Learning Approach Research Articles

Sell not, TEL not

An interdisciplinary group of researchers designed an innovative mobile application (app) to support bioscience language proficiency among pre-registration nursing students. The app's design, which incorporated features for audition and speaking prompts, was a key factor in its effectiveness. The same pedagogical approach was applied to alternative topics within the unit in Moodle, the university’s learning management system (LMS). The intervention in both forms (IC) was trialled and evaluated in a pilot study with two consecutive student cohorts in 2023 (n1=631, n2=218). Student access to each form of the condition was measured via technology-provided user analytics. The intervention condition (IC) uptake was higher in its LMS form compared to the app form. However, there were more initial attempts of the standard condition (SC) than the IC overall, indicating a tendency towards novelty avoidance. This article highlights and discusses the challenges of implementing novel technology-enhanced learning (TEL) approaches, focusing on students as stakeholders. Recommendations are proposed to improve user uptake of TEL approaches by focusing on strategies to enhance students’ perceptions of the TEL’s utility. A consistent ‘sales pitch’ to students as stakeholders is suggested.

Predicting Student Dropout Rates in Higher Education: A Comparative Study of Machine Learning Algorithms

Recent years have seen a significant increase in study interest in the areas of predicting student performance, avoiding failure, and identifying the variables affecting student dropout. One important indicator in online and open distance learning courses is the student dropout rate. We purpose the naive bayes classification method to construct the student dropout prediction using naive bayes. This work examines the critical topic of forecasting student dropout rates in higher education using machine learning approaches, with a particular emphasis on the random forest algorithm and the naive bayes algorithm. The study's goal is to properly anticipate dropout rates using data mining methods and machine learning algorithms after conducting a thorough evaluation of existing literature and approaches. The systematic method consists of data collection from a Kaggle dataset, data preparation to solve class imbalance via SMOTE oversampling, and algorithm selection. Random forest and naive Bayes approaches outperform other machine learning algorithms in terms of accuracy, sensitivity, specificity, and precision. The study underscores the importance of considering diverse factors such as demographic data, socioeconomic factors, and academic performance in dropout prediction models. The implications of this research extend beyond academia, with the potential to inform proactive interventions and support systems, ultimately leading to improved student outcomes and institutional effectiveness. According to this paper, the paper outputs that for the binary classification on the data set used in this project has best performed with Naive Bayes and Random Forest Algorithm with SMOTE oversampling. Keywords- SMOTE oversampling, machine learning, Random forest, naive bayes.

The Effect of Game-Based Learning on Improving the Basic English Speaking and Listening Ability of Children Aged 5-6 at Golden Cradle Kindergarten in High-Tech Zone of Nanchang, Jiangxi

Background and Aims: English language proficiency is increasingly becoming a vital skill in today's globalized world. As countries strive to enhance their competitiveness on the international stage, the ability to communicate effectively in English has become a critical aspect of education, particularly during early childhood development. Early exposure to English language learning has been shown to yield significant benefits, leading to increased linguistic abilities and intercultural understanding. Consequently, the implementation of effective language learning strategies during early childhood education becomes paramount. The purposes of this study were: 1) to compare the basic English listening ability of children aged 5–6 after using game-based learning activities to that before learning, 2) to compare the basic English-speaking ability of children aged 5–6 after using the game-based learning activities with that before learning. And 3) to study the children’s interest in the game-based learning activities. Methodology: The study, conducted at Golden Cradle Kindergarten in Nanchang City, evaluated the impact of a game-based teaching method on the English listening and speaking skills of 35 children aged 5-6 over one month. Initial and final assessments recorded in Excel showed significant improvements in both skills and increased learning interest. The data analysis includes percentages, means, standard deviation, and paired t-tests. Results: Results showed that the English listening ability of students was significantly higher than before they learned, indicating that the game teaching method can improve the English listening ability of children aged 5–6 years old, students had English-speaking ability was significantly higher than before they learned, indicating that the game teaching method can improve the English-speaking ability of 5–6-year-old children and the learning interest of students was significantly higher than before they learned. Conclude: Children between the ages of five and six have demonstrated a significant improvement in their English listening skills after using the game teaching method; their ability was significantly better than before. Additionally, it improves their ability to speak English, with a discernible increase observed following the application of the game-based learning approach. Additionally, this approach makes the kids more engaged and motivated than they were before by piquing their curiosity in learning.

Leveraging AI for Traffic Offense Prediction: A Deep Learning Approach

Leveraging AI for Traffic Offense Prediction: A Deep Learning Approach

Spatio‐temporal dynamic navigation for electric vehicle charging using deep reinforcement learning

AbstractThis paper considers the real‐time spatio‐temporal electric vehicle charging navigation problem in a dynamic environment by utilizing a shortest path‐based reinforcement learning approach. In a data sharing system including transportation network, an electric vehicle (EV) and EV charging stations (EVCSs), it is aimed to determine the most convenient EVCS and the optimal path for reducing the travel, charging and waiting costs. To estimate the waiting times at EVCSs, Gaussian process regression algorithm is integrated using a real‐time dataset comprising of state‐of‐charge and arrival‐departure times of EVs. The optimization problem is modelled as a Markov decision process with unknown transition probability to overcome the uncertainties arising from time‐varying variables. A recently proposed on‐policy actor–critic method, phasic policy gradient (PPG) which extends the proximal policy optimization algorithm with an auxiliary optimization phase to improve training by distilling features from the critic to the actor network, is used to make EVCS decisions on the network where EV travels through the optimal path from origin node to EVCS by considering dynamic traffic conditions, unit value of EV owner and time‐of‐use charging price. Three case studies are carried out for 24 nodes Sioux‐Falls benchmark network. It is shown that phasic policy gradient achieves an average of 9% better reward compared to proximal policy optimization and the total time decreases by 7–10% when EV owner cost is considered.

Resource Allocation in UAV-D2D Networks: A Scalable Heterogeneous Multi-Agent Deep Reinforcement Learning Approach

In unmanned aerial vehicle (UAV)-assisted device-to-device (D2D) caching networks, the uncertainty from unpredictable content demands and variable user positions poses a significant challenge for traditional optimization methods, often making them impractical. Multi-agent deep reinforcement learning (MADRL) offers significant advantages in optimizing multi-agent system decisions and serves as an effective and practical alternative. However, its application in large-scale dynamic environments is severely limited by the curse of dimensionality and communication overhead. To resolve this problem, we develop a scalable heterogeneous multi-agent mean-field actor-critic (SH-MAMFAC) framework. The framework treats ground users (GUs) and UAVs as distinct agents and designs cooperative rewards to convert the resource allocation problem into a fully cooperative game, enhancing global network performance. We also implement a mixed-action mapping strategy to handle discrete and continuous action spaces. A mean-field MADRL framework is introduced to minimize individual agent training loads while enhancing total cache hit probability (CHP). The simulation results show that our algorithm improves CHP and reduces transmission delay. A comparative analysis with existing mainstream deep reinforcement learning (DRL) algorithms shows that SH-MAMFAC significantly reduces training time and maintains high CHP as GU count grows. Additionally, by comparing with SH-MAMFAC variants that do not include trajectory optimization or power control, the proposed joint design scheme significantly reduces transmission delay.

Do Beliefs About Intelligence Drive Engagement in Online Learning?

This study investigates the relationship between students’ beliefs about intelligence (fixed vs. growth mindset) and their engagement and behaviour in an online learning platform. The data was collected from 28 students enrolled in COMP90082 – Software Project subject at The University of Melbourne. Students’ beliefs about intelligence were measured through a survey administered during the first week, while their online learning behaviours were captured through clickstream data from the learning management system. The study adopted a two-fold approach: first, students’ learning strategies were identified through cluster analysis of their online session duration. By examining over 41,000 learning actions —such as accessing course materials, participating in discussions, and completing assignments— three distinct learning strategies emerged: “light”, “light-intensive”, and “intensive.” Subsequently, the relationship between these learning strategies and students’ beliefs about intelligence was examined using data visualisation techniques, chi-square tests, and logistic regression models. The findings revealed that students with a growth mindset were more likely to adopt intensive and engaged learning approaches within the online platform. Conversely, those with a fixed mindset tended to exhibit lower engagement levels and less intensive learning strategies. The study may contribute to the growing literature on the psychology of online learning and highlight the potential for tailoring instructional interventions and platform design to foster a growth mindset, thereby enhancing student engagement and academic success in digital learning environments.

Enhancing High-Availability Database Systems: An AI-Driven Approach to Anomaly Detection

High-availability database systems are critical components in modern IT infrastructure, demanding robust mechanisms for ensuring continuous operation and data integrity. This article explores integrating artificial intelligence (AI) techniques into anomaly detection processes for such systems, addressing the limitations of traditional rule-based and statistical methods. We present a comprehensive analysis of machine learning and deep learning approaches, including supervised and unsupervised learning models, autoencoders, recurrent neural networks, and hybrid solutions that combine AI with conventional techniques. The article examines the challenges of implementing AI-powered anomaly detection in high-availability environments, such as scalability, real-time processing, and the balance between sensitivity and specificity. Through case studies in the financial and e-commerce sectors, we demonstrate these advanced detection methods' practical applications and benefits. Our findings indicate that AI-driven approaches significantly enhance the accuracy and efficiency of anomaly detection, leading to improved system reliability and performance. The article concludes by discussing emerging trends, including edge computing and explainable AI, and their potential impact on the future of database management and anomaly detection.

Enhancing Healthcare with Personalized Recommendations: A Machine Learning Approach to Symptom-Driven Diagnosis

This paper introduces the Personalized Medical Recommendation System, a machine learning-based platform that supports users in symptom analysis and personalized healthcare management. Users input symptoms through a simple interface, and machine learning models predict potential diseases. The system offers tailored recommendations, including suggested medications, prescription details, and exercise routines. Built on a Flask web application, it provides accessible healthcare support with a strong focus on data privacy and security. Continuous improvement through user data refines predictive accuracy, advancing personalized medicine. This study outlines the system's design, implementation, and performance, underscoring its potential to enhance health outcomes. Key Words: personalized healthcare, machine learning, disease prediction, medical recommendation system, symptom analysis, data privacy.

DeepFake Image Detection: Fake Image Detection using CNNs and GANs Algorithm

Deep learning is a powerful and versatile technique that has seen extensive applications in areas such as natural language processing, machine learning, and computer vision. Among its most recent applications is the generation of deepfakes, which are high-quality, realistic altered videos or images that have garnered significant attention. While innovative uses of deepfake technology are being explored, its potential for misuse has raised serious concerns. Harmful applications, such as spreading fake news, creating celebrity pornography, financial fraud, and revenge pornography, have become increasingly prevalent in the digital age. As a result, public figures, including celebrities and politicians, face heightened risks from deepfake content. In response, substantial research has been conducted to explore the mechanics behind deepfakes, leading to the development of various deep learning-based algorithms for their detection. This study provides a comprehensive review of deepfake creation and detection techniques, focusing on different deep learning approaches. Additionally, it discusses the limitations of existing methods and the availability of datasets for research. The lack of a highly accurate and fully automated deepfake detection system presents a significant challenge as the ease of generating and distributing such content continues to grow. Nonetheless, recent efforts in deep learning have shown promising results, surpassing traditional detection methods.

Correlation of The Application of Teaching At The Right Level (Tarl) With Students' Basic Literacy Outcomes

This study aims to identify the correlation between the implementation of the TaRL (Teaching at the Right Level) approach and the improvement of grade I students' basic literacy learning outcomes. Using a quantitative correlational method, the study involved 25 students grouped into four ability levels: Special Intervention, Basic, Proficient and Advanced. The test results showed that 17 students were at the Special Intervention level, 1 student at the Basic level, 7 students at the Advanced level and none at the Proficient level. Meanwhile, to get data on learning outcomes, students were given multiple choice questions to measure their abilities. Data analysis using the Pearson Product Moment test showed a correlation value of 0.812, which is higher than the critical value of 0.396 at the 5% significance level. This indicates a very strong correlation between the application of TaRL and the improvement of basic literacy learning outcomes based on correlation interpretation. Thus, the null hypothesis (H0: no correlation) is rejected, and the alternative hypothesis (H1 : correlation) is accepted. In conclusion, the application of the TaRL approach is positively related to the improvement of students' basic literacy learning outcomes, demonstrating the effectiveness of this approach in literacy learning.

Pep2TCR: Accurate prediction of CD4 T cell receptor binding specificity through transfer learning and ensemble approach

Pep2TCR: Accurate prediction of CD4 T cell receptor binding specificity through transfer learning and ensemble approach

An Integrative Framework for Healthcare Recommendation Systems: Leveraging the Linear Discriminant Wolf–Convolutional Neural Network (LDW-CNN) Model

In the evolving healthcare landscape, recommender systems have gained significant importance due to their role in predicting and anticipating a wide range of health-related data for both patients and healthcare professionals. These systems are crucial for delivering precise information while adhering to high standards of quality, reliability, and authentication. Objectives: The primary objective of this research is to address the challenge of class imbalance in healthcare recommendation systems. This is achieved by improving the prediction and diagnostic capabilities of these systems through a novel approach that integrates linear discriminant wolf (LDW) with convolutional neural networks (CNNs), forming the LDW-CNN model. Methods: The LDW-CNN model incorporates the grey wolf optimizer with linear discriminant analysis to enhance prediction accuracy. The model’s performance is evaluated using multi-disease datasets, covering heart, liver, and kidney diseases. Established error metrics are used to compare the effectiveness of the LDW-CNN model against conventional methods, such as CNNs and multi-level support vector machines (MSVMs). Results: The proposed LDW-CNN system demonstrates remarkable accuracy, achieving a rate of 98.1%, which surpasses existing deep learning approaches. In addition, the model improves specificity to 99.18% and sensitivity to 99.008%, outperforming traditional CNN and MSVM techniques in terms of predictive performance. Conclusions: The LDW-CNN model emerges as a robust solution for multidisciplinary disease prediction and recommendation, offering superior performance in healthcare recommender systems. Its high accuracy, alongside its improved specificity and sensitivity, positions it as a valuable tool for enhancing prediction and diagnosis across multiple disease domains.

Segmentation and classification of brain tumor using Taylor fire hawk optimization enabled deep learning approach

ABSTRACT The brain is a crucial organ that controls the body’s neural system. The tumor develops and spreads across the brain as a result of irregular cell generation. The provision of substantial treatment to patients requires the early diagnosis of malignancies. However, timely diagnosis and accurate classification were difficult in the conventional models. Thus, the Taylor Fire Hawk optimization (TFHO) is implemented here for effective segmentation and classification. The TFHO is the merging of the Taylor series and Fire Hawk Optimizer (FHO). The de-noising is accomplished by the adaptive median filter, and the segmentation is carried out using M-Net, which has been trained by TFHO. Subsequently, image augmentation is performed to increase the image dimension, followed by the extraction of effective features. Finally, DenseNet is used for the classification, and the training is done by TFHO. The introduced method obtained 94.86% accuracy, 92.83% Negative Predictive Values, 89.33% Positive Predictive Values (PPV), 95.91% True Positive Rate (TPR), 4.37% False Negative Rate (FNR), and 90.98% F1-score.

Defect classification of composite materials using transfer learning methods

ABSTRACT Nowadays, composite materials have become prevalent across various sectors, particularly finding usage in large-scale applications such as spaceships, automobiles, and aircrafts. The accurate detection of the defects in these materials is crucial, yet traditional methods often rely on human inspection, which is susceptible to errors. Recent advancements in machine learning have enabled defect detection using ultrasonic non-destructive testing methods. This paper introduces a new dataset named UNDT, which is obtained from the scans of 60 different composite materials, generating a total of 1150 images depicting both defective and non-defective areas. Several transfer learning methods are applied on the newly introduced UNDT dataset as well as the publicly available USimgAIST ultrasonic dataset. Comparative performance assessments illustrate the significance of utilising the transfer learning approach for defect classification on ultrasonic inspection images. Furthermore, the research emphasises the substantial benefits of employing these transfer learning methods. Notably, the DenseNet121 and VGG19 models achieve the highest accuracy rates, with 98.8% and 98.6% on the UNDT and USimgAIST datasets, respectively.

Unconfined Compressive Strength Prediction of Soils Improved with Biopolymers: Machine Learning Approach

Unconfined Compressive Strength Prediction of Soils Improved with Biopolymers: Machine Learning Approach

Cloud based adaptive learning system: virtual reality and augmented reality assisted educational pedagogy development on clinical simulation

An adaptive learning system provides support for student’s academic progress by varying learning routes and pedagogical contents to satisfy various students’ unique academic needs. The primary goal of this research work is to enhance and expand occupational efficiency through the use of technological pedagogical content knowledge (TPCK) frameworks in nursing education. The authors developed an adaptive federated cloud computing learning system in order to better prepare students for the demands of real-world professional practice. The article examined the use of virtual reality(VR) and augmented (AR) technologies in nursing education and developed an instrument that skillfully combines knowledge from the reputable literature on clinical competencies with the peculiarities and ingenuousness particular to virtual simulations. Designated for the remote sharing and utilization of combined educational resources, cloud computing is a distributed information communication technology (ICT) infrastructure built on the Internet of Things (IoT) framework. To increase the occupational efficiency among the nurses and midwives, an educational simulation applied in a way that gamifies the curriculum component for interactive educational pedagogy will be necessary. As a completely integrated part of the curriculum, simulation has proven to be beneficial for nursing students at all levels of instruction. Nursing educators have a plethora of options at their disposal to introduce simulation-based learning approaches into their curricula in the modern era. A three-dimensional (3D) interactive environment that provides users with a sense of spatial presence is created through computer technology, which is an immersive digital ecosystem of VR and AR utilized to improve teaching and learning in nursing education. In this study, the authors offered a cost- effective clinical simulation laboratory for clinical students, nurses, midwives, and practicing healthcare professionals in a manner that ensures constant learning and occupational efficiency by adopting the TPCK model in its central discussion and implementation. The TPCK model has demonstrated how to recognize the boundaries between technological components and pedagogical structures as well as how they are connected when creating facts in nursing education.

Quality Assessment of Edible Plant-Based Fixed Oils Using Different Analytical Techniques and Machine Learning Approaches

Plant-based carrier oils are constantly gaining popularity for their beneficial health effects on human organisms, thus shifting consumers’ preferences to alternative options in the oil market. The purpose of the study was to evaluate the quality characteristics of twenty-six edible plant-based fixed oils via chromatographic, chromatic, and spectroscopic techniques, suggesting their potential use as complementary edible oil sources. Palmitic, stearic, oleic, linoleic, and a-linolenic acids were found in considerable proportions, whereas the majority of oils possessed unsaturated (UFA)/saturated (SFA) fatty acid ratio greater than 1.6, suggesting their function in lowering blood pressure and preventing cardiovascular disease. Linseed, chia seed, macadamia, and canola oils provide a balanced intake of n-6 and n-3 polyunsaturated fatty acids (PUFA) within the range of 1:1 to 5:1. Oxidative stability was inversely related to oils’ PUFA content, with linseed, chia seed, pine cone, and walnut oils being the least stable oils against oxidation. Chlorophyll content in all oils was below the limit (50 mg/kg), preventing oxidation in the presence of light, whereas the highest values of b-carotene were noticed in soybean, linseed, and canola oils (61.18, 60.42, and 60.12 ppm, respectively). The application of machine learning algorithms for analyzing ATR-FTIR band intensities and FA proportions via discriminant analysis succeeded in discriminating pulp from seed oils, with a classification accuracy of 96.0% and 88.0%, respectively.

Machine Learning Approaches for Microorganism Identification, Virulence Assessment, and Antimicrobial Susceptibility Evaluation Using DNA Sequencing Methods: A Systematic Review

Machine Learning Approaches for Microorganism Identification, Virulence Assessment, and Antimicrobial Susceptibility Evaluation Using DNA Sequencing Methods: A Systematic Review

A Machine Learning Approach for Breast Cancer Risk Prediction in Digital Mammography

Breast cancer is among the most prevalent cancers in the female population globally. Therefore, screening campaigns as well as approaches to identify patients at risk are particularly important for the early detection of suspect lesions. This study aims to propose a workflow for the automatic classification of patients based on one of the most relevant risk factors in breast cancer, which is represented by breast density. The proposed classification methodology takes advantage of the features automatically extracted from mammographic images, as digital mammography represents the major screening tool in women. Textural features were extracted from the breast parenchyma through a radiomics approach, and they were used to train different machine learning algorithms and neural network models to classify the breast density according to the standard Breast Imaging Reporting and Data System (BI-RADS) guidelines. Both binary and multiclass tasks have been carried out and compared in terms of performance metrics. Preliminary results show interesting classification accuracy (93.55% for the binary task and 82.14% for the multiclass task), which are promising compared to the current literature. As the proposed workflow relies on straightforward and computationally efficient algorithms, it could serve as a basis for a fast-track protocol for the screening of mammograms to reduce the radiologists’ workload.