Advances In Risk Analysis Research Articles

Active deep learning on entity resolution by risk sampling

While the state-of-the-art performance on entity resolution (ER) has been achieved by deep learning, its effectiveness depends on large quantities of accurately labeled training data. To alleviate the data labeling burden, Active Learning (AL) presents itself as a feasible solution that focuses on data deemed useful for model training.Building upon the recent advances in risk analysis for ER, which can provide a more refined estimate on label misprediction risk than the simpler classifier outputs, we propose a novel AL approach of risk sampling for ER. Risk sampling leverages misprediction risk estimation for active instance selection. Based on the core-set characterization for AL, we theoretically derive an optimization model which aims to minimize core-set loss with non-uniform Lipschitz continuity. Since the defined weighted K-medoids problem is NP-hard, we then present an efficient heuristic algorithm. Finally, we empirically verify the efficacy of the proposed approach on real data by a comparative study. Our extensive experiments have shown that it outperforms the existing alternatives by considerable margins.

Advances in Risk Analysis with Big Data.

With cloud computing, Internet-of-things, wireless sensors, social media, fast storage and retrieval, etc., organizations and enterprises have access to unprecedented amounts and varieties of data. Current risk analysis methodology and applications are experiencing related advances and breakthroughs. For example, highway operations data are readily available, and making use of them reduces risks of traffic crashes and travel delays. Massive data of financial and enterprise systems support decision making under risk by individuals, industries, regulators, etc. In this introductory article, we first discuss the meaning of big data for risk analysis. We then examine recent advances in risk analysis with big data in several topic areas. For each area, we identify and introduce the relevant articles that are featured in the special issue. We conclude with a discussion on future research opportunities.

Erratum to "A model of exposure to rotavirus from nondietary ingestion iterated by simulated intermittent contacts." By T.R. Julian, R.A. Canales, J.O. Leckie, and A.B. Boehm, in Risk Analysis, 29(5):617-622.

Article: Risk Analysis 2009; 29: 617–632

Building Better Environmental Risk Assessments.

Risk assessment is a reasoned, structured approach to address uncertainty based on scientific and technical evidence. It forms the foundation for regulatory decision-making, which is bound by legislative and policy requirements, as well as the need for making timely decisions using available resources. In order to be most useful, environmental risk assessments (ERAs) for genetically modified (GM) crops should provide consistent, reliable, and transparent results across all types of GM crops, traits, and environments. The assessments must also separate essential information from scientific or agronomic data of marginal relevance or value for evaluating risk and complete the assessment in a timely fashion. Challenges in conducting ERAs differ across regulatory systems – examples are presented from Canada, Malaysia, and Argentina. One challenge faced across the globe is the conduct of risk assessments with limited resources. This challenge can be overcome by clarifying risk concepts, placing greater emphasis on data critical to assess environmental risk (for example, phenotypic and plant performance data rather than molecular data), and adapting advances in risk analysis from other relevant disciplines.

The Advances in Risk Analysis of Reservoir Operation Research

There are many uncertain factors in the process of reservoir’s operation. For the reasons, the "deterministic theory" must turn to" probability theory", and it becomes necessary to strengthen risk analysis. Summarize the connotation of reservoir risk analysis, progress in research of risk analysis for reservoir inflow forecasting, risk analysis for reservoir operation, risk analysis for flood protection and flood discharge , and reservoir multi-objective risk analysis. There are so much influence factors and much of them uncertainly, not enough research on multiple targets and multiple factors, not much apply on new risk analysis theories and methods. Therefore, it should research more on the fields of risk identification, reducing the impact of human factors, new analysis methods and coupling methods.

A review of recent advances in risk analysis for wildfire management

Risk analysis evolved out of the need to make decisions concerning highly stochastic events, and is well suited to analyse the timing, location and potential effects of wildfires. Over the past 10 years, the application of risk analysis to wildland fire management has seen steady growth with new risk-based analytical tools that support a wide range of fire and fuels management planning scales from individual incidents to national, strategic interagency programs. After a brief review of the three components of fire risk – likelihood, intensity and effects – this paper reviews recent advances in quantifying and integrating these individual components of fire risk. We also review recent advances in addressing temporal dynamics of fire risk and spatial optimisation of fuels management activities. Risk analysis approaches have become increasingly quantitative and sophisticated but remain quite disparate. We suggest several necessary and fruitful directions for future research and development in wildfire risk analysis.

Landslide risk assessment and management: an overview

Landslides can result in enormous casualties and huge economic losses in mountainous regions. In order to mitigate landslide hazard effectively, new methodologies are required to develop a better understanding of landslide hazard and to make rational decisions on the allocation of funds for management of landslide risk. Recent advances in risk analysis and risk assessment are beginning to provide systematic and rigorous processes to enhance slope management. In recent years, risk analysis and assessment has become an important tool in addressing uncertainty inherent in landslide hazards. This article reviews recent advances in landslide risk assessment and management, and discusses the applicability of a variety of approaches to assessing landslide risk. Firstly, a framework for landslide risk assessment and management by which landslide risk can be reduced is proposed. This is followed by a critical review of the current state of research on assessing the probability of landsliding, runout behavior, and vulnerability. Effective management strategies for reducing economic and social losses due to landslides are described. Problems in landslide risk assessment and management are also examined.