Online-Journals.org (International Association of Online Engineering)
Not a member yet
9805 research outputs found
Sort by
Sentiment Analysis and Topic Modelling for Academic Integrity in the Era of AI
Get PDFThis study explores the sentiments and discussion topics of X/Twitter users regarding academic integrity in the era of artificial intelligence (AI). The approach incorporates sentiment analysis and topic modelling to reveal the public perspective on academic integrity issues, including plagiarism, online exams, and AI usage. Our study aims to provide a framework for exploring topics and findings related to the trend of academic integrity in the era of AI. In sentiment classification, Naive Bayes, support vector machine (SVM), and Random Forest algorithms are combined with vectorization techniques such as Count Vectorizer, Word Level TF-IDF, N-Gram TF-IDF, and Character Level TF-IDF. The results show that Naive Bayes with Count Vectorizer provides the best performance on imbalanced data. For the topic modelling, NMF proved to be the most effective in generating specific topics, such as plagiarism and AI detection, with the highest coherence scores. This study also examines the crucial role of each preprocessing step in enhancing data quality, which significantly impacts classification and topic modelling performance. The findings are expected to provide new insights into sentiment analysis and a deeper understanding of academic integrity issues in the era of artificial intelligence
Exploration of the Employment Quality Evaluation Method for Local University Graduates Based on PROMETHEE
Get PDFCurrent evaluation methods for college graduate employment quality mainly employ techniques such as the analytic hierarchy process (AHP) and factor analysis (FA), which tend to be similar in analytical approach and have certain limitations. This paper utilizes the PROMETHEE method with priority function characteristics, introduces possibility comparison interval hesitant fuzzy numbers, and evaluates the employment quality of local college graduates from three dimensions: overall employment quality, social influence, and overall satisfaction. An improved PROMETHEE decision-making method considering attribute correlation is proposed, and its stability and superiority are demonstrated, providing support for the system of employment quality evaluation methods
Plithogenic Machine Learning Solutions to Material Selection in Renewable Energy Systems
Get PDFPlithogenic-based decision models are more effective in designing optimal solutions to intricate problems. This study work proposes an integrated decisioning model conjoining plithogeny and machine learning algorithms. This study considers the decision-making problem of selecting smart and sustainable materials for the effective functioning of renewable energy systems. The decisioning model has ten evaluation criteria and considers alternatives for materials subjected to five categories of photovoltaic, thermoelectric, piezoelectric, phase change, supercapacitor, and electrochromic. This work employs the algorithm of a random forest classifier in determining the most crucial criteria for selecting smart and sustainable materials. The plithogenic-based decision method of TOPSIS (Technique for Order of Preference by Similarity to Ideal Solution) is employed in ranking materials of each kind. The proposed decisioning approach is the combination of a machine learning algorithm and a plithogenic decision approach, which is further facilitated by the intervention of Python programming. The criteria selection accuracy is compared with a support vector machine algorithm to demonstrate the efficacy of this integrated decision approach in ranking the materials used in formulating robust renewable energy systems. Sensitivity analysis is also performed to exhibit the efficacy of this proposed model. This model has few limitations, as it considers a few selected materials under each of the categories
Understanding Student Engagement with Mobile Learning: A Structural Model for Interactive Education in Vietnam
Get PDFThe factors that impact the willingness of Vietnamese secondary school students to embrace mobile learning are examined in this study using the framework of the technology acceptance model (TAM). The study utilized a quantitative research approach, and data were gathered accordingly from 611 students via structured questionnaires. The research focused on critical variables, including perceived usefulness (PU), perceived ease of use (PEU), perceived mobility (PM), social influence (SI), self-efficacy (SE), learning autonomy (LA), and perceived enjoyment (PE). Covariance-based structural equation modeling (CB-SEM) was utilized to analyze the relationships among these variables. The findings revealed that PU, PEU, and PE significantly affect students’ behavioral intention (BI) to adopt mobile learning. At the same time, external factors such as PM, SI, SE, and LA indirectly shape PU and PEU. These results deepen the understanding of mobile learning adoption within the Vietnamese educational context and underscore the need for targeted policy development to foster student engagement with mobile technologies. This study contributes to the broader educational technology literature, offering valuable insights for educators and policymakers aiming to optimize learning experiences through mobile platforms
Integrating Human-Computer Interaction and Software Engineering for Enhanced Usability Using Support Vector Machines
Get PDFSoftware system development is evolving as human-computer interaction (HCI) concepts find application in software engineering. This shift emphasises more the need of knowing user requirements, behaviours, and preferences. This approach is especially important to meet the always rising demand for straightforward-to-run and-understand software systems. On the other hand, traditional methods of software development sometimes ignore the need for usability, which generates products that satisfy less than user expectations. This work suggests a new framework combining HCI methods with software engineering techniques based on the results. Support vector machine (SVM) application helps to improve process usability in development. There are many iterative prototyping, usability testing, and constant user feedback components inside the framework proposed for the software development lifecycle. Strong machine learning (ML) techniques, SVM, are applied for modelling and prediction of user satisfaction based on interaction data acquired during usability testing. Data analysis helps the SVM model find significant factors affecting usability. This enables developers to make sensible decisions exactly matched with consumer expectations. The results of the pertinent tests revealed the degree of improvement in software usability this approach brought about. Apart from a precision of 90.8%, a recall of 91.7%, and an F1-score of 91.2%, the SVM model was evaluated to have reached a classification accuracy of 92.3%. HCI, combined with software engineering, achieved by means of SVM, offers a strong approach for the development of software that not only serves but also strongly connects with users, resulting in a more significant impact on society
AI-Powered Chatbots for Mobile Teacher Training: Enhancing Interactive Learning Experiences
Get PDFThe study aims to explore the impact of artificial intelligence (AI)-powered chatbots in mobile teacher training, specifically in enhancing interactive learning. To achieve the objective, the researcher employed an experimental approach using a single-group (pre-post) design. The study sample consisted of 50 male and female secondary school teachers who received training based on chatbot applications over a four-week period. Following the training, the research instrument—a teaching skills scale was administered. This scale measured three key skills: interactive planning, the integration of technology in teaching, and interactive evaluation. The results revealed statistically significant differences between the participants’ scores on the teaching skills scale in the pre- and post-tests across all dimensions and the overall score. These differences favored the post-test results, indicating an improvement after exposure to the chatbot-based training program. Considering these findings, the study recommends the development of ongoing professional training programs for teachers, particularly those incorporating innovative models such as chatbot applications
MentAlzh: Mobile Application Prototype for Early Detection of Cognitive Impairment
Get PDFThe increasing incidence of neurodegenerative diseases has raised a pressing need for practical tools to detect mild cognitive impairment in its early stages. This study presents MentAlzh, a mobile app prototype that employs an interactive multimedia approach to cognitive assessment, adapting traditional methods. The app was developed using the design thinking methodology to engage the end-user; the prototype encompasses six cognitive tests inspired by established methods, focusing on orientation, clock drawing, fixation, language, calculation, and memory. A group of 21 participants, aged between 51 and 76, evaluated the app regarding usability, accuracy, and repeatability. Results demonstrated consistent performance on orientation and fixation tasks, while variability on memory and clock drawing tests indicated potential areas for further refinement. The MentAlzh prototype was found to have the potential to serve as an accessible and user-friendly tool for early cognitive screening
AI-Driven Personalized Mathematics Learning Through Interactive Mobile Platforms: Effects on Achievement and Motivation
Get PDFAbstract—The increasing use of mobile learning platforms shows that people need adapted experiences that address mobile device features and system capabilities. This research evaluates the powerful effect that mobile learning environments enabled through AI technology would have on mathematics education. The research investigates an AI-based mobile educational platform to enhance students' mathematics achievement and motivation. The researcher adopted a quasi-experimental approach and selected the participants purposefully, comprising 76 students. The participants were randomly divided into two groups of eleventh grade students. The experimental group consisted of (38) students who were taught mathematics using AI-based mobile educational platform. The control group included (38) students who were taught mathematics using traditional teaching methods. The findings revealed statistically significant differences at the significance level of (α = 0.05) between the mean scores of the experimental and control groups in the post-test. The adjusted mean score of the experimental group was 15.83, compared to 9.49 for the control group, demonstrating the positive impact of AI-based mobile educational platform in enhancing students' mathematics achievement. The finding confirmed that there are statistically significant differences at the significance level (a = 0.05) between the average ranks of the students in the experimental and control groups on the motivation instrument among high school students due to the use of AI-based mobile educational platform. Based on these findings, the study recommends implementing comprehensive training programs for teachers to ensure the effective integration of AI-based mobile educational platform in the classroom. There is an immediate need for additional studies in this domain, particularly within high schools, to enhance the student's capacity for self-reliance in skill developmentThe increasing use of mobile learning platforms shows that people need adapted experiences that address mobile device features and system capabilities. This study evaluates the powerful effect that mobile learning environments enabled through artificial intelligence (AI) technology would have on mathematics education. The researcher adopted a quasi-experimental approach. The participants were randomly divided into two groups of eleventh-grade students. The findings revealed statistically significant differences at the significance level of (α = 0.05) between the mean scores of the experimental and control groups in the post-test, demonstrating the positive impact of an AI-based mobile educational platform in enhancing students’ mathematics achievement. The finding confirmed that there are statistically significant differences at the significance level (a = 0.05) between the average ranks of the students in the experimental and control groups on the motivation instrument among high school students. Based on these findings, the study recommends implementing comprehensive training programs for teachers to ensure the effective integration of AI-based mobile educational platforms in the classroom. There is an immediate need for additional studies in this domain, particularly within high schools, to enhance the students’ capacity for self-reliance in skill development
Mobile Technology-Enabled Classroom Interaction: Advancing Interactive Learning in Higher Education
Get PDFAmid the rapid advancement of information technology, mobile technology—characterized by its portability, real-time functionality, and interactive capacity—has been increasingly leveraged to transform models of classroom interaction in higher education. Traditional lecture-based instruction has encountered growing demands for innovation, while the widespread adoption of mobile devices and applications has not only altered students’ learning behaviors but also introduced novel interaction scenarios between teachers and students. However, significant methodological limitations remain in existing studies. Many rely predominantly on traditional survey-based approaches to examine teacher-student interaction, which are insufficient for capturing the dynamic complexity and real-time characteristics of mobile-mediated interaction networks. Furthermore, most investigations into temporal features of interaction remain confined to descriptive statistics, lacking deeper insights into burstiness and memory effects. Focusing on interactive learning in higher education classrooms through the lens of mobile interaction networks, two principal dimensions were explored in this study. First, teacher-student interaction patterns were analyzed by examining overall and individual network structural characteristics, as well as interaction dynamics, through social network analysis to uncover node connectivity patterns and interaction intensity. Second, temporal characteristics of interaction were investigated via time series analysis, with an emphasis on identifying burstiness and memory-related patterns. By integrating structural and temporal dimensions, this study aims to establish a theoretical foundation for the development of mobile technology–enabled interactive learning models. The findings are intended to inform pedagogical strategy optimization and enhance classroom interaction effectiveness while also providing practical guidance for the improvement of mobile educational applications
Novel Framework for Robust Gene Selection and Accurate Multi-Cancer Classification
Get PDFThis study presents the ensemble adaptive gene selection and classification framework (EAGSCF), a novel method for cancer classification using high-dimensional gene expression data. EAGSCF integrates hybrid feature selection, adaptive dimensionality reduction, and ensemble deep learning to address challenges such as high dimensionality, class imbalance, and interpretability. By combining mutual information (MI), recursive feature elimination, and the least absolute shrinkage and selection operator (LASSO), the framework extracts a compact, biologically meaningful subset of features. Meanwhile, uniform manifold approximation projection and variation auto encoders (VAEs) enhance their capacity to capture nonlinear relationships, which are crucial for distinguishing complex cancer subtypes. With top accuracy across four cancer datasets—98.9% for lung, 98.5% for colon, 98.2% for prostate, and 97.8% for lymphoma—EAGSCF outperforms existing methods, demonstrating significant potential in biomarker discovery and clinical use