triển khai ít rủi ro nhất.

Modelling the balance of care: Impact of an evidence-informed policy on a mental health ecosystem

Major efforts worldwide have been made to provide balanced Mental Health (MH) care. Any integrated MH ecosystem includes hospital and community-based care, highlighting the role of outpatient care in reducing relapses and readmissions. This study aimed (i) to identify potential expert-based causal relationships between inpatient and outpatient care variables, (ii) to assess them by using statistical procedures, and finally (iii) to assess the potential impact of a specific policy enhancing the MH care balance on real ecosystem performance. Causal relationships (Bayesian network) between inpatient and outpatient care variables were defined by expert knowledge and confirmed by using multivariate linear regression (generalized least squares). Based on the Bayesian network and regression results, a decision support system that combines data envelopment analysis, Monte Carlo simulation and fuzzy inference was used to assess the potential impact of the designed policy. As expected, there were strong statistical relationships between outpatient and inpatient care variables, which preliminarily confirmed their potential and a priori causal nature. The global impact of the proposed policy on the ecosystem was positive in terms of efficiency assessment, stability and entropy. To the best of our knowledge, this is the first study that formalized expert-based causal relationships between inpatient and outpatient care variables. These relationships, structured by a Bayesian network, can be used for designing evidence-informed policies trying to balance MH care provision. By integrating causal models and statistical analysis, decision support systems are useful tools to support evidence-informed planning and decision making, as they allow us to predict the potential impact of specific policies on the ecosystem prior to its real application, reducing the risk and considering the population’s needs and scientific findings.