Computer Science articles list

Numerous initiatives to rely on new renewable energy sources, such solar electricity, have been sparked by the increased interest in global warming. With an increase in home photovoltaic (PV) panels that are available to the public, more precise calculations of energy generation are now possible. Segmenting satellite images offers a straightforward and inexpensive way to categorize solar panels..This work suggests a method for classifying and segmenting solar panels that combines the watershed algorithm with deep learning approaches. First, a Convolutional Neural Network (CNN) architecture with the ResNet, EfficientNet, and Inception architectures is used for classification. Through the fine-tuning of pre-trained networks on a heterogeneous dataset of solar panels, transfer learning improves performance. The categorization model recognizes solar panels in a variety of settings with accuracy, making maintenance and monitoring easier. After classification, the watershed method uses intensity gradients to precisely delineate solar panels from the background. Tasks like defect detection and layout optimization are made easier when deep learning-based classification and watershed segmentation are combined. The outcomes of the experiments show how well the suggested method performs in terms of segmenting and classifying solar panels under various circumstances. A flexible automated solar panel management solution is provided by the combination of deep learning and the watershed algorithm, which promotes increased sustainability and efficiency in solar energy systems.

BAKKA ARUN KUMAR

Online distance learning policies were formulated and implemented among some Malaysian universities long ago, but their value emerged since COVID- 19. Emanating from the diffusion of innovation theory, this study examined the perception of higher education students on the influence and relationship between six independent variables (compatibility, observability, relative advantage, complexity, trialability, and digital skills) and one dependent variable (digital literacy). A total of 524 respondents were sampled, comprising students from six public and private Malaysian universities. The findings from the correlation analysis show a significant positive relationship between the six independent variables and the dependent variable. Meanwhile, in the regression analysis, three of the independent variables (observability, trialability, and digital skill) have a significant and positive effect on digital literacy. This study placed the diffusion of innovation in a specific context that supports designing online distance learning and digital literacy policies

Mohammed Fadel Arandas

In the last decade, we have observed the usage of artificial intelligence algorithms and machine learning models in industry, education, healthcare, entertainment, and several other areas. In this paper, we focus on using machine learning algorithms in the loan approval process of financial institutions. First, we briefly review some prior research papers that dealt with loan approval predictions using machine learning models. Next, we analyze the loan approval prediction dataset we downloaded from Kaggle, which was used in this paper to compare several machine learning classification models. During this analysis, we observed that credit scores and loan terms are the attributes that probably most affect the result. Next, we divided the dataset into a training set (80%) and a test set (20%). We trained 27 various machine learning models in MATLAB. Three models were optimized with Bayesian optimization to find the best hyperparameters with minimum error. We used 5-fold cross-validation for the validations to prevent overfitting during the training. In the following step, we used the test set on trained models to measure the models’ accuracy on unseen data. The result showed that the best accuracy both on validation and test data, more than 98%, was reached with neural networks and ensemble classification models.

Ladislav Végh

For a long time, cardiovascular diseases have been the leading cause of death worldwide. Machine learning has found significant usage in the medical field as it can find patterns in data. Classification models can help cardiologists to diagnose heart diseases and minimize misdiagnosis accurately. In this paper, we explored a dataset related to heart disease and compared the accuracy of 43 machine learning classification models. The dataset for this research was downloaded from Kaggle; it contained 1190 observations, 11 features (age, sex, chest pain type, resting blood pressure, serum cholesterol, fasting blood sugar, resting electrocardiogram results, maximum heart rate achieved, exercise induced angina, oldpeak, the slope of the peak exercise ST segment) and a binary target variable (no heart disease or observed cardiovascular disease). For data exploration, preprocessing, training, testing, and predictor importance analysis, we used MATLAB R2004a software and the Classification Learner app included in this software. Before training machine learning classification models, we divided the dataset into a training set (90% of observations) and a test set (10% of observations). To prevent overfitting during the training of classification models, 10-fold cross-validation was used. The result showed that the best accuracy was reached with an optimized ensemble classification model (validation accuracy: 0.9262 and test accuracy: 0.9580). After calculating the permutation importance of each feature, we observed that the most important feature among all 11 features was the slope of the peak exercise ST segment.

Ladislav Végh

This paper compares different optimizable machine learning classification models to predict eight types of anemia from complete blood count (CBC) data. For the research, we used a publicly available Kaggle dataset containing 1281 observations, 14 predictors, and the diagnosis as the categorical target variable with nine categories (eight types of anemia and the healthy category). First, we examined the dataset and observed the histograms of some of the predictors. We compared the values of predictors of observations with no anemia to the observations where any anemia was diagnosed. Next, we used MATLAB R2024a to train and test nine optimizable machine-learning classification models. These models were Ensemble, Tree, SVM, Efficient Linear, Neural Network, Kernel, KNN, Naïve Bayes, and the Discriminant. Bayesian optimization was used to optimize the hyperparameters of all these models. We used 90% of observations for training and 10% of observations for testing. During the training, 10-fold cross-validation was used to prevent overfitting. The results showed the best accuracy was reached with the Ensemble classification model using the bag ensemble method (validation accuracy: 99.22%, test accuracy: 100%). Finally, we inspected our best classification model in more detail. We calculated the permutation feature importance to determine the contribution of each predictor to the final model. The results showed 6–7 important predictors, while the most important feature was the amount of hemoglobin.

Ladislav Végh

The “Green AI Revolution” distils a paradigm-shifting methodology for creating machine learning solutions for the design and enhancement of ecologically sustainable communication networks. To address sustainability concerns in communication infrastructures, this study presents a comprehensive architecture that emphasises the integration of machine learning (ML) and artificial intelligence (AI) techniques. With the fitting moniker “Green AI”, the suggested model aims to improve overall resource efficiency in communication networks while minimising energy usage and carbon footprints. The goal of Green AI is to transform conventional communication systems by utilising sophisticated algorithms, dynamic optimisation, and intelligent decision-making techniques. Higher energy efficiency, less of an impact on the environment, and better network performance are the main goals. The present study examines the fundamental elements of the Green AI architecture, encompassing intelligent routing, dynamic power management, and adaptive power distribution of resources. Furthermore, case studies and simulations highlight the real advantages of incorporating machine learning into communication networks, highlighting the technology’s potential to make a substantial contribution to a future that is more environmentally friendly and sustainable. The Green AI Revolution is a paradigm shift in the way we think about and use communication technology. It encourages innovation that is in line with environmental stewardship and technical progress.

Mrutyunjaya S Yalawar

Developing automated systems with a reasonable cost for long-term care for elders is a promising research direction. Such smart systems are based on realizing activities of daily living (ADLs) to enable aging in place while preserving the quality of life of all inhabitants in smart homes. One of the research directions is based on localizing items used by elders to monitor their activities with fine-grained details of the progress. In this paper, we shed the light on this issue by presenting an approach for localizing items in smart homes. The presented method is based on applying machine learning algorithms to Radio Frequency IDentification (RFID) tags readings. Our approach achieves the required task through two stages. The first stage detects in which room the selected object is located. Then, the second one determines the exact position of the selected object inside the detected room. Additionally, we present an efficient approach based on gradient boosted decision trees for detecting the location of the selected object in a real-world smart home. Moreover, we employ some techniques of over- and under-sampling with data clustering for improving the performance of the presented techniques. Many experiments are conducted in this work to evaluate the performance of the presented approach for localizing objects in a real smart home. The results of the experiments have shown that our approach provides remarkable performance.

Shadi Abudalfa

Computer-assisted disease diagnosis is cost-effective and time-saving, increasing accuracy and reducing the need for an additional workforce in medical decision-making. In our prior research, we trained, tested, and compared the accuracies of nine optimizable classification models to diagnose and predict eight anaemia types from Complete Blood Count (CBC) data. This study aimed to improve these classification models by oversampling the original imbalanced dataset with four algorithms related to the Synthetic Minority Over-sampling Technique (SMOTE). The results showed that the validation accuracy increased from 99.22% (Ensemble model) to 99.57% (Tree model), and most importantly, the False Discovery Rate (FDR) for the anaemia type with the highest FDR decreased from 23.1% to 1.5%.

Ladislav Végh

This research aims to evaluate the optimization of network security forensic tools on virtual router (VR) traffic. The methodology used includes the selection of several forensic tools on the Windows operating system such as Wireshark, Windump, and Network Miner, with testing in a virtual network environment. Testing, includes simulating various attack scenarios to assess the effectiveness of threat detection, performance of forensic tools, and impact on network performance. The main results show that the tools have varying detection capabilities with variations in resource usage and impact on network latency. Network traffic has been successfully recorded using the Win-dump tool in the static-forensics method, the Wireshark tool and Network Miner in the live-forensics method. The evaluation results of the meta-router network forensic recording tool recommend Win-dump as a recording tool that does not burden the Windows operating system with memory usage of 1696 kb while the Wireshark and Network Miner applications are recorded at more than 20MB. Based on this research, the static forensic method which have been built with meta-router objects can be used by investigators to detect cyber attacks. Proper selection and configuration of forensic tools is critical to achieving a balance between security and network performance, and specific adjustments to network requirements can increase the effectiveness of threat detection and mitigation.

Firmansyah Yasin

eveloping automated systems with a reasonable cost for long-term care for elders is a promising research direction. Such smart systems are based on realizing activities of daily living (ADLs) to enable aging in place while preserving the quality of life of all inhabitants in smart homes. One of the research directions is based on localizing items used by elders to monitor their activities with fine-grained details of the progress. In this paper, we shed the light on this issue by presenting an approach for localizing items in smart homes. The presented method is based on applying machine learning algorithms to Radio Frequency IDentification (RFID) tags readings. Our approach achieves the required task through two stages. The first stage detects in which room the selected object is located. Then, the second one determines the exact position of the selected object inside the detected room. Additionally, we present an efficient approach based on gradient boosted decision trees for detecting the location of the selected object in a real-world smart home. Moreover, we employ some techniques of over- and under- sampling with data clustering for improving the performance of the presented techniques. Many experiments are conducted in this work to evaluate the performance of the presented approach for localizing objects in a real smart home. The results of the experiments have shown that our approach provides remarkable performance.

Shadi Abudalfa

Formality transfer seeks to adjust text formality without altering its core meaning, which carries substantial implications across diverse domains like machine translation, dialogue systems, and social media content creation. This study provides an extensive overview of formality transfer specifically within Arabic text, an emerging domain within natural language processing. Particularly, we carried out a comprehensive review of literature on text formality transfer, focusing on studies published between July 2010 and April 2024. Our focus lies in treating formality transfer in Arabic as akin to a machine translation task, presenting synthesized insights. Despite advancements in formality transfer for English and other languages, Arabic’s distinct linguistic features present unique challenges and opportunities. Our investigation uncovers several research gaps necessitating future exploration, emphasizing persistent limitations. Moreover, we delve into text formality transfer as a promising avenue for forthcoming research initiatives in the realm of Arabic text processing.

Shadi Abudalfa

Assessing students' progress in introductory programming courses is crucial for identifying learning gaps and improving teaching methods. This study evaluates the effectiveness of Moodle-based tests with randomized questions in monitoring student progress in C programming courses at J. Selye University during the 2023/24 academic year. A series of ten tests were administered across two courses, covering essential programming topics such as data types, variables, conditional statements, loops, two-and three-dimensional arrays, recursion, and sorting algorithms. The results revealed significant variations in student performance, with recursion and the pretest/posttest loops presenting the greatest challenges. The correlation analysis of test scores showed strong relationships among related topics, confirming the structured progression of the curriculum. These findings suggest that Moodle-based assessments offer valuable insights into students' learning trajectories, enabling educators to adapt their instructional strategies accordingly. Such insights can help optimize introductory programming curricula, enhancing student engagement and understanding.

Ladislav Végh