Skip to main content
main-content

Über dieses Buch

This volume represents the 18th International Conference on Information Technology - New Generations (ITNG), 2021. ITNG is an annual event focusing on state of the art technologies pertaining to digital information and communications. The applications of advanced information technology to such domains as astronomy, biology, education, geosciences, security, and health care are the among topics of relevance to ITNG. Visionary ideas, theoretical and experimental results, as well as prototypes, designs, and tools that help the information readily flow to the user are of special interest. Machine Learning, Robotics, High Performance Computing, and Innovative Methods of Computing are examples of related topics. The conference features keynote speakers, a best student award, poster award, service award, a technical open panel, and workshops/exhibits from industry, government and academia. This publication is unique as it captures modern trends in IT with a balance of theoretical and experimental work. Most other work focus either on theoretical or experimental, but not both. Accordingly, we do not know of any competitive literature.

Inhaltsverzeichnis

Frontmatter

AI and Robotics

Frontmatter

Chapter 1. Conceptualisation of Breast Cancer Domain Using Ontology

The conceptualization of the breast cancer domain using ontology is an emerging area of intelligent decision support system. Eventhough this is not the replacement for clinicians, this intelligent system can support them in an effective way during diagnosis. As the system requires data from clinicians and patients, the unorganized data is gathered and processed. As the input data are unstructured, it is hard to gather information and share knowledge from that. An adaptive questionnaire is used to gather data to optimize the result of the system. The paper discusses the prototype model which uses various ontology as the knowledge base, java engine to provide information to modeller and a reasoner to take effective decision. SPARQL is used to retrieve required information as per the conditions. Protégé that supports OWL representation provides a platform to build concepts and relationships. How ontology is representing the details of Breast cancer guidelines and how instances of the class are identified using the query are shown in this paper.

Reshmy Krishnan, P. C. Sherimon, Menila James

Chapter 2. Traffic Light Control and Machine Learning: A Systematic Mapping Review

The global vehicle fleet has grown rapidly over the past decade, impacting the way traffic is to be managed. Vehicle traffic management and control through technology is a well-known and widely studied problem that continues to present challenges and opportunities for action, mainly due to the growing demand, the mentioned increase in the vehicle fleet, and inefficiency of current systems, generally based on fixed-time traffic lights. Solutions have been presented for this scenario, and among them, Artificial Intelligence (AI) and Machine Learning (ML) techniques have stood out. The AI/ML field, however, is vast and varied. This article proposes a survey of the most used AI/ML techniques in the management of vehicular traffic lights, and it does so through a Systematic Mapping Review (SMR), pointing out models that receive greater focus, research trends and gaps.

Dimitrius F. Borges, Edmilson M. Moreira, Adler D. de Souza, João Paulo R. R. Leite

Chapter 3. Human-in-the-Loop Flight Training of a Quadcopter for Autonomous Systems

A software framework was developed connecting a Parrot AR 2.0 quadcopter to a full motion flight simulator at the Michigan Aerospace Center for Simulations. The combination of a drone with a flight simulator provides for precise remote operations without putting a human pilot at risk. We use the motion capabilities of our flight simulator to keep the pilot oriented consistently with the quadcopter. The result was a responsive system utilizing telemetry data to synchronize a flight simulator to a drone’s movement with low latency. The proposed system was developed over the course of 30 weeks and put through its paces in our lab over two days. This paper outlines our methods, from the software architecture, to a detailed description of the hardware, to accomplish this aim and outlines some directions for further study.

Luke Rogers, Alex Redei

Chapter 4. COVID-19: The Importance of Artificial Intelligence and Digital Health During a Pandemic

The Covid-19 has brought about a major change in the way people live, work and interact. To face the challenges of the epidemic, health professionals and researchers have implemented several technologies from Industry 4.0. In order to elucidate the application of these technologies in the context of the pandemic, the objective of this article is to analyze the main research trends of the Technologies 4.0 from the main publications on the subject. Data collection was carried out in the Scopus database in September 2020 and 413 studies were identified. The gaps identified in this research were: Apply artificial intelligence and I4.0 technologies to support and speed up Covid-19 diagnosis, Implement Risk Management tools to prevent and mitigate new Covid-19 infection waves, Integrate I4.0 technologies into microbiology and clinical trials, Mapping and sharing data that identify transmission rates and Covid’19 diffusion routes, Search for treatment alternatives to Covid-19 through algorithms and artificial intelligence. The main academic contribution of this article was to systematize technological trends and understanding the influence of artificial intelligence and impact on the most urgent issues of the pandemic.

Maximilian Espuny, José S. da Motta Reis, Gabriel M. Monteiro Diogo, Thalita L. Reis Campos, Vitor H. de Mello Santos, Ana C. Ferreira Costa, Gildarcio S. Gonçalves, Paulo M. Tasinaffo, Luiz A. Vieira Dias, Adilson M. da Cunha, Nilo A. de Souza Sampaio, Andréia M. Rodrigues, Otávio J. de Oliveira

Chapter 5. CropWaterNeed: A Machine Learning Approach for Smart Agriculture

In this paper, we propose an approach CropWaterNeed in order to estimate and predict the future water needs and maximize the productivity in the irrigated areas. Unfortunately, we have not identified data available to be employed in such machine learning process in order to predict plants water needs. The proposed approach consists of extending the classic machine learning process. Particularly, we define a process to build dataset that contains plant water requirements. To collect data, we extract meteorological data from Climwat database and plants water requirements using Cropwat Tool. Then, we aggregate the extracted data into a dataset. Subsequently, we use the dataset to perform the learning process using XGBRegressor, Decision Tree, Random Forest and Gradiant Boost Regressor. Afterward, we evaluate the model generated by each algorithm by measuring the performance measures such as MSE, RMSE and MAE. Our work shows that the model generated by XGBRegressor is the most efficient in our case while Random Forest is the least efficient. As future work, we aim to apply the proposed process to test the performance of other regression algorithms and to test the impact of using deep learning techniques with the extracted data.

Malek Fredj, Rima Grati, Khouloud Boukadi

Chapter 6. Machine Learning: Towards an Unified Classification Criteria

In a broad sense, Machine Learning (ML) is the performance optimization in a certain task through computational means, following a certain criterion and using referential data and/or past results from previous iterations. ML is a subset of Artificial Intelligence (AI) and has attracted a substantial amount of research during the last decades. This blooming subject led to the statement of different definitions for classifications, criteria, algorithms and so on. This paper summarizes these different definitions and proposes a homologation between them, providing an unified vision for each definition.

Clara Burbano, David Reveló, Julio Mejía, Daniel Soto

Cybersecurity I

Frontmatter

Chapter 7. Classification and Update Proposal for Modern Computer Worms, Based on Obfuscation

Computer worms are a type of malware that have a complex technological structure and the ability to automatically create replicas of themselves without human interaction and distributing themselves to other computers connected to the network; they have a malicious code component that allows them to infect one computer and then use it to infect others. This cycle repeats itself, rapidly increasing the number of infected computers if action is not taken in time. Within this framework, the research is based on a systematic review of the methodology used to analyze scientific articles related to malware and specifically to computer worms. Through this review and the abstraction of important data, a synthesis of the results is made to support the research, resulting in a new proposal for the classification of computer worms according to their obfuscation capacity, dividing it into four levels: species, type, class and evasion. This classification allows a modern computer worm to be categorized in such a way that the main contribution is that it can serve as a model or as a complement to an Information Security Management System (ISMS), in the systems responsible for detecting and/or defending organizations against worms attacks.

Hernaldo Salazar, Cristian Barria

Chapter 8. Conceptual Model of Security Variables in Wi-Fi Wireless Networks: Review

Systems, data, users, and networks are essential in terms of information security. Systems, data, users and networks are essential in terms of information security. Wi-Fi wireless networks play a crucial role in increasing connectivity, as well as preventing and monitoring unauthorized access. Nonetheless, Wi-Fi wireless networks’ security is conditioned by different variables incorporated in standards, norms, good practices, and various investigations concerning this topic.In this way, the present research exposes an information survey of certain variables based on a narrative review, which allows their identification and possibly the incorporation of others. The results obtained from the survey will be shown through a conceptual model, which allows visualizing the different aspects required in the security that is applied to this technology.

Lorena Galeazzi, Cristian Barría, Julio Hurtado

Chapter 9. Cybersecurity Analysis in Nodes that Work on the DICOM Protocol, a Case Study

Currently the Internet is the main tool for interconnection of systems and data processing, SCADA systems, marketing, government; Even medical systems, among others, need to be connected to the internet to facilitate the development and processing of their data, the services that operate with the DICOM protocol (Digital Imaging and Communications in Medicine) work with medical equipment. This protocol is the universal format for the exchange of medical images, due to which it is used worldwide for communication between devices, the so-called PACS servers (Image Archiving and Communication System). These are information receptacles where medical centers store X-rays, files, personal information of patients, information of the treating physician, among others. The purpose of this research is to carry out a cybersecurity analysis in the operational and connected nodes in Chile that operate with the DICOM protocol among their services, with the execution of a modified experimental design that will allow the discovery of active nodes, discovery of exposed services and vulnerabilities, the analysis of said services as well as their vulnerabilities, their categorization and finally the validation of the vulnerabilities found. It seeks to know the current situation in cybersecurity issues of the nodes that use the DICOM protocol for communication, identifying the possible attack vectors that third parties may use in order to compromise the integrity, confidentiality, availability and authenticity of said systems.

David Cordero, Cristian Barría

Chapter 10. Hybrid Security Risk Assessment Model

Cybersecurity risk management often uses experience-based data to quantify the potential risks of new security technologies based on their exploitability and impact. However, use of such data may be limited and is rarely reusable because it often contains confidential information. This paper proposes a new approach using the Department of Homeland Security’s public National Vulnerability Database (NVD) for information on known vulnerabilities, and MITRE’s public Common Attack Pattern Enumeration and Classification (CAPEC™) tools as the basis of a risk scoring system.

Robert Banks, Jim Jones, Noha Hazzazi, Pete Garcia, Russell Zimmermann

Chapter 11. Enriching Financial Software Requirements Concerning Privacy and Security Aspects: A Semiotics Based Approach

Enriching software requirements with key security and privacy features requires professionals to have knowledge of requirements elicitation techniques, based on systematic processes and methods. We propose the Software Requirements Analysis Method for Improvement of Privacy and Security (SRAM-PS), which is based on concepts and techniques from Organizational Semiotics and on the analysis of information security and data privacy standards. SRAM-PS is a 7-steps systematic approach where an input set of software requirements is analyzed, processed, and then enriched with new security and privacy requirements. A case study with 4 experts was carried out, where SRAM-PS is used in a real world scenario: a bank sends a financial transaction receipt containing the customer’s personal data over the Internet. SRAM-PS is aimed at researchers and engineers who analyze and specify software requirements and need to systematize their methods and techniques.

Leonardo Manoel Mendes, Ferrucio de Franco Rosa, Rodrigo Bonacin

Chapter 12. Efficient Design of Underwater Acoustic Sensor Networks Communication for Delay Sensitive Applications over Multi-hop

Underwater Acoustic Sensor Networks (UASNs) play a critical role in the remote monitoring of a wide range of time-sensitive underwater applications, such as in the oil/gas pipeline to avoid oil spills. In this type of application, the transmission of collected information to the onshore infrastructure within a period of time is critical. Despite the advantages of UASNs over the limitations of Terrestrial Wireless Sensor Networks (TWSNs), the applicability of UASNs in different use-cases requires further investigation. In this paper, we investigate different MAC protocols and study the impact of non-environmental factors that may degrade performance. We simulate different MAC protocol approaches based on available underwater commercial modems to find the most efficient MAC protocol approach for the oil/gas industry based on core performance metrics. Our extensive simulation results show that the contention-based random access approach is the most suitable for time-sensitive application where the Network Size (NS) followed by Network Load (NL), Data Rate (DR), and Packet Size (PS), respectively have the strongest impact on delay.

Ahmed Al Guqhaiman, Oluwatobi Akanbi, Amer Aljaedi, C. Edward Chow

Cybersecurity II

Frontmatter

Chapter 13. Parallelized C Implementation of a Merkle Tree

Merkle trees are primarily known for being an attribute found in blockchain technology. They are used for encrypting data by hashing values multiple times to avoid incidents such as hash collisions, or the successful guessing of hash values. Merkle trees are not only a useful feature found on the blockchain but in the field of Cyber Security in general. This paper outlines the process of implementing a Merkle tree as a data structure in C++ and then parallelizing it using OpenMP. The final result is a Merkle tree password storing program with reduced running-time and the ability to operate on multiple processors. The validity of this program is tested by creating a Merkle tree of the correct passwords, storing the value of the root node, and then building a second tree where a single incorrect password is stored within that tree. The two trees are passed through an audit function that compares the root nodes of the two trees. If they are different, then the tree in question has been tampered with.

Andrew Flangas, Autumn Cuellar, Michael Reyes, Frederick C. Harris

Chapter 14. A Study on Ontologies of Vulnerabilities and Attacks on VLAN

Virtual Local Area Network (VLAN) is a technology capable of separating networks into specific domains. Attacks on VLANs could affect computing environments causing service interruptions. These attacks exploit vulnerabilities and operating characteristics of VLANs to gain access to critical information. Conceptual modeling of vulnerabilities and attacks related to VLANs is crucial to enable the construction of systematic methods and techniques for protecting critical infrastructures. Ontologies can contribute in this context, as they are modeling tools that enable the formalization of the main concepts and their relationships, in addition to enabling the creation of semantic rules that can be used by intelligent systems. We present a quasi-systematic literature review aiming at describing and classifying studies on ontologies of vulnerabilities and attacks on VLANs. The approach used in this review allowed for the verification and analysis of trends, as well as it uncovers the technological approaches adopted over the past 10 years. The main contributions of this review are: i) a description of the most recent ontologies, taxonomies, techniques and theories, in addition to the contributions and limitations of proposals in the literature; and ii) the identification of gaps in the literature and research challenges. Searches were carried out in the main scientific knowledge bases in the field of computing. Two hundred sixty-nine articles were found; 19 studies were analyzed according to their approaches, themes and related terms, pointing out contributions and research issues. This article is intended for researchers looking to conceptually model vulnerabilities and attacks on networks.

Marcio Silva Cruz, Ferrucio de Franco Rosa, Mario Jino

Chapter 15. Towards a Symmetric Crypto Algorithm: The HAJ

In symmetric encryption, the algorithm and secret key determine the security factor. This paper presents the idea to create a multiple (7 times) and block-separated encryption algorithm. To achieve this, we will use the Hill Cipher and Gauss-Jacques methods. In addition to the above, our most significant contribution will be to obtain large secret keys, which will allow us to obtain as a possible result a meaningful approximation to the Shannon perfect secrecy, as well as the reduction of computational complexity and the verification of security through anti-bot mechanisms such as code breakers.

Daniel Alarcón-Narváez, Fausto A. Jacques García

Chapter 16. A Comparative Study Between Two Numerical Methods for Symmetric Cryptography Uses and Applications

This document is focused in the comparison of two matrix numerical methods for symmetric cryptography, from a computational perspective in terms of memory, complexity and processing. The main task is to identify the most appropriate method along with Hill Cipher and form an improved cryptosystem. The methods are known as Gauss-Jacques and Gauss-Jordan with explicit modularization. Both of them could be used for the processing of secret keys in the approach to Shannon’s Perfect Secrecy, which is of vital importance in terms of security and information protection. The experimental method is used to evaluate and analyze the behavior of each method in RAM consumption, computational complexity and processing, through their implementation in a functional language.

Alba Nidia Martínez-Martínez, Fausto A. Jacques García

Chapter 17. Speed Up Over the Rainbow

Rainbow is a Digital Signature Algorithm (DSA) that is based on multivariate polynomials. It is one of the Round-3 candidates of the NIST’s Post-Quantum Cryptography Standardization project. Its computations rely heavily on GF(28) arithmetic and the Rainbow submission optimizes the code by using AVX2 shuffle and permute instructions. In this paper, we show a new optimization that leverages: (a) AVX512 architecture; (b) the latest processor capabilities Galois Field New Instructions (GF-NI), available on Intel “Ice Lake” processor. We achieved a speedup of 2.43 ×/3.13 ×/0.64 × for key generation/signing/verifying, respectively. We also propose a variation of Rainbow, with equivalent security, using a different representation of GF(28). With this variant, we achieve a speedup of 2.44 ×/4.7 ×/2.1 × for key generation/signing/verifying, respectively.

Nir Drucker, Shay Gueron

Chapter 18. Extending a Hybrid Security Risk Assessment Model with CWSS

Cybersecurity risk management is the foundation of business and organizational decisions involving digital technology. Various models have been proposed and are in use, but these apply to current technologies and use cases, and none are sufficient to evaluate new technologies. This paper builds upon prior work using CVSS to quantify potential security threats for which information is limited. That prior work merges CVSS data with MITRE’s Common Attack Pattern Enumeration and Classification (CAPEC™) tools to inform a new technology risk scoring system in a Bayesian Belief Network (BBN). This work extends this risk model to incorporate CWSS data to better reflect the environments’ weaknesses that may apply to new technologies. This approach enables a more accurate and trustworthy way of quantitatively estimating risk as a function of the Base Finding Subscore and Attack Surface Subscore for weaknesses most relevant to businesses, missions, and deployed technologies.

Robert Banks, Jim Jones, Noha Hazzazi, Pete Garcia, Russell Zimmermann

E-Health

Frontmatter

Chapter 19. Identifying and Prioritizing Applications of Internet of Things in the Supply Chain of Distribution and Sale of Health Care Products in Iran

Applying Internet of Things (IoT) leads to improvements in the quality of life. Supply chains are considered as the world’s most fundamental parts, for increasing productivity of which, countless efforts are applied. Although rare studies are conducted on the IoT’s application, no study was found on the distribution and sale part of the supply chain. Solving challenges in this area is lucrative for service provider companies since products of this industry have a considerable share in the family shopping basket. Firstly, properties and advantages of the IoT and its maturity level are studied. Since technology implementation depends on the context, the challenges of health care products’ supply chain in the distribution and sale sectors along with corresponding solutions to overcome such challenges are addressed through interviews with 23 experts in two groups of IoT and Health Care Products Industry. Interviews are analyzed by the content analysis method. As implementing such technologies are not practicable unless the technology is matured in the country, each application is prioritized according to its corresponding technology maturity. The result is applicable for the Health Care Industry and investors aiming to develop their new technology-based industry.

Niloofar AminiKalibar, Fatemeh Saghafi

Chapter 20. The Role of Information Technology in Patient Engagement

Patient Engagement (PE) promotes the patient’s interaction with and contribution to all aspects of care, where patients play an active and informed role in improving healthcare systems, enhancing health outcomes, and avoiding extra-costs, in addition to individual care decisions. Understanding the PE concept is essential for e-health professionals to adopt solutions to interacting intensely with patients. To identify the gaps in stakeholders’— particularly e-health people—worldviews, we conducted a scoping review of the evidence that has been published between 2010–2020. In this review, we included published PE articles that focused on the role of information technology. Our findings showed that stakeholders’ solutions have focused primarily on clinical records, communications, education, adherence, and recently, artificial intelligence to optimize the services. The authors focused their attention on the care’s aspects regarding cognitive, emotional, economic, behavioral, lifestyle, or wellness dimensions.Reviewed evidence rarely emphasizes the patients’ role in changing organizational policies, care redesign, or healthcare service improvements. We propose a model to develop PE by multi-stakeholder efforts and interrelated capabilities by coordinating diverse engagement tactics into a seamless orchestration, using versatile Information Technology (IT).

Sima Marzban, Paul Meade, Marziye Najafi, Hossein Zare

Chapter 21. Ambient Intelligence Technologies for Visually Impaired: A Mapping Study

We conducted a mapping study to investigate how the scientific community uses Ambient intelligence technologies to assist visually impaired people. Our initial search identified a total of 807 publications; after applying our selection criteria, we accepted 65 publications. We seek to show which technologies, methodologies, techniques, architectures, features, and evaluations are most used. Results indicated that only 15.53% have a specific focus on people who are visually impaired. Most of the results were published in 2015. Most authors used their own architectures. Many results used techniques such as Detection and recognition, Artificial Intelligence, Networking, and others.

Juliana Damasio Oliveira, João A. L. de Moraes Junior, Rafael H. Bordini

Chapter 22. Voice for the Voiceless: Developing a Low-Cost Open-Source Communication Device for the Speech Impaired

Advancements in the affordability and availability of Internet of Things (IoT) devices have led to incredible innovations in the field of speech-generating devices. The introduction of devices such as tablets, cellphones, and mobile computers has allowed individuals with speech disorders to have a medium in which to easily communicate without purchasing expensive and specialized medical equipment. Even though these devices increased access to this technology, it is still out of reach for many. Whether it is the cost of the device, cost of the speech generating software, or access to a reliable internet connection, this technology is inaccessible to some of the people who need it most.The focus of this project is bringing this voice technology to the oppressed and disadvantaged by creating a fully open-source device assembled with off the shelf parts that is a fraction of the cost of similar alternatives. The goal was to accomplish this with minimal compromises to quality and usability. This was achieved with a Raspberry Pi computer, touch screen, battery pack, and a plastic casing. It overall met the quality and usability expectations of an Alternative Communication (AAC) device for around $150 USD and shows that traditionally expensive AAC equipment can be made more accessible to people, without compromising usability. It hopefully will motivate others to research areas where off the shelf parts and open-source software can be used to increase the accessibility of otherwise expensive and specialized technologies to benefit the lives of others.

Travis Smith, Vasilios Pappademetriou

Chapter 23. Integration of Bioinformatics and Clinical Data to Personalized Precision Medicine

The incorporation of bioinformatics data in clinical practice is almost happening, mainly due to the rapid growth and access to next-generation sequencing. Thus, personalized precision medicine receives private and government incentives to improve the quality of life and the well-being of the population in the long term. We explore initiatives that seek to map the genome and integrate them with population clinical data, as well as its benefits. Then, we describe the need to develop advanced protocols for exchanging information and platforms capable of processing clinical and genomic data, seeking to increase the well-being of the patients.

Flavielle Blanco Marques, Gabriel Fernandes Leal, Giovani Nicolas Bettoni, Osmar Norberto de Souza

Chapter 24. Modeling the COVID-19 Epidemic in a Parallelized City Simulation

A pandemic can arise without warning, and it is important for those in charge of managing the outbreak to understand how diseases spread. Being able to simulate the spread of a disease in varying environments can help the world be more prepared when an outbreak occurs. The COVID-19 City Simulator allows the user to test the spread of the virus under multiple different scenarios. Parallel computing can help to make these simulations more efficient by allowing data to be gathered at a faster rate on a particle simulation. This paper shows how OpenMP and MPI can improve a pandemic simulation by cutting the runtime from over 25 s to under 10 s when 4 threads and 4 boxes are used. We also find that the speed of implementing a lockdown largely impacts the amount of cases and deaths in the city.

Derek Stratton, William Garner, Terra Williams, Frederick C. Harris

Management and Applications

Frontmatter

Chapter 25. Techniques and Tools for Selection and Strategic Alignment of Projects and Projects Portfolio Balancing: A Systematic Mapping

The optimization of resources used in projects, whether in the context of private companies or in the context of public management, is directly related to efficient the selection and strategic alignment of projects and portfolio balancing through appropriate techniques and tools. This work describes a systematic mapping carried out with the purpose of identifying which tools and techniques are used (or are more appropriate) for selection and strategic alignment of projects and project portfolio balancing. The research was conducted from the digital libraries Scopus and IEEE, resulting, initially, in a total of 128 articles. After applying the filters and the exclusion and inclusion criteria adopted, the study was restricted to 11 articles in Scopus and 11 in IEEE, in addition to 1 more article that was included by snowball. The research made it possible to verify that the vast majority of the techniques used are Multi-Criteria Decision Support Methods and models that use fuzzy logic, in addition to Evolutionary Algorithms as the main tool.

Djenane C. S. dos Santos, Adler D. de Souza, Flávio B. S. Mota

Chapter 26. Methods for Detecting Fraud in Civil and Military Service Examinations: A Systematic Mapping

Civil and military service examinations are carried out in several countries for the recruitment and admission of public servants in various spheres/levels of government. This is considered an effective and rational method for selection based on merit. Due to the constant economic variations and the stability provided by public offices, the interest in some offered positions can be huge. Criminals specialized in defrauding public examinations offer candidates the possibility of facilitated and illegal admission. Various types of information could be submitted to methods and techniques (e.g., application data, test performance, geodata, etc.) to detect fraud. We present a systematic mapping of the literature on fraud detection methods in several domains, which can be adapted and improved to detect fraud in public examinations. 31 articles were identified, and after analysis, 19 selected works were analyzed and classified. The usages of machine learning and data mining techniques were uppermost methods adopted in the analyzed papers. This work is aimed at researchers who seek to develop fraud detection techniques in admission exams.

Roberto Paulo Moreira Nunes, Rodrigo Bonacin, Ferrucio de Franco Rosa

Chapter 27. Citizens Engagement in Smart Cities: A Systematic Mapping Review

Although the term Smart City has a wide range of definitions, it is a consensus in the bibliography that public participation is essential. However, cities are not always successful in ensuring citizen involvement, which can compromise governance efficiency. This work presents a Literature Systematic Mapping that aims to understand what are the aspects that influence citizens’ interest to engage in public participation policies, how these policies are being implemented, and how citizens’ engagement affects the governance of a smart city. To accomplish these goals, 33 academic papers were selected through a rigorous search protocol.

Rafael Leite, Adler Diniz, Melise De Paula

Chapter 28. Use of Crowdsourcing Questionnaires to Validate the Requirements of an Application for Pet Management

Mobile applications usually fail to deliver the right set of features for its users, either by not offering necessary functionalities or by offering unnecessary ones. In the last few years, some modern Requirements Engineering methods have been created to better design mobile applications. One of these methodologies is the Crowdsourcing Requirements Engineering, based on short cycles of implementations and feedbacks by a large group of actual users. This work aims to validate the requirements of a mobile application for the management of domestic animals, through the use of crowndsourcing questionnaires. Two questionnaires were designed and implemented to assess the needs of the users. The first questionnaire verified functionalities provided by similar softwares, while the second one verified requirements established by the authors.From the analysis of the data gathered, all requirements, except for one, were validated. Finally, after the final functionalities of the software were defined, an alpha version of the software could be created.

Vitor S. Vidal, Marco Aurélio M. Suriani, Rodrigo A. S. Braga, Ana Carolina O. Santos, Otávio S. Silva, Roger J. Campos

Chapter 29. Discovery of Real World Context Event Patterns for Smartphone Devices Using Conditional Random Fields

Mobile applications are Event Driven Systems that react to user events and context events (e.g. changes in network connectivity, battery level, etc.) The large number of context events complicate the testing process. Context events may modify several context variables (e.g. screen orientation, connectivity status, etc.) that affect the behavior of an application. This work examines a data set of real-world context changes on Android phones. We collect every context event that occurs on the mobile devices of 58 Android users over 30 days to identify complex relationships and patterns. This work uses Machine Learning (ML) techniques including Conditional Random Fields (CRFs) and Deep Neural Networks (DNNs) to predict sequence labels for context events. These techniques are compared to Majority Baseline (MB). The trade-offs among these methods reveal that CRF is the most effective technique for sequence prediction/labeling of the data-set. Future work may apply the data collection strategy and ML techniques to domains for emerging technologies in areas such as Internet of Thing, smartwatches, and autonomous vehicles.

Shraddha Piparia, Md Khorrom Khan, Renée Bryce

Chapter 30. Computation at the Edge with WebAssembly

The introduction of WebAssembly in 2017 opened a new door for performing computation in the browser at 0.9 the speed of C/C++ code (Haas et al. ACM SIGPLAN Notices 52(6), 185–200, 2017). As browsers are the most ubiquitous software, it is now possible to build universal applications that run on every machine that has a web browser installed on it. In this paper we propose a design to build web applications that take advantage of the new performance capabilities in the browser. We also implemented this design and showed that it increases the overall performance of the web applications in our experiment.

Jebreel Alamari, C. Edward Chow

Chapter 31. Analysis of Traffic Based on Signals Using Different Feature Inputs

The Regional Transportation Commission of Southern Nevada (RTC) manages vehicle traffic in Clark County, Nevada. The RTC division that manages the roadway and freeway street signal devices is the division of Freeway Arterial System of Transportation (FAST). They capture and store their maintenance vehicle trucks’ GPS trip information into a publicly available data set. This data set brings a tremendous opportunity to analyze the traffic trends. This study uses machine learning algorithms to solve two types of problems, classification and regression. For classification, we have obtained 65% accuracy using signal types as the target and significantly low accuracy when we use the total stop time as the target variable. For regression, both Random Forest and Support Vector models performed poorly. However, the Random Forest model performed slightly better than the Support Vector model. Altogether these results help us determine the next steps to explore.

Baby Shalini Ravilla, Wolfgang Bein, Yazmin Elizabet Martinez

Theory and Computation

Frontmatter

Chapter 32. Effect of Boundary Approximation on Visibility

The problem of simplifying a complex shape with simpler ones is an important research area in computer science and engineering. In this paper we investigate the effect of visibility properties of polygons when their boundaries are approximated to have simpler shapes. The presented algorithm is expected to have wide applications in simplifying 1.5D terrain which are restricted class of simple polygons.

Laxmi Gewali, Samridhi Jha

Chapter 33. A Method for Improving Memory Efficiency of the Reachability Graph Generation Process in General Petri Nets

A Petri net is a mathematical model representing the behavior of adiscrete event system, and analysis of this model allows us to verify various properties of the system. When the state space is created for a net, the memory use increases when the number of states increases explosively. In this paper, we propose a method for reducing memory use during process execution by detecting and removing deletable (unused) state values among all state values in the hash map holding the state space. To judge whether a state can be deleted, it is necessary to detect routes by which the state can be reached. We developed a new method for detecting reachable paths by applying a reachability determination with a basic closed-loop matrix for the marking obtained by backward firing the transitions to the current state. A state generator was equipped with the proposed method and implemented in a Hierarchical Petri net Simulator developed at our university. We also tried to shorten the execution time by parallelizing generators equipped with the proposed method. We compared and evaluated the memory use and the execution time of the removable state deletion generator and conventional generators. The result shows suppressing runtime memory use during Petri net state space generation by detecting and deleting removable state values. Moreover, the state space generation algorithm was parallelized in an attempt to shorten execution time according to the proposed technique.

Kohei Fujimori, Katsumi Wasaki

Chapter 34. An Evaluation for Online Power Down Systems Using Piece-Wise Linear Strategies

Power-down mechanisms are well known and are widely used to save energy; these mechanisms are encountered on an everyday basis. We consider a device which has a set of states defined by a continuous function. The state of the device can be switched at any time. In any state the device consumes energy and a power up cost to switch the machine to full capacity. This project gives experimental results regarding power consumption to satisfy service request based on online competitive analysis. Competitive ratios, which show the effectiveness of the algorithms compare to the optimal solution.

James Andro-Vasko, Wolfgang Bein

Chapter 35. Mobile Test Suite Generation via Combinatorial Sequences

Mobile applications are event driven systems that are often driven primarily by user interactions through a GUI. The large event space for mobile applications poses challenges for testing. This work considers the architecture of modern mobile applications to generate test cases that systematically incorporate activities, elements, and events in different sequences for testing. In particular, this work optimizes the presence of different combinatorial-based element sequences of size t = 2 inside test suites. The test suites demonstrate increased exploration depth when compared to randomly generated tests, interacting with a wide range of elements and increase code coverage by 0.29–6.79%.

Ryan Michaels, Md Khorrom Khan, Renée Bryce

Chapter 36. Usability Smells: A Systematic Review

In software engineering, bad smells are symptoms that something may be wrong in the system design or code. The term bad smell usually describes a potential problem with known consequences, and also known solutions. In the case of usability, a bad smell is a hint of poor interface design, that makes it difficult for the end-user to accomplish common tasks. This paper presents the most relevant usability smells reported in the literature, as well as tools, methods and techniques that support the process of detecting usability anomalies.

Jonathan de Oliveira T. Souza, Adler Diniz de Souza, Leandro G. Vasconcelos, Laercio A. Baldochi

High Performance Computing Architectures

Frontmatter

Chapter 37. Motivating Computer Science Students in Lower-Division Undergraduate Curriculum

The student retention in undergraduate computer science degrees has been decreasing over the past 15 years where an attrition rate as high as 30% to 40% was observed during that period, with most students leaving the field after taking some introductory courses. Observing a similar trend or worse could be possible during or after the COVID-19 pandemic. A possible solution for this problem is to introduce highly motivating topics in the lower-division courses that will keep the students intrigued and wanting to learn more, and eventually, register in higher division courses. In this paper we provide a systematic review of the topics and subjects adopted in many institutions to motivate students of computer science in lower-division undergraduate curriculum. Then, we summarize the common attributes of these motivating subjects. In addition, we propose to leverage data processing subjects or big data problems to motivate college students to learn in lower-division courses. Based on the attributes of the existing motivating topics and the method of logic inference, we show that it is feasible and effective to motivate students’ learning by injecting data processing subjects or big data problems in lower-division computer science courses.

Yun Tian, Saqer Alhloul, Fangyang Shen, Yanqing Ji

Chapter 38. Evaluation of Power Consumption and Application Optimization for Adaptive-Ticks Feature in Linux Kernel

Scheduler timer architecture has significant impact on operating system performance and power consumption. The current generation of Linux kernel supports multiple timer implementations, including periodic ticks, Dyntick-idle and Adaptive-ticks. Adaptive-ticks kernel offers the benefits of previous generations with additional improvement in power consumption and performance. In this paper, we evaluate the impact of Adaptive-ticks on power consumption with Linux kernel version 5.4.0 on an Intel Core i9-9900K. The current generation of Adaptive-ticks feature does not support multiple tasks in a ready queue; however, with the increase in application parallelism, not having support for multiple tasks in the ready queue poses a significant disadvantage to this feature. To support multi-threaded applications, we propose an application optimization technique which splits threads into two main categories, lightweight and heavyweight, with proper affinity settings for better power consumption. In addition, this study proposes a possible implementation strategy to extend the Adaptive-ticks feature to support multiple tasks in the ready queue. Our tests use in-band “RAPL” for profiling power consumption, and synthetic benchmarks such as Livermore, RAMSpeed, and SysBench, as the workloads. For real-world application benchmarking, we use Linux kernel compilation. The study shows that Adaptive-ticks kernel can reduce power consumption by 1–2.7% and the application optimization technique provides a 2.4% enhancement in power consumption.

Abdullah Aljuhni, Shaji Yusuf, C. Edward Chow, Oluwatobi Akanbi, Amer Aljaedi

Chapter 39. Sequence Alignment Algorithms in Hardware Implementation: A Systematic Mapping of the Literature

One of the primary activities of the bioinformatics is the alignment of sequences, which can find similar patterns between two sequences and determine their structure, their control information, or even their own functions. The growth of databases increases the computational effort spent to execute sequence alignment algorithms. Consequently, it can take a considerably long time to run these algorithms on general purpose processors. This paper aims to map and analyze articles related to the implementation of sequence alignment algorithms in FPGAs and GPUs to identify the most recent findings on the subject, as well as possible gaps that may lead to further investigations. The systematic mapping led to the selection of twenty-three articles using FPGA and the GPU as the hardware platform. It also identified six sequence alignment algorithms: Needleman-Wunsch, Smith-Waterman, HMM, BLAST, BWA e FMIndex. The present work was able to evaluate how often these hardware and algorithms are being used in scientific researches and their benefits in terms of processing time and energy consumption.

Lucas S. M. Bragança, Adler D. Souza, Rodrigo A. S. Braga, Marco Aurélio M. Suriani, Rodrigo M. C. Dias

Chapter 40. Hardware Logic Library and High-Level Logic Synthesizer Combining LOTOS and a Functional Programming Language

LOTOS is a formal description language for parallel computing models based on process algebra. There is an arithmetic element library, DILL, for describing logic circuits on LOTOS. However, due to the LOTOS description’s rigor and the classical m4 macro language expansion method, full-scale circuit design has been complicated. This study aims to improve the hardware upper-level design’s descriptiveness by constructing a new arithmetic element library using syntax sugar written in OCaml, a functional language called HDCaml, and converting it LOTOS code while using the features of a high-level description language. We have developed a LOTOS code generator and the comprehensive library, as OpenDill, which consists of primary logic gate elements (AND, OR, and selector). Ocaml has various library construction features, such as class management, reusability, and data-driven dynamic generation of instances. In the future, we plan to build a library of complex circuit elements (fast adder, ALU) in HDCaml description.

Katsumi Wasaki

Chapter 41. GPU Acceleration of Sparse Neural Networks

In this paper, we use graphics processing units (GPU) to accelerate sparse and arbitrary structured neural networks. Sparse networks have nodes in the network that are not fully connected with nodes in preceding and following layers, and arbitrary structure neural networks have different number of nodes in each layers. Sparse Neural networks with arbitrary structures are generally created in the processes like neural network pruning and evolutionary machine learning strategies. We show that we can gain significant speedup for full activation of such neural networks using graphical processing units. We do a prepossessing step to determine dependency groups for all the nodes in a network, and use that information to guide the progression of activation in the neural network. Then we compute activation for each nodes in its own separate thread in the GPU, which allows for massive parallelization. We use CUDA framework to implement our approach and compare the results of sequential and GPU implementations. Our results show that the activation of sparse neural networks lends very well to GPU acceleration and can help speed up machine learning strategies which generate such networks or other processes that have similar structure.

Aavaas Gajurel, Sushil J. Louis, Rui Wu, Lee Barford, Frederick C. Harris

Chapter 42. Parallelizing the Slant Stack Transform with CUDA

In geophysics, the slant stack transform is a method used to align signals from different sensors. We focus on the use of the transform within passive refraction microtremor (ReMi) surveys, in order to produce high resolution slowness-frequency plots for use as samples in a machine learning model. Running on a single central processing unit (CPU) thread, this process takes approximately 45 min, 99.5% of which consists of the slant stack transform. In order to reduce the time taken to perform the transform, we use NVIDIA CUDA programming model. Using the same CPU, augmented with a GeForce RTX 2080 Ti we were able to reduce this time down to as little as 0.5 s.

Dustin Barnes, Andrew McIntyre, Sui Cheung, John Louie, Emily Hand, Frederick C. Harris

Social Computing/E-Learning

Frontmatter

Chapter 43. Recommender Systems Evaluator: A Framework for Evaluating the Performance of Recommender Systems

Recommender systems are filters that suggest products of interest to customers, which may positively impact sales. Nowadays, there is a multitude of algorithms for recommender systems, and their performance varies widely. So it is crucial to choose the most suitable option given a situation, but it is not a trivial task. In this context, we propose the Recommender Systems Evaluator (RSE): a framework aimed to accomplish an offline performance evaluation of recommender systems. We argue that the usage of a proper methodology is crucial when evaluating the available options. However, it is frequently overlooked, leading to inconsistent results. To help appraisers draw reliable conclusions, RSE is based on statistical concepts and displays results intuitively. A comparative study of classical recommendation algorithms is presented as an evaluation, highlighting RSE’s critical features.

Paulo V. G. dos Santos, Bruno Tardiole Kuehne, Bruno G. Batista, Dionisio M. Leite, Maycon L. M. Peixoto, Edmilson Marmo Moreira, Stephan Reiff-Marganiec

Chapter 44. Visualization of Georeferenced Data Through the Web: A Systematic Literature Review

The internet has the role of sharing data of the most diverse natures in a fast and accessible way. Some data carries position information within the geographic space. One of the ways used to disseminate this type of data is the Geographic Information System on Web (GIS-Web), which provides a visualization of the data based on its position in the geographic area. Currently, GIS-Webs are present in several scenarios, with an intention to plan or predict the future perspective for rated scenario. Given this, a search was performed in the bibliography on applications of this type of solution, a search was made in 3 repositories with a final result of 15 articles that provided information about which tools or techniques were used, and in which scenario the availability through GIS-Web was used. The illuminated studies of the growing search for the visualization of georeferenced data using elements in 3 dimensions to represent, for example, buildings and sewage or water pipes. The great use of libraries and APIs was identified when making the data available, but the biggest job is in the treatment of raw data that are particular to each application environment. This work presents a systematic literature review, where it is possible to identify possible strategies to be adopted in future works that use georeferenced data.

Lucas Lamounier Gonçalves Duarte, Adler Diniz de Souza

Chapter 45. Cognitive Issues in Intelligent Modeling of Pedagogical Task

The article describes the process of intelligent modeling of pedagogical situations using artificial neural networks, built on the basis of the analysis of cognitive patterns of human information processing, allows the development of effective decision support systems and forecasting learning systems. The experience of creating neural network systems is presented: to predict the success/failure of a student’s project activities with the development of recommendations for selecting a perspective project task; to predict student decision to attend/skip classes based on student personal qualities, aims and lesson schedules.

Marina Lapenok, Anna Lozinskaya, Vasilisa Likhacheva

Chapter 46. Immersive Virtual Reality and Its Use in Developing Empathy in Undergraduate Students

In recent years, Immersive Virtual Reality (IVR) has taken on great relevance in the academic and scientific world, especially in relation to the development of empathy. Virtual Reality (VR) allows taking perspective of others through virtual environments and, through this experience it is plausible to think we can promote empathic capacity. This work is part of an ongoing doctoral research and presents the initial findings about the topic and problem statement through a diagnosis and literature review on VR for the development of empathic responses. Findings include reassertion about the general potential of VR, continuous usage of the Interpersonal Reactivity Index (IRI) to quantify empathy and, the lack of rigorous evidence to prove VR could be the ultimate empathy machine.

Éder Estrada Villalba, Fausto Abraham Jacques-García

Chapter 47. E-NEST Remote Learning Transition in STEM Education

This research analyzes how remote learning models are utilized in STEM Education. The E-NEST project developed online teaching models to instruct teaching interns during the unprecedented times of the coronavirus pandemic including mentorships, internships and culturally responsive teaching summer workshops. Based on key findings from data collection and program evaluations from the National Science Foundation Robert Noyce Teacher Scholarship program, a comprehensive online learning classroom was created to teach cultural diversity in STEM Education with modified project management and recruitment approaches. As a result, E-NEST online internships and professional development workshops were effective and promoted student achievement during the transition to remote learning. The project team learned the functions of online apps to instruct students and gained experience in facilitating online learning classes.

Fangyang Shen, Janine Roccosalvo, Jun Zhang, Yun Tian, Yang Yi, Yanqing Ji, Ashwin Satyanarayana, Xiangdong Li, Ahmet Mete Kok, Annie Han, Hon Jie Teo

Chapter 48. Ethics and Human Values in the Software Design

Technology is changing the world, society, and people. For many, there is no conceivable living without it. Its evolution has had exponential growth, that is why many of the futuristic scenarios are no longer so and we are facing an inevitable human transformation; such as, the impact of these new technologies on human life. It is necessary to reflect what the design considerations have been or if there was an analysis that incorporated human and ethical values in its evaluation was carried out. This takes us a little further back to the training of the engineer, who was taught to evaluate technically and economically a technological project, but a methodology to evaluate the scope and impacts that this technological project has were not included in his training. Thus, it is vital to teach how to reflect carefully on the consequences and effects it causes to nature and society. The aim of this research is to offer a purposeful assessment of the incorporation of ethics and human values linked to the training of an engineer in the technological area.

Alejandra Acuña, César Collazos, Cristian Barría

Pandemic

Frontmatter

Chapter 49. Using UAV, IoMT and AI for Monitoring and Supplying of COVID-19 Patients

The Corona Virus Disease 2019 (COVID-19) is an infectious disease caused by a newly discovered corona virus SARS-CoV-2. It is similar to the flu and raises concerns about the alarming levels of spread and severity, resulting in a continuous pandemic worldwide. In eight months, it infected 90 million people worldwide and more than 2 million died. Unmanned Aerial Vehicles (UAVs) can be very useful in supporting logistical support for the COVID-19 pandemic. This work aims to investigate UAV-based applications for logistical support to situations caused by the COVID-19 pandemic and proposes an architecture to deal with pandemic situations in different scenarios using real-time case studies.

A. J. Dantas, L. D. Jesus, A. C. B. Ramos, P. Hokama, F. Mora-Camino, R. Katarya, O. P. Verma, P. K. Grupta, G. Singh, K. Ouahada

Chapter 50. A Comprehensive Analysis of SARS-CoV-2 in India

Modeling, Prediction and Vaccination

In this paper, we conduct mathematical and statistical analysis to answer the following questions for COVID-19 : (Q1) How has the pandemic progressed since March 2020? (Q2) How effective have the social distancing approaches been? (Q3) How has the pandemic spread differently in different Indian states? (Q4) What would be the impact of a vaccination campaign on the current state of the disease spread? (Q5) What are the economic and ethical nuances that need to be considered when undertaking such a campaign? (Q6) Whom do we prioritize for vaccination? (Q7) How long does the immunity from the vaccine likely to last? (Q8) How important is it to track the mutations of the virus?

Debdeep Dey, Sarangi Patel, Karasani Tharun Kumar Reddy, Siddharth Sen

Chapter 51. Virtual Hospital: A System for Remote Monitoring of Patients with COVID-19

Coronavirus disease represents a global public health concern. To minimize its damage is necessary to create technologies for the prevention and control of this emerging disease. However, there is still no equipment that can remotely monitor COVID-19 patients, regardless of where the patient is. Within the scope of the Internet of Things, this article presents the Virtual Hospital, a system for remote monitoring of patients with COVID-19. The device features biosensors that are configured to monitor the symptoms of COVID-19, taking into account the specification of the NEWS2 protocol, among them: temperature, heart rate, blood oxygenation, systolic pressure, level of consciousness, and respiratory rate. With the monitoring system, health professionals will continually monitor patients without the need for physical contact because any anomaly that may occur will cause the system to notify the person responsible for taking appropriate action immediately. We believe that, as a consequence, its use can decrease the number of critically ill patients in hospitals, reducing in-hospital mortality rates.

Vanessa Stangherlin Machado Paixão-Cortes, Walter Ritzel Paixão-Cortes, Dorval Thomaz, Felipe de Siqueira Zanella, Ricardo Luís Ravazzolo, Gerson Luis da Silva Laureano

Chapter 52. Single-Cell RNA Sequencing Data Imputation Using Deep Neural Network

Recent research in biology has shifted the focus toward single-cell data analysis. The new single-cell technologies have allowed us to monitor and characterize cells in early embryonic stage and in heterogeneous tumor tissue. However, current single-cell RNA sequencing (scRNA-seq) technologies still need to overcome significant challenges to ensure accurate measurement of gene expression. One critical challenge is to address the dropout event. Due to the low amount of starting material, a large portion of expression values in scRNA-seq data is missing and reported as zeros. These missing values can greatly affect the accuracy of downstream analysis. Here we introduce scIRN, a neural network-based approach, that can reliably recover the missing values in single-cell data and thus can effectively improve the performance of downstream analyses. To impute the dropouts in single-cell data, we build a neural network that consists of two sub-networks: imputation sub-network and quality assessment sub-network. We compare scIRN with state-of-the-art imputation methods using 10 scRNA-seq datasets. In our extensive analysis, scIRN outperforms existing imputation methods in improving the identification of cell sub-populations and the quality of visualizing transcriptome landscape.

Duc Tran, Frederick C. Harris, Bang Tran, Nam Sy Vo, Hung Nguyen, Tin Nguyen

Blockchain Technology

Frontmatter

Chapter 53. Blockchain and IoT: A Systematic Literature Review for Access Control Issues

IoT has been one of the emerging technologies for the past decade because of its facilities and it’s also becoming widely used for monitoring purposes. Therefore, many questions are rising about the security of devices that can operate either in an everyday life or industry. It was found that IoT networks are using low-safety communication protocols that might be questionable and this paper reveals fair alternatives to replace these technologies using blockchain and smart contracts that will enforce the protection of these devices from security violation by setting up access control policies and roles to every peer in the network.

André Mury de Carvalho, Bruno Guazzelli Batista, Adler Diniz de Souza

Chapter 54. A Bitcoin Wallet Security System (BWSS)

The Bitcoin technique faces several security challenges such as bitcoin wallet damage and attacks. This research presents a Bitcoin Wallet Security System (BWSS) to ensure the confidentiality and integrity of bitcoin wallet. The proposed system functions are distributed into two components and then implemented using Java programming language. A two-factor authentication is used to provide a more confidential bitcoin wallet while a backup mechanism is used to back up the wallet automatically or manually. Finally, the system is evaluated and refined after evaluation to come out with the required security features in a secure manner. The proposed system enables the bitcoin wallet owner to first protect his/her bitcoin wallet(s) with a second security layer using a two factor authentication and, second, back up his/her wallet in a secure and flexible ways. The system can be improved in the future to secure bitcoin wallets in mobile devices.

Ibrahım Alkhammash, Waleed Halboob

Chapter 55. Disruptive Technologies for Disruptive Innovations: Challenges and Opportunities

Disruptive technologies continuously and significantly alter the way people communicate and collaborate as well as the way industries operate today and in the future. To create new business models and opportunities, several combinations of disruptive technologies are being introduced nowadays. Among these technologies, cloud computing, IoT, Blockchain, artificial intelligence, social networks and media, big data, and 5G are mostly used. For instance, Blockchain technology made distributed solutions feasible and popular. On the other hand, big data and the social media are two contemporary technologies which lament significant impact on business and society. This paper presents a holistic approach to integration perspectives of these technologies considering many challenges like security and privacy. This paper also surveys the most relevant work in order to analyze how some of these technologies could potentially improve each other.

Amjad Gawanmeh, Jamal N. Al-Karaki

Chapter 56. Framework for Securing Automatic Meter Reading Using Blockchain Technology

The Automatic Meter Reader (AMR) for energy consumption is one of the most important issues in smart cities, as meters and electricity companies suffer from insecurity. The Internet of Things (IoT) can be used to achieve effective and reliable AMR in real time. Blockchain is a very advanced technology and technology that can be used to secure transactions such as meter readings and meter control. It is based on the idea of sequencing data blocks in a secure and distributed manner. Ensures the security of the blockchain representing the data in each block (reading the meter for example). The block is created and verified by many devices distributed on a network. Blockchain can be implemented in various ways and environments such as the Ethereum platform. In this paper, we will present a new automated meter reading platform using blockchain technology to meet the complete security requirements of AMR systems. When DoS attack was launched, requests did not affect the data itself, but the response speed of the blockchain network to incoming transactions from the servers was reduced very slightly. In addition, the results show that Blockchain could provide a promising technology that can participate in securing network meters.

Esraa Dbabseh, Radwan Tahboub

Biometrics, Pattern Recognition and Classification

Frontmatter

Chapter 57. Using Machine Learning to Process Filters and Mimic Instant Camera Effect

In this paper we use Machine Learning to convert images taken with an iPhone camera and visually alter them to appear as if taken with a Leica Sofort Instant Camera, more commonly known as the Polaroid look. While such image filters already exist and are highly effective, they function using ad-hoc techniques. Our goal is to achieve similar results by having a model learn what the Polaroid look is on its own and how many image pairs are required to train it. We found that using linear regression we need, on average, 800 images before the model began displaying good consistent results while using Pix2Pix (Isola et al., Image-to-image translation with conditional adversarial networks. In Proceedings of the IEEE conference on computer vision and pattern recognition, pp 1125–1134, 2017) (Conditional Adversarial Networks) and CycleGAN (Goodfellow et al., Generative adversarial nets. In Ghahramani Z, Welling M, Cortes C, Lawrence ND, Weinberger KQ (eds) Advances in neural information processing systems, vol 27, pp 2672–2680. Curran Associates, Inc., Red Hook, NY, 2014 [Online]. Available: http://papers.nips.cc/paper/5423-generative-adversarial-nets.pdf ) (Generative Adversarial Networks) only required 500 images.

Deirdre Chong, John Farhad Hanifzai, Hassan Adam, Jorge Garcia, Jorge Ramón Fonseca Cacho

Chapter 58. Benchmarking Accuracy and Precision of the Convolutional Neural Networks for Face Recognition on Makeup and Occluded Images

We benchmark established state-of-the-art Convolutional Neural Network (CNN) models for face recognition in the real-life conditions on the novel data set with makeup and occlusions. Strength and weaknesses of different CNN implementation architectures and degree of depth on particular types of makeup and occlusions are identified in the context of the selecting complementary models for the ensemble use. A practical approach of isolating uncertainty of the model verdicts trustworthiness in order to boost precision is investigated.

Stanislav Selitskiy, Nikolaos Christou, Natalya Selitskaya

Chapter 59. Combined Classification Models Applied to People Personality Identification

The popularization of social networks has considerably increased the volume of data generated from the interaction between people. Understanding this data can be useful both for companies and governments and for users. This work proposes to study how to infer the behavior of people on social networks from published comments, specifically using the Myers-Briggs Typological Indicator (MBTI) in a social network focused on discussions on behavioral issues. The analysis carried out employs Natural Language Processing (NLP) techniques, resampling of the data set and classification algorithms combined by Majority Vote. The results showed 90% efficiency of the combiner with the use of random oversampling. SVM and KNN were the best individual classifiers regardless of the resampling technique used. Although smaller compared to the best individual classifier, the combination approach shows a decrease in the misclassification for INFJ and INFP classes up to 11% and 34%, respectively.

Flávio Mota, Melise Paula, Isabela Drummond

Chapter 60. A Health Detection Model Based on Facial Data

The purpose of this study is to develop a model which employs the facial expressions and features of people to predict their health. Our objective is to find the best Machine learning approaches to develop a health model which utilizes the facial features. This report also discusses the available datasets of facial expressions. Here, we utilize such machine learning techniques as regression, neural network, and clustering to predict symptoms of sickness. To construct the model, we train our model with the healthy people images acquired from JAFFE database. After that, we ran the test dataset that includes an equal amount of sick and healthy people images. Utilizing the CCN (convolutional neural network) approach, our model has been able to predict the health of a person based on the facial features with an accuracy of 70%. This model could be utilized as the first level of diagnosis and can be implemented to distinguish between a healthy and sick person at the entrance of the public facilities. Such information could be crucial in the prevention and control of infectious diseases.

Sunil Manzoor, Shahram Latifi

Chapter 61. Performance Comparison of Algorithms Involving Automatic Learned Features and Hand-Crafted Features in Computer Vision

Three case studies in this report provide an analysis between two different machine learning algorithms in computer vision (CV) systems, with the difference in automatic feature learning as found in Deep Learning (DL), and manual hand-crafting feature extraction/selection, as found in traditional CV methods. Furthermore, this report focuses on state-of-the-art between Convolution Neural Networks (CNNs) and Scale-Invariant Feature Transform (SIFT) related-models.In the first case study, a CNN algorithm is trained by a set of 8000 images to binary classify 2000 test images. Adding more epochs and layers to the CNN indicates an increase of performance for the training set and testing set with a trade-off in increased system complexity. In the second case study, an image retrieval system is implemented using SIFT, SIFT-filtered, and CNN algorithms to evaluate an Oxford-5 k dataset. Results using a baseline SIFT and SIFT-filtered algorithm show the mean average precision (mAP) performance to be better than that performed by the CNN algorithm. In the third case study, different datasets are evaluated for image instance retrieval using the CNN and the Fisher Vector (FV) involving SIFT descriptors. Depending on the dataset being evaluated, a baseline FV shows the mAP to be comparable in performance to that of the CNN.

Rocky Y. Gonzalez, Shahram Latifi

Data Sciences

Frontmatter

Chapter 62. Big Data Analytics in Social Media: A Triple T (Types, Techniques, and Taxonomy) Study

Society 2.0; with the help of recent advancements in the internet and web 2.0 technology, makes the social media-based platform the most popular source for big data research. Big Data Analytics contributes by adjusting, analyzing, and forecasting insightful recommendations from this huge source of noisy & mostly unstructured “Big Social Data”. We present 10 mostly used big data analytics in the working domain of social media-based platforms. Different popular techniques or algorithms related to each big data analytic are also listed in this study. We show that “Text Analytics” is the most popular big data analytics in social media data analysis. Through this research, we try to explain the 10 Bigs of big data and introduce the “Sunflower Model of Big Data”. We also explain the reason why the social media-based platform is so significant and popular source of big data by analyzing the most recent statistics. This study will be a handful for all other researchers who want to work with big data in social media and in advance; make their work easy for selecting the best big data analytics method suitable for their research work.

Md. Saifur Rahman, Hassan Reza

Chapter 63. CARS: A Containerized Amazon Recommender System

With the big data boom, recommender systems that make intelligent recommendations for users have been playing an important role in today’s industry. However, existing recommender systems often overlook scalability, flexibility, and portability. They also commonly lack in-situ visualizations. To solve these problems, we present CARS: A Containerized Amazon Recommender System. CARS processes large Amazon data sets for analysis and makes product recommendations. However, its utility is not restricted to only prominent organizations like Amazon. CARS achieves scalability by taking advantage of industry-grade recommendation tools irrespective of available hardware resources. CARS runs in a completely isolated environment to promote flexibility and remote collaboration. The demonstrated implementation generates shopping recommendations from user ratings within product review data sets. CARS processes this review data using Apache Spark, a unified analytics engine for big data. The system complements recommendations with data-driven insights and interactive visualizations. In addition to these features, CARS contains a robust set of command line options to customize the results shown to the end-user, perform logging of processed data, and provide performance monitoring through Spark’s built-in web-interface. Highly portable and automated analysis of purchase data helps organizations understand the habits of their customers. CARS demonstrates the feasibility of such a system for a wide variety of users.

Adam Cassell, Andrew Muñoz, Brianna Blain-Castelli, Nikkolas Irwin, Feng Yan, Sergiu M. Dascalu, Frederick C. Harris

Chapter 64. Using Technologies to Uncover Patterns in Human Trafficking

In this paper, the researchers provide a background of human trafficking, review the use and implications of digital currency, and apply machine learning techniques to analyze publicly available trafficking datasets. The study also provides recommendations related to data collection, management, and analysis to aid the vital fight against individuals’ exploitation. The researchers conducted an exploratory data analysis using Python, RapidMiner, and Microsoft Excel towards an iterative review and interpretation of the dataset from the Counter Trafficking Data Collaborative (CTDC). The researchers found that there are more female victims of human trafficking in most age groups than male victims. However, for the age group between 39–47, there was a higher male victim count. Additionally, researchers found that the top five countries affected by human trafficking were the Philippines, Ukraine, Republic of Moldova, USA, and Cambodia. However, it must be noted that there are limitations to the overall data because they are provided voluntarily by organizations, and therefore, there is no equitable distribution of actual results from all countries and players. After mapping the country of origin and country of exploitation, it was made clear that there is a movement of victims from the country of origin to the country of exploitation. Lastly, researchers found that a complex combination of different variables is needed to provide accurate predictions for law enforcement and anti-trafficking organizations to aid them in fighting human trafficking, including country of exploitation and type of exploitation being the most important features in the prediction.

Annamaria Szakonyi, Harshini Chellasamy, Andreas Vassilakos, Maurice Dawson

Chapter 65. Data-Driven Identification of Pedagogical and Curricular Factors Conducive to Student Satisfaction

In this paper, we apply machine learning methods to the problem of identifying key pedagogical and curricular factors that play a critical role in determining student satisfaction. Using standard end-of-semester course evaluations, we employ two different approaches to address this problem, namely, training a variety of regression models of overall satisfaction based on specific characteristics of teaching style and course contents, and computing correlations between overall satisfaction metrics and these specific characteristics. To validate our approach, we present a case study using data from a public institution of higher education. Based on empirical evidence, we identify a key subset of the questions included in course evaluations that are the biggest determinants of student satisfaction.

Laura Sorto, Sourav Mukherjee, Vasudevan Janarthanan

Chapter 66. An Information Quality Framework for College and University Websites

Information Quality (IQ) of a university website plays a major role in the decision process for prospective students when selecting a university for their higher education. Furthermore, current students and others rely on university websites for many other purposes. In this paper we identify university website information quality dimensions relevant to prospective and current students and other site users. We discuss the rationales for identifying these IQ dimensions and propose a University Website Information Quality (UWebIQ) framework to quantify the individual IQ dimensions as well as a strategy for defining the composite IQ for such a website. The outcome of this research is expected to provide insight for universities that wish to maximize the fitness for use of their websites.

Joseph Elliot, Daniel Berleant

Chapter 67. An Agile MDD Method for Web Applications with Modeling Language

This paper presents the Agile and Collaborative Model-Driven Development method for Web applications named WebAC-MDD. It was conceptualized to transform agile models into Web application source-codes, using a novel Unified Modeling Language (UML) profile named Web Agile Modeling Language (Web-AML). It intends to represent a proposed solution for existing and inherent problems, regarding productivity in the development of Web applications and efforts for modeling and documentation, which do not add any value to clients. The main contributions of this paper are the WebAC-MDD Method, the Web-AML profile, and the proposed automatic generation of products, from the agile models, that are of value to stakeholders.

Breno Lisi Romano, Adilson Marques da Cunha

Backmatter

Weitere Informationen

Premium Partner