Machine learning methods for classification and prediction information security risk assessment

Purpose – Information security has become an essential entity for organizations across the globe to eliminate the possible risks in their organizations by conducting information security risk assessment (ISRA). However, the existence of numerous different types of risk assessment methods, standards, guidelines and specifications readily available causes the organizations to face the daunting tasks in determining the most suitable method that would augur well in meeting their needs. Therefore, to overcome this tedious process, this paper suggests collective information structure model for ISRA. Design/methodology/approach – The proposed ISRA model was developed by deploying a questionnaire using close-ended questions administrated to a group of information security practitioners in Malaysia (N = 80). The purpose of the survey was to strengthen and add more relevant additional features to the existing framework, as it was developed based on secondary data. Findings – Previous comparative and analyzed studies reveals that all the six types of ISRA methodologies have features of the same kind of information with a slight difference in form. Therefore, questionnaires were designed to insert additional features to the research framework. All the additional features chosen were based on high frequency of more than half percentage agreed responses from respondents. The analyses results inspire in generating a collective information structure model which more practical in the real environment of the workplace. Practical implications – Generally, organizations need to make comparisons between methodologies and decide on the best due to the inexistence of agreed reference benchmark in ISRA methodologies. This tedious process leads to unwarranted time, money and energy consumption. Originality/value – The collective information structure model for ISRA aims to assist organizations in getting a general view of ISRA flow and gathering information on the requirements to be met before risk assessment can be conducted successfully. This model can be conveniently used by organizations to complete all the required planning as well as to select the suitable methods to complete the ISRA.

In recent years, there have been rapid improvements in both remote sensing methods and satellite image availability that have the potential to massively improve burn severity assessments of the Alaskan boreal forest. In this study, we utilized recent pre- and post-fire Sentinel-2 satellite imagery of the 2019 Nugget Creek and Shovel Creek burn scars located in Interior Alaska to both assess burn severity across the burn scars and test the effectiveness of several remote sensing methods for generating accurate map products: Normalized Difference Vegetation Index (NDVI), Normalized Burn Ratio (NBR), and Random Forest (RF) and Support Vector Machine (SVM) supervised classification. We used 52 Composite Burn Index (CBI) plots from the Shovel Creek burn scar and 28 from the Nugget Creek burn scar for training classifiers and product validation. For the Shovel Creek burn scar, the RF and SVM machine learning (ML) classification methods outperformed the traditional spectral indices that use linear regression to separate burn severity classes (RF and SVM accuracy, 83.33%, versus NBR accuracy, 73.08%). However, for the Nugget Creek burn scar, the NDVI product (accuracy: 96%) outperformed the other indices and ML classifiers. In this study, we demonstrated that when sufficient ground truth data is available, the ML classifiers can be very effective for reliable mapping of burn severity in the Alaskan boreal forest. Since the performance of ML classifiers are dependent on the quantity of ground truth data, when sufficient ground truth data is available, the ML classification methods would be better at assessing burn severity, whereas with limited ground truth data the traditional spectral indices would be better suited. We also looked at the relationship between burn severity, fuel type, and topography (aspect and slope) and found that the relationship is site-dependent.

Smart infrastructure integrates cyberphysical systems, self-organizing networks of devices, fog computing to enhance the quality and convenience of manufacturing and social life, providing efficient resource management and increasing productivity and availability of traditional services. New technologies are characterized with a high degree of heterogeneity, structural complexity, and dynamic topology that result in a large number of new information and cyber security threats. Given a high degree of integration of digital cyber space into the processes of management, monitoring, aggregation, and data transferring, an assessment of its security level is a necessary requirement to maintain and control a safe and reliable environment for handling data and performing physical processes. Risk analysis and assessment is an extremely important task at all stages of the life cycle of subsystems and components of the smart environment. It allows us to assess a level of protection, identify vulnerable assets, and counteract the most likely directions of security threats. The known methods for security risk assessment do not consider the features of modern smart technologies, and this forces us to search for new approaches. This paper proposes a comprehensive method that includes a calculation of security indicators, risk assessment and selection of protective measures, based on attack graphs and thus allows us to consider the dynamics of changes in the components of the smart infrastructure. The automated system for assessing the security risks in the smart infrastructure and choosing the protective measures has been implemented.

Information Security (IS) Risk Assessment is a main part of risk analysis; it helps organizations make decisions to protect their Information Technology (IT) services and underlying IT assets from potentially adverse events. How to do assessment in this context, however, is not a well defined task. Some approaches provide guidelines but leave analysts to define how to implement them, leading to different mechanisms to identify input data, different procedures to process those inputs, and different results as a consequence. To address this problem, we present a semiautomatic procedure, based on data systematically obtained from modern IT Service Management (ITSM) tools used by IT staff to handle IT services’ assets and configurations. We argue that these tools handle actual data that may be used to collect inputs for a IS risk assessment procedure, thus reducing subjective values. We evaluated the procedure in a real case study and found that our approach actually reduces variability of some results. We also identified areas that must be addressed in future work.

Cyber-Security in the Internet of Things (IoT) is a major concern for information exploitation which hinder the growth of information system. To address security levels and issues, security risk assessment is considered an effective tool for system security, products, process, and readiness. Effective system vulnerabilities guidance is involved in the prioritization of security risk assessment. At present, the differential equation provides a significant tool for risk assessment. However, for second-order derivatives, the error rate is higher which impacts on overall risk assessment model. To overcome those limitations, this paper presented Decision Support Light Weight Risk Assessment Model (DSLiRAM). The proposed DSLiRAM is the domain-specific framework for security assessment. The proposed DSLiRAM is adopted in four stages for the specification of practices applied for cybersecurity and organizational characteristics. The proposed DSLiRAM includes a fuzzy differential equation with a second-order derivative. To minimize error rate Taylor series expansion is integrated with Fredholm for risk assessment. The proposed DSLiRAM is examined in three scenarios, RT server, BPCS, and HMI. Analysis of results stated that the proposed DSLiRAM significantly predicts risk and prevents the attack.

Internet of Things (IoT) applications are among the major trends of nowadays. Billions of connected devices are creating great business profits and performing a multitude of automated tasks in many daily human activities. In healthcare service delivery, IoT capabilities are difficult to overestimate, they are progressively becoming entangled and commonly coined Internet of Medical Things (IoMT). The participating nodes in IoMT networks generate, collect and exchange huge amounts of extremely private and sensitive data. Numerous security vulnerabilities arise due to the complexity and the heterogeneity of the technology. New risks, born out of IoMT systems, cannot easily be supported by existing risk management frameworks. The existing cyber-security risk assessment methods and approaches, deployed in several organizations, will not address the IoMT inherent risks properly. This study includes a comprehensive review of IoMT systems. Popular risk assessment methods are discussed and their suitability to IoMT is dealt with in detail. Based on this study, we propose a framework to enhance trust and help with decision making in e-healthcare environments given its high-risk exposure. The proposal is based on a quantified risk assessment approach. Our aim is to define a novel approach/model for improving trust and risk management in an e-health context.

Chapter Five - Cybersecurity risk management-I

The threat of cyber-attacks and wireless communication technologies are now affecting a number of private and public organizations around the world. It is a challenge to protect today's data from cyber-attacks since it is highly dependent on electronic technology. It is possible to mitigate or reduce these risks by identifying them. Risk assessments can help to develop a plan for responding to and recovering from a cyber attack. It is crucial to continuously monitor and assess the risk environment to identify any changes in the organization's context and to keep track of the entire risk management process. Researchers from all around the world have proposed a number of methods to thwart cyber-attacks or minimize the harm they do. The aim of this chapter is to present in-depth analysis of the typical improvements achieved in the field of cyber security risk assessment and management and examines the impact of artificial intelligence and machine learning in cyber security risk assessment, and management.

This paper constructs the index system and risk assessment model of information security risk in smart cities. According to the principles of integrity, pertinence, operability and completeness of index selection, and according to the key factors specified in the information security risk assessment standard, 20 risk factors are selected as evaluation indicators from five levels: smart city infrastructure, data service, information content, information management and public literacy, forming the final evaluation index system. By analyzing the requirements of smart city information security risk assessment, the common risk assessment methods are compared, and finally the decision tree algorithm is selected to build the information security risk assessment model of smart city.

The article is devoted to the assessment of information security risks of organizations for the purpose of leveling or reducing them. Approaches to the content, regulatory regulation, methods of analysis and assessment of information risks are considered. Analytical procedures for assessing and managing information security risks for a specific municipal enterprise were performed. The results obtained can be used by municipal enterprises to ensure the security of the information system and sustainable business development.

Photometric redshifts (photo- z ’s) are crucial for the cosmology, galaxy evolution, and transient science drivers of next-generation imaging facilities like the Euclid Mission, the Vera C. Rubin Observatory, and the Nancy Grace Roman Space Telescope. Previous work has shown that image-based deep learning photo- z methods produce smaller scatter than photometry-based classical machine learning (ML) methods on the Sloan Digital Sky Survey (SDSS. Main Galaxy Sample, a test bed photo- z dataset. However, global assessments can obscure local trends. To explore this possibility, we used a self-organizing map (SOM) to cluster SDSS galaxies based on their ugriz colors. Deep learning methods achieve lower photo- z scatter than classical ML methods for all SOM cells. The fractional reduction in scatter is roughly constant across most of color space with the exception of the most bulge-dominated and reddest cells where it is smaller in magnitude. Interestingly, classical ML photo- z ’s suffer from a significant color-dependent attenuation bias, where photo- z ’s for galaxies within an SOM cell are systematically biased towards the cell’s mean spectroscopic redshift and away from extreme values, which is not readily apparent when all objects are considered. In contrast, deep learning photo- z ’s suffer from very little color-dependent attenuation bias. The increased attenuation bias for classical ML photo- z methods is the primary reason why they exhibit larger scatter than deep learning methods. This difference can be explained by the deep learning methods weighting redshift information from the individual pixels of a galaxy image more optimally than integrated photometry.

With the development of information technology and the popular use of the information network system, the security of the information system becomes particularly important. To ensure the security of the information system, it is a key point to have risk assessment. This paper deals with the risk assessment in information security of distributed banking sector. The process of identification and analysis of various assets (both hardware and software), vulnerabilities, threats and risks in distributed banking sector has assumed utmost importance. This paper also presents collective information regarding various risk assessment methodologies like qualitative, quantitative and hybrid methods and tools such as OCTAVE and IRAM that are being employed in order to perform risk assessment in information security. Keywordsassets, authentication, concurrency, distributed computing, quantitative and qualitative methodologies, risk assessment, security, threat, vulnerabilities.

Contemporary methods for assessing information security risks have adopted mainly technical views on the information and technology assets. Organizational dynamics of information management and knowledge sharing have gained less attention. This article outlines how an information security risk assessment method can be elaborated using knowledge-centric analysis of information assets. For this purpose, we suggest the use of a genre-based analysis method for identifying organizational communication patterns, through which organizational knowledge is shared. Initial experiences of the method try-outs by three experienced information security professionals are discussed. The article concludes with a look at the implications of a genre-based analysis of knowledge assets for future research and practice.

This chapter summarizes innovations in security risk and capability assessment frameworks and methodologies to directly support dynamic decision making, forge a convergence with S&T, operations, intelligence and policy, and inform investment priorities of the Canadian Safety and Security Program (CSSP). Using a thematic approach, this paper reviews how multiple innovations were adopted as “best practices” into the conceptual design, development and implementation of security risk assessment frameworks and methodologies in the CSSP. These include the: use taxonomies to foster key stakeholder ‘buy in’; leveraging architecture frameworks to drive tools, methods and rating schema; the formulation of scenario based approaches for assessing high consequence all hazards risks; and, the application of an integrated risk and capability gap assessment model to guide investment in public safety and security S&T. In the CSSP, the innovations in security risk assessment identified best practices and common tools that supported numerous strategies and S&T portfolios (e.g., CBRNE, Critical Infrastructure Protection, all hazards safety and security; defence infrastructure/military bases), where risk assessment expertise/capability improvements that were adopted by practitioners, interdisciplinary teams and decision makers. Underpinning these innovations were strong partnerships and community-based approaches that enabled a credible risk-informed decision support environment, with consistent/scalable approaches to analysis and decision making under conditions of uncertainty.KeywordsSecurity risk assessmentVulnerability

Critical systems have very stringent requirements on both security and safety. Recent mishaps such as the missing MH370 aircraft and the sunk Korean Sewol ferry go to show that our technology in safety and security risk assessment still need a more integrated approach. Nuclear plant meltdown in the recent Fukushima accident is also a typical example of insufficient risk assessments. This work is a case study on how a unified security and safety risk assessment methodology may be applied to a High Pressure Core Flooder (HPCF) system in a nuclear power plant. Individual risk security or safety assessments may overlook the possible higher risk associated with such critical systems. The case study shows how the proposed method provides a more accurate risk assessment compared to individual assessments.