Next Issue
Volume 12, May
Previous Issue
Volume 12, March
 
 

Information, Volume 12, Issue 4 (April 2021) – 38 articles

Cover Story (view full-size image): With the heterogeneous nature of an IoT system, the number of active devices, the number of resources consumed at a given time, and the task duration will always change to adapt to the environment. Due to this problem, it is difficult to choose a single scheduling algorithm because of the trade-off between creating an optimized scheduler and the computation time required. Our approach utilizes multiple scheduling algorithms via switching to achieve pseudo-dynamic scheduler optimization that reacts with the computation deadline. Our solution relies on multi-task learning to identify the suitable scheduling algorithm based on the parameters given and a fallback scheduling algorithm as a failsafe to meet the computation deadline. 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 Readerexternal link to open them.
Order results
Result details
Section
Select all
Export citation of selected articles as:
Article
Evaluation of a Human–Machine Interface for Motion Sickness Mitigation Utilizing Anticipatory Ambient Light Cues in a Realistic Automated Driving Setting
Information 2021, 12(4), 176; https://doi.org/10.3390/info12040176 - 20 Apr 2021
Cited by 4 | Viewed by 2312
Abstract
Motion sickness (MS) is a syndrome associated with symptoms like nausea, dizziness, and other forms of physical discomfort. Automated vehicles (AVs) are potent at inducing MS because users are not adapted to this novel form of transportation, are provided with less information about [...] Read more.
Motion sickness (MS) is a syndrome associated with symptoms like nausea, dizziness, and other forms of physical discomfort. Automated vehicles (AVs) are potent at inducing MS because users are not adapted to this novel form of transportation, are provided with less information about the own vehicle’s trajectory, and are likely to engage in non-driving related tasks. Because individuals with an especially high MS susceptibility could be limited in their use of AVs, the demand for MS mitigation strategies is high. Passenger anticipation has been shown to have a modulating effect on symptoms, thus mitigating MS. To find an effective mitigation strategy, the prototype of a human–machine interface (HMI) that presents anticipatory ambient light cues for the AV’s next turn to the passenger was evaluated. In a realistic driving study with participants (N = 16) in an AV on a test track, an MS mitigation effect was evaluated based on the MS increase during the trial. An MS mitigation effect was found within a highly susceptible subsample through the presentation of anticipatory ambient light cues. The HMI prototype was proven to be effective regarding highly susceptible users. Future iterations could alleviate MS in field settings and improve the acceptance of AVs. Full article
(This article belongs to the Section Information Applications)
Show Figures

Figure 1

Editorial
Editorial for the Special Issue on “Fault Trees and Attack Trees: Extensions, Solution Methods, and Applications”
Information 2021, 12(4), 175; https://doi.org/10.3390/info12040175 - 20 Apr 2021
Viewed by 940
Abstract
Fault Trees are well-known models for the reliability analysis of systems, used to compute several kinds of qualitative and quantitative measures, such as minimal cut-sets, system failure probability, sensitivity (importance) indices, etc [...] Full article
Article
The Nature of Employee–Organization Relationships at Polish Universities under Pandemic Conditions
Information 2021, 12(4), 174; https://doi.org/10.3390/info12040174 - 19 Apr 2021
Cited by 2 | Viewed by 1617
Abstract
(1) Background: The aim of this study is to describe manager–employee and employee–employee relations during the COVID-19 pandemic and their impact on measures of the likely use of elements of remote teaching by university employees in the future. (2) Methods: The study used [...] Read more.
(1) Background: The aim of this study is to describe manager–employee and employee–employee relations during the COVID-19 pandemic and their impact on measures of the likely use of elements of remote teaching by university employees in the future. (2) Methods: The study used a descriptive-correlation research design with a survey as the primary instrument for data gathering. A total of 732 personnel took part in the survey, selected by a convenience sampling technique. The researchers used an adapted and modified instrument to gather data. The instrument underwent a reliability test. This study used structural equation modeling to confirm hypotheses. (3) Results: It was shown that manager–employee relations at Polish universities during the COVID-19 pandemic were of low quality. However, employee–employee relations were of above-average quality, and have a significant positive impact on intentions to use elements of remote working in the future. (4) Conclusions: Based on the results of the study, some general recommendations are presented for change management and relationship-building. Full article
(This article belongs to the Special Issue Business Process Management)
Show Figures

Figure 1

Article
Information Behavior on Video on Demand Services: User Motives and Their Selection Criteria for Content
Information 2021, 12(4), 173; https://doi.org/10.3390/info12040173 - 16 Apr 2021
Viewed by 2038
Abstract
Introduction. Are viewers of video-on-demand (VoD) services more intrinsically (i.e., preferentially self-determined) or extrinsically (i.e., externally determined) motivated when selecting movies and series? For extrinsic motivation, we distinguish between algorithmically generated suggestions from the services and personal recommendations from other users. Methods. [...] Read more.
Introduction. Are viewers of video-on-demand (VoD) services more intrinsically (i.e., preferentially self-determined) or extrinsically (i.e., externally determined) motivated when selecting movies and series? For extrinsic motivation, we distinguish between algorithmically generated suggestions from the services and personal recommendations from other users. Methods. We empirically investigated the information behavior on video streaming services of users from German-speaking countries with the help of an online survey (N = 1258). Results. Active VoD users watch videos online mainly on a daily basis. They are externally determined in the selection of their videos both by algorithmically generated recommendations from the systems and―to a higher extent―from personal suggestions from acquaintances, friends, and relatives. However, there is a clear indication that intrinsic motivation plays a major role in the selection of videos. Discussion. Users of VoD services move in a cycle between machine-generated recommendations, suggestions, and exchange of opinions from and with other people, and self-determined information behavior. Full article
Show Figures

Figure 1

Article
An Academic Text Recommendation Method Based on Graph Neural Network
Information 2021, 12(4), 172; https://doi.org/10.3390/info12040172 - 16 Apr 2021
Cited by 1 | Viewed by 1384
Abstract
Academic text recommendation, as a kind of text recommendation, has a wide range of application prospects. Predicting texts of interest to scholars in different fields based on anonymous sessions is a challenging problem. However, the existing session-based method only considers the sequential information, [...] Read more.
Academic text recommendation, as a kind of text recommendation, has a wide range of application prospects. Predicting texts of interest to scholars in different fields based on anonymous sessions is a challenging problem. However, the existing session-based method only considers the sequential information, and pays more attention to capture the session purpose. The relationship between adjacent items in the session is not noticed. Specifically in the field of session-based text recommendation, the most important semantic relationship of text is not fully utilized. Based on the graph neural network and attention mechanism, this paper proposes a session-based text recommendation model (TXT-SR) incorporating the semantic relations, which is applied to the academic field. TXT-SR makes full use of the tightness of semantic connections between adjacent texts. We have conducted experiments on two real-life academic datasets from CiteULike. Experimental results show that TXT-SR has better effectiveness than existing session-based recommendation methods. Full article
(This article belongs to the Special Issue Information Retrieval, Recommender Systems and Adaptive Systems)
Show Figures

Figure 1

Article
Cyberbullying Detection in Social Networks Using Bi-GRU with Self-Attention Mechanism
Information 2021, 12(4), 171; https://doi.org/10.3390/info12040171 - 16 Apr 2021
Cited by 15 | Viewed by 2349
Abstract
With the propagation of cyberbullying in social networks as a trending subject, cyberbullying detection has become a social problem that researchers are concerned about. Developing intelligent models and systems helps detect cyberbullying automatically. This work focuses on text-based cyberbullying detection because it is [...] Read more.
With the propagation of cyberbullying in social networks as a trending subject, cyberbullying detection has become a social problem that researchers are concerned about. Developing intelligent models and systems helps detect cyberbullying automatically. This work focuses on text-based cyberbullying detection because it is the commonly used information carrier in social networks and is the widely used feature in this regard studies. Motivated by the documented success of neural networks, we propose a complete model combining the bidirectional gated recurrent unit (Bi-GRU) and the self-attention mechanism. In detail, we introduce the design of a GRU cell and Bi-GRU’s advantage for learning the underlying relationships between words from both directions. Besides, we present the design of the self-attention mechanism and the benefit of this joining for achieving a greater performance of cyberbullying classification tasks. The proposed model could address the limitation of the vanishing and exploding gradient problems. We avoid using oversampling or downsampling on experimental data which could result in the overestimation of evaluation. We conduct a comparative assessment on two commonly used datasets, and the results show that our proposed method outperformed baselines in all evaluation metrics. Full article
(This article belongs to the Section Artificial Intelligence)
Show Figures

Figure 1

Article
Exploring the Value of Nodes with Multicommunity Membership for Classification with Graph Convolutional Neural Networks
Information 2021, 12(4), 170; https://doi.org/10.3390/info12040170 - 15 Apr 2021
Cited by 3 | Viewed by 1993
Abstract
Sampling is an important step in the machine learning process because it prioritizes samples that help the model best summarize the important concepts required for the task at hand. The process of determining the best sampling method has been rarely studied in the [...] Read more.
Sampling is an important step in the machine learning process because it prioritizes samples that help the model best summarize the important concepts required for the task at hand. The process of determining the best sampling method has been rarely studied in the context of graph neural networks. In this paper, we evaluate multiple sampling methods (i.e., ascending and descending) that sample based off different definitions of centrality (i.e., Voterank, Pagerank, degree) to observe its relation with network topology. We find that no sampling method is superior across all network topologies. Additionally, we find situations where ascending sampling provides better classification scores, showing the strength of weak ties. Two strategies are then created to predict the best sampling method, one that observes the homogeneous connectivity of the nodes, and one that observes the network topology. In both methods, we are able to evaluate the best sampling direction consistently. Full article
(This article belongs to the Section Artificial Intelligence)
Show Figures

Figure 1

Article
RDAMS: An Efficient Run-Time Approach for Memory Fault and Hardware Trojans Detection
Information 2021, 12(4), 169; https://doi.org/10.3390/info12040169 - 14 Apr 2021
Cited by 3 | Viewed by 1173
Abstract
Ensuring the security of IoT devices and chips at runtime has become an urgent task as they have been widely used in human life. Embedded memories are vital components of SoC (System on Chip) in these devices. If they are attacked or incur [...] Read more.
Ensuring the security of IoT devices and chips at runtime has become an urgent task as they have been widely used in human life. Embedded memories are vital components of SoC (System on Chip) in these devices. If they are attacked or incur faults at runtime, it will bring huge losses. In this paper, we propose a run-time detection architecture for memory security (RDAMS) to detect memory threats (fault and Hardware Trojans attack). The architecture consists of a Security Detection Core (SDC) that controls and enforces the detection procedure as a “security brain”, and a memory wrapper (MEM_wrapper) which interacts with memory to assist the detection. We also design a low latency response mechanism to solve the SoC performance degradation caused by run-time detection. A block-based multi-granularity detection approach is proposed to render the design flexible and reduce the cost in implementation using the FPGA’s dynamic partial reconfigurable (DPR) technology, which enables online detection mode reconfiguration according to the requirements. Experimental results show that RDAMS can correctly detect and identify 10 modeled memory faults and two types of Hardware Trojans (HTs) attacks without leading a great performance degradation to the system. Full article
Show Figures

Figure 1

Article
Diagnostic of Data Processing by Brazilian Organizations—A Low Compliance Issue
Information 2021, 12(4), 168; https://doi.org/10.3390/info12040168 - 14 Apr 2021
Cited by 7 | Viewed by 1563
Abstract
In order to guarantee the privacy of users’ data, the Brazilian government created the Brazilian General Data Protection Law (LGPD). This article made a diagnostic of Brazilian organizations in relation to their suitability for LGPD, based on the perception of Information Technology (IT) [...] Read more.
In order to guarantee the privacy of users’ data, the Brazilian government created the Brazilian General Data Protection Law (LGPD). This article made a diagnostic of Brazilian organizations in relation to their suitability for LGPD, based on the perception of Information Technology (IT) practitioners who work in these organizations. We used a survey with 41 questions to diagnose different Brazilian organizations, both public and private. The diagnostic questionnaire was answered by 105 IT practitioners. The results show that 27% of organizations process personal data of public access based on good faith and LGPD principles. In addition, our findings also revealed that 16.3% of organizations have not established a procedure or methodology to verify that the LGPD principles are being respected during the development of services that will handle personal data from the product or service design phase to its execution and 20% of the organizations did not establish a communication process to the personal data holders, regarding the possible data breaches. The result of the diagnostic allows organizations and data users to have an overview of how the treatment of personal data of their customers is being treated and which points of attention are in relation to the principles of LGPD. Full article
Show Figures

Figure 1

Article
Virtual Restoration and Virtual Reconstruction in Cultural Heritage: Terminology, Methodologies, Visual Representation Techniques and Cognitive Models
Information 2021, 12(4), 167; https://doi.org/10.3390/info12040167 - 13 Apr 2021
Cited by 24 | Viewed by 3449
Abstract
Today, the practice of making digital replicas of artworks and restoring and recontextualizing them within artificial simulations is widespread in the virtual heritage domain. Virtual reconstructions have achieved results of great realistic and aesthetic impact. Alongside the practice, a growing methodological awareness has [...] Read more.
Today, the practice of making digital replicas of artworks and restoring and recontextualizing them within artificial simulations is widespread in the virtual heritage domain. Virtual reconstructions have achieved results of great realistic and aesthetic impact. Alongside the practice, a growing methodological awareness has developed of the extent to which, and how, it is permissible to virtually operate in the field of restoration, avoid a false sense of reality, and preserve the reliability of the original content. However, there is not yet a full sharing of meanings in virtual restoration and reconstruction domains. Therefore, this article aims to clarify and define concepts, functions, fields of application, and methodologies. The goal of virtual heritage is not only producing digital replicas. In the absence of materiality, what emerges as a fundamental value are the interaction processes, the semantic values that can be attributed to the model itself. The cognitive process originates from this interaction. The theoretical discussion is supported by exemplar case studies carried out by the authors over almost twenty years. Finally, the concepts of uniqueness and authenticity need to be again pondered in light of the digital era. Indeed, real and virtual should be considered as a continuum, as they exchange information favoring new processes of interaction and critical thinking. Full article
(This article belongs to the Special Issue Virtual Reality Technologies and Applications for Cultural Heritage)
Show Figures

Figure 1

Addendum
Addendum: Dumouchel, S., et al. GOTRIPLE: A User-Centric Process to Develop a Discovery Platform. Information 2020, 11, 563
Information 2021, 12(4), 166; https://doi.org/10.3390/info12040166 - 13 Apr 2021
Viewed by 971
Abstract
The authors would like to add the following reference to the “Reference” section of their paper [...] Full article
Article
A 2D Convolutional Gating Mechanism for Mandarin Streaming Speech Recognition
Information 2021, 12(4), 165; https://doi.org/10.3390/info12040165 - 12 Apr 2021
Cited by 1 | Viewed by 1101
Abstract
Recent research shows recurrent neural network-Transducer (RNN-T) architecture has become a mainstream approach for streaming speech recognition. In this work, we investigate the VGG2 network as the input layer to the RNN-T in streaming speech recognition. Specifically, before the input feature is passed [...] Read more.
Recent research shows recurrent neural network-Transducer (RNN-T) architecture has become a mainstream approach for streaming speech recognition. In this work, we investigate the VGG2 network as the input layer to the RNN-T in streaming speech recognition. Specifically, before the input feature is passed to the RNN-T, we introduce a gated-VGG2 block, which uses the first two layers of the VGG16 to extract contextual information in the time domain, and then use a SEnet-style gating mechanism to control what information in the channel domain is to be propagated to RNN-T. The results show that the RNN-T model with the proposed gated-VGG2 block brings significant performance improvement when compared to the existing RNN-T model, and it has a lower latency and character error rate than the Transformer-based model. Full article
Show Figures

Figure 1

Article
Pedestrian Arching Mechanism at Bottleneck in Subway Transit Hub
Information 2021, 12(4), 164; https://doi.org/10.3390/info12040164 - 11 Apr 2021
Cited by 2 | Viewed by 1299
Abstract
Under the massive pedestrian flow, pedestrians arching phenomenon forms easily at bottleneck in subway hubs, which might stampede and crush. To explore pedestrian arching mechanism at bottleneck in subway transit hub, this paper conducts a series of simulation experiment. Firstly, movement preference characteristic [...] Read more.
Under the massive pedestrian flow, pedestrians arching phenomenon forms easily at bottleneck in subway hubs, which might stampede and crush. To explore pedestrian arching mechanism at bottleneck in subway transit hub, this paper conducts a series of simulation experiment. Firstly, movement preference characteristic in subway transit hubs was introduced into the social force model which considers multiple force. Then, after setting basic experiment scenario, unidirectional flow at different bottlenecks were simulated. Finally, the mechanism of pedestrian arching phenomenon at bottleneck was quantitative analyzed with the help of experimental data. Some main conclusions are summarized. Pedestrian arching phenomenon could be divided into four stages: Free, arching formation, arching stabilization and arching dissipation. In addition, the relationship between bottleneck scenario and passing time could be built to a function model. With the different of bottleneck width ratio, passing time presents positive correlation. The research results could give some helps for understanding the dynamic evolution process of unidirectional flow at bottleneck, improving the pedestrian efficiency at bottleneck and optimizing pedestrian facilities in subway transit hub. Full article
(This article belongs to the Section Information Theory and Methodology)
Show Figures

Figure 1

Article
Effects of COVID-19 Pandemic on University Students’ Learning
Information 2021, 12(4), 163; https://doi.org/10.3390/info12040163 - 11 Apr 2021
Cited by 14 | Viewed by 28814
Abstract
The risk of COVID-19 in higher education has affected all its degrees and forms of training. To assess the impact of the pandemic on the learning of university students, a new reference framework for educational data processing was proposed. The framework unifies the [...] Read more.
The risk of COVID-19 in higher education has affected all its degrees and forms of training. To assess the impact of the pandemic on the learning of university students, a new reference framework for educational data processing was proposed. The framework unifies the steps of analysis of COVID-19 effects on the higher education institutions in different countries and periods of the pandemic. It comprises both classical statistical methods and modern intelligent methods: machine learning, multi-criteria decision making and big data with symmetric and asymmetric information. The new framework has been tested to analyse a dataset collected from a university students’ survey, which was conducted during the second wave of COVID-19 at the end of 2020. The main tasks of this research are as follows: (1) evaluate the attitude and the readiness of students in regard to distance learning during the lockdown; (2) clarify the difficulties, the possible changes and the future expectations from distance learning in the next few months; (3) propose recommendations and measures for improving the higher education environment. After data analysis, the conclusions are drawn and recommendations are made for enhancement of the quality of distance learning of university students. Full article
(This article belongs to the Section Information Processes)
Show Figures

Figure 1

Review
Effects of User Interfaces on Take-Over Performance: A Review of the Empirical Evidence
Information 2021, 12(4), 162; https://doi.org/10.3390/info12040162 - 10 Apr 2021
Cited by 4 | Viewed by 1921
Abstract
In automated driving, the user interface plays an essential role in guiding transitions between automated and manual driving. This literature review identified 25 studies that explicitly studied the effectiveness of user interfaces in automated driving. Our main selection criterion was how the user [...] Read more.
In automated driving, the user interface plays an essential role in guiding transitions between automated and manual driving. This literature review identified 25 studies that explicitly studied the effectiveness of user interfaces in automated driving. Our main selection criterion was how the user interface (UI) affected take-over performance in higher automation levels allowing drivers to take their eyes off the road (SAE3 and SAE4). We categorized user interface (UI) factors from an automated vehicle-related information perspective. Short take-over times are consistently associated with take-over requests (TORs) initiated by the auditory modality with high urgency levels. On the other hand, take-over requests directly displayed on non-driving-related task devices and augmented reality do not affect take-over time. Additional explanations of take-over situation, surrounding and vehicle information while driving, and take-over guiding information were found to improve situational awareness. Hence, we conclude that advanced user interfaces can enhance the safety and acceptance of automated driving. Most studies showed positive effects of advanced UI, but a number of studies showed no significant benefits, and a few studies showed negative effects of advanced UI, which may be associated with information overload. The occurrence of positive and negative results of similar UI concepts in different studies highlights the need for systematic UI testing across driving conditions and driver characteristics. Our findings propose future UI studies of automated vehicle focusing on trust calibration and enhancing situation awareness in various scenarios. Full article
Show Figures

Figure 1

Article
Adaptive Machine Learning for Robust Diagnostics and Control of Time-Varying Particle Accelerator Components and Beams
Information 2021, 12(4), 161; https://doi.org/10.3390/info12040161 - 10 Apr 2021
Cited by 1 | Viewed by 1842
Abstract
Machine learning (ML) is growing in popularity for various particle accelerator applications including anomaly detection such as faulty beam position monitor or RF fault identification, for non-invasive diagnostics, and for creating surrogate models. ML methods such as neural networks (NN) are useful because [...] Read more.
Machine learning (ML) is growing in popularity for various particle accelerator applications including anomaly detection such as faulty beam position monitor or RF fault identification, for non-invasive diagnostics, and for creating surrogate models. ML methods such as neural networks (NN) are useful because they can learn input-output relationships in large complex systems based on large data sets. Once they are trained, methods such as NNs give instant predictions of complex phenomenon, which makes their use as surrogate models especially appealing for speeding up large parameter space searches which otherwise require computationally expensive simulations. However, quickly time varying systems are challenging for ML-based approaches because the actual system dynamics quickly drifts away from the description provided by any fixed data set, degrading the predictive power of any ML method, and limits their applicability for real time feedback control of quickly time-varying accelerator components and beams. In contrast to ML methods, adaptive model-independent feedback algorithms are by design robust to un-modeled changes and disturbances in dynamic systems, but are usually local in nature and susceptible to local extrema. In this work, we propose that the combination of adaptive feedback and machine learning, adaptive machine learning (AML), is a way to combine the global feature learning power of ML methods such as deep neural networks with the robustness of model-independent control. We present an overview of several ML and adaptive control methods, their strengths and limitations, and an overview of AML approaches. Full article
(This article belongs to the Special Issue Machine Learning and Accelerator Technology)
Show Figures

Figure 1

Article
Conversation Concepts: Understanding Topics and Building Taxonomies for Financial Services
Information 2021, 12(4), 160; https://doi.org/10.3390/info12040160 - 09 Apr 2021
Cited by 2 | Viewed by 1735
Abstract
Knowledge graphs are proving to be an increasingly important part of modern enterprises, and new applications of such enterprise knowledge graphs are still being found. In this paper, we report on the experience with the use of an automatic knowledge graph system called [...] Read more.
Knowledge graphs are proving to be an increasingly important part of modern enterprises, and new applications of such enterprise knowledge graphs are still being found. In this paper, we report on the experience with the use of an automatic knowledge graph system called Saffron in the context of a large financial enterprise and show how this has found applications within this enterprise as part of the “Conversation Concepts Artificial Intelligence” tool. In particular, we analyse the use cases for knowledge graphs within this enterprise, and this led us to a new extension to the knowledge graph system. We present the results of these adaptations, including the introduction of a semi-supervised taxonomy extraction system, which includes analysts in-the-loop. Further, we extend the kinds of relations extracted by the system and show how the use of the BERTand ELMomodels can produce high-quality results. Thus, we show how this tool can help realize a smart enterprise and how requirements in the financial industry can be realised by state-of-the-art natural language processing technologies. Full article
(This article belongs to the Collection Knowledge Graphs for Search and Recommendation)
Show Figures

Figure 1

Article
On Two-Stage Guessing
Information 2021, 12(4), 159; https://doi.org/10.3390/info12040159 - 09 Apr 2021
Cited by 1 | Viewed by 1515
Abstract
Stationary memoryless sources produce two correlated random sequences Xn and Yn. A guesser seeks to recover Xn in two stages, by first guessing Yn and then Xn. The contributions of this work are twofold: (1) We [...] Read more.
Stationary memoryless sources produce two correlated random sequences Xn and Yn. A guesser seeks to recover Xn in two stages, by first guessing Yn and then Xn. The contributions of this work are twofold: (1) We characterize the least achievable exponential growth rate (in n) of any positive ρ-th moment of the total number of guesses when Yn is obtained by applying a deterministic function f component-wise to Xn. We prove that, depending on f, the least exponential growth rate in the two-stage setup is lower than when guessing Xn directly. We further propose a simple Huffman code-based construction of a function f that is a viable candidate for the minimization of the least exponential growth rate in the two-stage guessing setup. (2) We characterize the least achievable exponential growth rate of the ρ-th moment of the total number of guesses required to recover Xn when Stage 1 need not end with a correct guess of Yn and without assumptions on the stationary memoryless sources producing Xn and Yn. Full article
(This article belongs to the Special Issue Statistical Communication and Information Theory)
Show Figures

Figure 1

Article
Colvis—A Structured Annotation Acquisition System for Data Visualization
Information 2021, 12(4), 158; https://doi.org/10.3390/info12040158 - 09 Apr 2021
Cited by 1 | Viewed by 1371
Abstract
Annotations produced by analysts during the exploration of a data visualization are a precious source of knowledge. Harnessing this knowledge requires a thorough structure of annotations, but also a means to acquire them without harming user engagement. The main contribution of this article [...] Read more.
Annotations produced by analysts during the exploration of a data visualization are a precious source of knowledge. Harnessing this knowledge requires a thorough structure of annotations, but also a means to acquire them without harming user engagement. The main contribution of this article is a method, taking the form of an interface, that offers a comprehensive “subject-verb-complement” set of steps for analysts to take annotations, and seamlessly translate these annotations within a prior classification framework. Technical considerations are also an integral part of this study: through a concrete web implementation, we prove the feasibility of our method, but also highlight some of the unresolved challenges that remain to be addressed. After explaining all concepts related to our work, from a literature review to JSON Specifications, we follow by showing two use cases that illustrate how the interface can work in concrete situations. We conclude with a substantial discussion of the limitations, the current state of the method and the upcoming steps for this annotation interface. Full article
(This article belongs to the Special Issue Trends and Opportunities in Visualization and Visual Analytics)
Show Figures

Figure 1

Article
A Distributed Approach to Speaker Count Problem in an Open-Set Scenario by Clustering Pitch Features
Information 2021, 12(4), 157; https://doi.org/10.3390/info12040157 - 09 Apr 2021
Viewed by 1132
Abstract
Counting the number of speakers in an audio sample can lead to innovative applications, such as a real-time ranking system. Researchers have studied advanced machine learning approaches for solving the speaker count problem. However, these solutions are not efficient in real-time environments, as [...] Read more.
Counting the number of speakers in an audio sample can lead to innovative applications, such as a real-time ranking system. Researchers have studied advanced machine learning approaches for solving the speaker count problem. However, these solutions are not efficient in real-time environments, as it requires pre-processing of a finite set of data samples. Another approach for solving the problem is via unsupervised learning or by using audio processing techniques. The research in this category is limited and does not consider the large-scale open set environment. In this paper, we propose a distributed clustering approach to address the speaker count problem. The separability of the speaker is computed using statistical pitch parameters. The proposed solution uses multiple microphones available in smartphones in a large geographical area to capture and extract statistical pitch features from the audio samples. These features are shared between the nodes to estimate the number of speakers in the neighborhood. One of the major challenges is to reduce the error count that arises due to the proximity of the users and multiple microphones. We evaluate the algorithm’s performance using real smartphones in a multi-group arrangement by capturing parallel conversations between the users in both indoor and outdoor scenarios. The average error count distance is 1.667 in a multi-group scenario. The average error count distances in indoor environments are 16% which is better than in the outdoor environment. Full article
Show Figures

Figure 1

Article
The Image Classification Method with CNN-XGBoost Model Based on Adaptive Particle Swarm Optimization
Information 2021, 12(4), 156; https://doi.org/10.3390/info12040156 - 09 Apr 2021
Cited by 6 | Viewed by 2475
Abstract
CNN is particularly effective in extracting spatial features. However, the single-layer classifier constructed by activation function in CNN is easily interfered by image noise, resulting in reduced classification accuracy. To solve the problem, the advanced ensemble model XGBoost is used to overcome the [...] Read more.
CNN is particularly effective in extracting spatial features. However, the single-layer classifier constructed by activation function in CNN is easily interfered by image noise, resulting in reduced classification accuracy. To solve the problem, the advanced ensemble model XGBoost is used to overcome the deficiency of a single classifier to classify image features. To further distinguish the extracted image features, a CNN-XGBoost image classification model optimized by APSO is proposed, where APSO optimizes the hyper-parameters on the overall architecture to promote the fusion of the two-stage model. The model is mainly composed of two parts: feature extractor CNN, which is used to automatically extract spatial features from images; feature classifier XGBoost is applied to classify features extracted after convolution. In the process of parameter optimization, to overcome the shortcoming that traditional PSO algorithm easily falls into a local optimal, the improved APSO guide the particles to search for optimization in space by two different strategies, which improves the diversity of particle population and prevents the algorithm from becoming trapped in local optima. The results on the image set show that the proposed model gets better results in image classification. Moreover, the APSO-XGBoost model performs well on the credit data, which indicates that the model has a good ability of credit scoring. Full article
(This article belongs to the Special Issue Data Modeling and Predictive Analytics)
Show Figures

Figure 1

Article
Information Bottleneck for a Rayleigh Fading MIMO Channel with an Oblivious Relay
Information 2021, 12(4), 155; https://doi.org/10.3390/info12040155 - 08 Apr 2021
Cited by 1 | Viewed by 1408
Abstract
This paper considers the information bottleneck (IB) problem of a Rayleigh fading multiple-input multiple-out (MIMO) channel with an oblivious relay. The relay is constrained to operating without knowledge of the codebooks, i.e., it performs oblivious processing. Moreover, due to the bottleneck constraint, it [...] Read more.
This paper considers the information bottleneck (IB) problem of a Rayleigh fading multiple-input multiple-out (MIMO) channel with an oblivious relay. The relay is constrained to operating without knowledge of the codebooks, i.e., it performs oblivious processing. Moreover, due to the bottleneck constraint, it is impossible for the relay to inform the destination node of the perfect channel state information (CSI) in each channel realization. To evaluate the bottleneck rate, we first provide an upper bound by assuming that the destination node can obtain a perfect CSI at no cost. Then, we provide four achievable schemes, where each scheme satisfies the bottleneck constraint and gives a lower bound to the bottleneck rate. In the first and second schemes, the relay splits the capacity of the relay–destination link into two parts and conveys both the CSI and its observation to the destination node. Due to CSI transmission, the performance of these two schemes is sensitive to the MIMO channel dimension, especially the channel input dimension. To ensure that it still performs well when the channel dimension grows large, in the third and fourth achievable schemes, the relay only transmits compressed observations to the destination node. Numerical results show that, with simple symbol-by-symbol oblivious relay processing and compression, the proposed achievable schemes work well and can demonstrate lower bounds that come quite close to the upper bound on a wide range of relevant system parameters. Full article
(This article belongs to the Special Issue Statistical Communication and Information Theory)
Show Figures

Figure 1

Review
Monitoring Real Time Security Attacks for IoT Systems Using DevSecOps: A Systematic Literature Review
Information 2021, 12(4), 154; https://doi.org/10.3390/info12040154 - 07 Apr 2021
Cited by 10 | Viewed by 3600
Abstract
In many enterprises and the private sector, the Internet of Things (IoT) has spread globally. The growing number of different devices connected to the IoT and their various protocols have contributed to the increasing number of attacks, such as denial-of-service (DoS) and remote-to-local [...] Read more.
In many enterprises and the private sector, the Internet of Things (IoT) has spread globally. The growing number of different devices connected to the IoT and their various protocols have contributed to the increasing number of attacks, such as denial-of-service (DoS) and remote-to-local (R2L) ones. There are several approaches and techniques that can be used to construct attack detection models, such as machine learning, data mining, and statistical analysis. Nowadays, this technique is commonly used because it can provide precise analysis and results. Therefore, we decided to study the previous literature on the detection of IoT attacks and machine learning in order to understand the process of creating detection models. We also evaluated various datasets used for the models, IoT attack types, independent variables used for the models, evaluation metrics for assessment of models, and monitoring infrastructure using DevSecOps pipelines. We found 49 primary studies, and the detection models were developed using seven different types of machine learning techniques. Most primary studies used IoT device testbed datasets, and others used public datasets such as NSL-KDD and UNSW-NB15. When it comes to measuring the efficiency of models, both numerical and graphical measures are commonly used. Most IoT attacks occur at the network layer according to the literature. If the detection models applied DevSecOps pipelines in development processes for IoT devices, they were more secure. From the results of this paper, we found that machine learning techniques can detect IoT attacks, but there are a few issues in the design of detection models. We also recommend the continued use of hybrid frameworks for the improved detection of IoT attacks, advanced monitoring infrastructure configurations using methods based on software pipelines, and the use of machine learning techniques for advanced supervision and monitoring. Full article
(This article belongs to the Special Issue Smart IoT Systems)
Show Figures

Figure 1

Article
Analysis of the Temporal and Spatial Characteristics of Material Cultural Heritage Driven by Big Data—Take Museum Relics as an Example
Information 2021, 12(4), 153; https://doi.org/10.3390/info12040153 - 06 Apr 2021
Cited by 3 | Viewed by 1404
Abstract
Museum cultural relics represent a special material cultural heritage, and modern interpretations of them are needed in current society. Based on the catalogue data of cultural relics published by the State Administration of Cultural Heritage, this paper analyzes the continuity and intermittentness of [...] Read more.
Museum cultural relics represent a special material cultural heritage, and modern interpretations of them are needed in current society. Based on the catalogue data of cultural relics published by the State Administration of Cultural Heritage, this paper analyzes the continuity and intermittentness of cultural relics in time series by using the method of continuity judgment of cultural relics, analyzes the aggregation and migration of cultural relics in space by using the method of spatial analysis, and then uses cosine similarity to explain the similarity distribution in space and time. The results show that the overall distribution of cultural relics exhibits the characteristics of class aggregation, dynasty aggregation and regional aggregation. From the perspective of a time scale, cultural relics have different “life cycles”, displaying continuity, intermittentness, and similarity. From the perspective of a spatial scale, the cultural relic distribution forms a small “cultural communication circle”, showing aggregation, migration, and similarity. The temporal and spatial distribution of cultural relics exhibited more similar characteristics among dynasties that were closer together than those that were far away. Full article
(This article belongs to the Section Information Processes)
Show Figures

Figure 1

Editorial
Editorial for Special Issue Indoor Navigation in Smart Cities
Information 2021, 12(4), 152; https://doi.org/10.3390/info12040152 - 03 Apr 2021
Viewed by 1027
Abstract
The lifecycle of indoor navigation includes various phases [...] Full article
(This article belongs to the Special Issue Indoor Navigation in Smart Cities)
Article
Student Thoughts on Virtual Reality in Higher Education—A Survey Questionnaire
Information 2021, 12(4), 151; https://doi.org/10.3390/info12040151 - 02 Apr 2021
Cited by 7 | Viewed by 5111
Abstract
This paper explores the benefits of using Virtual Reality (VR) technologies in higher education. The theoretical part investigates the classical education system and its features in order to compare advantages of using VR systems in education. VR technologies and its current state in [...] Read more.
This paper explores the benefits of using Virtual Reality (VR) technologies in higher education. The theoretical part investigates the classical education system and its features in order to compare advantages of using VR systems in education. VR technologies and its current state in industry and in education were explored in addition to which branches of higher education use these systems. A survey was conducted through an online questionnaire where respondents (N = 55) gave their opinion on VR and the implementation of VR technologies in education. Three hypotheses related to the use of VR technology, student interest, and learning outcomes as well as the effectiveness, immersiveness and the effect of VR systems on the users were tested through 27 questions. Full article
(This article belongs to the Collection Augmented Reality Technologies, Systems and Applications)
Article
Multi-Task Learning-Based Task Scheduling Switcher for a Resource-Constrained IoT System
Information 2021, 12(4), 150; https://doi.org/10.3390/info12040150 - 01 Apr 2021
Cited by 2 | Viewed by 1747
Abstract
In this journal, we proposed a novel method of using multi-task learning to switch the scheduling algorithm. With multi-task learning to change the scheduling algorithm inside the scheduling framework, the scheduling framework can create a scheduler with the best task execution optimization under [...] Read more.
In this journal, we proposed a novel method of using multi-task learning to switch the scheduling algorithm. With multi-task learning to change the scheduling algorithm inside the scheduling framework, the scheduling framework can create a scheduler with the best task execution optimization under the computation deadline. With the changing number of tasks, the number of types of resources taken, and computation deadline, it is hard for a single scheduling algorithm to achieve the best scheduler optimization while avoiding the worst-case time complexity in a resource-constrained Internet of Things (IoT) system due to the trade-off in computation time and optimization in each scheduling algorithm. Furthermore, different hardware specifications affect the scheduler computation time differently, making it hard to rely on Big-O complexity as a reference. With multi-task learning to profile the scheduling algorithm behavior on the hardware used to compute the scheduler, we can identify the best scheduling algorithm. Our benchmark result shows that it can achieve an average of 93.68% of accuracy in meeting the computation deadline, along with 23.41% of average optimization. Based on the results, our method can improve the scheduling of the resource-constrained IoT system. Full article
(This article belongs to the Special Issue Recent Advances in IoT and Cyber/Physical Security)
Show Figures

Figure 1

Article
A Social Media Mining and Ensemble Learning Model: Application to Luxury and Fast Fashion Brands
Information 2021, 12(4), 149; https://doi.org/10.3390/info12040149 - 31 Mar 2021
Viewed by 2825
Abstract
This research proposes a framework for the fashion brand community to explore public participation behaviors triggered by brand information and to understand the importance of key image cues and brand positioning. In addition, it reviews different participation responses (likes, comments, and shares) to [...] Read more.
This research proposes a framework for the fashion brand community to explore public participation behaviors triggered by brand information and to understand the importance of key image cues and brand positioning. In addition, it reviews different participation responses (likes, comments, and shares) to build systematic image and theme modules that detail planning requirements for community information. The sample includes luxury fashion brands (Chanel, Hermès, and Louis Vuitton) and fast fashion brands (Adidas, Nike, and Zara). Using a web crawler, a total of 21,670 posts made from 2011 to 2019 are obtained. A fashion brand image model is constructed to determine key image cues in posts by each brand. Drawing on the findings of the ensemble analysis, this research divides cues used by the six major fashion brands into two modules, image cue module and image and theme cue module, to understand participation responses in the form of likes, comments, and shares. The results of the systematic image and theme module serve as a critical reference for admins exploring the characteristics of public participation for each brand and the main factors motivating public participation. Full article
(This article belongs to the Section Information Applications)
Show Figures

Figure 1

Article
A Data-Driven Framework for Coding the Intent and Extent of Political Tweeting, Disinformation, and Extremism
Information 2021, 12(4), 148; https://doi.org/10.3390/info12040148 - 31 Mar 2021
Cited by 3 | Viewed by 1873
Abstract
Disinformation campaigns on online social networks (OSNs) in recent years have underscored democracy’s vulnerability to such operations and the importance of identifying such operations and dissecting their methods, intents, and source. This paper is another milestone in a line of research on political [...] Read more.
Disinformation campaigns on online social networks (OSNs) in recent years have underscored democracy’s vulnerability to such operations and the importance of identifying such operations and dissecting their methods, intents, and source. This paper is another milestone in a line of research on political disinformation, propaganda, and extremism on OSNs. A total of 40,000 original Tweets (not re-Tweets or Replies) related to the U.S. 2020 presidential election are collected. The intent, focus, and political affiliation of these political Tweets are determined through multiple discussions and revisions. There are three political affiliations: rightist, leftist, and neutral. A total of 171 different classes of intent or focus are defined for Tweets. A total of 25% of Tweets were left out while defining these classes of intent. The purpose is to assure that the defined classes would be able to cover the intent and focus of unseen Tweets (Tweets that were not used to determine and define these classes) and no new classes would be required. This paper provides these classes, their definition and size, and example Tweets from them. If any information is included in a Tweet, its factuality is verified through valid news sources and articles. If any opinion is included in a Tweet, it is determined that whether or not it is extreme, through multiple discussions and revisions. This paper provides analytics with regard to the political affiliation and intent of Tweets. The results show that disinformation and extreme opinions are more common among rightists Tweets than leftist Tweets. Additionally, Coronavirus pandemic is the topic of almost half of the Tweets, where 25.43% of Tweets express their unhappiness with how Republicans have handled this pandemic. Full article
(This article belongs to the Special Issue Decentralization and New Technologies for Social Media)
Article
On Training Knowledge Graph Embedding Models
Information 2021, 12(4), 147; https://doi.org/10.3390/info12040147 - 31 Mar 2021
Cited by 1 | Viewed by 2088
Abstract
Knowledge graph embedding (KGE) models have become popular means for making discoveries in knowledge graphs (e.g., RDF graphs) in an efficient and scalable manner. The key to success of these models is their ability to learn low-rank vector representations for knowledge graph entities [...] Read more.
Knowledge graph embedding (KGE) models have become popular means for making discoveries in knowledge graphs (e.g., RDF graphs) in an efficient and scalable manner. The key to success of these models is their ability to learn low-rank vector representations for knowledge graph entities and relations. Despite the rapid development of KGE models, state-of-the-art approaches have mostly focused on new ways to represent embeddings interaction functions (i.e., scoring functions). In this paper, we argue that the choice of other training components such as the loss function, hyperparameters and negative sampling strategies can also have substantial impact on the model efficiency. This area has been rather neglected by previous works so far and our contribution is towards closing this gap by a thorough analysis of possible choices of training loss functions, hyperparameters and negative sampling techniques. We finally investigate the effects of specific choices on the scalability and accuracy of knowledge graph embedding models. Full article
(This article belongs to the Collection Knowledge Graphs for Search and Recommendation)
Show Figures

Figure 1

Previous Issue
Next Issue
Back to TopTop