MIIAM: An Algorithmic Model for Predicting Multimedia Effectiveness in eLearning Systems

Ensemble of neural networks are widely studied and applied in machine learning. Ensemble reduces the generalization error by reducing either the bias or variance part of the error or both. For ensembles to perform better, the constituent base classifiers should be diverse and accurate. Diversity can be introduced by injecting randomness in the data or architecture. Random Vector Functional Link (RVFL) neural network with closed form solution is a randomized neural network suitable to use as base classifiers of the ensemble because of its extremely fast training time, good generalization and innate randomization in its architecture. Thus, we propose an ensemble of random vector functional link neural networks by introducing additional regularization or randomization in its architecture via well established regularization techniques in the literature namely Dropout and DropConnect. It is evident that stronger randomization helps ensembles generalize better. Based on the experiments on several datasets, we observe that our proposed ensemble performs better than other RVFL based ensembles in most of the datasets.

An algorithm that can predict the review rating of a potential business with only existing information about the location and business categories would be an invaluable tool in making investment decisions. Utilizing the Yelp business dataset, we built a model, that can do as such, by classifying whether a potential business belongs to a positively-reviewed class (star ratings greater than or equal to 4) or a negatively-reviewed class (star ratings less than 4) given its location in latitude and longitude and the categories the potential business belongs to. More specifically, we applied a feature engineering technique using Extremely Randomized Trees, and constructed a hybrid ensemble classifier using neural network, decision tree, and logistic regression. We compared our model with other popular ensemble algorithms such as random forest and neural network ensemble, and our hybrid ensemble model generates the best result with an accuracy rate of 67.37% and AUC of 0.7322.

The piece of research presents a conceptual overview on diverse cognitive styles reflections in adaptable Open Learning systems. The main goal of this approach is quantitative forecasting the performance of adaptable Open Learning (equivalently e-learning) Systems using cognitive Neural Network modelling. Furthermore, analysis of interactive two diverse learners' cognitive styles with a friendly adaptable teaching environment(e-courses material). Consequently, presented paper provides e-learning systems' designers with relevant guide for learning performance enhancement. Additionally, it supports e-learners in fulfilment of better learning achievements during face to face tutoring. Accordingly, quantitative analysis of e-learning adaptability performed herein, via assessment of matching between learning style preferences and the instructor's teaching style and/or e-courses material. Interestingly, application of two realistic cognitive models using Artificial Neural Network gives an opportunity to experience well assessment of adaptable e-learning features. Such as adaptability mismatching, adaptation time convergence, and individual differences of e-learners' adaptability.

This article proposes a novel conceptual framework that integrates Artificial Intelligence (AI) with Cognitive Load Theory (CLT) and the Cognitive Theory of Multimedia Learning (CTML) to enhance Open Distance eLearning (ODeL) systems. By bridging the gap between traditional cognitive theories and cutting-edge AI technologies, this framework supports adaptive cognitive load management, AI-mediated schema creation, and human-AI collaborative learning. A refined critique of Twabu's (2023) PhD study, based on conceptual analysis of its theoretical scope and limitations, highlights the absence of AI integration. This article presents a unique contribution by synthesizing established cognitive theories with the dynamic potential of AI, a perspective underrepresented in current literature. Practical examples, policy recommendations, and ethical considerations are consolidated to offer a comprehensive path forward. The study proposes an expanded framework incorporating AI-enhanced cognitive load management, AI-mediated schema creation, and human-AI collaborative learning. The Literature review and theoretical framework emphasises AI's role in simplifying cognitive processes by dynamically adjusting content presentation based on learner needs and performance. This includes AI-mediated adjustments to manage cognitive load and enhance schema development through personalised feedback and tailored content delivery. Such integration promises to create a more effective learning environment by aligning multimedia content with individual learner profiles, thus addressing the challenge of cognitive overload and improving knowledge retention and understanding.

From a system’s theoretical point of view, adaptive learning systems (ALS) for education and training contain in their core – in a simplified form – closed feedback control loops in which the control is determined by the measured users’ performance. Improving this performance can increase the learning outcome, especially for critical disciplines such as education or training for disaster risk management. However, for this special form of intelligent (e-learning) assistance systems, learning theories and behavioral models have to be considered, e.g., game flow theory, cognition models, or learning models. The research question is how adaptive interactive learning environments (ILE) such as serious games and computer simulations can be characterized and analyzed to determine optimal adaptation strategies. Adaptive learning environments should adapt to the context-related needs of the user in order to ensure and optimize learning success, especially for disaster management training. This contribution presents a concept for an interoperable, adaptive ILE framework which follows control theory and its models, contributing to the state of the art for adaptive games or simulations in disaster risk management.

This paper describes an investigation into the ways in which learning using a multimedia application can be supported and enhanced by means of a simple co-operative student model of learner characteristics. This paper reports the design, implementation and evaluation of an individually configurable multimedia learning application, based upon such a model. A multimedia learning application was developed that presented information differentially based upon the individual characteristics of learners, held in the student model. The characteristics employed in the model were language level, cognitive style, task and question levels, and help level. Small groups of learners followed the multimedia course in learning centres located in colleges in the UK. A Grounded Theory study was carried out in order to understand the many and complex interactions that took place between learners, tutors and the learning environment. Stages in the Grounded Theory method are described and some qualitative data is presented. It was possible to conclude from these, that the quality of learning for individuals was improved by the use of the co-operative student model. Quantitative data is presented to support this view and where possible, to relate performance on the multimedia learning application to the student model configuration.

As Internet rises fast in recent decades, teaching and learning tools based on Internet technology are rapidly applied in education. Learning through Internet can make learners absorb knowledge without the limitations on learning time and distance. Therefore, in academy, e-learning is one of the popular learning assistant instruments. Recently, “student-centered” instruction has become one of the primary approaches in education, and the e-learning system, which can provide the learning environment of personalization and adaptability, is more and more popular. By using e-learning system, teachers can adjust the learning schedule instantly for learners according to their learning achievements, and build more adaptive learning environments. However, in some cases, bias assessments are given for student achievements under specific uncontrollable conditions (i.e. tiredness, preference). In dire need of overcoming this predicament, a new model based on radial basis function neural networks (RBF-NN) and similarity filter to evaluate learning achievements is proposed. The proposed model includes three phases to reduce bias assessments: (1) preprocess: select important features (attributes) to enhance classification performance by feature selection methods and utilize minimal entropy principle approach (MEPA) to fuzzify the quantitative data, (2) similarity filter: select linguistic values for each feature and delete inconsistent data by the similarity threshold (similarity filter) and (3) construct classification model and accuracy evaluation: build the proposed model based on RBF-NN and evaluate model performance. To verify the proposed model, a practical achievement dataset, collected from e-learning online examination system in a university of Taiwan, is used as experiment dataset, and the performance of the proposed model is compared with the listing models in this paper. From the empirical study, it is shown that the proposed model provided more proper achievement evaluations than the listing models.

Identification of personal traits in adaptive learning environment: Systematic literature review

Multimedia learning is defined as building mental representations from words and pictures. In multimedia learning, the difference in cognitive style indicates different learning strategies. The cognitive styles of visual and verbal exert influence on behavior, preferences, and even learning outcomes. On the other hand, eye-tracking has been used to study cognitive aspects during multimedia learning. Unfortu-nately, previous studies on the identification of cognitive styles were limited to statistical descriptive analysis. The use of eye-tracking was limited merely for validation purposes. In addition, previous studies have yet to apply automatic classification of cognitive style based on eye-tracking data. Hence, this study proposes a method to automatically classify visual-verbal cogni-tive styles based on eye-tracking metrics. We implemented three shallow classifiers: K-Nearest Neighbors, Random Forest, and Support Vector Machine. Based on our experimental results, Random Forest—enhanced with two selected features from SelectKBest-gained 78% of classification accuracy. Our study has been the first investigation that reveals the possibility of implementing machine learning for automatic classification of cognitive styles based on eye-tracking data.

Relative to target species, priority conservation species occur rarely in fishery interactions, resulting in imbalanced, overdispersed data. We present Ensemble Random Forests (ERFs) as an intuitive extension of the Random Forest algorithm to handle rare event bias. Each Random Forest receives individual stratified randomly sampled training/test sets, then down-samples the majority class for each decision tree. Results are averaged across Random Forests to generate an ensemble prediction. Through simulation, we show that ERFs outperform Random Forest with and without down-sampling, as well as with the synthetic minority over-sampling technique, for highly class imbalanced to balanced datasets. Spatial covariance greatly impacts ERFs’ perceived performance, as shown through simulation and case studies. In case studies from the Hawaii deep-set longline fishery, giant manta rayMobula birostrissyn.Manta birostrisand scalloped hammerheadSphyrna lewinipresence had high spatial covariance and high model test performance, while false killer whalePseudorca crassidenshad low spatial covariance and low model test performance. Overall, we find ERFs have 4 advantages: (1) reduced successive partitioning effects; (2) prediction uncertainty propagation; (3) better accounting for interacting covariates through balancing; and (4) minimization of false positives, as the majority of Random Forests within the ensemble vote correctly. As ERFs can readily mitigate rare event bias without requiring large presence sample sizes or imparting considerable balancing bias, they are likely to be a valuable tool in bycatch and species distribution modeling, as well as spatial conservation planning, especially for protected species where presence can be rare.

Evaluation of Preceding Variables Affecting Behavioral Use and Acceptance of Chord-Enabled Keyboard Among Students

This study investigates the transformative potential of integrating Artificial Intelligence (AI), animation, and personalized learning in contemporary education. Employing a mixed-methods approach involving interviews and experimental manipulations, the research examines the interconnectedness of these three domains and their collective impact on student engagement and English reading comprehension learning outcomes. The study employed a quasi-experimental design. Participants engaged in an 8-week AI-driven personalized learning intervention that incorporated animated content, with pre- and post-test assessments evaluating reading comprehension outcomes. Qualitative data from interviews provided further insight into the impact of animation and personalized learning on student engagement and comprehension. Findings reveal that AI-driven personalized learning, coupled with engaging animations, leads to enhanced learner motivation, improved knowledge retention, and higher academic performance. The integration of these elements creates an inclusive and adaptable learning environment, addressing the diverse needs of students. This research highlights the significance of interdisciplinary collaboration in educational innovation and emphasizes the vast potential that lies at the intersection of AI, animation, and personalized learning to revolutionize 21st-century education.

The early diagnosis of Brain Tumors (BT) is a critical challenge in medical imaging. This study proposes an explainable machine learning (XAI) framework that integrates multimodal imaging, including Magnetic Resonance Imaging (MRI) and Computed Tomography (CT), for accurate and interpretable BT detection. A hybrid feature extraction strategy was employed, combining deep learning-based spatial features with handcrafted texture descriptors, including GLCM and LBP. These features are fused using an attention-based mechanism to enhance discriminative performance. The refined features are classified using an ensemble of Random Forest, XGBoost, and Deep Neural Networks. Explainability is incorporated using SHAP and Grad-CAM to visualize the model's decision rationale. Experiments on publicly available datasets demonstrate superior performance, achieving 97.3% accuracy, 96.4% precision, 96.0% recall, and 96.2% F1-score, outperforming existing methods while ensuring clinical interpretability.

Previous ensemble streamflow prediction (ESP) studies in Korea reported that modelling error significantly affects the accuracy of the ESP probabilistic winter and spring (i.e. dry season) forecasts, and thus suggested that improving the existing rainfall‐runoff model, TANK, would be critical to obtaining more accurate probabilistic forecasts with ESP. This study used two types of artificial neural network (ANN), namely the single neural network (SNN) and the ensemble neural network (ENN), to provide better rainfall‐runoff simulation capability than TANK, which has been used with the ESP system for forecasting monthly inflows to the Daecheong multipurpose dam in Korea. Using the bagging method, the ENN combines the outputs of member networks so that it can control the generalization error better than an SNN. This study compares the two ANN models with TANK with respect to the relative bias and the root‐mean‐square error. The overall results showed that the ENN performed the best among the three rainfall‐runoff models. The ENN also considerably improved the probabilistic forecasting accuracy, measured in terms of average hit score, half‐Brier score and hit rate, of the present ESP system that used TANK. Therefore, this study concludes that the ENN would be more effective for ESP rainfall‐runoff modelling than TANK or an SNN. Copyright © 2005 John Wiley & Sons, Ltd.

The integration of multimedia learning and Universal Design for Learning (UDL) in UK higher education has significantly enhanced student engagement, accessibility, and learning outcomes in asynchronous digital environments. Video lectures, interactive simulations, and AI-driven adaptive learning have improved knowledge retention and participation, yet challenges such as faculty preparedness, technological infrastructure, and accessibility disparities persist. This study examines the role of interactive content, gamification, and virtual reality (VR) in promoting active learning while addressing barriers related to the digital divide, institutional policies, and faculty training. Research indicates that students using adaptive learning technologies achieve higher retention and completion rates, while gamified learning environments foster greater motivation and participation. However, equitable access to digital resources and ethical AI governance remain critical concerns. Future strategies should focus on AI-driven personalization, immersive learning experiences, and expanded faculty development initiatives to ensure scalable and inclusive digital education. By leveraging emerging technologies, data-driven learning analytics, and open educational resources (OERs), UK universities can create more effective, accessible, and engaging learning experiences for diverse student populations.