Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
5.3
2.5
Journals
Applied Sciences
Editor’s Choice
applsci-logo

Journal Menu

► Journal Menu

Journal Browser

► Journal Browser
share announcement

Editor’s Choice Articles

Editor’s Choice articles are based on recommendations by the scientific editors of MDPI journals from around the world. Editors select a small number of articles recently published in the journal that they believe will be particularly interesting to readers, or important in the respective research area. The aim is to provide a snapshot of some of the most exciting work published in the various research areas of the journal.

Order results
Result details
Results per page
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
39 pages, 4414 KiB  
Review
Review on Wearable Technology in Sports: Concepts, Challenges and Opportunities
by Ahmet Çağdaş Seçkin, Bahar Ateş and Mine Seçkin
Appl. Sci. 2023, 13(18), 10399; https://doi.org/10.3390/app131810399 - 17 Sep 2023
Cited by 36 | Viewed by 27730
Abstract
Wearable technology is increasingly vital for improving sports performance through real-time data analysis and tracking. Both professional and amateur athletes rely on wearable sensors to enhance training efficiency and competition outcomes. However, further research is needed to fully understand and optimize their potential [...] Read more.
Wearable technology is increasingly vital for improving sports performance through real-time data analysis and tracking. Both professional and amateur athletes rely on wearable sensors to enhance training efficiency and competition outcomes. However, further research is needed to fully understand and optimize their potential in sports. This comprehensive review explores the measurement and monitoring of athletic performance, injury prevention, rehabilitation, and overall performance optimization using body wearable sensors. By analyzing wearables’ structure, research articles across various sports, and commercial sensors, the review provides a thorough analysis of wearable sensors in sports. Its findings benefit athletes, coaches, healthcare professionals, conditioners, managers, and researchers, offering a detailed summary of wearable technology in sports. The review is expected to contribute to future advancements in wearable sensors and biometric data analysis, ultimately improving sports performance. Limitations such as privacy concerns, accuracy issues, and costs are acknowledged, stressing the need for legal regulations, ethical principles, and technical measures for safe and fair use. The importance of personalized devices and further research on athlete comfort and performance impact is emphasized. The emergence of wearable imaging devices holds promise for sports rehabilitation and performance monitoring, enabling enhanced athlete health, recovery, and performance in the sports industry. Full article
(This article belongs to the Special Issue Advances in Wearable Devices for Sports)
►▼ Show Figures

Figure 1

20 pages, 1441 KiB  
Article
Crop Prediction Model Using Machine Learning Algorithms
by Ersin Elbasi, Chamseddine Zaki, Ahmet E. Topcu, Wiem Abdelbaki, Aymen I. Zreikat, Elda Cina, Ahmed Shdefat and Louai Saker
Appl. Sci. 2023, 13(16), 9288; https://doi.org/10.3390/app13169288 - 16 Aug 2023
Cited by 43 | Viewed by 28703
Abstract
Machine learning applications are having a great impact on the global economy by transforming the data processing method and decision making. Agriculture is one of the fields where the impact is significant, considering the global crisis for food supply. This research investigates the [...] Read more.
Machine learning applications are having a great impact on the global economy by transforming the data processing method and decision making. Agriculture is one of the fields where the impact is significant, considering the global crisis for food supply. This research investigates the potential benefits of integrating machine learning algorithms in modern agriculture. The main focus of these algorithms is to help optimize crop production and reduce waste through informed decisions regarding planting, watering, and harvesting crops. This paper includes a discussion on the current state of machine learning in agriculture, highlighting key challenges and opportunities, and presents experimental results that demonstrate the impact of changing labels on the accuracy of data analysis algorithms. The findings recommend that by analyzing wide-ranging data collected from farms, incorporating online IoT sensor data that were obtained in a real-time manner, farmers can make more informed verdicts about factors that affect crop growth. Eventually, integrating these technologies can transform modern agriculture by increasing crop yields while minimizing waste. Fifteen different algorithms have been considered to evaluate the most appropriate algorithms to use in agriculture, and a new feature combination scheme-enhanced algorithm is presented. The results show that we can achieve a classification accuracy of 99.59% using the Bayes Net algorithm and 99.46% using Naïve Bayes Classifier and Hoeffding Tree algorithms. These results will indicate an increase in production rates and reduce the effective cost for the farms, leading to more resilient infrastructure and sustainable environments. Moreover, the findings we obtained in this study can also help future farmers detect diseases early, increase crop production efficiency, and reduce prices when the world is experiencing food shortages. Full article
(This article belongs to the Special Issue Advances in Technology Applied in Agricultural Engineering)
►▼ Show Figures

Figure 1

20 pages, 3740 KiB  
Article
Variational Autoencoders for Data Augmentation in Clinical Studies
by Dimitris Papadopoulos and Vangelis D. Karalis
Appl. Sci. 2023, 13(15), 8793; https://doi.org/10.3390/app13158793 - 30 Jul 2023
Cited by 12 | Viewed by 3056
Abstract
Sample size estimation is critical in clinical trials. A sample of adequate size can provide insights into a given population, but the collection of substantial amounts of data is costly and time-intensive. The aim of this study was to introduce a novel data [...] Read more.
Sample size estimation is critical in clinical trials. A sample of adequate size can provide insights into a given population, but the collection of substantial amounts of data is costly and time-intensive. The aim of this study was to introduce a novel data augmentation approach in the field of clinical trials by employing variational autoencoders (VAEs). Several forms of VAEs were developed and used for the generation of virtual subjects. Various types of VAEs were explored and employed in the production of virtual individuals, and several different scenarios were investigated. The VAE-generated data exhibited similar performance to the original data, even in cases where a small proportion of them (e.g., 30–40%) was used for the reconstruction of the generated data. Additionally, the generated data showed even higher statistical power than the original data in cases of high variability. This represents an additional advantage for the use of VAEs in situations of high variability, as they can act as noise reduction. The application of VAEs in clinical trials can be a useful tool for decreasing the required sample size and, consequently, reducing the costs and time involved. Furthermore, it aligns with ethical concerns surrounding human participation in trials. Full article
(This article belongs to the Special Issue Advanced Artificial Intelligence in Medicine and Bioinformatics)
►▼ Show Figures

Figure 1

23 pages, 1433 KiB  
Review
Edible Packaging: A Technological Update for the Sustainable Future of the Food Industry
by Surya Sasikumar Nair, Joanna Trafiałek and Wojciech Kolanowski
Appl. Sci. 2023, 13(14), 8234; https://doi.org/10.3390/app13148234 - 15 Jul 2023
Cited by 14 | Viewed by 10276
Abstract
This review aims to address the current data on edible packaging systems used in food production. The growing global population, changes in the climate and dietary patterns, and the increasing need for environmental protection, have created an increasing demand for waste-free food production. [...] Read more.
This review aims to address the current data on edible packaging systems used in food production. The growing global population, changes in the climate and dietary patterns, and the increasing need for environmental protection, have created an increasing demand for waste-free food production. The need for durable and sustainable packaging materials has become significant in order to avoid food waste and environmental pollution. Edible packaging has emerged as a promising solution to extend the shelf life of food products and reduce dependence on petroleum-based resources. This review analyzes the history, production methods, barrier properties, types, and additives of edible packaging systems. The review highlights the advantages and importance of edible packaging materials and describes how they can improve sustainability measures. The market value of edible packaging materials is expanding. Further research on and developments in edible food packaging materials are needed to increase sustainable, eco-friendly packaging practices that are significant for environmental protection and food safety. Full article
(This article belongs to the Special Issue Feature Review Papers in ‘Food Science and Technology’ Section)
►▼ Show Figures

Figure 1

16 pages, 1713 KiB  
Review
Digital Twins: The New Frontier for Personalized Medicine?
by Michaela Cellina, Maurizio Cè, Marco Alì, Giovanni Irmici, Simona Ibba, Elena Caloro, Deborah Fazzini, Giancarlo Oliva and Sergio Papa
Appl. Sci. 2023, 13(13), 7940; https://doi.org/10.3390/app13137940 - 6 Jul 2023
Cited by 25 | Viewed by 7355
Abstract
Digital twins are virtual replicas of physical objects or systems. This new technology is increasingly being adopted in industry to improve the monitoring and efficiency of products and organizations. In healthcare, digital human twins (DHTs) represent virtual copies of patients, including tissues, organs, [...] Read more.
Digital twins are virtual replicas of physical objects or systems. This new technology is increasingly being adopted in industry to improve the monitoring and efficiency of products and organizations. In healthcare, digital human twins (DHTs) represent virtual copies of patients, including tissues, organs, and physiological processes. Their application has the potential to transform patient care in the direction of increasingly personalized data-driven medicine. The use of DHTs can be integrated with digital twins of healthcare institutions to improve organizational management processes and resource allocation. By modeling the complex multi-omics interactions between genetic and environmental factors, DHTs help monitor disease progression and optimize treatment plans. Through digital simulation, DHT models enable the selection of the most appropriate molecular therapy and accurate 3D representation for precision surgical planning, together with augmented reality tools. Furthermore, they allow for the development of tailored early diagnosis protocols and new targeted drugs. Furthermore, digital twins can facilitate medical training and education. By creating virtual anatomy and physiology models, medical students can practice procedures, enhance their skills, and improve their understanding of the human body. Overall, digital twins have immense potential to revolutionize healthcare, improving patient care and outcomes, reducing costs, and enhancing medical research and education. However, challenges such as data security, data quality, and data interoperability must be addressed before the widespread adoption of digital twins in healthcare. We aim to propose a narrative review on this hot topic to provide an overview of the potential applications of digital twins to improve treatment and diagnostics, but also of the challenges related to their development and widespread diffusion. Full article
(This article belongs to the Special Issue Methods, Applications and Developments in Biomedical Informatics)
►▼ Show Figures

Figure 1

14 pages, 2847 KiB  
Article
Teeth Segmentation in Panoramic Dental X-ray Using Mask Regional Convolutional Neural Network
by Giulia Rubiu, Marco Bologna, Michaela Cellina, Maurizio Cè, Davide Sala, Roberto Pagani, Elisa Mattavelli, Deborah Fazzini, Simona Ibba, Sergio Papa and Marco Alì
Appl. Sci. 2023, 13(13), 7947; https://doi.org/10.3390/app13137947 - 6 Jul 2023
Cited by 8 | Viewed by 4653
Abstract
Background and purpose: Accurate instance segmentation of teeth in panoramic dental X-rays is a challenging task due to variations in tooth morphology and overlapping regions. In this study, we propose a new algorithm, for instance, segmentation of the different teeth in panoramic dental [...] Read more.
Background and purpose: Accurate instance segmentation of teeth in panoramic dental X-rays is a challenging task due to variations in tooth morphology and overlapping regions. In this study, we propose a new algorithm, for instance, segmentation of the different teeth in panoramic dental X-rays. Methods: An instance segmentation model was trained using the architecture of a Mask Region-based Convolutional Neural Network (Mask-RCNN). The data for the training, validation, and testing were taken from the Tuft dental database (1000 panoramic dental radiographs). The number of the predicted label was 52 (20 deciduous and 32 permanent). The size of the training, validation, and test sets were 760, 190, and 70 images, respectively, and the split was performed randomly. The model was trained for 300 epochs, using a batch size of 10, a base learning rate of 0.001, and a warm-up multistep learning rate scheduler (gamma = 0.1). Data augmentation was performed by changing the brightness, contrast, crop, and image size. The percentage of correctly detected teeth and Dice in the test set were used as the quality metrics for the model. Results: In the test set, the percentage of correctly classified teeth was 98.4%, while the Dice score was 0.87. For both the left mandibular central and lateral incisor permanent teeth, the Dice index result was 0.91 and the accuracy was 100%. For the permanent teeth right mandibular first molar, mandibular second molar, and third molar, the Dice indexes were 0.92, 0.93, and 0.78, respectively, with an accuracy of 100% for all three different teeth. For deciduous teeth, the Dice indexes for the right mandibular lateral incisor, right mandibular canine, and right mandibular first molar were 0.89, 0.91, and 0.85, respectively, with an accuracy of 100%. Conclusions: A successful instance segmentation model for teeth identification in panoramic dental X-ray was developed and validated. This model may help speed up and automate tasks like teeth counting and identifying specific missing teeth, improving the current clinical practice. Full article
(This article belongs to the Special Issue Advanced Materials and Technology in Dental, Oral and Maxillofacial Health)
►▼ Show Figures

Figure 1

37 pages, 11612 KiB  
Review
New Trends in 4D Printing: A Critical Review
by Somayeh Vatanparast, Alberto Boschetto, Luana Bottini and Paolo Gaudenzi
Appl. Sci. 2023, 13(13), 7744; https://doi.org/10.3390/app13137744 - 30 Jun 2023
Cited by 26 | Viewed by 5035
Abstract
In a variety of industries, Additive Manufacturing has revolutionized the whole design–fabrication cycle. Traditional 3D printing is typically employed to produce static components, which are not able to fulfill dynamic structural requirements and are inappropriate for applications such as soft grippers, self-assembly systems, [...] Read more.
In a variety of industries, Additive Manufacturing has revolutionized the whole design–fabrication cycle. Traditional 3D printing is typically employed to produce static components, which are not able to fulfill dynamic structural requirements and are inappropriate for applications such as soft grippers, self-assembly systems, and smart actuators. To address this limitation, an innovative technology has emerged, known as “4D printing”. It processes smart materials by using 3D printing for fabricating smart structures that can be reconfigured by applying different inputs, such as heat, humidity, magnetism, electricity, light, etc. At present, 4D printing is still a growing technology, and it presents numerous challenges regarding materials, design, simulation, fabrication processes, applied strategies, and reversibility. In this work a critical review of 4D printing technologies, materials, and applications is provided. Full article
(This article belongs to the Special Issue Recent Advances in Design of Additive Manufacturing: Materials and Production Processes)
►▼ Show Figures

Figure 1

22 pages, 4939 KiB  
Review
Modelling and Control Methods in Path Tracking Control for Autonomous Agricultural Vehicles: A Review of State of the Art and Challenges
by Quanyu Wang, Jin He, Caiyun Lu, Chao Wang, Han Lin, Hanyu Yang, Hang Li and Zhengyang Wu
Appl. Sci. 2023, 13(12), 7155; https://doi.org/10.3390/app13127155 - 15 Jun 2023
Cited by 9 | Viewed by 3269
Abstract
This paper provides a review of path-tracking strategies used in autonomous agricultural vehicles, mainly from two aspects: vehicle model construction and the development and improvement of path-tracking algorithms. Vehicle models are grouped into numerous types based on the structural characteristics and working conditions, [...] Read more.
This paper provides a review of path-tracking strategies used in autonomous agricultural vehicles, mainly from two aspects: vehicle model construction and the development and improvement of path-tracking algorithms. Vehicle models are grouped into numerous types based on the structural characteristics and working conditions, including wheeled tractors, tracked tractors, rice transplanters, high clearance sprays, agricultural robots, agricultural tractor–trailers, etc. The application and improvement of path-tracking control methods are summarized based on the different working scenes and types of agricultural machinery. This study explores each of these methods in terms of accuracy, stability, robustness, and disadvantages/advantages. The main challenges in the field of agricultural vehicle path tracking control are defined, and future research directions are offered based on critical reviews. This review aims to provide a reference for determining which controllers to use in path-tracking control development for an autonomous agricultural vehicle. Full article
(This article belongs to the Special Issue Feature Review Papers in Agricultural Science and Technology)
►▼ Show Figures

Figure 1

18 pages, 9277 KiB  
Article
Solar Sail Orbit Raising with Electro-Optically Controlled Diffractive Film
by Alessandro A. Quarta and Giovanni Mengali
Appl. Sci. 2023, 13(12), 7078; https://doi.org/10.3390/app13127078 - 13 Jun 2023
Cited by 12 | Viewed by 1659
Abstract
The aim of this paper is to analyze the transfer performance of a spacecraft whose primary propulsion system is a diffractive solar sail with active, switchable panels. The spacecraft uses a propellantless thruster that converts the solar radiation pressure into propulsive acceleration by [...] Read more.
The aim of this paper is to analyze the transfer performance of a spacecraft whose primary propulsion system is a diffractive solar sail with active, switchable panels. The spacecraft uses a propellantless thruster that converts the solar radiation pressure into propulsive acceleration by taking advantage of the diffractive property of an electro-optically controlled (binary) metamaterial. The proposed analysis considers a heliocentric mission scenario where the spacecraft is required to perform a two-dimensional transfer between two concentric and coplanar circular orbits. The sail attitude is assumed to be Sun-facing, that is, with its sail nominal plane perpendicular to the incoming sunlight. This is possible since, unlike a more conventional solar sail concept that uses metalized highly reflective thin films to reflect the photons, a diffractive sail is theoretically able to generate a component of the thrust vector along the sail nominal plane also in a Sun-facing configuration. The electro-optically controlled sail film is used to change the in-plane component of the thrust vector to accomplish the transfer by minimizing the total flight time without changing the sail attitude with respect to an orbital reference frame. This work extends the mathematical model recently proposed by the authors by including the potential offered by an active control of the diffractive sail film. The paper also thoroughly analyzes the diffractive sail-based spacecraft performance in a set of classical circle-to-circle heliocentric trajectories that model transfers from Earth to Mars, Venus and Jupiter. Full article
(This article belongs to the Special Issue Recent Advances in Space Propulsion Technology)
►▼ Show Figures

Figure 1

16 pages, 9511 KiB  
Article
Materials and Technique: The First Look at Saturnino Gatti
by Letizia Bonizzoni, Simone Caglio, Anna Galli, Luca Lanteri and Claudia Pelosi
Appl. Sci. 2023, 13(11), 6842; https://doi.org/10.3390/app13116842 - 5 Jun 2023
Cited by 11 | Viewed by 2162
Abstract
As part of the study project of the pictorial cycle, attributed to Saturnino Gatti, in the church of San Panfilo at Villagrande di Tornimparte (AQ), image analyses were performed in order to document the general conservation conditions of the surfaces, and to map [...] Read more.
As part of the study project of the pictorial cycle, attributed to Saturnino Gatti, in the church of San Panfilo at Villagrande di Tornimparte (AQ), image analyses were performed in order to document the general conservation conditions of the surfaces, and to map the different painting materials to be subsequently examined using spectroscopic techniques. To acquire the images, radiation sources, ranging from ultraviolet to near infrared, were used; analyses of ultraviolet fluorescence (UVF), infrared reflectography (IRR), infrared false colors (IRFC), and optical microscopy in visible light (OM) were carried out on all the panels of the mural painting of the apsidal conch. The Hypercolorimetric Multispectral Imaging (HMI) technique was also applied in selected areas of two panels. Due to the accurate calibration system, this technique is able to obtain high-precision colorimetric and reflectance measurements, which can be repeated for proper surface monitoring. The integrated analysis of the different wavelengths’ images—in particular, the ones processed in false colors—made it possible to distinguish the portions affected by retouching or repainting and to recover the legibility of some figures that showed chromatic alterations of the original pictorial layers. The IR reflectography, in addition to highlighting the portions that lost materials and were subject to non-original interventions, emphasized the presence of the underdrawing, which was detected using the spolvero technique. UVF photography led to a preliminary mapping of the organic and inorganic materials that exhibited characteristic induced fluorescence, such as a binder in correspondence with the original azurite painting or the wide use of white zinc in the retouched areas. The collected data made it possible to form a better iconographic interpretation. Moreover, it also enabled us to accurately select the areas to be investigated using spectroscopic analyses, both in situ and on micro-samples, in order to deepen our knowledge of the techniques used by the artist to create the original painting, and to detect subsequent interventions. Full article
(This article belongs to the Special Issue Results of the II National Research Project of AIAr: Archaeometric Study of the Frescoes by Saturnino Gatti and Workshop at the Church of San Panfilo in Tornimparte (AQ, Italy))
►▼ Show Figures

Figure 1

37 pages, 6325 KiB  
Review
Structural Health Monitoring and Management of Cultural Heritage Structures: A State-of-the-Art Review
by Michela Rossi and Dionysios Bournas
Appl. Sci. 2023, 13(11), 6450; https://doi.org/10.3390/app13116450 - 25 May 2023
Cited by 19 | Viewed by 3713
Abstract
In recent decades, the urgency to protect and upgrade cultural heritage structures (CHS) has become of primary importance due to their unique value and potential areas of impact (economic, social, cultural, and environmental). Structural health monitoring (SHM) and the management of CHS are [...] Read more.
In recent decades, the urgency to protect and upgrade cultural heritage structures (CHS) has become of primary importance due to their unique value and potential areas of impact (economic, social, cultural, and environmental). Structural health monitoring (SHM) and the management of CHS are emerging as decisive safeguard measures aimed at assessing the actual state of the conservation and integrity of the structure. Moreover, the data collected from SHM are essential to plan cost-effective and sustainable maintenance solutions, in compliance with the basic preservation principles for historic buildings, such as minimum intervention. It is evident that, compared to new buildings, the application of SHM to CHS is even more challenging because of the uniqueness of each monitored structure and the need to respect its architectural and historical value. This paper aims to present a state-of-the-art evaluation of the current traditional and innovative SHM techniques adopted for CHS and to identify future research trends. First, a general introduction regarding the use of monitoring strategies and technologies for CHS is presented. Next, various traditional SHM techniques currently used in CHS are described. Then, attention is focused on the most recent technologies, such as fibre optic sensors and smart-sensing materials. Finally, an overview of innovative methods and tools for managing and analysing SHM data, including IoT-SHM systems and the integration of BIM in heritage structures, is provided. Full article
(This article belongs to the Collection Nondestructive Testing (NDT))
►▼ Show Figures

Figure 1

17 pages, 1171 KiB  
Review
Comparing Vision Transformers and Convolutional Neural Networks for Image Classification: A Literature Review
by José Maurício, Inês Domingues and Jorge Bernardino
Appl. Sci. 2023, 13(9), 5521; https://doi.org/10.3390/app13095521 - 28 Apr 2023
Cited by 79 | Viewed by 21033
Abstract
Transformers are models that implement a mechanism of self-attention, individually weighting the importance of each part of the input data. Their use in image classification tasks is still somewhat limited since researchers have so far chosen Convolutional Neural Networks for image classification and [...] Read more.
Transformers are models that implement a mechanism of self-attention, individually weighting the importance of each part of the input data. Their use in image classification tasks is still somewhat limited since researchers have so far chosen Convolutional Neural Networks for image classification and transformers were more targeted to Natural Language Processing (NLP) tasks. Therefore, this paper presents a literature review that shows the differences between Vision Transformers (ViT) and Convolutional Neural Networks. The state of the art that used the two architectures for image classification was reviewed and an attempt was made to understand what factors may influence the performance of the two deep learning architectures based on the datasets used, image size, number of target classes (for the classification problems), hardware, and evaluated architectures and top results. The objective of this work is to identify which of the architectures is the best for image classification and under what conditions. This paper also describes the importance of the Multi-Head Attention mechanism for improving the performance of ViT in image classification. Full article
(This article belongs to the Special Issue Artificial Intelligence in Complex Networks)
►▼ Show Figures

Figure 1

16 pages, 928 KiB  
Article
HDLNIDS: Hybrid Deep-Learning-Based Network Intrusion Detection System
by Emad Ul Haq Qazi, Muhammad Hamza Faheem and Tanveer Zia
Appl. Sci. 2023, 13(8), 4921; https://doi.org/10.3390/app13084921 - 14 Apr 2023
Cited by 32 | Viewed by 5506
Abstract
Attacks on networks are currently the most pressing issue confronting modern society. Network risks affect all networks, from small to large. An intrusion detection system must be present for detecting and mitigating hostile attacks inside networks. Machine Learning and Deep Learning are currently [...] Read more.
Attacks on networks are currently the most pressing issue confronting modern society. Network risks affect all networks, from small to large. An intrusion detection system must be present for detecting and mitigating hostile attacks inside networks. Machine Learning and Deep Learning are currently used in several sectors, particularly the security of information, to design efficient intrusion detection systems. These systems can quickly and accurately identify threats. However, because malicious threats emerge and evolve regularly, networks need an advanced security solution. Hence, building an intrusion detection system that is both effective and intelligent is one of the most cognizant research issues. There are several public datasets available for research on intrusion detection. Because of the complexity of attacks and the continually evolving detection of an attack method, publicly available intrusion databases must be updated frequently. A convolutional recurrent neural network is employed in this study to construct a deep-learning-based hybrid intrusion detection system that detects attacks over a network. To boost the efficiency of the intrusion detection system and predictability, the convolutional neural network performs the convolution to collect local features, while a deep-layered recurrent neural network extracts the features in the proposed Hybrid Deep-Learning-Based Network Intrusion Detection System (HDLNIDS). Experiments are conducted using publicly accessible benchmark CICIDS-2018 data, to determine the effectiveness of the proposed system. The findings of the research demonstrate that the proposed HDLNIDS outperforms current intrusion detection approaches with an average accuracy of 98.90% in detecting malicious attacks. Full article
(This article belongs to the Collection Innovation in Information Security)
►▼ Show Figures

Figure 1

21 pages, 5162 KiB  
Article
Quality Assessment of Banana Ripening Stages by Combining Analytical Methods and Image Analysis
by Vassilia J. Sinanoglou, Thalia Tsiaka, Konstantinos Aouant, Elizabeth Mouka, Georgia Ladika, Eftichia Kritsi, Spyros J. Konteles, Alexandros-George Ioannou, Panagiotis Zoumpoulakis, Irini F. Strati and Dionisis Cavouras
Appl. Sci. 2023, 13(6), 3533; https://doi.org/10.3390/app13063533 - 10 Mar 2023
Cited by 14 | Viewed by 7845
Abstract
Currently, the evaluation of fruit ripening progress in relation to physicochemical and texture-quality parameters has become an increasingly important issue, particularly when considering consumer acceptance. Therefore, the purpose of the present study was the application of rapid, nondestructive, and conventional methods to assess [...] Read more.
Currently, the evaluation of fruit ripening progress in relation to physicochemical and texture-quality parameters has become an increasingly important issue, particularly when considering consumer acceptance. Therefore, the purpose of the present study was the application of rapid, nondestructive, and conventional methods to assess the quality of banana peels and flesh in terms of ripening and during storage in controlled temperatures and humidity. For this purpose, we implemented various analytical techniques, such as attenuated total reflection-Fourier transform infrared (ATR-FTIR) spectroscopy for texture, colorimetrics, and physicochemical features, along with image-analysis methods and discriminant as well as statistical analysis. Image-analysis outcomes showed that storage provoked significant degradation of banana peels based on the increased image-texture dissimilarity and the loss of the structural order of the texture. In addition, the computed features were sufficient to discriminate four ripening stages with high accuracy. Moreover, the results revealed that storage led to significant changes in the color parameters and dramatic decreases in the texture attributes of banana flesh. The combination of image and chemical analyses pinpointed that storage caused water migration to the flesh and significant starch decomposition, which was then converted into soluble sugars. The redness and yellowness of the peel; the flesh moisture content; the texture attributes; Brix; and the storage time were all strongly interrelated. The combination of these techniques, coupled with statistical tools, to monitor the physicochemical and organoleptic quality of bananas during storage could be further applied for assessing the quality of other fruits and vegetables under similar conditions. Full article
(This article belongs to the Special Issue Innovative Technologies in Food Detection)
►▼ Show Figures

Figure 1

20 pages, 1537 KiB  
Review
Use of Machine Learning and Remote Sensing Techniques for Shoreline Monitoring: A Review of Recent Literature
by Chrysovalantis-Antonios D. Tsiakos and Christos Chalkias
Appl. Sci. 2023, 13(5), 3268; https://doi.org/10.3390/app13053268 - 3 Mar 2023
Cited by 22 | Viewed by 5279
Abstract
Climate change and its effects (i.e., sea level rise, extreme weather events) as well as anthropogenic activities, determine pressures to the coastal environments and contribute to shoreline retreat and coastal erosion phenomena. Coastal zones are dynamic and complex environments consisting of heterogeneous and [...] Read more.
Climate change and its effects (i.e., sea level rise, extreme weather events) as well as anthropogenic activities, determine pressures to the coastal environments and contribute to shoreline retreat and coastal erosion phenomena. Coastal zones are dynamic and complex environments consisting of heterogeneous and different geomorphological features, while exhibiting different scales and spectral responses. Thus, the monitoring of changes in the coastal land classes and the extraction of coastlines/shorelines can be a challenging task. Earth Observation data and the application of spatiotemporal analysis methods can facilitate shoreline change analysis and detection. Apart from remote sensing methods, the advent of machine learning-based techniques presents an emerging trend, being capable of supporting the monitoring and modeling of coastal ecosystems at large scales. In this context, this study aims to provide a review of the relevant literature falling within the period of 2015–2022, where different machine learning approaches were applied for cases of coast-line/shoreline extraction and change analysis, and/or coastal dynamic monitoring. Particular emphasis is given on the analysis of the selected studies, including details about their performances, as well as their advantages and weaknesses, and information about the different environmental data employed. Full article
(This article belongs to the Special Issue GIS and Spatial Planning for Natural Hazards Mitigation)
►▼ Show Figures

Figure 1

19 pages, 7110 KiB  
Article
Comparative Analysis of Primary Photosynthetic Reactions Assessed by OJIP Kinetics in Three Brassica Crops after Drought and Recovery
by Jasenka Antunović Dunić, Selma Mlinarić, Iva Pavlović, Hrvoje Lepeduš and Branka Salopek-Sondi
Appl. Sci. 2023, 13(5), 3078; https://doi.org/10.3390/app13053078 - 27 Feb 2023
Cited by 9 | Viewed by 1849
Abstract
Plant drought tolerance depends on adaptations of the photosynthetic apparatus to changing environments triggered by water deficit. The seedlings of three Brassica crops differing in drought sensitivity, Brassica oleracea L. var. capitata—white cabbage, Brassica oleracea L. var. acephala—kale, and Brassica rapa [...] Read more.
Plant drought tolerance depends on adaptations of the photosynthetic apparatus to changing environments triggered by water deficit. The seedlings of three Brassica crops differing in drought sensitivity, Brassica oleracea L. var. capitata—white cabbage, Brassica oleracea L. var. acephala—kale, and Brassica rapa L. var. pekinensis—Chinese cabbage, were exposed to drought by withholding water. Detailed insight into the photosynthetic machinery was carried out when the seedling reached a relative water content of about 45% and after re-watering by analyzing the OJIP kinetics. The key objective of this study was to find reliable parameters for distinguishing drought−tolerant and drought-sensitive varieties before permanent structural and functional changes in the photosynthetic apparatus occur. According to our findings, an increase in the total performance index (PItotal) and structure–function index (SFI), positive L and K bands, total driving forces (ΔDF), and drought resistance index (DRI) suggest drought tolerance. At the same time, susceptible varieties can be distinguished based on negative L and K bands, PItotal, SFI, and the density of reaction centers (RC/CS0). Kale proved to be the most tolerant, Chinese cabbage was moderately susceptible, and white cabbage showed high sensitivity to the investigated drought stress. The genetic variation revealed among the selected Brassica crops could be used in breeding programs and high-precision crop management. Full article
(This article belongs to the Special Issue Biophysical Properties of Agricultural Crops)
►▼ Show Figures

Figure 1

24 pages, 2558 KiB  
Article
Comparison of the Spreadability of Butter and Butter Substitutes
by Małgorzata Ziarno, Dorota Derewiaka, Anna Florowska and Iwona Szymańska
Appl. Sci. 2023, 13(4), 2600; https://doi.org/10.3390/app13042600 - 17 Feb 2023
Cited by 10 | Viewed by 4891
Abstract
There are many types of butter, soft margarine, and blends, e.g., a mixture of butter and vegetable fats, on the market as bread spreads. Among these, butter and blends of butter with vegetable fats are very popular. The consumer’s choice of product is [...] Read more.
There are many types of butter, soft margarine, and blends, e.g., a mixture of butter and vegetable fats, on the market as bread spreads. Among these, butter and blends of butter with vegetable fats are very popular. The consumer’s choice of product is often determined by functional properties, such as texture, and the physicochemical composition of butter and butter substitutes. The aim of this study was to compare sixteen market samples of butter and butter substitutes in terms of spreadability and other selected structural (spreadability, hardness, adhesive force, and adhesiveness) and physicochemical parameters (water content, water distribution, plasma pH, color, acid value, peroxide number, saponification number, and instrumentally measured fatty acid profile) to investigate their correlation with spreadability. The parameters determined here were correlated with factors such as the type of sample, measuring temperature, and physicochemical composition. The statistical analysis revealed a very strong positive correlation between hardness and spreadability for all samples tested at 4 °C, as well as between hardness and spreadability for all samples tested 30 min after removal from the refrigerator; however, the interpretation of the results was different if the butter and butter substitute samples were subjected to a multivariate analysis separately. Full article
(This article belongs to the Special Issue Unconventional Raw Materials for Food Products)
►▼ Show Figures

Figure 1

19 pages, 1225 KiB  
Review
Review of Studies on Emotion Recognition and Judgment Based on Physiological Signals
by Wenqian Lin and Chao Li
Appl. Sci. 2023, 13(4), 2573; https://doi.org/10.3390/app13042573 - 16 Feb 2023
Cited by 37 | Viewed by 6222
Abstract
People’s emotions play an important part in our daily life and can not only reflect psychological and physical states, but also play a vital role in people’s communication, cognition and decision-making. Variations in people’s emotions induced by external conditions are accompanied by variations [...] Read more.
People’s emotions play an important part in our daily life and can not only reflect psychological and physical states, but also play a vital role in people’s communication, cognition and decision-making. Variations in people’s emotions induced by external conditions are accompanied by variations in physiological signals that can be measured and identified. People’s psychological signals are mainly measured with electroencephalograms (EEGs), electrodermal activity (EDA), electrocardiograms (ECGs), electromyography (EMG), pulse waves, etc. EEG signals are a comprehensive embodiment of the operation of numerous neurons in the cerebral cortex and can immediately express brain activity. EDA measures the electrical features of skin through skin conductance response, skin potential, skin conductance level or skin potential response. ECG technology uses an electrocardiograph to record changes in electrical activity in each cardiac cycle of the heart from the body surface. EMG is a technique that uses electronic instruments to evaluate and record the electrical activity of muscles, which is usually referred to as myoelectric activity. EEG, EDA, ECG and EMG have been widely used to recognize and judge people’s emotions in various situations. Different physiological signals have their own characteristics and are suitable for different occasions. Therefore, a review of the research work and application of emotion recognition and judgment based on the four physiological signals mentioned above is offered. The content covers the technologies adopted, the objects of application and the effects achieved. Finally, the application scenarios for different physiological signals are compared, and issues for attention are explored to provide reference and a basis for further investigation. Full article
(This article belongs to the Special Issue Recent Advances in Biological Science and Technology)
►▼ Show Figures

Figure 1

26 pages, 6738 KiB  
Article
Tannin Extraction from Chestnut Wood Waste: From Lab Scale to Semi-Industrial Plant
by Clelia Aimone, Giorgio Grillo, Luisa Boffa, Samuele Giovando and Giancarlo Cravotto
Appl. Sci. 2023, 13(4), 2494; https://doi.org/10.3390/app13042494 - 15 Feb 2023
Cited by 13 | Viewed by 5232
Abstract
The chestnut tree (Castanea sativa, Mill.) is a widespread plant in Europe whose fruits and wood has a relevant economic impact. Chestnut wood (CW) is rich in high-value compounds that exhibit various biological activities, such as antioxidant as well as anticarcinogenic [...] Read more.
The chestnut tree (Castanea sativa, Mill.) is a widespread plant in Europe whose fruits and wood has a relevant economic impact. Chestnut wood (CW) is rich in high-value compounds that exhibit various biological activities, such as antioxidant as well as anticarcinogenic and antimicrobial properties. These metabolites can be mainly divided into monomeric polyphenols and tannins. In this piece of work, we investigated a sustainable protocol to isolate enriched fractions of the above-mentioned compounds from CW residues. Specifically, a sequential extraction protocol, using subcritical water, was used as a pre-fractionation step, recovering approximately 88% of tannins and 40% of monomeric polyphenols in the first and second steps, respectively. The optimized protocol was also tested at pre-industrial levels, treating up to 13.5 kg CW and 160 L of solution with encouraging results. Ultra- and nanofiltrations were used to further enrich the recovered fractions, achieving more than 98% of the tannin content in the heavy fraction, whilst the removed permeate achieved up to 752.71 mg GAE/gext after the concentration (75.3%). Samples were characterized by means of total phenolic content (TPC), antioxidant activity (DPPH· and ABTS·), and tannin composition (hydrolysable and condensed). In addition, LC-MS-DAD was used for semiqualitative purposes to detect vescalagin/castalagin and vescalin/castalin, as well as gallic acid and ellagic acid. The developed valorization protocol allows the efficient fractionation and recovery of the major polyphenolic components of CW with a sustainable approach that also evaluates pre-industrial scaling-up. Full article
(This article belongs to the Section Green Sustainable Science and Technology)
►▼ Show Figures

Figure 1

14 pages, 5185 KiB  
Article
Speech Emotion Recognition Based on Two-Stream Deep Learning Model Using Korean Audio Information
by A-Hyeon Jo and Keun-Chang Kwak
Appl. Sci. 2023, 13(4), 2167; https://doi.org/10.3390/app13042167 - 8 Feb 2023
Cited by 15 | Viewed by 3798
Abstract
Identifying a person’s emotions is an important element in communication. In particular, voice is a means of communication for easily and naturally expressing emotions. Speech emotion recognition technology is a crucial component of human–computer interaction (HCI), in which accurately identifying emotions is key. [...] Read more.
Identifying a person’s emotions is an important element in communication. In particular, voice is a means of communication for easily and naturally expressing emotions. Speech emotion recognition technology is a crucial component of human–computer interaction (HCI), in which accurately identifying emotions is key. Therefore, this study presents a two-stream-based emotion recognition model based on bidirectional long short-term memory (Bi-LSTM) and convolutional neural networks (CNNs) using a Korean speech emotion database, and the performance is comparatively analyzed. The data used in the experiment were obtained from the Korean speech emotion recognition database built by Chosun University. Two deep learning models, Bi-LSTM and YAMNet, which is a CNN-based transfer learning model, were connected in a two-stream architecture to design an emotion recognition model. Various speech feature extraction methods and deep learning models were compared in terms of performance. Consequently, the speech emotion recognition performance of Bi-LSTM and YAMNet was 90.38% and 94.91%, respectively. However, the performance of the two-stream model was 96%, which was a minimum of 1.09% and up to 5.62% improved compared with a single model. Full article
(This article belongs to the Special Issue Deep Learning and Machine Learning in Biomedical Signal and Image Processing)
►▼ Show Figures

Figure 1

19 pages, 12841 KiB  
Article
Design of a Smart Factory Based on Cyber-Physical Systems and Internet of Things towards Industry 4.0
by Mutaz Ryalat, Hisham ElMoaqet and Marwa AlFaouri
Appl. Sci. 2023, 13(4), 2156; https://doi.org/10.3390/app13042156 - 8 Feb 2023
Cited by 80 | Viewed by 11296
Abstract
The rise of Industry 4.0, which employs emerging powerful and intelligent technologies and represents the digital transformation of manufacturing, has a significant impact on society, industry, and other production sectors. The industrial scene is witnessing ever-increasing pressure to improve its agility and versatility [...] Read more.
The rise of Industry 4.0, which employs emerging powerful and intelligent technologies and represents the digital transformation of manufacturing, has a significant impact on society, industry, and other production sectors. The industrial scene is witnessing ever-increasing pressure to improve its agility and versatility to accommodate the highly modularized, customized, and dynamic demands of production. One of the key concepts within Industry 4.0 is the smart factory, which represents a manufacturing/production system with interconnected processes and operations via cyber-physical systems, the Internet of Things, and state-of-the-art digital technologies. This paper outlines the design of a smart cyber-physical system that complies with the innovative smart factory framework for Industry 4.0 and implements the core industrial, computing, information, and communication technologies of the smart factory. It discusses how to combine the key components (pillars) of a smart factory to create an intelligent manufacturing system. As a demonstration of a simplified smart factory model, a smart manufacturing case study with a drilling process is implemented, and the feasibility of the proposed method is demonstrated and verified with experiments. Full article
(This article belongs to the Section Mechanical Engineering)
►▼ Show Figures

Figure 1

51 pages, 3550 KiB  
Review
Solid Lipid Nanoparticles (SLNs) and Nanostructured Lipid Carriers (NLCs) as Food-Grade Nanovehicles for Hydrophobic Nutraceuticals or Bioactives
by Chuan-He Tang, Huan-Le Chen and Jin-Ru Dong
Appl. Sci. 2023, 13(3), 1726; https://doi.org/10.3390/app13031726 - 29 Jan 2023
Cited by 25 | Viewed by 5840
Abstract
Although solid lipid nanoparticles (SLNs) and nanostructured lipid carriers (NLCs) have been successfully used as drug delivery systems for about 30 years, the usage of these nanoparticles as food-grade nanovehicles for nutraceuticals or bioactive compounds has been, relatively speaking, scarcely investigated. With fast-increasing [...] Read more.
Although solid lipid nanoparticles (SLNs) and nanostructured lipid carriers (NLCs) have been successfully used as drug delivery systems for about 30 years, the usage of these nanoparticles as food-grade nanovehicles for nutraceuticals or bioactive compounds has been, relatively speaking, scarcely investigated. With fast-increasing interest in the incorporation of a wide range of bioactives in food formulations, as well as health awareness of consumers, there has been a renewed urge for the development of food-compatible SLNs and/or NLCs as nanovehicles for improving water dispersibility, stability, bioavailability, and bioactivities of many lipophilic nutraceuticals or poorly soluble bioactives. In this review, the development of food-grade SLNs and NLCs, as well as their utilization as nanosized delivery systems for lipophilic or hydrophobic nutraceuticals, was comprehensively reviewed. First, the structural composition and preparation methods of food-grade SLNs and NLCs were simply summarized. Next, some key issues about the usage of such nanoparticles as oral nanovehicles, e.g., incorporation and release of bioactives, oxidative stability, lipid digestion and absorption, and intestinal transport, were critically discussed. Then, recent advances in the utilization of SLNs and NLCs as nanovehicles for encapsulation and delivery of different liposoluble or poorly soluble nutraceuticals or bioactives were comprehensively reviewed. The performance of such nanoparticles as nanovehicles for improving stability, bioavailability, and bioactivities of curcuminoids (and curcumin in particular) was also highlighted. Lastly, some strategies to improve the oral bioavailability and delivery of loaded nutraceuticals in such nanoparticles were presented. The review will be relevant, providing state-of-the-art knowledge about the development of food-grade lipid-based nanovehicles for improving the stability and bioavailability of many nutraceuticals. Full article
(This article belongs to the Special Issue Editorial Board Members' Collection Series: Functional Foods)
►▼ Show Figures

Graphical abstract

18 pages, 1715 KiB  
Article
TeleFE: A New Tool for the Tele-Assessment of Executive Functions in Children
by Carlotta Rivella, Costanza Ruffini, Clara Bombonato, Agnese Capodieci, Andrea Frascari, Gian Marco Marzocchi, Alessandra Mingozzi, Chiara Pecini, Laura Traverso, Maria Carmen Usai and Paola Viterbori
Appl. Sci. 2023, 13(3), 1728; https://doi.org/10.3390/app13031728 - 29 Jan 2023
Cited by 11 | Viewed by 2658
Abstract
In recent decades, the utility of cognitive tele-assessment has increasingly been highlighted, both in adults and in children. The present study aimed to present TeleFE, a new tool for the tele-assessment of EF in children aged 6–13. TeleFE consists of a web platform [...] Read more.
In recent decades, the utility of cognitive tele-assessment has increasingly been highlighted, both in adults and in children. The present study aimed to present TeleFE, a new tool for the tele-assessment of EF in children aged 6–13. TeleFE consists of a web platform including four tasks based on robust neuropsychological paradigms to evaluate inhibition, interference suppression, working memory, cognitive flexibility, and planning. It also includes questionnaires on EF for teachers and parents, to obtain information on the everyday functioning of the children. As TeleFE allows the assessment of EF both remotely and in-person, a comparison of the two modalities was conducted by administering TeleFE to 1288 Italian primary school children. A series of ANOVA was conducted, showing no significant effect of assessment modality (p > 0.05 for all the measures). In addition, significant differences by class emerged for all the measures (p < 0.001 for all the measures except p = 0.008 for planning). Finally, a significant sex effect emerged for inhibition (p < 0.001) and for the reaction times in both interference control (p = 0.013) and cognitive flexibility (p < 0.001), with boys showing a lower inhibition and faster reaction times. The implications of these results along with the indications for the choice of remote assessment are discussed. Full article
(This article belongs to the Special Issue New Digital Technologies for Diagnosis and Rehabilitation of Neurodevelopmental Disorders)
►▼ Show Figures

Figure 1

17 pages, 7391 KiB  
Article
Data Augmentation Method for Plant Leaf Disease Recognition
by Byeongjun Min, Taehyun Kim, Dongil Shin and Dongkyoo Shin
Appl. Sci. 2023, 13(3), 1465; https://doi.org/10.3390/app13031465 - 22 Jan 2023
Cited by 16 | Viewed by 3477
Abstract
Recently, several plant pathogens have become more active due to temperature increases arising from climate change, which has caused damage to various crops. If climate change continues, it will likely be very difficult to maintain current crop production, and the problem of a [...] Read more.
Recently, several plant pathogens have become more active due to temperature increases arising from climate change, which has caused damage to various crops. If climate change continues, it will likely be very difficult to maintain current crop production, and the problem of a shortage of expert manpower is also deepening. Fortunately, research on various early diagnosis systems based on deep learning is actively underway to solve these problems, but the problem of lack of diversity in some hard-to-collect disease samples remains. This imbalanced data increases the bias of machine learning models, causing overfitting problems. In this paper, we propose a data augmentation method based on an image-to-image translation model to solve the bias problem by supplementing these insufficient diseased leaf images. The proposed augmentation method performs translation between healthy and diseased leaf images and utilizes attention mechanisms to create images that reflect more evident disease textures. Through these improvements, we generated a more plausible diseased leaf image compared to existing methods and conducted an experiment to verify whether this data augmentation method could further improve the performance of a classification model for early diagnosis of plants. In the experiment, the PlantVillage dataset was used, and the extended dataset was built using the generated images and original images, and the performance of the classification models was evaluated through the test set. Full article
(This article belongs to the Special Issue Applications of Machine Learning in Agriculture)
►▼ Show Figures

Figure 1

16 pages, 951 KiB  
Review
Pasteurization of Food and Beverages by High Pressure Processing (HPP) at Room Temperature: Inactivation of Staphylococcus aureus, Escherichia coli, Listeria monocytogenes, Salmonella, and Other Microbial Pathogens
by Filipa Vinagre M. Silva and Evelyn
Appl. Sci. 2023, 13(2), 1193; https://doi.org/10.3390/app13021193 - 16 Jan 2023
Cited by 22 | Viewed by 8229
Abstract
Vegetative pathogens actively grow in foods, metabolizing and dividing their cells. They have consequently become a focus of concern for the food industry, food regulators and food control agencies. Although much has been done by the food industry and food regulatory agencies, foodborne [...] Read more.
Vegetative pathogens actively grow in foods, metabolizing and dividing their cells. They have consequently become a focus of concern for the food industry, food regulators and food control agencies. Although much has been done by the food industry and food regulatory agencies, foodborne outbreaks are still reported globally, causing illnesses, hospitalizations, and in certain cases, deaths, together with product recalls and subsequent economic losses. Major bacterial infections from raw and processed foods are caused by Escherichia coli serotype O157:H7, Salmonella enteritidis, and Listeria monocytogenes. High pressure processing (HPP) (also referred to as high hydrostatic pressure, HHP) is a non-thermal pasteurization technology that relies on very high pressures (400–600 MPa) to inactivate pathogens, instead of heat, thus causing less negative impact in the food nutrients and quality. HPP can be used to preserve foods, instead of chemical food additives. In this study, a review of the effect of HPP treatments on major vegetative bacteria in specific foods was carried out. HPP at 600 MPa, commonly used by the food industry, can achieve the recommended 5–8-log reductions in E. coli, S. enteritidis, L. monocytogenes, and Vibrio. Staphylococcus aureus presented the highest resistance to HPP among the foodborne vegetative pathogens investigated, followed by E. coli. More susceptible L. monocytogenes and Salmonella spp. bacteria were reduced by 6 logs at pressures within 500–600 MPa. Vibrio spp. (e.g., raw oysters), Campylobacter jejuni, Yersinia enterocolitica, Citrobacter freundii and Aeromonas hydrophila generally required lower pressures (300–400 MPa) for inactivation. Bacterial species and strain, as well as the food itself, with a characteristic composition, affect the microbial inactivation. This review demonstrates that HPP is a safe pasteurization technology, which is able to achieve at least 5-log reduction in major food bacterial pathogens, without the application of heat. Full article
(This article belongs to the Special Issue Non-thermal Technologies for Food Processing)
►▼ Show Figures

Figure 1

27 pages, 1989 KiB  
Review
A Review of Recent Progress of Carbon Capture, Utilization, and Storage (CCUS) in China
by Jia Yao, Hongdou Han, Yang Yang, Yiming Song and Guihe Li
Appl. Sci. 2023, 13(2), 1169; https://doi.org/10.3390/app13021169 - 15 Jan 2023
Cited by 46 | Viewed by 7360
Abstract
The continuous temperature rise has raised global concerns about CO2 emissions. As the country with the largest CO2 emissions, China is facing the challenge of achieving large CO2 emission reductions (or even net-zero CO2 emissions) in a short period. [...] Read more.
The continuous temperature rise has raised global concerns about CO2 emissions. As the country with the largest CO2 emissions, China is facing the challenge of achieving large CO2 emission reductions (or even net-zero CO2 emissions) in a short period. With the strong support and encouragement of the Chinese government, technological breakthroughs and practical applications of carbon capture, utilization, and storage (CCUS) are being aggressively pursued, and some outstanding accomplishments have been realized. Based on the numerous information from a wide variety of sources including publications and news reports only available in Chinese, this paper highlights the latest CCUS progress in China after 2019 by providing an overview of known technologies and typical projects, aiming to provide theoretical and practical guidance for achieving net-zero CO2 emissions in the future. Full article
(This article belongs to the Special Issue Applications and Challenges in Carbon Capture, Utilization and Storage)
►▼ Show Figures

Figure 1

21 pages, 1011 KiB  
Article
A Deep Learning Method for Lightweight and Cross-Device IoT Botnet Detection
by Marta Catillo, Antonio Pecchia and Umberto Villano
Appl. Sci. 2023, 13(2), 837; https://doi.org/10.3390/app13020837 - 7 Jan 2023
Cited by 17 | Viewed by 3618
Abstract
Ensuring security of Internet of Things (IoT) devices in the face of threats and attacks is a primary concern. IoT plays an increasingly key role in cyber–physical systems. Many existing intrusion detection systems (IDS) proposals for the IoT leverage complex machine learning architectures, [...] Read more.
Ensuring security of Internet of Things (IoT) devices in the face of threats and attacks is a primary concern. IoT plays an increasingly key role in cyber–physical systems. Many existing intrusion detection systems (IDS) proposals for the IoT leverage complex machine learning architectures, which often provide one separate model per device or per attack. These solutions are not suited to the scale and dynamism of modern IoT networks. This paper proposes a novel IoT-driven cross-device method, which allows learning a single IDS model instead of many separate models atop the traffic of different IoT devices. A semi-supervised approach is adopted due to its wider applicability for unanticipated attacks. The solution is based on an all-in-one deep autoencoder, which consists of training a single deep neural network with the normal traffic from different IoT devices. Extensive experimentation performed with a widely used benchmarking dataset indicates that the all-in-one approach achieves within 0.9994–0.9997 recall, 0.9999–1.0 precision, 0.0–0.0071 false positive rate and 0.9996–0.9998 F1 score, depending on the device. The results obtained demonstrate the validity of the proposal, which represents a lightweight and device-independent solution with considerable advantages in terms of transferability and adaptability. Full article
(This article belongs to the Collection Innovation in Information Security)
►▼ Show Figures

Figure 1

22 pages, 33375 KiB  
Article
Using UAS-Aided Photogrammetry to Monitor and Quantify the Geomorphic Effects of Extreme Weather Events in Tectonically Active Mass Waste-Prone Areas: The Case of Medicane Ianos
by Evelina Kotsi, Emmanuel Vassilakis, Michalis Diakakis, Spyridon Mavroulis, Aliki Konsolaki, Christos Filis, Stylianos Lozios and Efthymis Lekkas
Appl. Sci. 2023, 13(2), 812; https://doi.org/10.3390/app13020812 - 6 Jan 2023
Cited by 9 | Viewed by 1975
Abstract
Extreme weather events can trigger various hydrogeomorphic phenomena and processes including slope failures. These shallow instabilities are difficult to monitor and measure due to the spatial and temporal scales in which they occur. New technologies such as unmanned aerial systems (UAS), photogrammetry and [...] Read more.
Extreme weather events can trigger various hydrogeomorphic phenomena and processes including slope failures. These shallow instabilities are difficult to monitor and measure due to the spatial and temporal scales in which they occur. New technologies such as unmanned aerial systems (UAS), photogrammetry and the structure-from-motion (SfM) technique have recently demonstrated capabilities useful in performing accurate terrain observations that have the potential to provide insights into these geomorphic processes. This study explores the use of UAS-aided photogrammetry and change detection, using specialized techniques such as the digital elevation model (DEM) of differences (DoD) and cloud-to-cloud distance (C2C) to monitor and quantify geomorphic changes before and after an extreme medicane event in Myrtos, a highly visited touristic site on Cephalonia Island, Greece. The application demonstrates that the combination of UAS with photogrammetry allows accurate delineation of instabilities, volumetric estimates of morphometric changes, insights into erosion and deposition processes and the delineation of higher-risk areas in a rapid, safe and practical way. Overall, the study illustrates that the combination of tools facilitates continuous monitoring and provides key insights into geomorphic processes that are otherwise difficult to observe. Through this deeper understanding, this approach can be a stepping stone to risk management of this type of highly-visited sites, which in turn is a key ingredient to sustainable development in high-risk areas. Full article
(This article belongs to the Section Earth Sciences)
►▼ Show Figures

Figure 1

10 pages, 626 KiB  
Article
Knowing Knowledge: Epistemological Study of Knowledge in Transformers
by Leonardo Ranaldi and Giulia Pucci
Appl. Sci. 2023, 13(2), 677; https://doi.org/10.3390/app13020677 - 4 Jan 2023
Cited by 40 | Viewed by 3070
Abstract
Statistical learners are leading towards auto-epistemic logic, but is it the right way to progress in artificial intelligence (AI)? Ways to discover AI fit the senses and the intellect. The structure of symbols–the operations by which the intellectual solution is realized–and the search [...] Read more.
Statistical learners are leading towards auto-epistemic logic, but is it the right way to progress in artificial intelligence (AI)? Ways to discover AI fit the senses and the intellect. The structure of symbols–the operations by which the intellectual solution is realized–and the search for strategic reference points evoke essential issues in the analysis of AI. Studying how knowledge can be represented through methods of theoretical generalization and empirical observation is only the latest step in a long process of evolution. In this paper, we try to outline the origin of knowledge and how modern artificial minds have inherited it. Full article
(This article belongs to the Special Issue Deep Learning Based on Neural Network Design)
►▼ Show Figures

Figure 1

16 pages, 925 KiB  
Review
A Review of the Relationship between Gut Microbiome and Obesity
by Dorottya Zsálig, Anikó Berta, Vivien Tóth, Zoltán Szabó, Klára Simon, Mária Figler, Henriette Pusztafalvi and Éva Polyák
Appl. Sci. 2023, 13(1), 610; https://doi.org/10.3390/app13010610 - 2 Jan 2023
Cited by 26 | Viewed by 13369
Abstract
Obesity is a rapidly growing problem of public health on a worldwide scale, responsible for more than 60% of deaths associated with high body mass index. Recent studies underpinned the augmenting importance of the gut microbiota in obesity. Gut microbiota alterations affect the [...] Read more.
Obesity is a rapidly growing problem of public health on a worldwide scale, responsible for more than 60% of deaths associated with high body mass index. Recent studies underpinned the augmenting importance of the gut microbiota in obesity. Gut microbiota alterations affect the energy balance of the host organism; namely, as a factor affecting energy production from the diet and as a factor affecting host genes regulating energy expenditure and storage. Gut microbiota composition is characterised by constant variability, and is affected by several dietary factors, suggesting the probability that manipulation of the gut microbiota may promote leaning or prevent obesity. Our narrative review summarizes the results of recent years that stress the effect of gut microbiota in the development of obesity. It investigates the factors (diet, dietary components, lifestyle, and environment) that might affect the gut microbiota composition. Possible strategies for the prevention and/or treatment of obesity include restoring or modifying the composition of the microbiota by consuming prebiotics and probiotics, fermented foods, fruits, vegetables, and avoiding foods of animal origin high in saturated fat and sugar. Full article
(This article belongs to the Special Issue Functional Foods in Disease Prevention and Health Promotion: Volume II)
►▼ Show Figures

Figure 1

16 pages, 939 KiB  
Review
Review: Renewable Energy in an Increasingly Uncertain Future
by Patrick Moriarty and Damon Honnery
Appl. Sci. 2023, 13(1), 388; https://doi.org/10.3390/app13010388 - 28 Dec 2022
Cited by 14 | Viewed by 3999
Abstract
A number of technical solutions have been proposed for tackling global climate change. However, global climate change is not the only serious global environmental challenge we face demanding an urgent response, even though atmospheric CO2 ppm have risen from 354 in 1990 [...] Read more.
A number of technical solutions have been proposed for tackling global climate change. However, global climate change is not the only serious global environmental challenge we face demanding an urgent response, even though atmospheric CO2 ppm have risen from 354 in 1990 to 416 in 2020. The rise of multiple global environmental challenges makes the search for solutions more difficult, because all technological solutions give rise to some unwanted environmental effects. Further, not only must these various problems be solved in the same short time frame, but they will need to be tackled in a time of rising international tensions, and steady global population increase. This review looks particularly at how all these environmental problems impact the future prospects for renewable energy (RE), given that RE growth must not exacerbate the other equally urgent problems, and must make a major difference in a decade or so. The key finding is that, while the world must shift to RE in the longer run, in the short term what is more important is to improve Earth’s ecological sustainability by the most effective means possible. It is shown that reducing both the global transport task and agricultural production (while still providing an adequate diet for all) can be far more effective than converting the energy used in these sectors to RE. Full article
(This article belongs to the Special Issue New Developments and Prospects in Clean and Renewable Energies)
►▼ Show Figures

Figure 1

26 pages, 2225 KiB  
Article
Nature-Based Solutions in Urban Areas: A European Analysis
by Sara Bona, Armando Silva-Afonso, Ricardo Gomes, Raquel Matos and Fernanda Rodrigues
Appl. Sci. 2023, 13(1), 168; https://doi.org/10.3390/app13010168 - 23 Dec 2022
Cited by 19 | Viewed by 5904
Abstract
Currently, the world is facing resource scarcity as the environmental impacts of human intervention continue to intensify. To facilitate the conservation and recovery of ecosystems and to transform cities into more sustainable, intelligent, regenerative, and resilient environments, the concepts of circularity and nature-based [...] Read more.
Currently, the world is facing resource scarcity as the environmental impacts of human intervention continue to intensify. To facilitate the conservation and recovery of ecosystems and to transform cities into more sustainable, intelligent, regenerative, and resilient environments, the concepts of circularity and nature-based solutions (NbS) are applied. The role of NbS within green infrastructure in urban resilience is recognised, and considerable efforts are being made by the European Commission (EC) to achieve the European sustainability goals. However, it is not fully evidenced, in an integrated way, which are the main NbS implemented in the urban environment and their effects. This article aims to identify the main and most recent NbS applied in urban environments at the European level and to analyse the integration of different measures as an innovative analysis based on real cases. For this purpose, this work presents a literature review of 69 projects implemented in 24 European cities, as well as 8 urban actions and 3 spatial scales of implementation at the district level. Therefore, there is great potential for NbS adoption in buildings and their surroundings, which are still not prioritized, given the lack of effective monitoring of the effects of NbS. Full article
(This article belongs to the Special Issue Sustainability in Energy and Buildings: Future Perspectives and Challenges)
►▼ Show Figures

Figure 1

30 pages, 3754 KiB  
Review
A Review of Deep Reinforcement Learning Approaches for Smart Manufacturing in Industry 4.0 and 5.0 Framework
by Alejandro del Real Torres, Doru Stefan Andreiana, Álvaro Ojeda Roldán, Alfonso Hernández Bustos and Luis Enrique Acevedo Galicia
Appl. Sci. 2022, 12(23), 12377; https://doi.org/10.3390/app122312377 - 3 Dec 2022
Cited by 26 | Viewed by 6993
Abstract
In this review, the industry’s current issues regarding intelligent manufacture are presented. This work presents the status and the potential for the I4.0 and I5.0’s revolutionary technologies. AI and, in particular, the DRL algorithms, which are a perfect response to the unpredictability and [...] Read more.
In this review, the industry’s current issues regarding intelligent manufacture are presented. This work presents the status and the potential for the I4.0 and I5.0’s revolutionary technologies. AI and, in particular, the DRL algorithms, which are a perfect response to the unpredictability and volatility of modern demand, are studied in detail. Through the introduction of RL concepts and the development of those with ANNs towards DRL, the potential and variety of these kinds of algorithms are highlighted. Moreover, because these algorithms are data based, their modification to meet the requirements of industry operations is also included. In addition, this review covers the inclusion of new concepts, such as digital twins, in response to an absent environment model and how it can improve the performance and application of DRL algorithms even more. This work highlights that DRL applicability is demonstrated across all manufacturing industry operations, outperforming conventional methodologies and, most notably, enhancing the manufacturing process’s resilience and adaptability. It is stated that there is still considerable work to be carried out in both academia and industry to fully leverage the promise of these disruptive tools, begin their deployment in industry, and take a step closer to the I5.0 industrial revolution. Full article
(This article belongs to the Special Issue Smart Machines and Intelligent Manufacturing)
►▼ Show Figures

Figure 1

27 pages, 2011 KiB  
Review
A Study of Network Intrusion Detection Systems Using Artificial Intelligence/Machine Learning
by Patrick Vanin, Thomas Newe, Lubna Luxmi Dhirani, Eoin O’Connell, Donna O’Shea, Brian Lee and Muzaffar Rao
Appl. Sci. 2022, 12(22), 11752; https://doi.org/10.3390/app122211752 - 18 Nov 2022
Cited by 35 | Viewed by 11780
Abstract
The rapid growth of the Internet and communications has resulted in a huge increase in transmitted data. These data are coveted by attackers and they continuously create novel attacks to steal or corrupt these data. The growth of these attacks is an issue [...] Read more.
The rapid growth of the Internet and communications has resulted in a huge increase in transmitted data. These data are coveted by attackers and they continuously create novel attacks to steal or corrupt these data. The growth of these attacks is an issue for the security of our systems and represents one of the biggest challenges for intrusion detection. An intrusion detection system (IDS) is a tool that helps to detect intrusions by inspecting the network traffic. Although many researchers have studied and created new IDS solutions, IDS still needs improving in order to have good detection accuracy while reducing false alarm rates. In addition, many IDS struggle to detect zero-day attacks. Recently, machine learning algorithms have become popular with researchers to detect network intrusion in an efficient manner and with high accuracy. This paper presents the concept of IDS and provides a taxonomy of machine learning methods. The main metrics used to assess an IDS are presented and a review of recent IDS using machine learning is provided where the strengths and weaknesses of each solution is outlined. Then, details of the different datasets used in the studies are provided and the accuracy of the results from the reviewed work is discussed. Finally, observations, research challenges and future trends are discussed. Full article
(This article belongs to the Special Issue Information Security and Privacy)
►▼ Show Figures

Figure 1

18 pages, 7267 KiB  
Article
Machine Learning-Assisted Prediction of Oil Production and CO2 Storage Effect in CO2-Water-Alternating-Gas Injection (CO2-WAG)
by Hangyu Li, Changping Gong, Shuyang Liu, Jianchun Xu and Gloire Imani
Appl. Sci. 2022, 12(21), 10958; https://doi.org/10.3390/app122110958 - 29 Oct 2022
Cited by 12 | Viewed by 3398
Abstract
In recent years, CO2 flooding has emerged as an efficient method for improving oil recovery. It also has the advantage of storing CO2 underground. As one of the promising types of CO2 enhanced oil recovery (CO2-EOR), CO2 [...] Read more.
In recent years, CO2 flooding has emerged as an efficient method for improving oil recovery. It also has the advantage of storing CO2 underground. As one of the promising types of CO2 enhanced oil recovery (CO2-EOR), CO2 water-alternating-gas injection (CO2-WAG) can suppress CO2 fingering and early breakthrough problems that occur during oil recovery by CO2 flooding. However, the evaluation of CO2-WAG is strongly dependent on the injection parameters, which in turn renders numerical simulations computationally expensive. So, in this work, machine learning is used to help predict how well CO2-WAG will work when different injection parameters are used. A total of 216 models were built by using CMG numerical simulation software to represent CO2-WAG development scenarios of various injection parameters where 70% of them were used as training sets and 30% as testing sets. A random forest regression algorithm was used to predict CO2-WAG performance in terms of oil production, CO2 storage amount, and CO2 storage efficiency. The CO2-WAG period, CO2 injection rate, and water–gas ratio were chosen as the three main characteristics of injection parameters. The prediction results showed that the predicted value of the test set was very close to the true value. The average absolute prediction deviations of cumulative oil production, CO2 storage amount, and CO2 storage efficiency were 1.10%, 3.04%, and 2.24%, respectively. Furthermore, it only takes about 10 s to predict the results of all 216 scenarios by using machine learning methods, while the CMG simulation method spends about 108 min. It demonstrated that the proposed machine-learning method can rapidly predict CO2-WAG performance with high accuracy and high computational efficiency under conditions of various injection parameters. This work gives more insights into the optimization of the injection parameters for CO2-EOR. Full article
(This article belongs to the Special Issue Artificial Intelligence Applications in Petroleum Exploration and Production)
►▼ Show Figures

Figure 1

17 pages, 4661 KiB  
Article
Forecast of Airblast Vibrations Induced by Blasting Using Support Vector Regression Optimized by the Grasshopper Optimization (SVR-GO) Technique
by Lihua Chen, Panagiotis G. Asteris, Markos Z. Tsoukalas, Danial Jahed Armaghani, Dmitrii Vladimirovich Ulrikh and Mojtaba Yari
Appl. Sci. 2022, 12(19), 9805; https://doi.org/10.3390/app12199805 - 29 Sep 2022
Cited by 16 | Viewed by 1929
Abstract
Air overpressure (AOp) is an undesirable environmental effect of blasting. To date, a variety of empirical equations have been developed to forecast this phenomenon and prevent its negative impacts with accuracy. However, the accuracy of these methods is not sufficient. In addition, they [...] Read more.
Air overpressure (AOp) is an undesirable environmental effect of blasting. To date, a variety of empirical equations have been developed to forecast this phenomenon and prevent its negative impacts with accuracy. However, the accuracy of these methods is not sufficient. In addition, they are resource-consuming. This study employed support vector regression (SVR) optimized with the grasshopper optimizer (GO) algorithm to forecast AOp resulting from blasting. Additionally, a novel input selection technique, the Boruta algorithm (BFS), was applied. A new algorithm, the SVR-GA-BFS7, was developed by combining the models mentioned above. The findings showed that the SVR-GO-BFS7 model was the best technique (R2 = 0.983, RMSE = 1.332). The superiority of this model means that using the seven most important inputs was enough to forecast the AOp in the present investigation. Furthermore, the performance of SVR-GO-BFS7 was compared with various machine learning techniques, and the model outperformed the base models. The GO was compared with some other optimization techniques, and the superiority of this algorithm over the others was confirmed. Therefore, the suggested method presents a framework for accurate AOp prediction that supports the resource-saving forecasting methods. Full article
(This article belongs to the Special Issue Blast and Impact Engineering on Structures and Materials)
►▼ Show Figures

Figure 1

26 pages, 5272 KiB  
Article
Benchmarking 4G and 5G-Based Cellular-V2X for Vehicle-to-Infrastructure Communication and Urban Scenarios in Cooperative Intelligent Transportation Systems
by Tibor Petrov, Peter Pocta and Tatiana Kovacikova
Appl. Sci. 2022, 12(19), 9677; https://doi.org/10.3390/app12199677 - 26 Sep 2022
Cited by 15 | Viewed by 3360
Abstract
Vehicle-to-Infrastructure (V2I) communication is expected to bring tremendous benefits in terms of increased road safety, improved traffic efficiency and decreased environmental impact. In 2017, The 3rd Generation Partnership Project (3GPP) released 3GPP Release 14, which introduced Cellular Vehicle-to-Everything communication (C-V2X), bringing Vehicle-to-Everything (V2X) [...] Read more.
Vehicle-to-Infrastructure (V2I) communication is expected to bring tremendous benefits in terms of increased road safety, improved traffic efficiency and decreased environmental impact. In 2017, The 3rd Generation Partnership Project (3GPP) released 3GPP Release 14, which introduced Cellular Vehicle-to-Everything communication (C-V2X), bringing Vehicle-to-Everything (V2X) communication capabilities to cellular networks, hence creating an alternative to Dedicated Short-Range Communications (DSRC) technology. Since then, every new 3GPP Release including Release 15, a first full set of 5G standards, offered V2X capabilities. In this paper, we present a complex simulation study, which benchmarks the performance of LTE-based and 5G-based C-V2X technologies deployed for V2I communication in an urban setting. The study compares LTE and 5G deployed both in the Device-to-Device in mode 3 and in infrastructural mode. Target performance indicators used for comparison are average end-to-end (E2E) latency and Packet Delivery Ratio (PDR). The performance of those technologies is studied under varying communication conditions realized by a variation of vehicle traffic intensity, communication perimeter and message generation frequency. Furthermore, the effects of infrastructure deployment density on the performance of selected C-V2X communication technologies are explored by comparing the performance of the investigated technologies for three infrastructure density scenarios, i.e., involving two, four and eight base stations (BSs). The performance results are put into a context of the connectivity requirements of the most popular V2I communication services. The results indicate that both C-V2X technologies can support all the considered V2I services without any limitations in terms of the communication perimeter, traffic intensity and message generation frequency. When it comes to the infrastructure density deployment, the results show that increasing the density of the infrastructure deployment from two BSs to four BSs offers a remarkable performance improvement for all the considered V2I services as well as investigated technologies and their modes. Further infrastructure density increase (from four BSs to eight BSs) does not yield any practical benefits in the investigated urban scenario. Full article
(This article belongs to the Special Issue 5G Vehicle-to-Everything (V2X): Latest Advances and Prospects)
►▼ Show Figures

Figure 1

15 pages, 6052 KiB  
Article
Scalability of Mach Number Effects on Noise Emitted by Side-by-Side Propellers
by Caterina Poggi, Giovanni Bernardini, Massimo Gennaretti and Roberto Camussi
Appl. Sci. 2022, 12(19), 9507; https://doi.org/10.3390/app12199507 - 22 Sep 2022
Cited by 14 | Viewed by 1499
Abstract
This paper presents a numerical investigation of noise radiated by two side-by-side propellers, suitable for Distributed-Electric-Propulsion concepts. The focus is on the assessment of the variation of the effects of blade tip Mach number on the radiated noise for variations of the direction [...] Read more.
This paper presents a numerical investigation of noise radiated by two side-by-side propellers, suitable for Distributed-Electric-Propulsion concepts. The focus is on the assessment of the variation of the effects of blade tip Mach number on the radiated noise for variations of the direction of rotation, hub relative position, and the relative phase angle between the propeller blades. The aerodynamic analysis is performed through a potential-flow-based boundary integral formulation, which is able to model severe body–wake interactions.The noise field is evaluated through a boundary-integral formulation for the solution of the Ffowcs Williams and Hawkings equation. The numerical investigation shows that: the blade tip Mach number strongly affects the magnitude and directivity of the radiated noise; the increase of the tip-clearance increases the spatial frequency of the noise directivity at the two analyzed tip Mach numbers for both co-rotating and counter-rotating configurations; for counter-rotating propellers, the relative phase angle between the propeller blades provides a decrease of the averaged emitted noise, regardless the tip Mach number. One of the main results achieved is the scalability with the blade tip Mach number of the influence on the emitted noise of the considered design parameters. Full article
(This article belongs to the Special Issue Aerodynamic Aeroelasticity and Aeroacoustics of Rotorcraft)
►▼ Show Figures

Figure 1

13 pages, 2503 KiB  
Article
An Improved Algorithm of Drift Compensation for Olfactory Sensors
by Siyu Lu, Jialiang Guo, Shan Liu, Bo Yang, Mingzhe Liu, Lirong Yin and Wenfeng Zheng
Appl. Sci. 2022, 12(19), 9529; https://doi.org/10.3390/app12199529 - 22 Sep 2022
Cited by 81 | Viewed by 3069
Abstract
This research mainly studies the semi-supervised learning algorithm of different domain data in machine olfaction, also known as sensor drift compensation algorithm. Usually for this kind of problem, it is difficult to obtain better recognition results by directly using the semi-supervised learning algorithm. [...] Read more.
This research mainly studies the semi-supervised learning algorithm of different domain data in machine olfaction, also known as sensor drift compensation algorithm. Usually for this kind of problem, it is difficult to obtain better recognition results by directly using the semi-supervised learning algorithm. For this reason, we propose a domain transformation semi-supervised weighted kernel extreme learning machine (DTSWKELM) algorithm, which converts the data through the domain and uses SWKELM algorithmic classification to transform the semi-supervised classification problem of different domain data into a semi-supervised classification problem of the same domain data. Full article
(This article belongs to the Special Issue Advances in Artificial Intelligence for Perception Augmentation and Reasoning)
►▼ Show Figures

Figure 1

36 pages, 617 KiB  
Review
Federated Learning for Edge Computing: A Survey
by Alexander Brecko, Erik Kajati, Jiri Koziorek and Iveta Zolotova
Appl. Sci. 2022, 12(18), 9124; https://doi.org/10.3390/app12189124 - 11 Sep 2022
Cited by 34 | Viewed by 10260
Abstract
New technologies bring opportunities to deploy AI and machine learning to the edge of the network, allowing edge devices to train simple models that can then be deployed in practice. Federated learning (FL) is a distributed machine learning technique to create a global [...] Read more.
New technologies bring opportunities to deploy AI and machine learning to the edge of the network, allowing edge devices to train simple models that can then be deployed in practice. Federated learning (FL) is a distributed machine learning technique to create a global model by learning from multiple decentralized edge clients. Although FL methods offer several advantages, including scalability and data privacy, they also introduce some risks and drawbacks in terms of computational complexity in the case of heterogeneous devices. Internet of Things (IoT) devices may have limited computing resources, poorer connection quality, or may use different operating systems. This paper provides an overview of the methods used in FL with a focus on edge devices with limited computational resources. This paper also presents FL frameworks that are currently popular and that provide communication between clients and servers. In this context, various topics are described, which include contributions and trends in the literature. This includes basic models and designs of system architecture, possibilities of application in practice, privacy and security, and resource management. Challenges related to the computational requirements of edge devices such as hardware heterogeneity, communication overload or limited resources of devices are discussed. Full article
(This article belongs to the Special Issue Edge Computing Communications)
►▼ Show Figures

Figure 1

18 pages, 1753 KiB  
Article
A Novel Hybrid Method for Short-Term Wind Speed Prediction Based on Wind Probability Distribution Function and Machine Learning Models
by Rabin Dhakal, Ashish Sedai, Suhas Pol, Siva Parameswaran, Ali Nejat and Hanna Moussa
Appl. Sci. 2022, 12(18), 9038; https://doi.org/10.3390/app12189038 - 8 Sep 2022
Cited by 14 | Viewed by 2402
Abstract
The need to deliver accurate predictions of renewable energy generation has long been recognized by stakeholders in the field and has propelled recent improvements in more precise wind speed prediction (WSP) methods. Models such as Weibull-probability-density-based WSP (WEB), Rayleigh-probability-density-based WSP (RYM), autoregressive integrated [...] Read more.
The need to deliver accurate predictions of renewable energy generation has long been recognized by stakeholders in the field and has propelled recent improvements in more precise wind speed prediction (WSP) methods. Models such as Weibull-probability-density-based WSP (WEB), Rayleigh-probability-density-based WSP (RYM), autoregressive integrated moving average (ARIMA), Kalman filter and support vector machines (SVR), artificial neural network (ANN), and hybrid models have been used for accurate prediction of wind speed with various forecast horizons. This study intends to incorporate all these methods to achieve a higher WSP accuracy as, thus far, hybrid wind speed predictions are mainly made by using multivariate time series data. To do so, an error correction algorithm for the probability-density-based wind speed prediction model is introduced. Moreover, a comparative analysis of the performance of each method for accurately predicting wind speed for each time step of short-term forecast horizons is performed. All the models studied are used to form the prediction model by optimizing the weight function for each time step of a forecast horizon for each model that contributed to forming the proposed hybrid prediction model. The National Oceanic and Atmospheric Administration (NOAA) and System Advisory Module (SAM) databases were used to demonstrate the accuracy of the proposed models and conduct a comparative analysis. The results of the study show the significant improvement on the performance of wind speed prediction models through the development of a proposed hybrid prediction model. Full article
(This article belongs to the Special Issue Very Short/Short/Medium/Long Term Load Forecasting and Renewables Forecasting)
►▼ Show Figures

Figure 1

15 pages, 3237 KiB  
Article
Machine Learning and Deep Learning Models Applied to Photovoltaic Production Forecasting
by Moisés Cordeiro-Costas, Daniel Villanueva, Pablo Eguía-Oller and Enrique Granada-Álvarez
Appl. Sci. 2022, 12(17), 8769; https://doi.org/10.3390/app12178769 - 31 Aug 2022
Cited by 15 | Viewed by 2391
Abstract
The increasing trend in energy demand is higher than the one from renewable generation, in the coming years. One of the greatest sources of consumption are buildings. The energy management of a building by means of the production of photovoltaic energy in situ [...] Read more.
The increasing trend in energy demand is higher than the one from renewable generation, in the coming years. One of the greatest sources of consumption are buildings. The energy management of a building by means of the production of photovoltaic energy in situ is a common alternative to improve sustainability in this sector. An efficient trade-off of the photovoltaic source in the fields of Zero Energy Buildings (ZEB), nearly Zero Energy Buildings (nZEB) or MicroGrids (MG) requires an accurate forecast of photovoltaic production. These systems constantly generate data that are not used. Artificial Intelligence methods can take advantage of this missing information and provide accurate forecasts in real time. Thus, in this manuscript a comparative analysis is carried out to determine the most appropriate Artificial Intelligence methods to forecast photovoltaic production in buildings. On the one hand, the Machine Learning methods considered are Random Forest (RF), Extreme Gradient Boost (XGBoost), and Support Vector Regressor (SVR). On the other hand, Deep Learning techniques used are Standard Neural Network (SNN), Recurrent Neural Network (RNN), and Convolutional Neural Network (CNN). The models are checked with data from a real building. The models are validated using normalized Mean Bias Error (nMBE), normalized Root Mean Squared Error (nRMSE), and the coefficient of variation (R2). Standard deviation is also used in conjunction with these metrics. The results show that the models forecast the test set with errors of less than 2.00% (nMBE) and 7.50% (nRMSE) in the case of considering nights, and 4.00% (nMBE) and 11.50% (nRMSE) if nights are not considered. In both situations, the R2 is greater than 0.85 in all models. Full article
(This article belongs to the Collection Improvements in the Production, Monitoring, Management and Impact on the Grid of Photovoltaic Installations)
►▼ Show Figures

Figure 1

19 pages, 3118 KiB  
Article
Zero-Shot Emotion Detection for Semi-Supervised Sentiment Analysis Using Sentence Transformers and Ensemble Learning
by Senait Gebremichael Tesfagergish, Jurgita Kapočiūtė-Dzikienė and Robertas Damaševičius
Appl. Sci. 2022, 12(17), 8662; https://doi.org/10.3390/app12178662 - 29 Aug 2022
Cited by 34 | Viewed by 5400
Abstract
We live in a digitized era where our daily life depends on using online resources. Businesses consider the opinions of their customers, while people rely on the reviews/comments of other users before buying specific products or services. These reviews/comments are usually provided in [...] Read more.
We live in a digitized era where our daily life depends on using online resources. Businesses consider the opinions of their customers, while people rely on the reviews/comments of other users before buying specific products or services. These reviews/comments are usually provided in the non-normative natural language within different contexts and domains (in social media, forums, news, blogs, etc.). Sentiment classification plays an important role in analyzing such texts collected from users by assigning positive, negative, and sometimes neutral sentiment values to each of them. Moreover, these texts typically contain many expressed or hidden emotions (such as happiness, sadness, etc.) that could contribute significantly to identifying sentiments. We address the emotion detection problem as part of the sentiment analysis task and propose a two-stage emotion detection methodology. The first stage is the unsupervised zero-shot learning model based on a sentence transformer returning the probabilities for subsets of 34 emotions (anger, sadness, disgust, fear, joy, happiness, admiration, affection, anguish, caution, confusion, desire, disappointment, attraction, envy, excitement, grief, hope, horror, joy, love, loneliness, pleasure, fear, generosity, rage, relief, satisfaction, sorrow, wonder, sympathy, shame, terror, and panic). The output of the zero-shot model is used as an input for the second stage, which trains the machine learning classifier on the sentiment labels in a supervised manner using ensemble learning. The proposed hybrid semi-supervised method achieves the highest accuracy of 87.3% on the English SemEval 2017 dataset. Full article
(This article belongs to the Special Issue Natural Language Processing (NLP) and Applications)
►▼ Show Figures

Figure 1

21 pages, 3152 KiB  
Article
Malware Detection Using Memory Analysis Data in Big Data Environment
by Murat Dener, Gökçe Ok and Abdullah Orman
Appl. Sci. 2022, 12(17), 8604; https://doi.org/10.3390/app12178604 - 27 Aug 2022
Cited by 35 | Viewed by 6948
Abstract
Malware is a significant threat that has grown with the spread of technology. This makes detecting malware a critical issue. Static and dynamic methods are widely used in the detection of malware. However, traditional static and dynamic malware detection methods may fall short [...] Read more.
Malware is a significant threat that has grown with the spread of technology. This makes detecting malware a critical issue. Static and dynamic methods are widely used in the detection of malware. However, traditional static and dynamic malware detection methods may fall short in advanced malware detection. Data obtained through memory analysis can provide important insights into the behavior and patterns of malware. This is because malwares leave various traces on memories. For this reason, the memory analysis method is one of the issues that should be studied in malware detection. In this study, the use of memory data in malware detection is suggested. Malware detection was carried out by using various deep learning and machine learning approaches in a big data environment with memory data. This study was carried out with Pyspark on Apache Spark big data platform in Google Colaboratory. Experiments were performed on the balanced CIC-MalMem-2022 dataset. Binary classification was made using Random Forest, Decision Tree, Gradient Boosted Tree, Logistic Regression, Naive Bayes, Linear Vector Support Machine, Multilayer Perceptron, Deep Feed Forward Neural Network, and Long Short-Term Memory algorithms. The performances of the algorithms used have been compared. The results were evaluated using the Accuracy, F1-score, Precision, Recall, and AUC performance metrics. As a result, the most successful malware detection was obtained with the Logistic Regression algorithm, with an accuracy level of 99.97% in malware detection by memory analysis. Gradient Boosted Tree follows the Logistic Regression algorithm with 99.94% accuracy. The Naive Bayes algorithm showed the lowest performance in malware analysis with memory data, with an accuracy of 98.41%. In addition, many of the algorithms used have achieved very successful results. According to the results obtained, the data obtained from memory analysis is very useful in detecting malware. In addition, deep learning and machine learning approaches were trained with memory datasets and achieved very successful results in malware detection. Full article
(This article belongs to the Section Computing and Artificial Intelligence)
►▼ Show Figures

Figure 1

19 pages, 1657 KiB  
Review
VR Games in Cultural Heritage: A Systematic Review of the Emerging Fields of Virtual Reality and Culture Games
by Anastasios Theodoropoulos and Angeliki Antoniou
Appl. Sci. 2022, 12(17), 8476; https://doi.org/10.3390/app12178476 - 25 Aug 2022
Cited by 39 | Viewed by 8375
Abstract
In recent years, the use of VR games in cultural heritage has been growing. VR Games have increasingly found their way into museums and exhibitions, highlighting the increasing cultural value associated with games and the institutionalization of game culture. In particular, serious VR [...] Read more.
In recent years, the use of VR games in cultural heritage has been growing. VR Games have increasingly found their way into museums and exhibitions, highlighting the increasing cultural value associated with games and the institutionalization of game culture. In particular, serious VR games have a variety of benefits for educational purposes. There are several studies that deployed VR games to improve visitor experiences in several contexts. However, there are not sufficient studies in the field that examine the benefits and drawbacks of VR gaming. This lack of classification studies is regarded as an obstacle to developing more effective games and proposing guidance on the best way of using them in cultural heritage. This review aims to analyze how VR games are used in cultural heritage settings, to explore the evolution and opportunities of this emerging field, the challenges and tensions these innovations present, and to collectively advance this work to benefit visitor experiences. Full article
(This article belongs to the Special Issue Advanced Technologies in Digitizing Cultural Heritage)
►▼ Show Figures

Figure 1

16 pages, 838 KiB  
Article
Using Chatbots as AI Conversational Partners in Language Learning
by Jose Belda-Medina and José Ramón Calvo-Ferrer
Appl. Sci. 2022, 12(17), 8427; https://doi.org/10.3390/app12178427 - 24 Aug 2022
Cited by 55 | Viewed by 18607
Abstract
Recent advances in Artificial Intelligence (AI) and machine learning have paved the way for the increasing adoption of chatbots in language learning. Research published to date has mostly focused on chatbot accuracy and chatbot–human communication from students’ or in-service teachers’ perspectives. This study [...] Read more.
Recent advances in Artificial Intelligence (AI) and machine learning have paved the way for the increasing adoption of chatbots in language learning. Research published to date has mostly focused on chatbot accuracy and chatbot–human communication from students’ or in-service teachers’ perspectives. This study aims to examine the knowledge, level of satisfaction and perceptions concerning the integration of conversational AI in language learning among future educators. In this mixed method research based on convenience sampling, 176 undergraduates from two educational settings, Spain (n = 115) and Poland (n = 61), interacted autonomously with three conversational agents (Replika, Kuki, Wysa) over a four-week period. A learning module about Artificial Intelligence and language learning was specifically designed for this research, including an ad hoc model named the Chatbot–Human Interaction Satisfaction Model (CHISM), which was used by teacher candidates to evaluate different linguistic and technological features of the three conversational agents. Quantitative and qualitative data were gathered through a pre-post-survey based on the CHISM and the TAM2 (technology acceptance) models and a template analysis (TA), and analyzed through IBM SPSS 22 and QDA Miner software. The analysis yielded positive results regarding perceptions concerning the integration of conversational agents in language learning, particularly in relation to perceived ease of use (PeU) and attitudes (AT), but the scores for behavioral intention (BI) were more moderate. The findings also unveiled some gender-related differences regarding participants’ satisfaction with chatbot design and topics of interaction. Full article
(This article belongs to the Special Issue Technologies and Environments of Intelligent Education)
►▼ Show Figures

Figure 1

22 pages, 1132 KiB  
Review
Where Are Smart Cities Heading? A Meta-Review and Guidelines for Future Research
by João Reis, Pedro Alexandre Marques and Pedro Carmona Marques
Appl. Sci. 2022, 12(16), 8328; https://doi.org/10.3390/app12168328 - 20 Aug 2022
Cited by 26 | Viewed by 3989
Abstract
(1) Background: Smart cities have been gaining attention in the community, both among researchers and professionals. Although this field of study is gaining some maturity, no academic manuscript yet offers a unique holistic view of the phenomenon. In fact, the existing systematic reviews [...] Read more.
(1) Background: Smart cities have been gaining attention in the community, both among researchers and professionals. Although this field of study is gaining some maturity, no academic manuscript yet offers a unique holistic view of the phenomenon. In fact, the existing systematic reviews make it possible to gather solid and relevant knowledge, but still dispersed; (2) Method: through a meta-review it was possible to provide a set of data, which allows the dissemination of the main theoretical and managerial contributions to enthusiasts and critics of the area; (3) Results: this research identified the most relevant topics for smart cities, namely, smart city dimensions, digital transformation, sustainability and resilience. In addition, this research emphasizes that the natural sciences have dominated scientific production, with greater attention being paid to megacities of developed nations. Recent empirical research also suggests that it is crucial to overcome key cybersecurity and privacy challenges in smart cities; (4) Conclusions: research on smart cities can be performed as multidisciplinary studies of small and medium-sized cities in developed or underdeveloped countries. Furthermore, future research should highlight the role played by cybersecurity in the development of smart cities and analyze the impact of smart city development on the link between the city and its stakeholders. Full article
(This article belongs to the Special Issue Sustainable Smart Cities: Building New Services and Products with Digital Technologies)
►▼ Show Figures

Figure 1

41 pages, 4861 KiB  
Review
Analysis of Technologies for Carbon Dioxide Capture from the Air
by Grazia Leonzio, Paul S. Fennell and Nilay Shah
Appl. Sci. 2022, 12(16), 8321; https://doi.org/10.3390/app12168321 - 19 Aug 2022
Cited by 22 | Viewed by 6761
Abstract
The increase in CO2 concentration in the atmosphere has prompted the research community to find solutions for this environmental problem, which causes climate change and global warming. CO2 removal through the use of negative emissions technologies could lead to global emission [...] Read more.
The increase in CO2 concentration in the atmosphere has prompted the research community to find solutions for this environmental problem, which causes climate change and global warming. CO2 removal through the use of negative emissions technologies could lead to global emission levels becoming net negative towards the end of this century. Among these negative emissions technologies, direct air capture (DAC), in which CO2 is captured directly from the atmosphere, could play an important role. The captured CO2 can be removed in the long term and through its storage can be used for chemical processes, allowing closed carbon cycles in the short term. For DAC, different technologies have been suggested in the literature, and an overview of these is proposed in this work. Absorption and adsorption are the most studied and mature technologies, but others are also under investigation. An analysis of the main key performance indicators is also presented here and it is suggested that more efforts should be made to develop DAC at a large scale by reducing costs and improving efficiency. An additional discussion, addressing the social concern, is indicated as well. Full article
(This article belongs to the Special Issue Advances in Carbon Dioxide Removal Technologies)
►▼ Show Figures

Figure 1

13 pages, 2730 KiB  
Article
2D/3D Multimode Medical Image Alignment Based on Spatial Histograms
by Yuxi Ban, Yang Wang, Shan Liu, Bo Yang, Mingzhe Liu, Lirong Yin and Wenfeng Zheng
Appl. Sci. 2022, 12(16), 8261; https://doi.org/10.3390/app12168261 - 18 Aug 2022
Cited by 65 | Viewed by 3301
Abstract
The key to image-guided surgery (IGS) technology is to find the transformation relationship between preoperative 3D images and intraoperative 2D images, namely, 2D/3D image registration. A feature-based 2D/3D medical image registration algorithm is investigated in this study. We use a two-dimensional weighted spatial [...] Read more.
The key to image-guided surgery (IGS) technology is to find the transformation relationship between preoperative 3D images and intraoperative 2D images, namely, 2D/3D image registration. A feature-based 2D/3D medical image registration algorithm is investigated in this study. We use a two-dimensional weighted spatial histogram of gradient directions to extract statistical features, overcome the algorithm’s limitations, and expand the applicable scenarios under the premise of ensuring accuracy. The proposed algorithm was tested on CT and synthetic X-ray images, and compared with existing algorithms. The results show that the proposed algorithm can improve accuracy and efficiency, and reduce the initial value’s sensitivity. Full article
(This article belongs to the Special Issue Advances in Artificial Intelligence for Perception Augmentation and Reasoning)
►▼ Show Figures

Figure 1

40 pages, 42877 KiB  
Review
Metal–Organic Frameworks as Powerful Heterogeneous Catalysts in Advanced Oxidation Processes for Wastewater Treatment
by Antía Fdez-Sanromán, Emilio Rosales, Marta Pazos and Angeles Sanroman
Appl. Sci. 2022, 12(16), 8240; https://doi.org/10.3390/app12168240 - 17 Aug 2022
Cited by 11 | Viewed by 4776
Abstract
Nowadays, the contamination of wastewater by organic persistent pollutants is a reality. These pollutants are difficult to remove from wastewater with conventional techniques; hence, it is necessary to go on the hunt for new, innovative and environmentally sustainable ones. In this context, advanced [...] Read more.
Nowadays, the contamination of wastewater by organic persistent pollutants is a reality. These pollutants are difficult to remove from wastewater with conventional techniques; hence, it is necessary to go on the hunt for new, innovative and environmentally sustainable ones. In this context, advanced oxidation processes have attracted great attention and have developed rapidly in recent years as promising technologies. The cornerstone of advanced oxidation processes is the selection of heterogeneous catalysts. In this sense, the possibility of using metal–organic frameworks as catalysts has been opened up given their countless physical–chemical characteristics, which can overcome several disadvantages of traditional catalysts. Thus, this review provides a brief review of recent progress in the research and practical application of metal–organic frameworks to advanced oxidation processes, with a special emphasis on the potential of Fe-based metal–organic frameworks to reduce the pollutants present in wastewater or to render them harmless. To do that, the work starts with a brief overview of the different types and pathways of synthesis. Moreover, the mechanisms of the generation of radicals, as well as their action on the organic pollutants and stability, are analysed. Finally, the challenges of this technology to open up new avenues of wastewater treatment in the future are sketched out. Full article
(This article belongs to the Section Environmental Sciences)
►▼ Show Figures

Figure 1

Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying article 1-50 on page 1 of 6.

Go to page    1 2 3 4 5 chevron_right last_page
Back to TopTop