Nothing Special   »   [go: up one dir, main page]
Next Issue
Volume 13, August
Previous Issue
Volume 13, June
You seem to have javascript disabled. Please note that many of the page functionalities won't work as expected without javascript enabled.
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
6.9
2.4
Journals
Information
Volume 13
Issue 7
share announcement

Information, Volume 13, Issue 7 (July 2022) – 46 articles

Cover Story (view full-size image): The rapid development of effective vaccines against COVID-19 is an extraordinary achievement. Several countries have a pharmacovigilance system that detects, assesses, understands, and prevents the possible adverse effects of a drug. To benefit from such huge data sources, specialists and researchers can start observing small datasets to “predict” the behavior of the big picture. This paper defines a general framework for a pharmaceutical data analysis application to collect, filter, enrich, analyze, and visualize data, from small to big. Our evaluation of COVID-19 vaccination side effects shows that most adverse events can be classed as non-serious, and they concern muscle/joint pain, chills and nausea, headache, and fatigue. View this paper
  • Issues are regarded as officially published after their release is announced to the table of contents alert mailing list.
  • You may sign up for e-mail alerts to receive table of contents of newly released issues.
  • PDF is the official format for papers published in both, html and pdf forms. To view the papers in pdf format, click on the "PDF Full-text" link, and use the free Adobe Reader to open them.
Order results
Result details
Section
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
19 pages, 2101 KiB  
Article
Sustainable Mobility as a Service: Supply Analysis and Test Cases
by Corrado Rindone
Information 2022, 13(7), 351; https://doi.org/10.3390/info13070351 - 21 Jul 2022
Cited by 27 | Viewed by 3968
Abstract
Urban mobility is one of the main issues in the pursuit of sustainability. The United Nations 2030 Agenda assigns mobility and transport central roles in sustainable development and its components: economic, social, and environment. In this context, the emerging concept of Mobility as [...] Read more.
Urban mobility is one of the main issues in the pursuit of sustainability. The United Nations 2030 Agenda assigns mobility and transport central roles in sustainable development and its components: economic, social, and environment. In this context, the emerging concept of Mobility as a Service (MaaS) offers an alternative to unsustainable mobility, often based on private car use. From the point of view of sustainable mobility, the MaaS paradigm implies greater insights into the transport system and its components (supply, demand, and reciprocal interactions). This paper proposes an approach to the transport system aimed at overcoming the current barriers to the implementation of the paradigm. The focus is on the implications for the transport supply subsystem. The investigation method is based on the analysis of the main components of such subsystem (governance, immaterial, material, equipment) and its role in the entire transport system. Starting with the first experiences of Finnish cities, the paper investigates some real case studies, which are experimenting with MaaS, to find common and uncommon elements. From the analyses, it emerges that the scientific literature and real experiences mainly focus on the immaterial components alone. To address the challenges related to sustainable mobility, this paper underlines the need to consider all components within a transport system approach. The findings of the paper are useful in several contexts. In the context of research, the paper offers an analysis of the transport supply system from the point of view of the MaaS paradigm. In the real context, the paper offers further useful insights for operators and decision-makers who intend to increase the knowledge and skills necessary to face challenges related to the introduction of MaaS. Full article
(This article belongs to the Special Issue Emerging Research in Urban Computing and Intelligent Transport Systems)
►▼ Show Figures

Figure 1

Figure 1
<p>Transport supply components.</p> Full article ">Figure 2
<p>Governance: institutional components.</p> Full article ">Figure 3
<p>Governance: managerial components.</p> Full article ">Figure 4
<p>Immaterial components.</p> Full article ">Figure 5
<p>Material components.</p> Full article ">Figure 6
<p>Equipment components.</p> Full article ">Figure 7
<p>Supply models inside the TSM framework.</p> Full article ">Figure 8
<p>Spatiotemporal connections of an interchange node.</p> Full article ">
31 pages, 68508 KiB  
Article
Optimal Energy Management Scheme of Battery Supercapacitor-Based Bidirectional Converter for DC Microgrid Applications
by Srinivas Punna, Sujatha Banka and Surender Reddy Salkuti
Information 2022, 13(7), 350; https://doi.org/10.3390/info13070350 - 21 Jul 2022
Cited by 3 | Viewed by 3093
Abstract
Because of the splendid front of sustainable energy reassets in a DC Microgrid, it is profoundly willing to variances in energy age. A hybrid energy storage system (HESS) which includes a battery and a supercapacitor (SC) is used to decrease in-built fluctuations. The [...] Read more.
Because of the splendid front of sustainable energy reassets in a DC Microgrid, it is profoundly willing to variances in energy age. A hybrid energy storage system (HESS) which includes a battery and a supercapacitor (SC) is used to decrease in-built fluctuations. The two different characteristics of the battery and supercapacitor make it a great match for HESS applications. The HESS is connected to the DC Microgrid through a bidirectional converter, which allows energy to be exchanged between the battery and supercapacitor. This paper discusses a converter presenting an approach for a double-input bidirectional converter. Related to this, a regulator was designed for use as a voltage regulation in a DC Microgrid. The designed controllers accelerated PV generation and load disturbance DC link voltage restoration, in addition to effective power balancing among the battery and the SC. The conventional PI, proposed PI, and predictive PI control techniques are effectively validated using MATLAB Simulink. Experimental findings with low power have been used to validate the operation of the predictive PI control technique. The DC grid voltage profile showed substantial improvement while using the predictive PI control in comparison with the proposed and conventional PI control techniques in terms of setting time and maximum peak overshoot. Full article
(This article belongs to the Special Issue Intelligent Manufacturing and Informatization)
►▼ Show Figures

Figure 1

Figure 1
<p>Various configurations of a HESS consisting of a battery connected to a DC Microgrid and an SC. (<b>a</b>) Two segregated bidirectional converter modules. (<b>b</b>) Bidirectional converter with single double-input.</p> Full article ">Figure 2
<p>HESS-enhanced DC Microgrid with PV source.</p> Full article ">Figure 3
<p>HESS with double-input bidirectional converter topology.</p> Full article ">Figure 4
<p>Consistent state waveforms for HESS Discharging Sequence.</p> Full article ">Figure 5
<p>Steady state waveforms for HESS charging mode of operation.</p> Full article ">Figure 6
<p>HESS mode of energy exchange. (<b>a</b>) Current flows across an active circuit. (<b>b</b>) Waveforms in a steady state.</p> Full article ">Figure 7
<p>A flow chart depicts transitions between distinct modes.</p> Full article ">Figure 8
<p>Overall management mechanism for current bifurcation between SC and battery units for conventional PI.</p> Full article ">Figure 9
<p>Overall management mechanism for current bifurcation between SC and battery units for proposed PI.</p> Full article ">Figure 10
<p>Logic for controlling a supercapacitor and a battery. (<b>a</b>) The supercapacitor control scheme. (<b>b</b>) Battery control system scheme.</p> Full article ">Figure 11
<p>Bode plot of SC’s inner current controller of logic for control, both with and without controller.</p> Full article ">Figure 12
<p>Bode plot of SC’s outer voltage controller of logic for control, both with and without controller.</p> Full article ">Figure 13
<p>Bode plot of battery control logic’s current controller.</p> Full article ">Figure 14
<p>Block diagram of predictive PI control scheme for a HESS.</p> Full article ">Figure 15
<p>Supercapacitor SoC charge controller. (<b>a</b>) Charging and discharging logic; logic for the charging and discharging process. (<b>b</b>) Flowchart of the control logic.</p> Full article ">Figure 16
<p>Simulation results for step change in PV generation for conventional PI controller.</p> Full article ">Figure 17
<p>Simulation results for step change in PV generation for proposed PI control scheme.</p> Full article ">Figure 18
<p>Simulation results for step change in load demand for conventional PI control scheme.</p> Full article ">Figure 19
<p>Simulation results for step change in load demand for proposed PI control scheme.</p> Full article ">Figure 20
<p>Simulation results for step change in PV generation for predictive PI control scheme.</p> Full article ">Figure 21
<p>Simulation results for step change in load demand for predictive PI control scheme.</p> Full article ">Figure 22
<p>Comparative analysis of conventional PI, proposed PI, and predictive PI control schemes. (<b>a</b>,<b>b</b>) DC grid voltage for step change in PV and load, (<b>c</b>) battery current, (<b>d</b>) SC current.</p> Full article ">Figure 23
<p>Low-power hardware developed a prototype model.</p> Full article ">Figure 24
<p>Experimental results for HESS charging mode.</p> Full article ">Figure 25
<p>Experimental results for HESS discharging mode.</p> Full article ">Figure 26
<p>Experimental results for energy exchange mode.</p> Full article ">
16 pages, 611 KiB  
Article
Drivers Influencing the Adoption Intention towards Mobile Fintech Services: A Study on the Emerging Bangladesh Market
by Md. Sharif Hassan, Md. Aminul Islam, Farid Ahammad Sobhani, Hussen Nasir, Imroz Mahmud and Fatema Tuz Zahra
Information 2022, 13(7), 349; https://doi.org/10.3390/info13070349 - 20 Jul 2022
Cited by 25 | Viewed by 6424
Abstract
People’s acceptance of technological changes has escalated with time. However, the acceptance and adoption of fintech services hiked after the outbreak of the virulent coronavirus. With this breakout, the adoption of mobile fintech services (MFS) increased among general citizens and business sectors around [...] Read more.
People’s acceptance of technological changes has escalated with time. However, the acceptance and adoption of fintech services hiked after the outbreak of the virulent coronavirus. With this breakout, the adoption of mobile fintech services (MFS) increased among general citizens and business sectors around the world, including in developed, emerging, and developing economies. This study aimed to identify the factors that impact the adoption intention of consumers to embrace and enhance the use of mobile fintech services in an emerging market, Bangladesh. A research model was developed to strengthen the objective of this paper. A total of 218 respondents responded to the questionnaire. The study utilized structural equation modeling to analyze the results in SmartPLS software. The results showed significant positive effects of social influence, trust, perceived benefit, and facilitating conditions on the adoption intention towards MFS. Mobile fintech service providers must keep their users’ needs and literacy rates in mind when designing the user interface (UI). Moreover, they should also cater more efficient services to the users and work based on the feedback received. The customers’ satisfaction will ultimately lead to customers conducting more digital transactions and will contribute to the escalation of fintech transactions, resulting in more financial inclusion. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Research framework.</p> Full article ">Figure 2
<p>Path coefficient (T-values) results from SmartPLS.</p> Full article ">
16 pages, 781 KiB  
Article
Computational Offloading of Service Workflow in Mobile Edge Computing
by Shuang Fu, Chenyang Ding and Peng Jiang
Information 2022, 13(7), 348; https://doi.org/10.3390/info13070348 - 19 Jul 2022
Cited by 5 | Viewed by 2592
Abstract
Mobile edge computing (MEC) sinks the functions and services of cloud computing to the edge of the network to provide users with storage and computing resources. For workflow tasks, the interdependency and the sequence constraint being among the tasks make the offloading strategy [...] Read more.
Mobile edge computing (MEC) sinks the functions and services of cloud computing to the edge of the network to provide users with storage and computing resources. For workflow tasks, the interdependency and the sequence constraint being among the tasks make the offloading strategy more complicated. To obtain the optimal offloading and scheduling scheme for workflow tasks to minimize the total energy consumption of the system, a workflow task offloading and scheduling scheme based on an improved genetic algorithm is proposed in an MEC network with multiple users and multiple virtual machines (VMs). Firstly, the system model of the offloading and scheduling of workflow tasks in a multi-user and multi-VMs MEC network is built. Then, the problem of how to determine the optimal offloading and scheduling scheme of workflow to minimize the total energy consumption of the system while meeting the deadline constraint is formulated. To solve this problem, the improved genetic algorithm is adopted to obtain the optimal offloading strategy and scheduling. Finally, the simulation results show that the proposed scheme can achieve a lower energy consumption than other benchmark schemes. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>System model.</p> Full article ">Figure 2
<p>Workflow directed acyclic graph.</p> Full article ">Figure 3
<p>The algorithm flowchart.</p> Full article ">Figure 4
<p>Coding scheme in the genetic algorithm.</p> Full article ">Figure 5
<p>The single−point crossover operation of task execution order.</p> Full article ">Figure 6
<p>Single−point crossover operation of task offloading position.</p> Full article ">Figure 7
<p>Single−point mutation operation of task offloading position.</p> Full article ">Figure 8
<p>Task execution order single−point mutation operation.</p> Full article ">Figure 9
<p>The total system energy consumption versus the number of MDs.</p> Full article ">Figure 10
<p>The total system energy consumption versus the number of workflow.</p> Full article ">Figure 11
<p>The total system energy consumption versus the number of virtual servers.</p> Full article ">Figure 12
<p>The total system energy consumption versus the average workload size.</p> Full article ">
9 pages, 278 KiB  
Article
More Constructions of Light MDS Transforms Based on Known MDS Circulant Matrices
by Jin-Bo Wang, You Wu and Yu Zhou
Information 2022, 13(7), 347; https://doi.org/10.3390/info13070347 - 18 Jul 2022
Cited by 2 | Viewed by 1537
Abstract
Maximum distance separable (MDS) codes have the maximum branch number in cryptography, and they are generally used in diffusion layers of symmetric ciphers. The diffusion layer of the Advanced Encryption Standard (AES) uses the circulant MDS matrix with the row element of [...] Read more.
Maximum distance separable (MDS) codes have the maximum branch number in cryptography, and they are generally used in diffusion layers of symmetric ciphers. The diffusion layer of the Advanced Encryption Standard (AES) uses the circulant MDS matrix with the row element of {2;3;1;1} in F28. It is the simplest MDS matrix in  F2n4, recorded as A=Circ(2;3;1;1). In this paper, we study the more extensive MDS constructions of A in F2n4. By transforming the element multiplication operation in the finite field into the bit-level operation, we propose a multivariable operation definition based on simple operations, such as cyclic shift, shift, and XOR. We apply this multivariable operation to more lightweight MDS constructions of A and discuss the classification of the MDS clusters. We also give an example of the MDS cluster of A. Without changing the structure, elements, and the implementation cost of the known MDS matrix, the number of existing MDS transformations is expanded to n2/2 times that of its original. The constructions in this paper provide rich component materials for the design of lightweight cryptographic algorithms. Full article
15 pages, 2061 KiB  
Article
Sustainable Mobility as a Service: Framework and Transport System Models
by Antonino Vitetta
Information 2022, 13(7), 346; https://doi.org/10.3390/info13070346 - 16 Jul 2022
Cited by 30 | Viewed by 4482
Abstract
Passenger mobility plays an important role in today’s society and optimized transport services are a priority. In recent years, MaaS (Mobility as a Service) has been studied and tested as new integrated services for users. In this paper, MaaS is studied considering the [...] Read more.
Passenger mobility plays an important role in today’s society and optimized transport services are a priority. In recent years, MaaS (Mobility as a Service) has been studied and tested as new integrated services for users. In this paper, MaaS is studied considering the sustainability objectives and goals to be achieved with particular reference to the consolidated methodologies adopted in the transport systems engineering for design, management, and monitoring of transport services; it is defined as Sustainable MaaS (S-MaaS). This paper considers the technological and communication platform essential and assumed to be a given considering that it has been proposed in many papers and it has been tested in some areas together with MaaS. Starting from the MaaS platform, the additional components and models necessary for the implementation of an S-MaaS are analyses in relation to: a Decision Support System (DSS) that supports MaaS public administrations and MaaS companies for the design of the service and demand management; a system for the evaluation of intervention policies; and also considers smart planning for a priori and a posteriori evaluation of sustainability objectives and targets. Full article
(This article belongs to the Special Issue Emerging Research in Urban Computing and Intelligent Transport Systems)
►▼ Show Figures

Figure 1

Figure 1
<p>Involved MaaS actors.</p> Full article ">Figure 2
<p>MaaS in relation to the integration, methods adopted for design, management and monitoring of the transport services.</p> Full article ">Figure 3
<p>Architecture of MaaS 1.0, 2.0 and 3.0 and actors involved.</p> Full article ">Figure 4
<p>MaaS Transportation System Model platform (level 2, Intelligent MaaS).</p> Full article ">Figure 5
<p>Sustainable objectives and goals and users acceptance (level 3, Sustainable MaaS; level 4, MaaS for Users).</p> Full article ">Figure 6
<p>Strait of Messina: study area.</p> Full article ">
19 pages, 501 KiB  
Article
Influencer Marketing on Instagram: A Sequential Mediation Model of Storytelling Content and Audience Engagement via Relatability and Trust
by Madiha Atiq, Ghulam Abid, Aizza Anwar and Muhammad Fazal Ijaz
Information 2022, 13(7), 345; https://doi.org/10.3390/info13070345 - 16 Jul 2022
Cited by 13 | Viewed by 18791
Abstract
Storytelling content is where the facts are conveyed by emotion and that make people more engaged and want to take action or change their surroundings. Stories fascinate people and can easily be remembered compared to the facts alone. The much-hyped feature “stories” of [...] Read more.
Storytelling content is where the facts are conveyed by emotion and that make people more engaged and want to take action or change their surroundings. Stories fascinate people and can easily be remembered compared to the facts alone. The much-hyped feature “stories” of Instagram, a trendy social media platform, has become a game-changer for influencer marketing. The present study extends reactance theory in the context of Instagram’s millennial users. Previous researchers have tested the effectiveness of the stories feature of this particular social media platform. Therefore, in line with the earlier studies, we propose a sequential mediation model that investigates the effect of storytelling content (made by Instagram Influencers) on audience engagement using two sequential mediation mechanisms of relatability and trust. Data were obtained using a cross-sectional study design from 273 millennial users of Instagram. Our results justify the direct and indirect hypothesized relationship through Process Macros. We found that relatability and trust play a significant role in building a strong relationship between storytelling content and audience engagement. Ultimately, the research findings suggest that professionals should be more creative while making the content on Instagram to engage the millennial market. Moreover, this research has tried to fill the gap in the literature on Instagram “stories” as an advertising platform. Full article
(This article belongs to the Special Issue Knowledge Management and Digital Humanities)
►▼ Show Figures

Figure 1

Figure 1
<p>Research framework.</p> Full article ">
21 pages, 12084 KiB  
Article
An Empirical Study on the Differences between Online Picture Reviews and Text Reviews
by Hanyang Luo, Wanhua Zhou, Wugang Song and Xiaofu He
Information 2022, 13(7), 344; https://doi.org/10.3390/info13070344 - 15 Jul 2022
Cited by 2 | Viewed by 2818
Abstract
In the context of e-commerce, online travel agencies often derive useful information from online reviews to improve transactions. Based on the dispute on the usefulness of different types of reviews and social exchange theory, this study investigates how the characteristics of pictures and [...] Read more.
In the context of e-commerce, online travel agencies often derive useful information from online reviews to improve transactions. Based on the dispute on the usefulness of different types of reviews and social exchange theory, this study investigates how the characteristics of pictures and text influence review reading and review posting behaviors and thus influencing the efficiency of online review systems. By analyzing crawled data of online hotels and conducting experiments, we first find that picture reviews are more useful than text reviews, and high-quality pictures in reviews have a significant impact on review usefulness. Second, posting pictures requires review posters to pay more perceived costs. Third, negative review posters have higher perceived costs, so they are more unwilling to post pictures, especially high-quality pictures. Our results indicate that review platforms need to add incentives to encourage consumers to post high-quality picture reviews and design workable interfaces to reduce the burden of negative reviewers to speed up the purchase decision process for review readers. This study provides theoretical implications by demonstrating how the adoption of the picture in review systems influences both review readers’ and review posters’ behaviors. Additionally, our findings also provide useful managerial insights for online travel suppliers in terms of building an effective review system to promote sales. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Screenshot of a hotel homepage from <a href="http://Ctrip.com" target="_blank">Ctrip.com</a> (accessed on 13 July 2022).</p> Full article ">Figure 2
<p>Screenshot of a text review.</p> Full article ">Figure 3
<p>Screenshot of a picture review.</p> Full article ">Figure 4
<p>Distribution of quality of pictures.</p> Full article ">Figure 5
<p>The relationship between review scores and the quality of pictures.</p> Full article ">
17 pages, 27245 KiB  
Article
A Multi-Sensory Guidance System for the Visually Impaired Using YOLO and ORB-SLAM
by Zaipeng Xie, Zhaobin Li, Yida Zhang, Jianan Zhang, Fangming Liu and Wei Chen
Information 2022, 13(7), 343; https://doi.org/10.3390/info13070343 - 15 Jul 2022
Cited by 13 | Viewed by 4126
Abstract
Guidance systems for visually impaired persons have become a popular topic in recent years. Existing guidance systems on the market typically utilize auxiliary tools and methods such as GPS, UWB, or a simple white cane that exploits the user’s single tactile or auditory [...] Read more.
Guidance systems for visually impaired persons have become a popular topic in recent years. Existing guidance systems on the market typically utilize auxiliary tools and methods such as GPS, UWB, or a simple white cane that exploits the user’s single tactile or auditory sense. These guidance methodologies can be inadequate in a complex indoor environment. This paper proposes a multi-sensory guidance system for the visually impaired that can provide tactile and auditory advice using ORB-SLAM and YOLO techniques. Based on an RGB-D camera, the local obstacle avoidance system is realized at the tactile level through point cloud filtering that can inform the user via a vibrating motor. Our proposed method can generate a dense navigation map to implement global obstacle avoidance and path planning for the user through the coordinate transformation. Real-time target detection and a voice-prompt system based on YOLO are also incorporated at the auditory level. We implemented the proposed system as a smart cane. Experiments are performed using four different test scenarios. Experimental results demonstrate that the impediments in the walking path can be reliably located and classified in real-time. Our proposed system can function as a capable auxiliary to help visually impaired people navigate securely by integrating YOLO with ORB-SLAM. Full article
(This article belongs to the Special Issue Intelligence Computing and Systems)
►▼ Show Figures

Figure 1

Figure 1
<p>Algorithm flow of the proposed multi-sensory guidance system.</p> Full article ">Figure 2
<p>A block diagram of the data flow in the ORB-SLAM process.</p> Full article ">Figure 3
<p>An illustration of the Orientation Correcting Algorithm. In situation (<b>a</b>), there is no intersection between the circle around the user and the planned path, thus an optimal path is recalculated. In situation (<b>b</b>), there is at least one intersection, and then we choose the closest intersection to the planned path’s destination as the current goal and provide the user with the corresponding orientation.</p> Full article ">Figure 4
<p>The physical diagram of proposed hardware system.</p> Full article ">Figure 5
<p>The experimental scenes. (<b>A</b>) U-shaped hallway; (<b>B</b>) H-shaped hallway; (<b>C</b>) Loop-shaped hallway; and (<b>D</b>) T-shaped hallway.</p> Full article ">Figure 6
<p>Example diagram of field experiment.</p> Full article ">Figure 7
<p>Example of the feature points extraction for (<b>A</b>) the washing machine and other objects in U-shaped hallway scene; (<b>B</b>) the obstacles in H-shaped hallway scene; (<b>C</b>) the seats and desks in the Loop-shaped hallway scene; (<b>D</b>) the walls in T-shaped hallway scene.</p> Full article ">Figure 8
<p>Example of the 3D point cloud of (<b>A</b>) the U-shaped hallway; (<b>B</b>) the H-shaped hallway; (<b>C</b>) the Loop-shaped hallway; (<b>D</b>) the T-shaped hallway.</p> Full article ">Figure 9
<p>Example of the dense point cloud map for (<b>A</b>) the U-shaped hallway; (<b>B</b>) the H-shaped hallway; (<b>C</b>) the Loop-shaped hallway; (<b>D</b>) the T-shaped hallway.</p> Full article ">Figure 10
<p>Example of the generated occupancy grid map of (<b>A</b>) the U-shaped hallway; (<b>B</b>) the H-shaped hallway; (<b>C</b>) the Loop-shaped hallway; (<b>D</b>) the T-shaped hallway.</p> Full article ">Figure 11
<p>An illustration of the path planning and navigation in a (<b>A</b>) the U-shaped hallway; (<b>B</b>) the H-shaped hallway; (<b>C</b>) the Loop-shaped hallway; (<b>D</b>) the T-shaped hallway.</p> Full article ">
13 pages, 1110 KiB  
Article
Coded Parallel Transmission for Half-Duplex Distributed Computing
by Qixuan Zai, Kai Yuan and Youlong Wu
Information 2022, 13(7), 342; https://doi.org/10.3390/info13070342 - 15 Jul 2022
Viewed by 1835
Abstract
This work studies a general distributed coded computing system based on the MapReduce-type framework, where distributed computing nodes within a half-duplex network wish to compute multiple output functions. We first introduce a definition of communication delay to characterize the time cost during the [...] Read more.
This work studies a general distributed coded computing system based on the MapReduce-type framework, where distributed computing nodes within a half-duplex network wish to compute multiple output functions. We first introduce a definition of communication delay to characterize the time cost during the date shuffle phase, and then propose a novel coding strategy that enables parallel transmission among the computation nodes by delicately designing the data placement, message symbols encoding, data shuffling, and decoding. Compared to the coded distributed computing (CDC) scheme proposed by Li et al., the proposed scheme significantly reduces the communication delay, in particular when the computation load is relatively smaller than the number of computing nodes K. Moreover, the communication delay of CDC is a monotonically increasing function of K, while the communication delay of our scheme decreases as K increases, indicating that the proposed scheme can make better use of the computing resources. Full article
(This article belongs to the Special Issue Advanced Technologies in Storage, Computing, and Communication)
►▼ Show Figures

Figure 1

Figure 1
<p>The MapReduce system where multiple users connect with each other via a switch.</p> Full article ">Figure 2
<p>Variation trend of communication delay <span class="html-italic">D</span> with computation load <span class="html-italic">r</span>: the computing node <span class="html-italic">K</span> and output function <span class="html-italic">Q</span> in a given network are 50. Number of file N and output function <span class="html-italic">Q</span> are both 50. In addition, the transmission rate <span class="html-italic">C</span> is 100 Mbps, while the length of intermediate value T is 100 Mbits. Compare the variation of communication delay with computing load for uncoded, CDC and proposed scheme.</p> Full article ">Figure 3
<p>Variation trend of communication delay <span class="html-italic">D</span> with the number of computing nodes <span class="html-italic">K</span>. The computation load <span class="html-italic">r</span> is 2, and the number of file <span class="html-italic">N</span> and output function <math display="inline"><semantics> <mrow> <mi>Q</mi> <mo>=</mo> <mn>50</mn> </mrow> </semantics></math> are both 50. In addition, transmission rate <span class="html-italic">C</span> is 100 Mbps while the length of intermediate value <span class="html-italic">T</span> is 100 Mbits. The communication delay of uncoded computing, coded distributed computing and parallel coded computing changes with the number of computing nodes <span class="html-italic">K</span> in the network.</p> Full article ">Figure 4
<p>Variation trend of communication delay <span class="html-italic">D</span> with the number of computing nodes <span class="html-italic">K</span> and computation load <span class="html-italic">r</span>. The number of file <span class="html-italic">N</span> and output function <span class="html-italic">Q</span> are both 50. In addition, transmission rate <span class="html-italic">C</span> is 100 Mbps while the length of the intermediate value <span class="html-italic">T</span> is 100 Mbits. The communication delay of the uncoded computing, coded distributed computing and parallel coded computing changes with the number of computing nodes <span class="html-italic">K</span> in the network.</p> Full article ">Figure 5
<p>Example of coding strategy for simultaneous transmission of two broadcast groups: a distributed computing network composed of <math display="inline"><semantics> <mrow> <mi>K</mi> <mo>=</mo> <mn>6</mn> </mrow> </semantics></math> nodes, including <math display="inline"><semantics> <mrow> <mi>N</mi> <mo>=</mo> <mn>15</mn> </mrow> </semantics></math> input files, and each file is stored by <math display="inline"><semantics> <mrow> <mi>r</mi> <mo>=</mo> <mn>2</mn> </mrow> </semantics></math> nodes. A total of <math display="inline"><semantics> <mrow> <mi>Q</mi> <mo>=</mo> <mn>6</mn> </mrow> </semantics></math> output functions need to be processed.</p> Full article ">Figure 6
<p>An example of coding strategy for the simultaneous transmission of two broadcast groups. At the same time, both nodes can send messages without interference. Solid lines from the same point represent multicast messages.</p> Full article ">
14 pages, 1712 KiB  
Article
Vertical Integration Decision Making in Information Technology Management
by Menekse Gizem Gorgun, Seckin Polat and Umut Asan
Information 2022, 13(7), 341; https://doi.org/10.3390/info13070341 - 15 Jul 2022
Cited by 1 | Viewed by 2874
Abstract
Vertical integration, also known as make-or-buy, defines whether activities are conducted by company or provided by external parties. There are different models to support decision making for vertical integration in the literature. However, they ignore the uncertainty aspect of vertical integration. As a [...] Read more.
Vertical integration, also known as make-or-buy, defines whether activities are conducted by company or provided by external parties. There are different models to support decision making for vertical integration in the literature. However, they ignore the uncertainty aspect of vertical integration. As a strategic decision, vertical integration is multidimensional and less frequent. This study contributes a new data-driven model that includes all these characteristics of vertical integration decisions. In this study, a methodology is suggested that benefits from the models in the literature and assesses the results with data obtained from real IT cases. Different methodologies were followed to reach a model that accurately predicts make-or-buy decisions in IT projects at a retail company. Firstly, three different knowledge-based generic models derived from the literature were applied to predict decisions for twenty-one different make-or-buy cases in IT. The highest accuracy rate reached among these knowledge-based models was 76%. Secondly, the same cases were also analyzed with Naïve Bayes using factors originally introduced by these generic models. The Naïve Bayes algorithm can represent the uncertainty inherent in the decision model. The highest accuracy rate obtained was 67%. Thirdly, a new data-driven model based on Naïve Bayes using IT-related factors was proposed for the decision problem of vertical integration. The data-driven model correctly classified 86% of the decisions. Full article
(This article belongs to the Special Issue Evaluating Methods and Decision Making)
►▼ Show Figures

Figure 1

Figure 1
<p>A summary for methodology.</p> Full article ">Figure 2
<p>Knowledge-based Model 1: Process Technology [<a href="#B9-information-13-00341" class="html-bibr">9</a>].</p> Full article ">Figure 3
<p>Knowledge-based Model 2: Competitiveness, Capability and Opportunism [<a href="#B36-information-13-00341" class="html-bibr">36</a>].</p> Full article ">Figure 4
<p>Knowledge-based Model 3: Task Programmability, Separability and Asset Specificity [<a href="#B8-information-13-00341" class="html-bibr">8</a>].</p> Full article ">
27 pages, 3299 KiB  
Communication
Integrating Human Factors in the Visualisation of Usable Transparency for Dynamic Risk Assessment
by Anastasija Collen, Ioan-Cosmin Szanto, Meriem Benyahya, Bela Genge and Niels Alexander Nijdam
Information 2022, 13(7), 340; https://doi.org/10.3390/info13070340 - 14 Jul 2022
Cited by 5 | Viewed by 2391
Abstract
Modern technology and the digitisation era accelerated the pace of data generation and collection for various purposes. The orchestration of such data is a daily challenge faced by even experienced professional users in the context of Internet of Things (IoT)-enabled environments, especially when [...] Read more.
Modern technology and the digitisation era accelerated the pace of data generation and collection for various purposes. The orchestration of such data is a daily challenge faced by even experienced professional users in the context of Internet of Things (IoT)-enabled environments, especially when it comes to cybersecurity and privacy risks. This article presents the application of a user-centric process for the visualisation of automated decision making security interventions. The user interface (UI) development was guided by iterative feedback collection from user studies on the visualisation of a dynamic risk assessment (DRA)-based security solution for regular lay users. The methodology we applied starts with the definition of the methodological process to map possible technical actions to related usable actions. The definition and refinement of the user interface (UI) was controlled by the survey feedback loop from end user studies on their general technological knowledge, experience with smart homes, cybersecurity awareness and privacy preservation needs. We continuously improved the visualisation interfaces for configuring a cybersecurity solution and adjusting usable transparency of the control and monitoring of the dynamic risk assessment (DRA). For this purpose, we have designed, developed and validated a decision tree workflow and showed the evolution of the interfaces through various stages of the real-life trials executed under European H2020 project GHOST. Full article
(This article belongs to the Special Issue Information Retrieval, Recommender Systems and Adaptive Systems)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Methodology of User Actions Mapping.</p> Full article ">Figure 2
<p>Decision tree conceptualisation with coloured highlighting of the main branches.</p> Full article ">Figure 3
<p>Missing communication.</p> Full article ">Figure 4
<p>Whitelisting.</p> Full article ">Figure 5
<p>Data type.</p> Full article ">Figure 6
<p>Frequency.</p> Full article ">Figure 7
<p>Time.</p> Full article ">Figure 8
<p>Blacklisting.</p> Full article ">Figure 9
<p>Privacy preferences.</p> Full article ">Figure 10
<p>Notification preferences.</p> Full article ">Figure 11
<p>Security preferences.</p> Full article ">Figure 12
<p>Pending action.</p> Full article ">Figure 13
<p>Pending decision.</p> Full article ">Figure 14
<p>Notification of automated decision.</p> Full article ">
16 pages, 1299 KiB  
Article
A Virtual Tour for the Promotion of Tourism of the City of Bari
by Valerio De Luca, Giorgia Marcantonio, Maria Cristina Barba and Lucio Tommaso De Paolis
Information 2022, 13(7), 339; https://doi.org/10.3390/info13070339 - 13 Jul 2022
Cited by 11 | Viewed by 4187
Abstract
The use of information technology in the field of cultural heritage makes it possible to involve more and more people in the promotion of cultural heritage, fostering social, cultural, economic and community growth. This work stems from the interest in using Virtual Reality [...] Read more.
The use of information technology in the field of cultural heritage makes it possible to involve more and more people in the promotion of cultural heritage, fostering social, cultural, economic and community growth. This work stems from the interest in using Virtual Reality (VR) in the field of cultural heritage, creating a tour of the city of Bari that tells its evolution over the years. To this end, a low-cost VR360 application has been developed which, by means of a cardboard, allows the user to experience a virtual journey through time. It tells the story of the city, focusing on its urban expansion and the evolution of its architectural styles, influenced by various dominations over the centuries, up to the current state. The virtual environment was created from spherical images of the city, captured through 360° cameras and enriched with various types of information content. The user experience was assessed by means of a questionnaire derived from previous work that was generalised and adapted to the considered scenario: the results showed a very good level of satisfaction, usability, engagement, immersion and sense of presence; the highest score was obtained for the visual quality of the images of the virtual environment. Full article
(This article belongs to the Special Issue eXtended Reality for Social Inclusion and Educational Purpose)
►▼ Show Figures

Figure 1

Figure 1
<p>Different types of interaction buttons.</p> Full article ">Figure 2
<p>Opening scene of the tour.</p> Full article ">Figure 3
<p>Scene of Bari Vecchia (the old town).</p> Full article ">Figure 4
<p>Scene of Borgo Nuovo (the new town).</p> Full article ">Figure 5
<p>Initial screen of the postcard game.</p> Full article ">Figure 6
<p>Density plot of the ages of the users involved in the test.</p> Full article ">Figure 7
<p>Distributions of the scores of the user experience factors.</p> Full article ">Figure 8
<p>Variable correlation plot for the user experience factors.</p> Full article ">Figure 9
<p>Scree plot of user experience factors.</p> Full article ">Figure 10
<p>Quality of representation of user experience factors.</p> Full article ">
22 pages, 1687 KiB  
Article
Improving Performance and Quantifying Uncertainty of Body-Rocking Detection Using Bayesian Neural Networks
by Rafael Luiz da Silva, Boxuan Zhong, Yuhan Chen and Edgar Lobaton
Information 2022, 13(7), 338; https://doi.org/10.3390/info13070338 - 12 Jul 2022
Viewed by 1838
Abstract
Body-rocking is an undesired stereotypical motor movement performed by some individuals, and its detection is essential for self-awareness and habit change. We envision a pipeline that includes inertial wearable sensors and a real-time detection system for notifying the user so that they are [...] Read more.
Body-rocking is an undesired stereotypical motor movement performed by some individuals, and its detection is essential for self-awareness and habit change. We envision a pipeline that includes inertial wearable sensors and a real-time detection system for notifying the user so that they are aware of their body-rocking behavior. For this task, similarities of body rocking to other non-related repetitive activities may cause false detections which prevent continuous engagement, leading to alarm fatigue. We present a pipeline using Bayesian Neural Networks with uncertainty quantification for jointly reducing false positives and providing accurate detection. We show that increasing model capacity does not consistently yield higher performance by itself, while pairing it with the Bayesian approach does yield significant improvements. Disparities in uncertainty quantification are better quantified by calibrating them using deep neural networks. We show that the calibrated probabilities are effective quality indicators of reliable predictions. Altogether, we show that our approach provides additional insights on the role of Bayesian techniques in deep learning as well as aids in accurate body-rocking detection, improving our prior work on this subject. Full article
(This article belongs to the Special Issue Biomedical Signal Processing and Data Analytics in Healthcare Systems)
►▼ Show Figures

Figure 1

Figure 1
<p>Illustration of the data collection, prediction, and uncertainty quantification pipelines. (<b>a</b>) Body-rocking movement illustration with sensor placed on the arm. The arrows indicate the forward and backward body rocking. (<b>b</b>) A sample gyroscope measurement with corresponding annotations. (<b>c</b>) Our pipeline for prediction and uncertainty, which uses a Bayesian Neural Network framework based on Monte Carlo (MC) Sampling from dropout. The uncertainty measures include a variance <math display="inline"><semantics> <msup> <mi>σ</mi> <mn>2</mn> </msup> </semantics></math> indicative of observation noise, and entropy <span class="html-italic">H</span> and mutual information <span class="html-italic">I</span> obtained from the MC samples. This framework improves the performance of prediction and yields a calibrated probability <math display="inline"><semantics> <mi>ρ</mi> </semantics></math> that is reliably indicating our confidence in a prediction.</p> Full article ">Figure 2
<p>Examples of the deep learning architectures explored in this manuscript. (<b>a</b>) Rad’s model [<a href="#B26-information-13-00338" class="html-bibr">26</a>], (<b>b</b>) WiderNet 2×, FCN 128, (<b>c</b>) Bayesian Approach to Rad’s model. The WiderNet type of architecture is also explored with variants containing eight nodes in the FCN layers in addition to the variant with 128 nodes.</p> Full article ">Figure 3
<p>Performance of the presented architectures on the datasets. (<b>Top</b>) EDAQA dataset, study 1. (<b>Middle</b>) EDAQA dataset, study 2. (<b>Bottom</b>) ESDB dataset. The architectures are evaluated under different metrics to aid the elicitation of Bayesian approach on a regular CNN. The columns represent the averages of AUC (<b>left</b>), F1-score (<b>middle</b>) and precision (<b>right</b>), across all subjects and 10 runs. <span class="html-italic">Rad Original</span> is the baseline represented by a blue dot, while the legends identifies the WiderNet variants. We observe an improvement in all models when considering their Bayesian variant. This improvement is more noticeable in the ESDB dataset. Best in color.</p> Full article ">Figure 4
<p>ROC curves for each dataset for Rad’s model and WiderNet 8×, FCN 128 (i.e., with 128 neurons in their fully connected layer). (<b>a</b>) EDAQA dataset, study 1, (<b>b</b>) EDAQA dataset, study 2, (<b>c</b>) ESDB dataset. The Bayesian model with the higher capacity performs best in all datasets. Best viewed in color.</p> Full article ">Figure 5
<p>Area from the diagonal of reliability plots with and without transfer learning (TF). TF prevents this error metric from increasing as the model capacity increases. Best in color.</p> Full article ">Figure 6
<p>Distribution of correct and incorrect predictions for WiderNet 8×, FCN-128 on the ESDB dataset. Please note that that the entropy has different distributions for the (<b>a</b>) correct vs. (<b>b</b>) incorrect predictions. This indicate that it is a good feature for quantifying the uncertainty (i.e., higher entropy means higher likelihood of having an incorrect prediction). Best in color.</p> Full article ">Figure 7
<p>F1-Score vs. Precision plot using uncertainty selection for ESDB dataset. The performance is obtained by setting the predictions of the model to a negative detection if their uncertainty follows below the specified calibrated probability or entropy thresholds. The original model is WiderNet 8×, FCN 128 with Bayesian approach and no selection. Please note that the curve produced by the cal. prob. is better than the entropy curve. Furthermore, a threshold of 0.65 on the cal. prob. yields almost no drop in F1-score but and 7% increase in precision. Best viewed in color.</p> Full article ">Figure 8
<p>Precision vs. Recall plot for uncertainty selection on ESDB dataset. The original model is WiderNet 8×, FCN 128 with Bayesian approach and no selection. Please note that a threshold of 0.65 on the cal. prob. yields about an equal percentage of increase in precision and decrease in recall. Best in color.</p> Full article ">Figure 9
<p>AUC score vs. Percentage of samples kept unchanged for uncertainty selection on ESDB dataset. Each curve shows the corresponding threshold values used for the data points. The original model is WiderNet 8×, FCN 128 with Bayesian approach. Best viewed in color.</p> Full article ">Figure 10
<p>Comparison for ESDB dataset considering a Utopian case where ground truth values are also removed if the associated calibrated probability is smaller than the selection threshold. Best viewed in color.</p> Full article ">
14 pages, 31666 KiB  
Article
Sequential Normalization: Embracing Smaller Sample Sizes for Normalization
by Neofytos Dimitriou and Ognjen Arandjelović
Information 2022, 13(7), 337; https://doi.org/10.3390/info13070337 - 12 Jul 2022
Cited by 1 | Viewed by 2488
Abstract
Normalization as a layer within neural networks has over the years demonstrated its effectiveness in neural network optimization across a wide range of different tasks, with one of the most successful approaches being that of batch normalization. The consensus is that better estimates [...] Read more.
Normalization as a layer within neural networks has over the years demonstrated its effectiveness in neural network optimization across a wide range of different tasks, with one of the most successful approaches being that of batch normalization. The consensus is that better estimates of the BatchNorm normalization statistics (μ and σ2) in each mini-batch result in better optimization. In this work, we challenge this belief and experiment with a new variant of BatchNorm known as GhostNorm that, despite independently normalizing batches within the mini-batches, i.e., μ and σ2 are independently computed and applied to groups of samples in each mini-batch, outperforms BatchNorm consistently. Next, we introduce sequential normalization (SeqNorm), the sequential application of the above type of normalization across two dimensions of the input, and find that models trained with SeqNorm consistently outperform models trained with BatchNorm or GhostNorm on multiple image classification data sets. Our contributions are as follows: (i) we uncover a source of regularization that is unique to GhostNorm, and not simply an extension from BatchNorm, and illustrate its effects on the loss landscape, (ii) we introduce sequential normalization (SeqNorm) a new normalization layer that improves the regularization effects of GhostNorm, (iii) we compare both GhostNorm and SeqNorm against BatchNorm alone as well as with other regularization techniques, (iv) for both GhostNorm and SeqNorm models, we train models whose performance is consistently better than our baselines, including ones with BatchNorm, on the standard image classification data sets of CIFAR–10, CIFAR-100, and ImageNet ((+0.2%, +0.7%, +0.4%), and (+0.3%, +1.7%, +1.1%) for GhostNorm and SeqNorm, respectively). Full article
(This article belongs to the Topic Advances in Artificial Neural Networks)
►▼ Show Figures

Figure 1

Figure 1
<p>The input tensor with <math display="inline"><semantics> <mrow> <mo>(</mo> <mi>M</mi> <mo>,</mo> <mi>C</mi> <mo>,</mo> <mi>F</mi> <mo>)</mo> </mrow> </semantics></math> dimensions is divided into a number of line (1D) or plane (2D) slices. Each normalization technique slices the input tensor differently and each slice is normalized independently of the other slices. For ease of visualization, both <math display="inline"><semantics> <msub> <mi>G</mi> <mi>C</mi> </msub> </semantics></math> and <math display="inline"><semantics> <msub> <mi>G</mi> <mi>M</mi> </msub> </semantics></math> are set to 4.</p> Full article ">Figure 2
<p>Comparison of the loss landscape on MNIST between BatchNorm, GhostNorm, and the baseline.</p> Full article ">Figure 3
<p>Comparison of the loss landscape on CIFAR-10 between BatchNorm, GhostNorm, and the baseline.</p> Full article ">Figure A1
<p>Python code for GhostNorm in PyTorch.</p> Full article ">Figure A2
<p>Python code for accumulating gradients in PyTorch.</p> Full article ">Figure A3
<p>Comparison of the loss landscape on MNIST between the baseline, BatchNorm, and GhostNorm with different <math display="inline"><semantics> <msub> <mi>G</mi> <mi>M</mi> </msub> </semantics></math> values. The last figure (bottom right) depicts the misclassification error on the testing set during training.</p> Full article ">Figure A4
<p>Comparison of the loss landscape on CIFAR–10 between BatchNorm and GhostNorm with different <math display="inline"><semantics> <msub> <mi>G</mi> <mi>M</mi> </msub> </semantics></math> values. The last figure (bottom right) depicts the misclassification error on the testing set during training.</p> Full article ">
27 pages, 29731 KiB  
Article
Using Augmented Reality in K-12 Education: An Indicative Platform for Teaching Physics
by Christina Volioti, Euclid Keramopoulos, Theodosios Sapounidis, Konstantinos Melisidis, Maria Zafeiropoulou, Charalampos Sotiriou and Vladimiros Spiridis
Information 2022, 13(7), 336; https://doi.org/10.3390/info13070336 - 12 Jul 2022
Cited by 16 | Viewed by 5123
Abstract
Augmented Reality (AR) could provide key benefits in education and create a richer user experience by increasing the motivation and engagement of the students. To this end, the current paper presents a system with three AR applications for teaching physics in the fifth [...] Read more.
Augmented Reality (AR) could provide key benefits in education and create a richer user experience by increasing the motivation and engagement of the students. To this end, the current paper presents a system with three AR applications for teaching physics in the fifth and sixth grades of primary school and in the first grade of secondary school, and the ultimate goal is the development of a unified platform that covers the subject of physics in all classes of K-12 education. The platform provides a useful tool to familiarize both teachers and pupils with AR technologies, aiming to improve the learning and teaching experience and to enhance their skills. The developed system is evaluated in terms of usability, gamification and willingness of the teachers to incorporate this technology into the teaching process. A total of 314 users participated in the research, where they were divided into three user groups: (i) teachers (N = 15), (ii) pupils (N = 189) and (iii) computer science students (N = 110). The outcomes were satisfactory, revealing that the gamified AR applications are easy to use, and teachers are interested in using these AR applications in their classrooms. Full article
(This article belongs to the Collection Augmented Reality Technologies, Systems and Applications)
►▼ Show Figures

Figure 1

Figure 1
<p>System Overview.</p> Full article ">Figure 2
<p>The process of the “Material bodies” experiment.</p> Full article ">Figure 3
<p>The process of the “Energy” experiment.</p> Full article ">Figure 4
<p>The process of the “Digestive system” experiment.</p> Full article ">Figure 5
<p>The process of the “Heat” experiment.</p> Full article ">Figure 6
<p>The process of the “Electricity” experiment.</p> Full article ">Figure 7
<p>The process of the “Light” experiment.</p> Full article ">Figure 8
<p>The process of the “Energy” experiment.</p> Full article ">Figure 9
<p>The process of the “Heat” experiment.</p> Full article ">Figure 10
<p>The process of the “Plant” experiment.</p> Full article ">Figure 11
<p>The process of the “Ecosystem” experiment.</p> Full article ">Figure 12
<p>The process of the “Magnetism” experiment.</p> Full article ">Figure 13
<p>The process of the “Acids” experiment.</p> Full article ">Figure 14
<p>The process of the “Length Measurements—The Average Value” experiment.</p> Full article ">Figure 15
<p>The process of the “Time Measurements—Accuracy” experiment.</p> Full article ">Figure 16
<p>The process of the “Mass Measurements—Charts” experiment.</p> Full article ">Figure 17
<p>The process of the “Temperature Measurements—The Calibration” experiment.</p> Full article ">Figure 18
<p>The process of the “From Heat to Temperature—The Thermal Balance” experiment.</p> Full article ">Figure 19
<p>The process of the “Water Expansion and Contraction—A Natural Abnormality” experiment.</p> Full article ">Figure 20
<p>The process of the “Light Heats—Cold and Warm Colors” experiment.</p> Full article ">Figure 21
<p>The process of the “The Greenhouse Effect Overheats” experiment.</p> Full article ">Figure 22
<p>The process of the “The Electric Short-Circuit—Dangers and Safety” experiment.</p> Full article ">Figure 23
<p>The process for the “Electric Motor” experiment.</p> Full article ">
22 pages, 994 KiB  
Article
A Comprehensive Assessment of Human Factors in Cyber Security Compliance toward Enhancing the Security Practice of Healthcare Staff in Paperless Hospitals
by Prosper Kandabongee Yeng, Muhammad Ali Fauzi and Bian Yang
Information 2022, 13(7), 335; https://doi.org/10.3390/info13070335 - 12 Jul 2022
Cited by 9 | Viewed by 6613
Abstract
Recent reports indicate that over 85% of data breaches are still caused by a human element, of which healthcare is one of the organizations that cyber criminals target. As healthcare IT infrastructure is characterized by a human element, this study comprehensively examined the [...] Read more.
Recent reports indicate that over 85% of data breaches are still caused by a human element, of which healthcare is one of the organizations that cyber criminals target. As healthcare IT infrastructure is characterized by a human element, this study comprehensively examined the effect of psycho-socio-cultural and work factors on security behavior in a typical hospital. A quantitative approach was adopted where we collected responses from 212 healthcare staff through an online questionnaire survey. A broad range of constructs was selected from psychological, social, cultural perception, and work factors based on earlier review work. These were related with some security practices to assess the information security (IS) knowledge, attitude and behavior gaps among healthcare staff in a comprehensive way. The study revealed that work emergency (WE) has a positive correlation with IS conscious care behavior (ISCCB) risk. Conscientiousness also had a positive correlation with ISCCB risk, but agreeableness was negatively correlated with information security knowledge (ISK) risk and information security attitude (ISA) risk. Based on these findings, intrinsic and extrinsic motivation methods combined with cutting-edge technologies can be explored to discourage IS risks behaviors while enhancing conscious care security practice. Full article
(This article belongs to the Section Information Security and Privacy)
►▼ Show Figures

Figure 1

Figure 1
<p>The study model.</p> Full article ">Figure 2
<p>Distribution of knowledge, attitude and behavior of security practice.</p> Full article ">Figure 3
<p>Role categories of respondents.</p> Full article ">Figure 4
<p>Comparison of KAB security practice risk among healthcare staff.</p> Full article ">Figure 5
<p>Position distribution by gender.</p> Full article ">
14 pages, 1211 KiB  
Review
Toward Trust-Based Recommender Systems for Open Data: A Literature Review
by Chenhao Li, Jiyin Zhang, Amruta Kale, Xiang Que, Sanaz Salati and Xiaogang Ma
Information 2022, 13(7), 334; https://doi.org/10.3390/info13070334 - 12 Jul 2022
Cited by 3 | Viewed by 2916
Abstract
In recent years, the concept of “open data” has received increasing attention among data providers and publishers. For some data portals in public sectors, such as data.gov, the openness enables public oversight of governmental proceedings. For many other data portals, especially those in [...] Read more.
In recent years, the concept of “open data” has received increasing attention among data providers and publishers. For some data portals in public sectors, such as data.gov, the openness enables public oversight of governmental proceedings. For many other data portals, especially those in academia, open data has shown its potential for driving new scientific discoveries and creating opportunities for multidisciplinary collaboration. While the number of open data portals and the volume of shared data have increased significantly, most open data portals still use keywords and faceted models as their primary methods for data search and discovery. There should be opportunities to incorporate more intelligent functions to facilitate the data flow between data portals and end-users. To find more theoretical and empirical evidence for that proposition, in this paper, we conduct a systematic literature review of open data, social trust, and recommender systems to explain the fundamental concepts and illustrate the potential of using trust-based recommender systems for open data portals. We hope this literature review can benefit practitioners in the field of open data and facilitate the discussion of future work. Full article
(This article belongs to the Section Information Systems)
►▼ Show Figures

Figure 1

Figure 1
<p>Linear plot for annual article numbers between 2007 and 2022.</p> Full article ">Figure 2
<p>Cumulative trend of authors’ keywords among the 1661 articles.</p> Full article ">Figure 3
<p>Analysis of keyword co-relationship based on authors’ keywords.</p> Full article ">
15 pages, 7410 KiB  
Article
What Attracts the Driver’s Eye? Attention as a Function of Task and Events
by Yke Bauke Eisma, Dirk J. Eijssen and Joost C. F. de Winter
Information 2022, 13(7), 333; https://doi.org/10.3390/info13070333 - 11 Jul 2022
Cited by 5 | Viewed by 2483
Abstract
This study explores how drivers of an automated vehicle distribute their attention as a function of environmental events and driving task instructions. Twenty participants were asked to monitor pre-recorded videos of a simulated driving trip while their eye movements were recorded using an [...] Read more.
This study explores how drivers of an automated vehicle distribute their attention as a function of environmental events and driving task instructions. Twenty participants were asked to monitor pre-recorded videos of a simulated driving trip while their eye movements were recorded using an eye-tracker. The results showed that eye movements are strongly situation-dependent, with areas of interest (windshield, mirrors, and dashboard) attracting attention when events (e.g., passing vehicles) occurred in those areas. Furthermore, the task instructions provided to participants (i.e., speed monitoring or hazard monitoring) affected their attention distribution in an interpretable manner. It is concluded that eye movements while supervising an automated vehicle are strongly ‘top-down’, i.e., based on an expected value. The results are discussed in the context of the development of driver availability monitoring systems. Full article
(This article belongs to the Special Issue Test and Evaluation Methods for Human-Machine Interfaces of Automated Vehicles II)
►▼ Show Figures

Figure 1

Figure 1
<p>Experimental setup. Shown from back to front is the monitor for displaying the videos, the eye-tracker camera with infrared illuminator, the keyboard used for pressing the spacebar and arrow keys, and the head support.</p> Full article ">Figure 2
<p>Heatmap of gaze points for the three videos aggregated for the Baseline Task (<b>top</b>), Speed Task (<b>middle</b>), and Hazard Task (<b>bottom</b>). The heatmap was divided into 10 × 10-pixel squares and normalized. The colors are on a base-10 logarithmic scale. For example, a dark red color (value = 5) means there was 100 times more attention to that region than a yellow color (value = 3).</p> Full article ">Figure 3
<p>Attention distribution across four areas of interest (AOIs) for Video 1 (<b>top</b>), Video 2 (<b>middle</b>), and Video 3 (<b>bottom</b>) in the Baseline task. The colored circular markers at the top of each figure represent the moments where the maximal number of participants glanced at the AOI.</p> Full article ">Figure 4
<p>Moment in Video 3 where most participants attended to the windshield (Baseline Task). The green markers represent the gaze coordinates of individual participants. White lines are drawn around the AOIs.</p> Full article ">Figure 5
<p>Moment in Video 3 where most participants attended to the left mirror (Baseline Task). The green markers represent the gaze coordinates of individual participants. White lines are drawn around the AOIs.</p> Full article ">Figure 6
<p>Moment in Video 3 where most participants attended to the right mirror (Baseline Task). The green markers represent the gaze coordinates of individual participants. White lines are drawn around the AOIs.</p> Full article ">Figure 7
<p>Moment in Video 3 where most participants attended to the dashboard (Baseline Task). The green markers represent the gaze coordinates of individual participants. White lines are drawn around the AOIs.</p> Full article ">Figure 8
<p>Left axis (black): Percentage of participants with their eyes on the left mirror (Video 2; three tasks averaged). Right axis (magenta): Width of the bounding box of the detected vehicle in the left mirror. The shaded background indicates whether the ego-vehicle was driving on the right lane (dark gray) or left lane (light gray).</p> Full article ">Figure 9
<p>Percentage of participants with their eyes on the dashboard (Video 2, Speed Task). The dotted vertical lines represent the moments of 100 km/h threshold crossings.</p> Full article ">Figure 10
<p>Percentage of participants with their eyes on the dashboard as a function of vehicle speed in the Baseline Task, Speed Task, and Hazard Task (Videos 1–3 combined, first 5 s of each video excluded).</p> Full article ">Figure 11
<p>Percentage of participants with their eyes on the left mirror when another vehicle overtook the ego-vehicle in the left lane (left panel) and with their eyes on the right mirror when the ego-vehicle overtook another vehicle in the right lane (right panel) (Videos 1–3 combined, 8 and 15 overtakes, respectively). The vertical line at <span class="html-italic">t</span> = 0 s represents the first moment when part of the overtaking car became visible in the left side window (<b>left panel</b>) or the last moment when part of the overtaken car was visible in the front window (<b>right panel</b>).</p> Full article ">
14 pages, 8791 KiB  
Article
An Effective Method for Detection and Recognition of Uyghur Texts in Images with Backgrounds
by Mayire Ibrayim, Ahmatjan Mattohti and Askar Hamdulla
Information 2022, 13(7), 332; https://doi.org/10.3390/info13070332 - 11 Jul 2022
Cited by 8 | Viewed by 1963
Abstract
Uyghur text detection and recognition in images with simple backgrounds is still a challenging task for Uyghur image content analysis. In this paper, we propose a new effective Uyghur text detection method based on channel-enhanced MSERs and the CNN classification model. In order [...] Read more.
Uyghur text detection and recognition in images with simple backgrounds is still a challenging task for Uyghur image content analysis. In this paper, we propose a new effective Uyghur text detection method based on channel-enhanced MSERs and the CNN classification model. In order to extract more complete text components, a new text candidate region extraction algorithm is put forward, which is based on the channel-enhanced MSERs according to the characteristics of Uyghur text. In order to effectively prune the non-text regions, we design a CNN classification network according to the LeNet-5, which gains the description characteristics automatically and avoids the tedious and low efficiency artificial characteristic extraction work. For Uyghur text recognition in images, we improved the traditional CRNN network, and to verify its effectiveness, the networks trained on a synthetic dataset and evaluated on the text recognition datasets. The experimental results indicated that the Uyghur text detection method in this paper is robust and applicable, and the recognition result by improvedCRNN was better than the original CRNN network. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>The characteristics of Uyghur text.</p> Full article ">Figure 2
<p>Flowchart of the proposed method and corresponding results of each step.</p> Full article ">Figure 3
<p>The network architecture of Uyghur text recognition based on improvement CRNN.</p> Full article ">Figure 4
<p>The correspondence of Uyghur language and Latin.</p> Full article ">Figure 5
<p>Examples of the detection dataset. (<b>a</b>) examples of color images; (<b>b</b>) positive samples in training set; (<b>c</b>) negative samples in training set.</p> Full article ">Figure 6
<p>Examples of the recognition dataset. (<b>a</b>) examples of random synthetic texts images; (<b>b</b>) examples of arbitrary length text in the natural scene images.</p> Full article ">Figure 7
<p>The network architecture of CNN classification model.</p> Full article ">Figure 8
<p>The performance of the CNN classification model in the Validation set. Subfigure (<b>a</b>) the accuracy changing curve of the CNN classification model; Subfigure (<b>b</b>) the average loss changing curve of the CNN classification model.</p> Full article ">
21 pages, 624 KiB  
Article
A Reinforcement Learning Approach to Speech Coding
by Jerry Gibson and Hoontaek Oh
Information 2022, 13(7), 331; https://doi.org/10.3390/info13070331 - 11 Jul 2022
Cited by 2 | Viewed by 1889
Abstract
Speech coding is an essential technology for digital cellular communications, voice over IP, and video conferencing systems. For more than 25 years, the main approach to speech coding for these applications has been block-based analysis-by-synthesis linear predictive coding. An alternative approach that has [...] Read more.
Speech coding is an essential technology for digital cellular communications, voice over IP, and video conferencing systems. For more than 25 years, the main approach to speech coding for these applications has been block-based analysis-by-synthesis linear predictive coding. An alternative approach that has been less successful is sample-by-sample tree coding of speech. We reformulate this latter approach as a multistage reinforcement learning problem with L step lookahead that incorporates exploration and exploitation to adapt model parameters and to control the speech analysis/synthesis process on a sample-by-sample basis. The minimization of the spectrally shaped reconstruction error to finite depth manages complexity and serves as an effective stand in for the overall subjective evaluation of reconstructed speech quality and intelligibility. Different control policies that attempt to persistently excite the system states and that encourage exploration are studied and evaluated. The resulting methods produce reconstructed speech quality competitive with the most popular speech codec utilized today. This new reinforcement learning formulation provides new insights and opens up new directions for system design and performance improvement. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Reinforcement learning for speech coding.</p> Full article ">Figure 2
<p>Lattice Predictor.</p> Full article ">Figure 3
<p>Perceptual spectral error shaping example.</p> Full article ">Figure 4
<p>A 4-2 multitree control segment.</p> Full article ">Figure 5
<p>The 40 sample 4-2 randomized control multitree sequence.</p> Full article ">Figure 6
<p>The 5-level control tree segment.</p> Full article ">Figure 7
<p>Control tree gain adaptation for 2-level tree segments.</p> Full article ">Figure A1
<p>Classical Differential Pulse Code Modulation (DPCM).</p> Full article ">
28 pages, 3820 KiB  
Article
Supervised Learning Models for the Preliminary Detection of COVID-19 in Patients Using Demographic and Epidemiological Parameters
by Aditya Pradhan, Srikanth Prabhu, Krishnaraj Chadaga, Saptarshi Sengupta and Gopal Nath
Information 2022, 13(7), 330; https://doi.org/10.3390/info13070330 - 10 Jul 2022
Cited by 23 | Viewed by 3895
Abstract
The World Health Organization labelled the new COVID-19 breakout a public health crisis of worldwide concern on 30 January 2020, and it was named the new global pandemic in March 2020. It has had catastrophic consequences on the world economy and well-being of [...] Read more.
The World Health Organization labelled the new COVID-19 breakout a public health crisis of worldwide concern on 30 January 2020, and it was named the new global pandemic in March 2020. It has had catastrophic consequences on the world economy and well-being of people and has put a tremendous strain on already-scarce healthcare systems globally, particularly in underdeveloped countries. Over 11 billion vaccine doses have already been administered worldwide, and the benefits of these vaccinations will take some time to appear. Today, the only practical approach to diagnosing COVID-19 is through the RT-PCR and RAT tests, which have sometimes been known to give unreliable results. Timely diagnosis and implementation of precautionary measures will likely improve the survival outcome and decrease the fatality rates. In this study, we propose an innovative way to predict COVID-19 with the help of alternative non-clinical methods such as supervised machine learning models to identify the patients at risk based on their characteristic parameters and underlying comorbidities. Medical records of patients from Mexico admitted between 23 January 2020 and 26 March 2022, were chosen for this purpose. Among several supervised machine learning approaches tested, the XGBoost model achieved the best results with an accuracy of 92%. It is an easy, non-invasive, inexpensive, instant and accurate way of forecasting those at risk of contracting the virus. However, it is pretty early to deduce that this method can be used as an alternative in the clinical diagnosis of coronavirus cases. Full article
(This article belongs to the Topic Advances in Data Analytics with Applications to Health Care)
►▼ Show Figures

Figure 1

Figure 1
<p>Statistical distribution of patients.</p> Full article ">Figure 2
<p>Pearson’s correlation matrix which indicates the strength of the relationship among variables.</p> Full article ">Figure 3
<p>Process workflow of COVID-19 diagnosis using machine learning.</p> Full article ">Figure 4
<p>The architecture of ANN for this research.</p> Full article ">Figure 5
<p>(<b>a</b>) Imbalanced classes, (<b>b</b>) balanced classes after using Borderline-SMOTE.</p> Full article ">Figure 6
<p>Confusion matrix of XGBoost algorithm.</p> Full article ">Figure 7
<p>AUCs of various classifiers that diagnose COVID-19.</p> Full article ">Figure 8
<p>Training and testing accuracies of various classifiers.</p> Full article ">Figure 9
<p>Performance metrics of all the classifiers during the testing phase.</p> Full article ">Figure 10
<p>(<b>a</b>) Bar chart using SHAP (<b>b</b>) beeswarm plot indicating Shapley values.</p> Full article ">Figure 11
<p>(<b>a</b>) COVID-19 positive diagnosis using LIME. (<b>b</b>) COVID-19 negative diagnosis using LIME.</p> Full article ">
19 pages, 1150 KiB  
Article
Comparing Worldwide, National, and Independent Notifications about Adverse Drug Reactions Due to COVID-19 Vaccines
by Francesco Branda and Davide Tosi
Information 2022, 13(7), 329; https://doi.org/10.3390/info13070329 - 8 Jul 2022
Viewed by 3060
Abstract
The rapid development of effective vaccines against COVID-19 is an extraordinary achievement. However, no medical product can ever be considered risk-free. Several countries have a pharmacovigilance system that detects, assesses, understands, and prevents possible adverse effects of a drug. To benefit from such [...] Read more.
The rapid development of effective vaccines against COVID-19 is an extraordinary achievement. However, no medical product can ever be considered risk-free. Several countries have a pharmacovigilance system that detects, assesses, understands, and prevents possible adverse effects of a drug. To benefit from such huge data sources, specialists and researchers need advanced big data analysis tools able to extract value and find valuable insights. This paper defines a general framework for a pharmaceutical data analysis application that provides a predefined (but extensible) set of functions for each data processing step (i.e., data collection, filtering, enriching, analysis, and visualization). As a case study, we present here an analysis of the potential side effects observed following the administration of the COVID-19 vaccines. The experimental evaluation shows that: (i) most adverse events can be classified as non-serious and concern muscle/joint pain, chills and nausea, headache, and fatigue; (ii) the notification rate is higher in the age group 20–39 years and decreases in older age groups and in very young people. Full article
(This article belongs to the Special Issue Advanced Information Technology, Big Data and Artificial Intelligence)
►▼ Show Figures

Figure 1

Figure 1
<p>Workflow of the proposed framework.</p> Full article ">Figure 2
<p>Main results of pharmacovigilance for <span class="html-italic">Comirnaty</span> vaccine: number of reports received over time (<b>a</b>), and their distribution by age group and sex (<b>b</b>) and by seriousness (<b>c</b>).</p> Full article ">Figure 3
<p>Main results of pharmacovigilance for <span class="html-italic">Spikevax</span> vaccine: number of reports received over time (<b>a</b>), and their distribution by age group and sex (<b>b</b>) and by seriousness (<b>c</b>).</p> Full article ">Figure 4
<p>Main results of pharmacovigilance for <span class="html-italic">Vaxzevria</span> vaccine: number of reports received over time (<b>a</b>), and their distribution by age group and sex (<b>b</b>) and by seriousness (<b>c</b>).</p> Full article ">Figure 5
<p>Main results of pharmacovigilance for <span class="html-italic">Janssen</span> vaccine: number of reports received over time (<b>a</b>), and their distribution by age group and sex (<b>b</b>) and by seriousness (<b>c</b>).</p> Full article ">Figure 6
<p>Distribution of reports by onset time of symptoms from vaccination date (<b>a</b>), by 1st dose (<b>b</b>), and by 2nd dose (<b>c</b>).</p> Full article ">Figure 7
<p>Main results of pharmacovigilance activities of AIFA: reporting rate vs. time (<b>a</b>), and distribution by sex (<b>b</b>) and by age (<b>c</b>).</p> Full article ">Figure 8
<p>An example of online reporting of an adverse event on our platform.</p> Full article ">Figure 9
<p>Distribution of reported adverse events on our platform by type (<b>a</b>) and by severity (<b>b</b>).</p> Full article ">Figure 10
<p>Main results of our pharmacovigilance activity: distribution of number of reported cases by sex (<b>a</b>), by age (<b>b</b>), and in “Other” category (<b>c</b>).</p> Full article ">
12 pages, 1826 KiB  
Article
Distributed Edge Computing for Resource Allocation in Smart Cities Based on the IoT
by Omar Abdulkareem Mahmood, Ali R. Abdellah, Ammar Muthanna and Andrey Koucheryavy
Information 2022, 13(7), 328; https://doi.org/10.3390/info13070328 - 7 Jul 2022
Cited by 22 | Viewed by 3624
Abstract
Smart cities using the Internet of Things (IoT) can operate various IoT systems with better services that provide intelligent and efficient solutions for various aspects of urban life. With the rapidly growing number of IoT systems, the many smart city services, and their [...] Read more.
Smart cities using the Internet of Things (IoT) can operate various IoT systems with better services that provide intelligent and efficient solutions for various aspects of urban life. With the rapidly growing number of IoT systems, the many smart city services, and their various quality of service (QoS) constraints, servers face the challenge of allocating limited resources across all Internet-based applications to achieve an efficient per-formance. The presence of a cloud in the IoT system of a smart city results in high energy con-sumption and delays in the network. Edge computing is based on a cloud computing framework where computation, storage, and network resources are moved close to the data source. The IoT framework is identical to cloud computing. The critical issue in edge computing when executing tasks generated by IoT systems is the efficient use of energy while maintaining delay limitations. In this paper, we study a multicriteria optimization approach for resource allocation with distributed edge computing in IoT-based smart cities. We present a three-layer network architecture for IoT-based smart cities. An edge resource allocation scheme based on an auctionable approach is proposed to ensure efficient resource computation for delay-sensitive tasks. Full article
(This article belongs to the Special Issue Advances in Wireless Communications Systems)
►▼ Show Figures

Figure 1

Figure 1
<p>The network structure for IoT-enabled smart city.</p> Full article ">Figure 2
<p>Controller elements.</p> Full article ">Figure 3
<p>Performance vs. tasks—average delay.</p> Full article ">Figure 4
<p>Performance vs. tasks—energy consumption.</p> Full article ">Figure 5
<p>Performance vs. VMs—average delay.</p> Full article ">Figure 6
<p>Performance vs. VMs—energy consumption.</p> Full article ">
24 pages, 443 KiB  
Article
GaSubtle: A New Genetic Algorithm for Generating Subtle Higher-Order Mutants
by Fadi Wedyan, Abdullah Al-Shishani and Yaser Jararweh
Information 2022, 13(7), 327; https://doi.org/10.3390/info13070327 - 7 Jul 2022
Cited by 2 | Viewed by 2183
Abstract
Mutation testing is an effective, yet costly, testing approach, as it requires generating and running large numbers of faulty programs, called mutants. Mutation testing also suffers from a fundamental problem, which is having a large percentage of equivalent mutants. These are mutants that [...] Read more.
Mutation testing is an effective, yet costly, testing approach, as it requires generating and running large numbers of faulty programs, called mutants. Mutation testing also suffers from a fundamental problem, which is having a large percentage of equivalent mutants. These are mutants that produce the same output as the original program, and therefore, cannot be detected. Higher-order mutation is a promising approach that can produce hard-to-detect faulty programs called subtle mutants, with a low percentage of equivalent mutants. Subtle higher-order mutants contribute a small set of the large space of mutants which grows even larger as the order of mutation becomes higher. In this paper, we developed a genetic algorithm for finding subtle higher-order mutants. The proposed approach uses a new mechanism in the crossover phase and uses five selection techniques to select mutants that go to the next generation in the genetic algorithm. We implemented a tool, called GaSubtle that automates the process of creating subtle mutants. We evaluated the proposed approach by using 10 subject programs. Our evaluation shows that the proposed crossover generates more subtle mutants than the technique used in a previous genetic algorithm with less execution time. Results vary on the selection strategies, suggesting a dependency relation with the tested code. Full article
(This article belongs to the Special Issue Information for Business and Management–Software Development for Data Processing and Management)
►▼ Show Figures

Figure 1

Figure 1
<p>Proposed crossover example.</p> Full article ">Figure 2
<p>Original program.</p> Full article ">Figure 3
<p>Parent 1: Third-order mutant selected for crossover.</p> Full article ">Figure 4
<p>Parent 2: Second-order mutant selected for crossover.</p> Full article ">Figure 5
<p>Child 1: Third-order mutant constructed by using the least fit FOMs from parent 1 and parent 2.</p> Full article ">Figure 6
<p>Child 2: Second-order mutant, constructed by using the most fit FOMs from parent 1 and parent 2.</p> Full article ">Figure 7
<p>GaSubtle architecture.</p> Full article ">
17 pages, 5469 KiB  
Article
Enhancing Inference on Physiological and Kinematic Periodic Signals via Phase-Based Interpretability and Multi-Task Learning
by Reza Soleimani and Edgar Lobaton
Information 2022, 13(7), 326; https://doi.org/10.3390/info13070326 - 7 Jul 2022
Cited by 19 | Viewed by 2723
Abstract
Physiological and kinematic signals from humans are often used for monitoring health. Several processes of interest (e.g., cardiac and respiratory processes, and locomotion) demonstrate periodicity. Training models for inference on these signals (e.g., detection of anomalies, and extraction of biomarkers) require large amounts [...] Read more.
Physiological and kinematic signals from humans are often used for monitoring health. Several processes of interest (e.g., cardiac and respiratory processes, and locomotion) demonstrate periodicity. Training models for inference on these signals (e.g., detection of anomalies, and extraction of biomarkers) require large amounts of data to capture their variability, which are not readily available. This hinders the performance of complex inference models. In this work, we introduce a methodology for improving inference on such signals by incorporating phase-based interpretability and other inference tasks into a multi-task framework applied to a generative model. For this purpose, we utilize phase information as a regularization term and as an input to the model and introduce an interpretable unit in a neural network, which imposes an interpretable structure on the model. This imposition helps us in the smooth generation of periodic signals that can aid in data augmentation tasks. We demonstrate the impact of our framework on improving the overall inference performance on ECG signals and inertial signals from gait locomotion. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Illustration of multi-task and interpretability pipeline. <span class="html-italic">f</span> represents the encoder and maps the input to <math display="inline"><semantics> <msub> <mi>z</mi> <mi>t</mi> </msub> </semantics></math>. The decoder is split into <math display="inline"><semantics> <msub> <mi>g</mi> <mi>r</mi> </msub> </semantics></math> for reconstructing the input signal and <math display="inline"><semantics> <msub> <mi>g</mi> <mi>f</mi> </msub> </semantics></math> for forecasting. <span class="html-italic">k</span> is the network for classification. <span class="html-italic">h</span> takes care of estimation of the phase summary <math display="inline"><semantics> <msub> <mi>s</mi> <mi>t</mi> </msub> </semantics></math> for regularization purposes.</p> Full article ">Figure 2
<p>Showing IMU (only a single channel for illustration) and ECG signal with their peaks and extracted phase (in radians) information.</p> Full article ">Figure 3
<p>Comparing the cosine component of two phase modalities. (<b>Bottom</b>) The mean phase shows less variability compared to the middle phase. (<b>Top</b>) The histogram of the mean phases (<b>right</b>) maps most values near zero while the middle phase (<b>left</b>) has a greater spread distribution.</p> Full article ">Figure 4
<p>Forecasting for IMU signals. The first 80 samples (2 s) are given to the model as an input. The reconstructed signal consists of the first 80 samples, and the forecasted signal is the last 80.</p> Full article ">Figure 5
<p>Showing different variants for <math display="inline"><semantics> <mfrac> <msub> <mi>λ</mi> <mn>1</mn> </msub> <msub> <mi>λ</mi> <mn>2</mn> </msub> </mfrac> </semantics></math>. Each plot is within the range [−2,2] for both <span class="html-italic">x</span> and <span class="html-italic">y</span> axis. The rows (from top to bottom) refers to the ratio being greater than one, equal to one and less than one respectively. Note that the full cyclic structure is not captured for <math display="inline"><semantics> <mrow> <mfrac> <msub> <mi>λ</mi> <mn>1</mn> </msub> <msub> <mi>λ</mi> <mn>2</mn> </msub> </mfrac> <mo>&lt;</mo> <mn>1</mn> </mrow> </semantics></math>. The last column does not capture the desired structure due to a low magnitude on the <math display="inline"><semantics> <mi>λ</mi> </semantics></math> regularization weights.</p> Full article ">Figure 6
<p>Fitting a circle in 3D on the circular part of the latent space PCA projection.</p> Full article ">Figure 7
<p>Forecasting for ECG signals. The first 512 samples (2 s) are used as an input to the model and for the reconstruction task, while the second half of the signal is predicted by the model for forecasting.</p> Full article ">Figure A1
<p>Down sampler and decoder for model 1. <math display="inline"><semantics> <mrow> <mi>d</mi> <mi>i</mi> <msub> <mi>m</mi> <mi>e</mi> </msub> </mrow> </semantics></math> is 8 if we use encoding.</p> Full article ">Figure A2
<p>Down sampler and decoder for model 2. <math display="inline"><semantics> <mrow> <mi>d</mi> <mi>i</mi> <msub> <mi>m</mi> <mi>e</mi> </msub> </mrow> </semantics></math> is 8 if we use encoding.</p> Full article ">Figure A3
<p>IMU signal generation using the mapped circle in 3D space. Each point on the circle corresponds to a time series. In total, we have 14 sequence, which are sequentially related. This shows that our intuition about the phase information is valid. Just by having the mapped circle, as it can be observed, we successfully generated a continuous time variant IMU signal. The resolution can be increased, and it depends on the number samples on the mapped circle. This signal generation can be used for augmentation to compensate for data imbalance. The sequence starts at the top of the first column to the bottom and continues to the top of the second column to the bottom.</p> Full article ">
26 pages, 8005 KiB  
Article
Intelligent Video Surveillance Systems for Vehicle Identification Based on Multinet Architecture
by Jacobo González-Cepeda, Álvaro Ramajo and José María Armingol
Information 2022, 13(7), 325; https://doi.org/10.3390/info13070325 - 6 Jul 2022
Cited by 5 | Viewed by 4286
Abstract
Security cameras have been proven to be particularly useful in preventing and combating crime through identification tasks. Here, two areas can be mainly distinguished: person and vehicle identification. Automatic license plate readers are the most widely used tool for vehicle identification. Although these [...] Read more.
Security cameras have been proven to be particularly useful in preventing and combating crime through identification tasks. Here, two areas can be mainly distinguished: person and vehicle identification. Automatic license plate readers are the most widely used tool for vehicle identification. Although these systems are very effective, they are not reliable enough in certain circumstances. For example, due to traffic jams, vehicle position or weather conditions, the sensors cannot capture an image of the entire license plate. However, there is still a lot of additional information in the image which may also be of interest, and that needs to be analysed quickly and accurately. The correct use of the processing mechanisms can significantly reduce analysis time, increasing the efficiency of video cameras significantly. To solve this problem, we have designed a solution based on two technologies: license plate recognition and vehicle re-identification. For its development and testing, we have also created several datasets recreating a real environment. In addition, during this article, it is also possible to read about some of the main artificial intelligence techniques for these technologies, as they have served as the starting point for this research. Full article
(This article belongs to the Special Issue Computer Vision for Security Applications)
►▼ Show Figures

Figure 1

Figure 1
<p>Example of an access control camera for license plate reading [<a href="#B8-information-13-00325" class="html-bibr">8</a>].</p> Full article ">Figure 2
<p>Example of DORI concept: (<b>a</b>) Detection, (<b>b</b>) Observation, (<b>c</b>) Recognition and (<b>d</b>) Identification.</p> Full article ">Figure 3
<p>Representation of main stages in a multi-stage license plate recognition system.</p> Full article ">Figure 4
<p>License plate detection methods.</p> Full article ">Figure 5
<p>Representation of the “anchor” (<b>a</b>) and a network that applies the triplet loss function (<b>b</b>).</p> Full article ">Figure 6
<p>Representation of a GAN.</p> Full article ">Figure 7
<p>Example representation of attention maps [<a href="#B53-information-13-00325" class="html-bibr">53</a>].</p> Full article ">Figure 8
<p>Example of Images from Stanford Cars Dataset [<a href="#B65-information-13-00325" class="html-bibr">65</a>].</p> Full article ">Figure 9
<p>Examples of images from the VeRi dataset [<a href="#B67-information-13-00325" class="html-bibr">67</a>] and VeRi-Wild [<a href="#B55-information-13-00325" class="html-bibr">55</a>].</p> Full article ">Figure 10
<p>Proposed multi-neural network architecture.</p> Full article ">Figure 11
<p>Character recognition.</p> Full article ">Figure 12
<p>Highway Gantry Dataset sample.</p> Full article ">Figure 13
<p>Operational Urban Dataset sample.</p> Full article ">Figure 14
<p>Spanish ALPR Dataset examples.</p> Full article ">Figure 15
<p>Dropout and learning rate effect.</p> Full article ">Figure 16
<p>EfficientNet B0-B3-B7 comparison in Stanford-Cars training.</p> Full article ">Figure 17
<p>EfficientNet B0-B3-B7 comparison in VeRi-776 training.</p> Full article ">Figure 18
<p>License plate detection training graph.</p> Full article ">Figure 19
<p>Character recognition training graph.</p> Full article ">
33 pages, 1079 KiB  
Review
Interactive Search on the Web: The Story So Far
by Sareh Aghaei, Kevin Angele, Elwin Huaman, Geni Bushati, Mathias Schiestl and Anna Fensel
Information 2022, 13(7), 324; https://doi.org/10.3390/info13070324 - 4 Jul 2022
Cited by 4 | Viewed by 4723
Abstract
Search on the web, specifically fetching of the relevant content, has been paid attention to since the advent of the web and particularly in recent years due to the tremendous growth in the volume of data and web pages. This paper categorizes the [...] Read more.
Search on the web, specifically fetching of the relevant content, has been paid attention to since the advent of the web and particularly in recent years due to the tremendous growth in the volume of data and web pages. This paper categorizes the search services from the early days of the web to the present into keyword search engines, semantic search engines, question answering systems, dialogue systems and chatbots. As the first generation of search engines, keyword search engines have adopted keyword-based techniques to find the web pages containing the query keywords and ranking search results. In contrast, semantic search engines try to find meaningful and accurate results on the meaning and relations of things. Question-answering systems aim to find precise answers to natural language questions rather than returning a ranked list of relevant sources. As a subset of question answering systems, dialogue systems target to interact with human users through a dialog expressed in natural language. As a subset of dialogue systems, chatbots try to simulate human-like conversations. The paper provides an overview of the typical aspects of the studied search services, including process models, data preparation and presentation, common methodologies and categories. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>The interaction of different types of the search services.</p> Full article ">Figure 2
<p>The common categories of different types of the search services.</p> Full article ">Figure 3
<p>Keyword search engine process model.</p> Full article ">Figure 4
<p>Semantic search engines process model.</p> Full article ">Figure 5
<p>The document-oriented process model.</p> Full article ">Figure 6
<p>The process model of Dialogue systems.</p> Full article ">Figure 7
<p>Chatbot process model.</p> Full article ">
14 pages, 559 KiB  
Article
Digital Transformation Strategy in Post-COVID Era: Innovation Performance Determinants and Digital Capabilities in Driving Schools
by Evangelia Nousopoulou, Maria Kamariotou and Fotis Kitsios
Information 2022, 13(7), 323; https://doi.org/10.3390/info13070323 - 4 Jul 2022
Cited by 20 | Viewed by 6222
Abstract
Businesses affected by the pandemic have realized the importance of incorporating digital transformation into their operations. However, as a result of the market lockdown, they realized that they needed to digitalize their firms immediately and make greater attempts to enhance their economic situation [...] Read more.
Businesses affected by the pandemic have realized the importance of incorporating digital transformation into their operations. However, as a result of the market lockdown, they realized that they needed to digitalize their firms immediately and make greater attempts to enhance their economic situation by integrating a greater number of technological components. While there have been numerous studies conducted on the adoption of digital transformation in small–medium enterprises, there has been no research carried out on the implementation of digital transformation in the specific industry of driving schools. This paper investigates the significance of digital transformation, as well as the potential for its application in this industry’s business setting and the ways in which it can be utilized to improve innovation capabilities and performance. The data for this study came from 300 driving instructors in Greece and Cyprus. Multivariate regression analysis was used to analyze the data. The outcomes suggest that driving schools have a generally positive reaction to and acknowledgement of the increasing speed of digital transformation. The results also give driving school owners useful information that helps them show how important digital transformation is to their businesses. Using the findings of this study, driving schools will be able to improve their operational capabilities and accelerate their development in the post-COVID era. Full article
(This article belongs to the Special Issue Knowledge Management and Digital Humanities)
►▼ Show Figures

Figure 1

Figure 1
<p>Research model.</p> Full article ">
16 pages, 3021 KiB  
Article
An Intrusion Detection Method for Industrial Control System Based on Machine Learning
by Yixin Cao, Lei Zhang, Xiaosong Zhao, Kai Jin and Ziyi Chen
Information 2022, 13(7), 322; https://doi.org/10.3390/info13070322 - 3 Jul 2022
Cited by 8 | Viewed by 3758
Abstract
The integration of communication networks and the internet of industrial control in Industrial Control System (ICS) increases their vulnerability to cyber attacks, causing devastating outcomes. Traditional Intrusion Detection Systems (IDS) largely rely on predefined models and are trained mostly on specific cyber attacks, [...] Read more.
The integration of communication networks and the internet of industrial control in Industrial Control System (ICS) increases their vulnerability to cyber attacks, causing devastating outcomes. Traditional Intrusion Detection Systems (IDS) largely rely on predefined models and are trained mostly on specific cyber attacks, which means the traditional IDS cannot cope with unknown attacks. Additionally, most IDS do not consider the imbalanced nature of ICS datasets, thus suffering from low accuracy and high False Positive Rates when being put to use. In this paper, we propose the NCO–double-layer DIFF_RF–OPFYTHON intrusion detection method for ICS, which consists of NCO modules, double-layer DIFF_RF modules, and OPFYTHON modules. Detected traffic will be divided into three categories by the double-layer DIFF_RF module: known attacks, unknown attacks, and normal traffic. Then, the known attacks will be classified into specific attacks by the OPFYTHON module according to the feature of attack traffic. Finally, we use the NCO module to improve the model input and enhance the accuracy of the model. The results show that the proposed method outperforms traditional intrusion detection methods, such as XGboost and SVM. The detection of unknown attacks is also considerable. The accuracy of the dataset used in this paper reaches 98.13%. The detection rates for unknown attacks and known attacks reach 98.21% and 95.1%, respectively. Moreover, the method we proposed has achieved suitable results on other public datasets. Full article
(This article belongs to the Special Issue Advances in Computing, Communication & Security)
►▼ Show Figures

Figure 1

Figure 1
<p>The structure of the NCO–double-layer DIFF_RF–OPFYTHON model.</p> Full article ">Figure 2
<p>The flow chart of the NCO.</p> Full article ">Figure 3
<p>The flow of the double-layer DIFF_RF model.</p> Full article ">Figure 4
<p>Detection process of the final model.</p> Full article ">Figure 5
<p><span class="html-italic">Acc</span> and DR_unknown under different threshold parameters. (<b>a</b>) <span class="html-italic">Acc</span> under different threshold parameters. (<b>b</b>) DR_unknown under different threshold parameters.</p> Full article ">Figure 6
<p>The results of the four models on experimental dataset l.</p> Full article ">Figure 7
<p>The heatmap of the correlation coefficients of input variables.</p> Full article ">Figure 8
<p>The performance of final model on datasets of different sizes.</p> Full article ">
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-46
Previous Issue
Next Issue
Back to TopTop