Advances in Core Computer Science-Based Technologies

The recognition of emotion/mood in groups of people is particularly interesting as it constitutes a whole new research area that can provide solutions to various high level problems and applications, such as, for example, the development of educational technologies for groups of learners in which teaching is adjusted to the mood of the group. The purpose of this chapter is to summarize and present research on the recognition via computer of emotions and/or the mood of a single person or groups of people by using a visual-facial modality combined with a keyboard-based modality. Additionally, this chapter presents research about a program that recognizes the emotions/mood of groups of people in live pictures collected directly from webcams. New algorithmic approaches are proposed and conclusions are drawn for use in similar systems. The implemented system successfully recognizes emotions in the following situations: The chapter is complemented with illustrations of use of the implemented system, as well as suggestions for future related work in this area.

Nowadays there is a wide range of digital information available to educational institutions regarding learners, including performance records, educational resources, student attendance, feedback on the course material, evaluations of courses and social network data. Although collecting, using, and sharing educational data do offer substantial potential, the privacy-sensitivity of the data raises legitimate privacy concerns. The sharing of data among education organizations has become an increasingly common procedure. However, any organization will most likely try to keep some patterns hidden if it must share its datasets with others. This chapter focuses on preserving the privacy of sensitive patterns when inducing decision trees and demonstrates the application of a heuristic to an educational data set. The employed heuristic hiding method allows the sanitized raw data to be readily available for public use and, thus, is preferable over other heuristic solutions, like output perturbation or cryptographic techniques, which limit the usability of the data.

Teaching various topics using gamification elements or Game-Based Learning (GBL) methods is a top trend nowadays. Gamification has shown great results towards this direction, however, the usage of GBL methods has not been sufficiently studied for the effectiveness of the learning process. This study examines how instructional design could be applied and how computer games could be a learning environment for acquiring the basic skills and experience in fundamental cybersecurity topics. Towards this direction, this research aspires to discover how specific computer games, designed as simulations, could be converted into virtual learning environments and enhance the learning process, by increasing the levels of motivation and engagement of undergraduate students in the topics of cybersecurity. Computer games are appropriate for creating effective virtual learning environments specific to cybersecurity, providing positive learning outcomes. More specifically, in this study a commercial computer game is evaluated for the effectiveness of using GBL to the learning process. The result of this approach is a learning experience, featuring positive outcomes in terms of engagement and distinct impact in terms of perceived learning. For this study, the ARCS motivation model was used, for evaluating motivation levels and for investigating potential attributes which are related to perceived learning, knowledge and skill acquisition.

This paper proposes to validate the structural relationships between the measurements of the academic agency components of student self-report and those observed in a mathematical learning episode, in order to provide evidence of how the metacognitive processes involved in success interact before and during an academic learning process. It is proposed to analyze the way in which learning styles, epistemic beliefs, and reading comprehension correlate with the degree of self-regulation and the ability to self-direct one’s cognitive resources in a conscious and voluntary way in the achievement of learning goals in tasks that involve solving mathematical problems around the notion of infinite process, mediated by the use of a computational tool. The findings underline the importance and role that the use of technology can play in developing teaching strategies and assessing learning outcomes that result in the promotion of metacognitive and self-regulated learning skills.

Cloud computing is a type of service that allows the use of computing resources from a distance, rather than a new technology. Various services exist on-demand, ranging from data storage and processing to software as a service, like email and developing platforms. Cloud computing enables ubiquitous, on-demand access over the net to a shared pool of configurable resources, like servers, applications, etc. that can be accessed, altered or even restored rapidly with minimal service provider interaction or management effort. Still, due to the vast growth of cloud computing, new security issues have been introduced. Key factors are the loss of control over any outsourced resources and cloud’s computing inherent security vulnerabilities. Managing these risks requires the adoption of an effective risk management method, capable of involving both the Cloud customer and the Cloud Service Provider. Risk assessment methods are common tools amongst IT security consultants for managing the risk of entire companies. Still, traditional risk management methodologies are having trouble managing cloud services. Extending our previous work, the purpose of this paper is to compare and examine whether popular risk management methods and tools (e.g. NIST SP800, EBIOS, MEHARI, OCTAVE, IT-Grundschutz, MAGERIT, CRAMM, HTRA, Risk-Safe Assessment, CORAS) are suitable for cloud computing environments. Specifically, based upon existing literature, this paper points out the essential characteristics that any risk assessment method addressed to cloud computing should incorporate, and suggests three new ones that are more appropriate based on their features.

In this paper, we present challenges and risks concerning cyber security in a supply chain environment. In particular, we focus on the MITIGATE (Multidimensional, IntegraTed, rIsk assessment framework and dynamic, collaborative risk manaGement tools for critical information infrAstrucTrurEs) Supply Chain Risk Assessment methodology, which is in compliance with ISO28001 and can be applied in order to assess the security risks of all the organizations involved in a supply chain. To validate the MITIGATE approach, we provide use cases based on real-life maritime scenarios and real-world data collection. To this end, a number of best practices in the form of guidelines for a successful application of the MITIGATE risk management system in supply chain environments are presented. The main advantages of the Mitigate Risk Assessment approach over existing maritime initiatives and efforts are also highlighted.

Internet of Things (IoT) technologies have enabled Cyber-Physical Systems (CPS) to become fully interconnected. This connectivity however has radically changed their threat landscape. Existing risk assessment methodologies often fail to identify various attack paths that stem from the new connectivity/functionality features of IoT-enabled CPS. Even worse, due to their inherent characteristics, IoT systems are usually the weakest link in the security chain and thus many attacks utilize IoT technologies as their key enabler. In this paper we review risk assessment methodologies for IoT-enabled CPS. In addition, based on our previous work (Stellios et al. in IEEE Commun Surv Tutor 20:3453–3495, 2018, [47]) on modeling IoT-enabled cyberattacks, we present a high-level risk assessment approach, specifically suited for IoT-enabled CPS. The mail goal is to enable an assessor to identify and assess non-obvious (indirect or subliminal) attack paths introduced by IoT technologies, that usually target mission critical components of an CPS.

Within Information and Knowledge Society the concept of Privacy has been enriched including aspects related to digital life, while the right to online Privacy gains more and more attention daily due to several cases of privacy breaches. Privacy is associated with the control, access and use or misuse of personal information by others, including governments, companies and other users as well. Social Network Sites as a part of digital space have altered the way that people communicate and have contributed to the construction of online social networks. During online interaction, users disclose information about them or others, while at the same time they express their concerns about Privacy infringement that may come up due to self-disclosure practices, not restricting or reversing though their disclosure behavior. Thus the “Privacy paradox” phenomenon is recorded since users cannot balance between Privacy concerns and their need for disclosure. Privacy’s circumvention destabilizes the trust between social actors, increases the feelings of insecurity and puts into risk social cohesion which is a prerequisite for the sustainability of our society. Legislation as well as technology may protect us, but sometimes they are not user friendly and sufficient. Users should protect themselves and other people in order to preserve their Privacy as a fundamental human right. In this paper, based on a literature review, we present the issue of Privacy in Social Network Sites focusing on factors that affect people’s Privacy concerns and behavior while relating these to social cohesion.

The Internet of Things (IoT) has changed the traditional computing models. While it has enabled multiple new computing applications, it has also raised significant issues regarding security and privacy. We are gradually shifting to using extended computing architectures, the nodes of which may be lightweight devices limited in hardware resources, scattered in terms of network topology and too diverse in terms of hardware and software to be efficiently administered and managed. Additionally, such nodes usually store, process and transmit sensitive private data of their users; thus, the risk of a security breach is significantly high. Blockchain technology, introduced through Bitcoin, enables the development of secure decentralized systems. It offers guarantees regarding data integrity, application logic integrity and service availability, while it lags behind in terms of privacy and efficiency. Because of the decentralized architecture of blockchain systems, there seems to be a good fit between blockchain and the IoT. Blockchain systems can be employed to develop solutions to some of the main security and privacy issues encountered in the IoT domain. In this chapter we discuss the convergence of the two technologies, we analyze possible use cases, where blockchain technology can enhance internet of things security and privacy, and we propose enhancements of blockchain technology to make it appropriate for application in the IoT domain.

The modern day society driven by a variety of electronic devices and high-speed internet is changing its perception and practice of paper currency, mode of economic and financial transactions, and so on. The usage of cash is increasingly reducing because of the ease of payments facilitated by cards, mobile phone apps and contact-less chips, online payment systems, etc. Furthermore, cryptotokens (cryptocurrencies), such as bitcoin, have fueled the interest of society and policymakers in investigating the usefulness and limitations of a central bank-backed electronic fiat currency. Also, blockchain/distributed ledger technology, the technology enabling the cryptotokens, has gained a lot of attention from almost all the sections of the society for its ability in providing a decentralized transaction verification process while maintaining the features similar to the traditional cash currency. This chapter presents an overview of the concepts and potential features, potential primary models of issuing electronic fiat currency, as well as the key design principles from a technical perspective and a high-level architecture of a central bank-backed electronic fiat currency. We present the key aspects of currency in modern society, the advancements in the field of financial technologies and how could these be harnessed to launch a central bank-backed electronic fiat currency. We also discuss the position of various central banks and governments on cryptocurrencies, blockchain technology, and the initiatives related to issuing electronic fiat currency.

Mobile devices have become an indispensable part of our daily lives. Practically, most of our everyday communication is performed through mobile devices which host third party apps and provide for various means of interaction with diverse levels of security. Android is by far the most widely used mobile operating system, with a user base in the scale of billions. However, while Android Open Source Project (AOSP) is paving the way for all manufacturers, Android market is so fragmented that those who are using the latest version are only a small minority. Moreover, Android comes in several flavours as manufacturers tailor it to their needs. However, this tailoring often prevents users from getting the latest updates. In fact, as we show, manufacturers may not follow the security and privacy guidelines of AOSP, exposing their users to unexpected threats. In this work we study a yet unpatched vulnerability by most major manufacturers, and partially fixed in AOSP, which allows for an adversary to extract important information from the victim’s device. To this end, we showcase that unprivileged apps, without actually using any permissions, can harvest a considerable amount of valuable user information. This is achieved by monitoring and exploiting the file and folder metadata of the most well-known messaging apps in Android, which have been hitherto considered secure, deriving thereby usage statistics in order to elicit user profiles, social connections, credentials or other sensitive information.

Task scheduling is crucial for offering the users of an operating system the impression that the system’s response is direct and in real time. However, it is a complicated process that is characterized by vagueness and uncertainty. A solution to this problem is the usage of fuzzy logic. Therefore, in this paper a rule-based fuzzy scheduling method, which considers both the execution time and the waiting time for each task, is presented. Fuzzy sets are used to describe both criteria. The operation of the presented scheduling method is based on a rule-based reasoner that decides dynamically about the priority of the tasks that wait to be executed. This reasoner is triggered each time a change occurs (i.e. the execution of a task ends, a new task arrives etc.). It has been compared with Short Job First (SJF) and First Come First Served (FCFS) scheduling algorithms for a large number of different sets of tasks. The results showed that the presented fuzzy scheduling algorithm in all the cases has the same or almost the same average waiting time with SJF, which is considered as the scheduling algorithm with the best (minimum) average waiting time. However, the presented fuzzy scheduling algorithm ensures that priority is given not only to tasks with short execution time, but also to tasks that remain into the queue waiting to be executed for a long time. In this way, the starvation, which may be caused by SJF algorithm, is eliminated.

The purpose of this chapter is to highlight the use of computational intelligence algorithms for solving a special class of combinatorial optimization problems in transportation science called routing problems. Classical routing problems, such as the Traveling Salesman Problem and the Vehicle Routing Problem, as well as highly relevant extensions of classical routing problems like Vehicle Routing Problem with Time Windows and Inventory Routing Problem have received a great deal of attention from academics, consultants and practitioners in the field of Supply Chain Management. The contribution of this study is fourfold: (i) it provides a comprehensive review of various solution algorithms that have been proposed in literature as possible solutions to many of the complex issues surrounding routing problem management, (ii) it presents formulation schemes related to basic routing problems and their extensions, (iii) it promotes the use of selected computational algorithms called meta-heuristics and sim-heuristics by providing various graphical presentation formats so as to simplify complicated issues and convey meaningful insights into the routing problems and (iv) it points out interesting research directions for further development in routing problems.

Biological systems are defined by their complexity and nonlinearity and thus provide fertile ground for the development of nonlinear deterministic models for predicting aspects of their behavior. This approach motivated the introduction of the concept named Homeodynamics which we will outline and then proceed to present a case study on the application of bifurcation theory and stability analysis, both topological approaches of dynamical systems, on three biological mechanisms of great importance in the context of Homeodynamics. These will be protein folding, protein dynamics and epigenetics.

In this paper, the problem of developing appropriate information search and retrieve mechanisms and tools in the web environment, is investigated. This problem is of great interest to those in information technology, since a vast amount of heterogeneous data are available, end so, are not interoperable on the Web to researchers or other interest groups. The problem is addressed here using, as, effective encoding for locating and sharing a very specific class of data, that of uniform diagnostic EEG features. In this study is proposed a suitable metadata schema, based on knowledge of medical EEG signal processing. The defined schema tries to initiate a dialog for further development of metadata specific formats of EEG recordings. The final aim of this study is to offer a web searching tool for data recorded and stored in a different operational structure or using several software and hardware systems, in a uniform EEG data collection for research and research purposes.

We present some basic results concerning the evaluation of the absolute values of roots of univariate polynomials with complex or real coefficients. There are discussed classical and modern algorithmic methods and their computational efficiency.