Next Issue
Volume 12, July
Previous Issue
Volume 12, May
 
 

Computers, Volume 12, Issue 6 (June 2023) – 17 articles

Cover Story (view full-size image): This paper addresses the complex task of introducing programming to secondary school students for the first time to improve the students’ computational thinking (CT) skills by using a visual execution environment (VEE). This introduced concepts that should be covered in any introductory course, namely: variables, input and output, conditionals, loops, arrays, functions, and files. This study explores two approaches to achieve this goal: visual programming (Scratch) and text programming (Java). Additionally, it proposes an AI recommendation model into the VEE to further improve the effectiveness of developing CT among secondary education students. This integrated model combines the capabilities of an AI learning system module and a personalized learning module. View this paper
  • Issues are regarded as officially published after their release is announced to the table of contents alert mailing list.
  • You may sign up for e-mail alerts to receive table of contents of newly released issues.
  • PDF is the official format for papers published in both, html and pdf forms. To view the papers in pdf format, click on the "PDF Full-text" link, and use the free Adobe Reader to open them.
Order results
Result details
Section
Select all
Export citation of selected articles as:
21 pages, 2284 KiB  
Article
Arabic Sentiment Analysis Based on Word Embeddings and Deep Learning
by Nasrin Elhassan, Giuseppe Varone, Rami Ahmed, Mandar Gogate, Kia Dashtipour, Hani Almoamari, Mohammed A. El-Affendi, Bassam Naji Al-Tamimi, Faisal Albalwy and Amir Hussain
Computers 2023, 12(6), 126; https://doi.org/10.3390/computers12060126 - 19 Jun 2023
Cited by 9 | Viewed by 3316
Abstract
Social media networks have grown exponentially over the last two decades, providing the opportunity for users of the internet to communicate and exchange ideas on a variety of topics. The outcome is that opinion mining plays a crucial role in analyzing user opinions [...] Read more.
Social media networks have grown exponentially over the last two decades, providing the opportunity for users of the internet to communicate and exchange ideas on a variety of topics. The outcome is that opinion mining plays a crucial role in analyzing user opinions and applying these to guide choices, making it one of the most popular areas of research in the field of natural language processing. Despite the fact that several languages, including English, have been the subjects of several studies, not much has been conducted in the area of the Arabic language. The morphological complexities and various dialects of the language make semantic analysis particularly challenging. Moreover, the lack of accurate pre-processing tools and limited resources are constraining factors. This novel study was motivated by the accomplishments of deep learning algorithms and word embeddings in the field of English sentiment analysis. Extensive experiments were conducted based on supervised machine learning in which word embeddings were exploited to determine the sentiment of Arabic reviews. Three deep learning algorithms, convolutional neural networks (CNNs), long short-term memory (LSTM), and a hybrid CNN-LSTM, were introduced. The models used features learned by word embeddings such as Word2Vec and fastText rather than hand-crafted features. The models were tested using two benchmark Arabic datasets: Hotel Arabic Reviews Dataset (HARD) for hotel reviews and Large-Scale Arabic Book Reviews (LARB) for book reviews, with different setups. Comparative experiments utilized the three models with two-word embeddings and different setups of the datasets. The main novelty of this study is to explore the effectiveness of using various word embeddings and different setups of benchmark datasets relating to balance, imbalance, and binary and multi-classification aspects. Findings showed that the best results were obtained in most cases when applying the fastText word embedding using the HARD 2-imbalance dataset for all three proposed models: CNN, LSTM, and CNN-LSTM. Further, the proposed CNN model outperformed the LSTM and CNN-LSTM models for the benchmark HARD dataset by achieving 94.69%, 94.63%, and 94.54% accuracy with fastText, respectively. Although the worst results were obtained for the LABR 3-imbalance dataset using both Word2Vec and FastText, they still outperformed other researchers’ state-of-the-art outcomes applying the same dataset. Full article
Show Figures

Figure 1

12 pages, 2498 KiB  
Article
An Analysis of Neighbor Discovery Protocol Attacks
by Firas Najjar, Qusay Bsoul and Hasan Al-Refai
Computers 2023, 12(6), 125; https://doi.org/10.3390/computers12060125 - 19 Jun 2023
Cited by 3 | Viewed by 4741
Abstract
Neighbor Discovery Protocol (NDP) is a network protocol used in IPv6 networks to manage communication between neighboring devices. NDP is responsible for mapping IPv6 addresses to MAC addresses and discovering the availability of neighboring devices on the network. The main risk of deploying [...] Read more.
Neighbor Discovery Protocol (NDP) is a network protocol used in IPv6 networks to manage communication between neighboring devices. NDP is responsible for mapping IPv6 addresses to MAC addresses and discovering the availability of neighboring devices on the network. The main risk of deploying NDP on public networks is the potential for hackers or attackers to launch various types of attacks, such as address spoofing attacks, denial-of-service attacks, and man-in-the-middle attacks. Although Secure Neighbor Discovery (SEND) is implemented to secure NDP, its complexity and cost hinder its widespread deployment. This research emphasizes the potential hazard of deploying IPv6 networks in public spaces, such as airports, without protecting NDP messages. These risks have the potential to crash the entire local network. To demonstrate these risks, the GNS3 testbed environment is used to generate NDP attacks and capture the resulting packets using Wireshark for analysis. The analysis results reveal that with just a few commands, attackers can execute various NDP attacks. This highlights the need to protect against the potential issues that come with deploying IPv6 on widely accessible public networks. In addition, the analysis result shows that NDP attacks have behavior that can be used to define various NDP attacks. Full article
(This article belongs to the Section ICT Infrastructures for Cybersecurity)
Show Figures

Figure 1

22 pages, 1104 KiB  
Article
Exploring Clustering Techniques for Analyzing User Engagement Patterns in Twitter Data
by Andreas Kanavos, Ioannis Karamitsos and Alaa Mohasseb
Computers 2023, 12(6), 124; https://doi.org/10.3390/computers12060124 - 19 Jun 2023
Cited by 7 | Viewed by 2732
Abstract
Social media platforms have revolutionized information exchange and socialization in today’s world. Twitter, as one of the prominent platforms, enables users to connect with others and express their opinions. This study focuses on analyzing user engagement levels on Twitter using graph mining and [...] Read more.
Social media platforms have revolutionized information exchange and socialization in today’s world. Twitter, as one of the prominent platforms, enables users to connect with others and express their opinions. This study focuses on analyzing user engagement levels on Twitter using graph mining and clustering techniques. We measure user engagement based on various tweet attributes, including retweets, replies, and more. Specifically, we explore the strength of user connections in Twitter networks by examining the diversity of edges. Our approach incorporates graph mining models that assign different weights to evaluate the significance of each connection. Additionally, clustering techniques are employed to group users based on their engagement patterns and behaviors. Statistical analysis was conducted to assess the similarity between user profiles, as well as attributes, such as friendship, followings, and interactions within the Twitter social network. The findings highlight the discovery of closely linked user groups and the identification of distinct clusters based on engagement levels. This research emphasizes the importance of understanding both individual and group behaviors in comprehending user engagement dynamics on Twitter. Full article
Show Figures

Figure 1

17 pages, 1144 KiB  
Article
Optimizing Water Distribution through Explainable AI and Rule-Based Control
by Enrico Ferrari, Damiano Verda, Nicolò Pinna and Marco Muselli
Computers 2023, 12(6), 123; https://doi.org/10.3390/computers12060123 - 18 Jun 2023
Cited by 3 | Viewed by 1780
Abstract
Optimizing water distribution both from an energy-saving perspective and from a quality of service perspective is a challenging task since it involves a complex system with many nodes, many hidden variables and many operational constraints. For this reason, water distribution systems need to [...] Read more.
Optimizing water distribution both from an energy-saving perspective and from a quality of service perspective is a challenging task since it involves a complex system with many nodes, many hidden variables and many operational constraints. For this reason, water distribution systems need to handle a delicate trade-off between the effectiveness and computational time of the solution. In this paper, we propose a new computationally efficient method, named rule-based control, to optimize water distribution networks without the need for a rigorous formulation of the optimization problem. As a matter of fact, since it is based on a machine learning approach, the proposed method employs only a set of historical data, where the configuration can be labeled according to a quality criterion. Since it is a data-driven approach, it could be applied to any complex network where historical labeled data are available. In particular, rule-based control exploits a rule-based classification method that allows us to retrieve the rules leading to good or bad performances of the system, even without any information about its physical laws. The evaluation of the results on some simulated scenarios shows that the proposed approach is able to reduce energy consumption while ensuring a good quality of the service. The proposed approach is currently used in the water distribution system of the Milan (Italy) water main. Full article
(This article belongs to the Special Issue System-Integrated Intelligence and Intelligent Systems 2023)
Show Figures

Figure 1

22 pages, 2266 KiB  
Article
A Learning Framework for Supporting Digital Innovation Hubs
by Joao Sarraipa, Majid Zamiri, Elsa Marcelino-Jesus, Andreia Artifice, Ricardo Jardim-Goncalves and Néjib Moalla
Computers 2023, 12(6), 122; https://doi.org/10.3390/computers12060122 - 15 Jun 2023
Cited by 3 | Viewed by 2160
Abstract
With the increasing demand for digital transformation and (digital) technology transfer (TT), digital innovation hubs (DIHs) are the new piece of the puzzle of our economy and industries’ landscapes. Evidence shows that DIHs can provide good opportunities to access needed innovations, technologies, and [...] Read more.
With the increasing demand for digital transformation and (digital) technology transfer (TT), digital innovation hubs (DIHs) are the new piece of the puzzle of our economy and industries’ landscapes. Evidence shows that DIHs can provide good opportunities to access needed innovations, technologies, and resources at a higher level than other organizations that can normally access them. However, it is critically important to note that DIHs are still evolving, under research, and under development. That is, there are many substantial aspects of DIHs that should be considered. For example, DIHs must cater to a wide spectrum of needs for TT. From this perspective, the contribution of this work is proposing a generic and flexible learning framework, aiming to assist DIHs in providing suitable education, training, and learning services that support the process of (digital) TT to companies. The proposed learning framework was designed, evaluated, and improved with the support of two EU projects, and these processes are discussed in brief. The primary and leading results gained in this way show that the learning framework has immense potential for application to similar cases, and it can facilitate and expedite the process of TT to companies. The study is concluded with some directions for future works. Full article
Show Figures

Figure 1

16 pages, 4539 KiB  
Article
Proactive Ransomware Detection Using Extremely Fast Decision Tree (EFDT) Algorithm: A Case Study
by Ibrahim Ba’abbad and Omar Batarfi
Computers 2023, 12(6), 121; https://doi.org/10.3390/computers12060121 - 15 Jun 2023
Cited by 5 | Viewed by 2302
Abstract
Several malware variants have attacked systems and data over time. Ransomware is among the most harmful malware since it causes huge losses. In order to get a ransom, ransomware is software that locks the victim’s machine or encrypts his personal information. Numerous research [...] Read more.
Several malware variants have attacked systems and data over time. Ransomware is among the most harmful malware since it causes huge losses. In order to get a ransom, ransomware is software that locks the victim’s machine or encrypts his personal information. Numerous research has been conducted to stop and quickly recognize ransomware attacks. For proactive forecasting, artificial intelligence (AI) techniques are used. Traditional machine learning/deep learning (ML/DL) techniques, however, take a lot of time and decrease the accuracy and latency performance of network monitoring. In this study, we utilized the Hoeffding trees classifier as one of the stream data mining classification techniques to detect and prevent ransomware attacks. Three Hoeffding trees classifier algorithms are selected to be applied to the Resilient Information Systems Security (RISS) research group dataset. After configuration, Massive Online Analysis (MOA) software is utilized as a testing framework. The results of Hoeffding tree classifier algorithms are then assessed to choose the enhanced model with the highest accuracy and latency performance. In conclusion, the 99.41% classification accuracy was the highest result achieved by the EFDT algorithm in 66 ms. Full article
Show Figures

Figure 1

22 pages, 9858 KiB  
Article
High-Capacity Reversible Data Hiding Based on Two-Layer Embedding Scheme for Encrypted Image Using Blockchain
by Arun Kumar Rai, Hari Om, Satish Chand and Chia-Chen Lin
Computers 2023, 12(6), 120; https://doi.org/10.3390/computers12060120 - 12 Jun 2023
Cited by 3 | Viewed by 1555
Abstract
In today’s digital age, ensuring the secure transmission of confidential data through various means of communication is crucial. Protecting the data from malicious attacks during transmission poses a significant challenge. To achieve this, reversible data hiding (RDH) and encryption methods are often used [...] Read more.
In today’s digital age, ensuring the secure transmission of confidential data through various means of communication is crucial. Protecting the data from malicious attacks during transmission poses a significant challenge. To achieve this, reversible data hiding (RDH) and encryption methods are often used in combination to safeguard confidential data from intruders. However, existing secure reversible hybrid hiding techniques are facing challenges related to low data embedding capacity. To address these challenges, the proposed research presents a solution that utilizes block-wise encryption and a two-layer embedding scheme to enhance the embedding capacity of the cover image. Additionally, this technique incorporates a blockchain-enabled RDH method to ensure traceability and integrity by storing confidential data alongside the hash value of the stego image. The proposed work is divided into three phases. First, the cover image is encrypted. Second, the data are embedded in the encrypted cover image using a two-layer embedding scheme. Finally, the stego image along with the hash value are deployed through blockchain technology. The proposed method reduces challenges associated with traceability and integrity while increasing the embedding capacity of images compared to traditional methods. Full article
Show Figures

Figure 1

17 pages, 428 KiB  
Article
A Query Expansion Benchmark on Social Media Information Retrieval: Which Methodology Performs Best and Aligns with Semantics?
by Evangelos A. Stathopoulos, Anastasios I. Karageorgiadis, Alexandros Kokkalas, Sotiris Diplaris, Stefanos Vrochidis and Ioannis Kompatsiaris
Computers 2023, 12(6), 119; https://doi.org/10.3390/computers12060119 - 10 Jun 2023
Cited by 1 | Viewed by 1898
Abstract
This paper presents a benchmarking survey on query expansion techniques for social media information retrieval, with a focus on comparing the performance of methods using semantic web technologies. The study evaluated query expansion techniques such as generative AI models and semantic matching algorithms [...] Read more.
This paper presents a benchmarking survey on query expansion techniques for social media information retrieval, with a focus on comparing the performance of methods using semantic web technologies. The study evaluated query expansion techniques such as generative AI models and semantic matching algorithms and how they are integrated in a semantic framework. The evaluation was based on cosine similarity metrics, including the Discounted Cumulative Gain (DCG), Ideal Discounted Cumulative Gain (IDCG), and normalized Discounted Cumulative Gain (nDCG), as well as the Mean Average Precision (MAP). Additionally, the paper discusses the use of semantic web technologies as a component in a pipeline for building thematic knowledge graphs from retrieved social media data with extended ontologies integrated for the refugee crisis. The paper begins by introducing the importance of query expansion in information retrieval and the potential benefits of incorporating semantic web technologies. The study then presents the methodologies and outlines the specific procedures for each query expansion technique. The results of the evaluation are presented, as well as the rest semantic framework, and the best-performing technique was identified, which was the curie-001 generative AI model. Finally, the paper summarizes the main findings and suggests future research directions. Full article
Show Figures

Figure 1

30 pages, 667 KiB  
Article
Unbalanced Web Phishing Classification through Deep Reinforcement Learning
by Antonio Maci, Alessandro Santorsola, Antonio Coscia and Andrea Iannacone
Computers 2023, 12(6), 118; https://doi.org/10.3390/computers12060118 - 9 Jun 2023
Cited by 15 | Viewed by 2658
Abstract
Web phishing is a form of cybercrime aimed at tricking people into visiting malicious URLs to exfiltrate sensitive data. Since the structure of a malicious URL evolves over time, phishing detection mechanisms that can adapt to such variations are paramount. Furthermore, web phishing [...] Read more.
Web phishing is a form of cybercrime aimed at tricking people into visiting malicious URLs to exfiltrate sensitive data. Since the structure of a malicious URL evolves over time, phishing detection mechanisms that can adapt to such variations are paramount. Furthermore, web phishing detection is an unbalanced classification task, as legitimate URLs outnumber malicious ones in real-life cases. Deep learning (DL) has emerged as a promising technique to minimize concept drift to enhance web phishing detection. Deep reinforcement learning (DRL) combines DL with reinforcement learning (RL); that is, a sequential decision-making paradigm in which the problem to be addressed is expressed as a Markov decision process (MDP). Recent studies have proposed an ad hoc MDP formulation to tackle unbalanced classification tasks called the imbalanced classification Markov decision process (ICMDP). In this paper, we exploit the ICMDP to present a double deep Q-Network (DDQN)-based classifier to address the unbalanced web phishing classification problem. The proposed algorithm is evaluated on a Mendeley web phishing dataset, from which three different data imbalance scenarios are generated. Despite a significant training time, it results in better geometric mean, index of balanced accuracy, F1 score, and area under the ROC curve than other DL-based classifiers combined with data-level sampling techniques in all test cases. Full article
(This article belongs to the Special Issue Using New Technologies on Cyber Security Solutions)
Show Figures

Figure 1

19 pages, 3655 KiB  
Article
Combining MAS-GiG Model and Related Problems to Optimization in Emergency Evacuation
by Dinh Thi Hong Huyen, Hoang Thi Thanh Ha and Michel Occello
Computers 2023, 12(6), 117; https://doi.org/10.3390/computers12060117 - 2 Jun 2023
Viewed by 1270
Abstract
Emergency evacuation is of paramount importance in protecting human lives and property while enhancing the effectiveness and preparedness of organizations and management agencies in responding to emergencies. In this paper, we propose a method for evacuating passengers to safe places with the shortest [...] Read more.
Emergency evacuation is of paramount importance in protecting human lives and property while enhancing the effectiveness and preparedness of organizations and management agencies in responding to emergencies. In this paper, we propose a method for evacuating passengers to safe places with the shortest possible evacuation time. The proposed method is based on a multi-level multi-agent MAS-GiG model combined with three problems. First, constructing a path map to select the shortest path; second, dividing the space of the experimental environment into smaller areas for efficient management, monitoring, and guiding evacuation; the third, adjusting the speed to handle collision issues and maintain distance between two or more groups of evacuees while moving. We extend our previous study by establishing groups based on the location of passengers and using a MAS-GiG model to guide evacuation. We compare the proposed method with our previous method to provide specific evaluations for the research and research in the future. We tested two methods in the departure hall, first floor, Danang International Airport, Vietnam. Full article
Show Figures

Figure 1

19 pages, 8953 KiB  
Article
The Effects of Individuals’ Opinion and Non-Opinion Characteristics on the Organization of Influence Networks in the Online Domain
by Vladislav N. Gezha and Ivan V. Kozitsin
Computers 2023, 12(6), 116; https://doi.org/10.3390/computers12060116 - 2 Jun 2023
Cited by 4 | Viewed by 1539
Abstract
The opinion dynamics literature argues that the way people perceive social influence depends not only on the opinions of interacting individuals, but also on the individuals’ non-opinion characteristics, such as age, education, gender, or place of residence. The current paper advances this line [...] Read more.
The opinion dynamics literature argues that the way people perceive social influence depends not only on the opinions of interacting individuals, but also on the individuals’ non-opinion characteristics, such as age, education, gender, or place of residence. The current paper advances this line of research by studying longitudinal data that describe the opinion dynamics of a large sample (~30,000) of online social network users, all citizens of one city. Using these data, we systematically investigate the effects of users’ demographic (age, gender) and structural (degree centrality, the number of common friends) properties on opinion formation processes. We revealed that females are less easily influenced than males. Next, we found that individuals that are characterized by similar ages have more chances to reach a consensus. Additionally, we report that individuals who have many common peers find an agreement more often. We also demonstrated that the impacts of these effects are virtually the same, and despite being statistically significant, are far less strong than that of opinion-related features: knowing the current opinion of an individual and, what is even more important, the distance in opinions between this individual and the person that attempts to influence the individual is much more valuable. Next, after conducting a series of simulations with an agent-based model, we revealed that accounting for non-opinion characteristics may lead to not very sound but statistically significant changes in the macroscopic predictions of the populations of opinion camps, primarily among the agents with radical opinions (≈3% of all votes). In turn, predictions for the populations of neutral individuals are virtually the same. In addition, we demonstrated that the accumulative effect of non-opinion features on opinion dynamics is seriously moderated by whether the underlying social network correlates with the agents’ characteristics. After applying the procedure of random shuffling (in which the agents and their characteristics were randomly scattered over the network), the macroscopic predictions have changed by ≈9% of all votes. What is interesting is that the population of neutral agents was again not affected by this intervention. Full article
(This article belongs to the Special Issue Computational Modeling of Social Processes and Social Networks)
Show Figures

Figure 1

17 pages, 2095 KiB  
Article
Machine Learning-Based Dynamic Attribute Selection Technique for DDoS Attack Classification in IoT Networks
by Subhan Ullah, Zahid Mahmood, Nabeel Ali, Tahir Ahmad and Attaullah Buriro
Computers 2023, 12(6), 115; https://doi.org/10.3390/computers12060115 - 29 May 2023
Cited by 12 | Viewed by 2942
Abstract
The exponential growth of the Internet of Things (IoT) has led to the rapid expansion of interconnected systems, which has also increased the vulnerability of IoT devices to security threats such as distributed denial-of-service (DDoS) attacks. In this paper, we propose a machine [...] Read more.
The exponential growth of the Internet of Things (IoT) has led to the rapid expansion of interconnected systems, which has also increased the vulnerability of IoT devices to security threats such as distributed denial-of-service (DDoS) attacks. In this paper, we propose a machine learning pipeline that specifically addresses the issue of DDoS attack detection in IoT networks. Our approach comprises of (i) a processing module to prepare the data for further analysis, (ii) a dynamic attribute selection module that selects the most adaptive and productive features and reduces the training time, and (iii) a classification module to detect DDoS attacks. We evaluate the effectiveness of our approach using the CICI-IDS-2018 dataset and five powerful yet simple machine learning classifiers—Decision Tree (DT), Gaussian Naive Bayes, Logistic Regression (LR), K-Nearest Neighbor (KNN), and Random Forest (RF). Our results demonstrate that DT outperforms its counterparts and achieves up to 99.98% accuracy in just 0.18 s of CPU time. Our approach is simple, lightweight, and accurate for detecting DDoS attacks in IoT networks. Full article
(This article belongs to the Special Issue Software-Defined Internet of Everything)
Show Figures

Figure 1

11 pages, 567 KiB  
Review
To Wallet or Not to Wallet: The Debate over Digital Health Information Storage
by Jasna Karacic Zanetti and Rui Nunes
Computers 2023, 12(6), 114; https://doi.org/10.3390/computers12060114 - 28 May 2023
Cited by 5 | Viewed by 2383
Abstract
The concept of the health wallet, a digital platform that consolidates health-related information, has garnered significant attention in the past year. Electronic health data storage and transmission have become increasingly prevalent in the healthcare industry, with the potential to revolutionize healthcare delivery. This [...] Read more.
The concept of the health wallet, a digital platform that consolidates health-related information, has garnered significant attention in the past year. Electronic health data storage and transmission have become increasingly prevalent in the healthcare industry, with the potential to revolutionize healthcare delivery. This paper emphasizes the significance of recognizing and addressing the ethical implications of digital health technologies and prioritizes ethical considerations in their development. The adoption of health wallets has theoretical contributions, including the development of personalized medicine through comprehensive data collection, reducing medical errors through consolidated information, and enabling research for the improvement of existing treatments and interventions. Health wallets also empower individuals to manage their own health by providing access to their health data, allowing them to make informed decisions. The findings herein emphasize the importance of informing patients about their rights to control their health data and have access to it while protecting their privacy and confidentiality. This paper stands out by presenting practical recommendations for healthcare organizations and policymakers to ensure the safe and effective implementation of health wallets. Full article
(This article belongs to the Special Issue e-health Pervasive Wireless Applications and Services (e-HPWAS'22))
Show Figures

Figure 1

17 pages, 3810 KiB  
Article
Prototype of a Recommendation Model with Artificial Intelligence for Computational Thinking Improvement of Secondary Education Students
by Raquel Hijón-Neira, Cornelia Connolly, Celeste Pizarro and Diana Pérez-Marín
Computers 2023, 12(6), 113; https://doi.org/10.3390/computers12060113 - 26 May 2023
Cited by 2 | Viewed by 2565
Abstract
There is a growing interest in finding new ways to address the difficult task of introducing programming to secondary students for the first time to improve students’ computational thinking (CT) skills. Therefore, extensive research is required in this field. Worldwide, new ways to [...] Read more.
There is a growing interest in finding new ways to address the difficult task of introducing programming to secondary students for the first time to improve students’ computational thinking (CT) skills. Therefore, extensive research is required in this field. Worldwide, new ways to address this difficult task have been developed: visual execution environments and approaches by text programming or visual programming are among the most popular. This paper addresses the complex task by using a visual execution environment (VEE) to introduce the first programming concepts that should be covered in any introductory programming course. These concepts include variables, input and output, conditionals, loops, arrays, functions, and files. This study explores two approaches to achieve this goal: visual programming (using Scratch) and text programming (using Java) to improve CT. Additionally, it proposes an AI recommendation model into the VEE to further improve the effectiveness of developing CT among secondary education students. This integrated model combines the capabilities of an AI learning system module and a personalized learning module to better address the task at hand. To pursue this task, an experiment has been carried out among 23 preservice secondary teachers’ students in two universities, one in Madrid, Spain, and the other in Galway, Ireland. The overall results showed a significant improvement in the Scratch group. However, when analyzing the results based on specific programming concepts, significance was observed only in the Scratch group, specifically for the Loop concept. Full article
Show Figures

Figure 1

16 pages, 4977 KiB  
Article
Image Denoising by Deep Convolution Based on Sparse Representation
by Shengqin Bian, Xinyu He, Zhengguang Xu and Lixin Zhang
Computers 2023, 12(6), 112; https://doi.org/10.3390/computers12060112 - 24 May 2023
Cited by 2 | Viewed by 2738
Abstract
Noise filtering is a crucial task in digital image processing, performing the function of preprocessing. In this paper, we propose an algorithm that employs deep convolution and soft thresholding iterative algorithms to extract and learn the features of noisy images. The extracted features [...] Read more.
Noise filtering is a crucial task in digital image processing, performing the function of preprocessing. In this paper, we propose an algorithm that employs deep convolution and soft thresholding iterative algorithms to extract and learn the features of noisy images. The extracted features are acquired through prior and sparse representation theory for image reconstruction. Effective separation of the image and noise is achieved using an end-to-end network of dilated convolution and fully connected layers. Several experiments were performed on public images subject to various levels of Gaussian noise, in order to evaluate the effectiveness of the proposed approach. The results indicated that our algorithm achieved a high peak signal-to-noise ratio (PSNR) and significantly improved the visual effects of the images. Our study supports the effectiveness of our approach and substantiates its potential to be applied to a broad spectrum of image processing tasks. Full article
Show Figures

Figure 1

26 pages, 595 KiB  
Article
Vocational Domain Identification with Machine Learning and Natural Language Processing on Wikipedia Text: Error Analysis and Class Balancing
by Maria Nefeli Nikiforos, Konstantina Deliveri, Katia Lida Kermanidis and Adamantia Pateli
Computers 2023, 12(6), 111; https://doi.org/10.3390/computers12060111 - 24 May 2023
Viewed by 1859
Abstract
Highly-skilled migrants and refugees finding employment in low-skill vocations, despite professional qualifications and educational backgrounds, has become a global tendency, mainly due to the language barrier. Employment prospects for displaced communities are mostly decided by their knowledge of the sublanguage of the vocational [...] Read more.
Highly-skilled migrants and refugees finding employment in low-skill vocations, despite professional qualifications and educational backgrounds, has become a global tendency, mainly due to the language barrier. Employment prospects for displaced communities are mostly decided by their knowledge of the sublanguage of the vocational domain they are interested in working. Common vocational domains include agriculture, cooking, crafting, construction, and hospitality. The increasing amount of user-generated content in wikis and social networks provides a valuable source of data for data mining, natural language processing, and machine learning applications. This paper extends the contribution of the authors’ previous research on automatic vocational domain identification by further analyzing the results of machine learning experiments with a domain-specific textual data set while considering two research directions: a. prediction analysis and b. data balancing. Wrong prediction analysis and the features that contributed to misclassification, along with correct prediction analysis and the features that were the most dominant, contributed to the identification of a primary set of terms for the vocational domains. Data balancing techniques were applied on the data set to observe their impact on the performance of the classification model. A novel four-step methodology was proposed in this paper for the first time, which consists of successive applications of SMOTE oversampling on imbalanced data. Data oversampling obtained better results than data undersampling in imbalanced data sets, while hybrid approaches performed reasonably well. Full article
Show Figures

Figure 1

15 pages, 983 KiB  
Article
Persistence Landscapes—Implementing a Dataset Verification Method in Resource-Scarce Embedded Systems
by Sérgio Branco, Ertugrul Dogruluk, João G. Carvalho, Marco S. Reis and Jorge Cabral
Computers 2023, 12(6), 110; https://doi.org/10.3390/computers12060110 - 23 May 2023
Viewed by 1573
Abstract
As more and more devices are being deployed across networks to gather data and use them to perform intelligent tasks, it is vital to have a tool to perform real-time data analysis. Data are the backbone of Machine Learning models, the core of [...] Read more.
As more and more devices are being deployed across networks to gather data and use them to perform intelligent tasks, it is vital to have a tool to perform real-time data analysis. Data are the backbone of Machine Learning models, the core of intelligent systems. Therefore, verifying whether the data being gathered are similar to those used for model building is essential. One fantastic tool for the performance of data analysis is the 0-Dimensional Persistent Diagrams, which can be computed in a Resource-Scarce Embedded System (RSES), a set of memory and processing-constrained devices that are used in many IoT applications because they are cost-effective and reliable. However, it is challenging to compare Persistent Diagrams, and Persistent Landscapes are used because they allow Persistent Diagrams to be passed to a space where the mean concept is well-defined. The following work shows how one can perform a Persistent Landscape analysis in an RSES. It also shows that the distance between two Persistent Landscapes makes it possible to verify whether two devices collect the same data. The main contribution of this work is the implementation of Persistent Landscape analysis in an RSES, which is not provided in the literature. Moreover, it shows that devices can now verify, in real-time, whether they can trust the data being collected to perform the intelligent task they were designed to, which is essential in any system to avoid bugs or errors. Full article
(This article belongs to the Special Issue Edge and Fog Computing for Internet of Things Systems 2023)
Show Figures

Figure 1

Previous Issue
Next Issue
Back to TopTop