Learning System Research Articles

Autonomous AI Systems for Continuous Learning and Real-Time Decision Making

Autonomous AI systems are upsetting navigation by constantly learning and pursuing ongoing choices without human oversight. These frameworks have applications across businesses like medical services, money, and assembling, where ongoing information is fundamental for functional achievement. This paper investigates the parts of independent simulated intelligence frameworks, including constant learning models, continuous information handling structures, and dynamic procedures. We likewise look at the difficulties such frameworks face, including information security, reasonableness, and taking care of edge cases. Through contextual analyses in various enterprises, we show the capability of independent computer-based intelligence frameworks to reshape the fate of savvy navigation. At last, we propose future bearings for research, zeroing in on moral structures and half and half learning models to further develop flexibility and straightforwardness in these frameworks.

Reinforcement Learning in AI-Driven Assessments : Enhancing Continuous Learning and Accessibility

This article explores the transformative potential of Reinforcement Learning (RL) in AI-driven assessments, examining its impact on personalized learning experiences and continuous skill development. We investigate how RL algorithms enable adaptive learning systems to create dynamic, individualized learning paths, provide real-time feedback, and optimize long-term educational outcomes. The article delves into key aspects such as personalized learning trajectories, adaptive difficulty adjustments, and the integration of gamification elements to enhance engagement. We discuss the application of these technologies in various contexts, from traditional educational settings to professional development and certification preparation. The article also addresses the challenges and limitations of implementing RL in educational technology, including concerns about algorithmic bias and the complexity of modeling human learning processes. By analyzing current research and potential future directions, this paper provides a comprehensive overview of how RL is reshaping the landscape of education and assessment, offering insights into the future of adaptive learning systems and their role in fostering lifelong learning in an increasingly digital world.

Developing an Adaptive Learning Recommendation Algorithm and System for MOOCs

Massive Open Online Courses (MOOC) based learning platform had totally changed the educational environment by providing easy and accessible learning opportunities for global learners. But even such environment display high dropout and low learner engagement which remain a significant challenge to be addressed. To handle the challenge of this study, propose an Adaptive Learning Recommendation System (ALRS) that is designed to personalize learning paths based on individual preferences and performance metrics. The study employed Open University Learning Analytics Dataset (OULAD) and build recommendation model that combine k-means Clustering, Content-based Filtering, Collaborative Filtering, and Random Forest (RF) classifiers to make course recommendations. The proposed model have shown better recommendation when compared to other models with Precision of 0.92, Recall of 0.89, F1 Score of 0.90, and AUC of 0.95. Also the proposed model had shown the lowest Mean Squared Error (MSE) and Root Mean Squared Error (RMSE) at 0.042 and 0.205, respectively.

A Study on Technology Integration Practices in Arts Instructors: Exploring the Mediating role of Value and Ability Beliefs

Teachers are the crux of every developing mind of newly blooming pupils in the education institutes and it is a universal factor in the learning process of every nature or aspect of study. Nowadays, the use of innovative technologies has become an essential practice in learning and teaching systems of every subject including art studies which indicates the constant incorporation of technology in instructional communication and reflects an advancement in the teaching process. Quantitative research design was implemented to assess the cause-and-effect relationship among variables. Primary data was collected through the survey strategy and a close-ended questions comprised questionnaire was designed. The researcher disbursed 350 questionnaires and finalized 331 questionnaires for the data analysis. The study encountered significant results for all synthesized associations except for the mediating role of ability beliefs. The present result hold numerous theoretical and practical implications. This study extends the growing body of literature regarding technology integration practices through the professional development, ICT knowledge and technical assistance. However, the study also provides practical implications as beneficial insights can be provided through which professional development can be fostered through technology integration practices. Research limitations and future indications have also been explained to increase the generalizability of this research.

Rapid shear capacity prediction of TRM-strengthened unreinforced masonry walls through interpretable machine learning deployed in a web app

The presented study provides an efficient and reliable tool for rapid shear capacity estimation of TRM-strengthened unreinforced masonry walls. For this purpose, a data-driven methodology based on a machine learning system is proposed using a dataset constituted of experimental results selected from the bibliography. The outlier points of the dataset were detected using the Cook’s distance methodology and removed from the raw dataset, which consisted of 113 examples and 11 input variables. In the processed dataset, 17 machine learning methods were trained, optimized through hyperparameter tuning, and compared on the test set. The most effective models were the optimized instances of XGBoost and CatBoost methods, which combined into a voting model to leverage the predictive capacity of more than a single regressor. The final blended model shows remarkable predicting capacity with the determination factor (R2) equal to 0.95 and the mean absolute percentage error equal to 8.03%. Also, the model’s predictions are compared with those of existing analytical relationships, and it is found to perform the best of all. In sequence, machine learning interpretation methods are applied to find how the predictors influence the target output. Am, ft, and n⋅tf were identified as the most significant predictors with a mainly positive influence on the shear capacity. Finally, the built machine learning system is employed in a user-friendly web app for easy access and usage by professionals and researchers.

Automated EEG-based language detection using directed quantum pattern technique

Electroencephalogram (EEG) signals contain complex useful information about brain activities. These EEG signals are noisy, highly varying and nonstationary in nature. Hence, extracting meaningful information from these signals is challenging. The existing machine learning systems struggle to capture the minute changes from the signals and yield high performance.This study introduces a novel quantum-inspired feature extraction technique called Directed Quantum Pattern (DQP), designed to address these challenges by using a lattice structure to capture directional binary features. These directions (paths) are computed using a maximum function providing a dynamic and adaptive feature representation.This paper presents a novel DQP-LangNet developed using DQP for automated classification of two- languages using EEG signals. We have proposed a hybrid approach, combining DQP, statistical features, and multi-level discrete wavelet transform (MDWT) to extract salient features similar to the deep learning approach. The EEG dataset consisting of 14 channels, produces 7 feature vectors per channel, yielding 98 feature vectors. Neighborhood component analysis and Chi-square (Chi2) feature selection approaches generated 196 feature vectors.In addition to the innovative feature extraction a new classification structure called “t” is proposed k-nearest neighbor (tkNN) and support vector machine (tSVM) classifiers are employed. Using the proposed tkNN and tSVM classifiers, 392 (=196×2) classifier-based outcomes are obtained. To further improve classification performance, we applied the iterative majority voting (IMV) technique to automatically select the best result.Our DQP-based model achieved a classification accuracy of 95.68 %using EEG language dataset with leave-one-subject-out (LOSO) cross-validation strategy. Also, an explainable feature engineering (XFE) structure of DQP-LangNet is employed to obtain channel-specific explainable results. Our proposed DQP-LangNet model can be employed for other applications in neuroscience.

Revolutionizing Education: Harnessing AI for Personalized Learning Pathways and Student Success

The integration of Artificial Intelligence (AI) into education is rapidly transforming traditional learning environments by enabling personalized learning experiences. This paper explores the role of AI in developing adaptive learning systems that tailor educational content and pathways to individual students' needs. By leveraging machine learning algorithms and learning analytics, AI-driven platforms dynamically adjust instructional materials based on students’ performance, cognitive abilities, and engagement levels. The study examines the potential of AI to enhance student outcomes by creating more interactive and engaging educational experiences, thus fostering better retention and comprehension. Key challenges, such as data privacy, algorithmic bias, and equitable access, are also addressed to ensure the effective deployment of AI in diverse learning environments. Through case studies and real-world examples, this research highlights how AI is reshaping education by making it more flexible, student-centered, and accessible.

An intelligent learning system based on electronic health records for unbiased stroke prediction

Stroke has a negative impact on people’s lives and is one of the leading causes of death and disability worldwide. Early detection of symptoms can significantly help predict stroke and promote a healthy lifestyle. Researchers have developed several methods to predict strokes using machine learning (ML) techniques. However, the proposed systems have suffered from the following two main problems. The first problem is that the machine learning models are biased due to the uneven distribution of classes in the dataset. Recent research has not adequately addressed this problem, and no preventive measures have been taken. Synthetic Minority Oversampling (SMOTE) has been used to remove bias and balance the training of the proposed ML model. The second problem is to solve the problem of lower classification accuracy of machine learning models. We proposed a learning system that combines an autoencoder with a linear discriminant analysis (LDA) model to increase the accuracy of the proposed ML model for stroke prediction. Relevant features are extracted from the feature space using the autoencoder, and the extracted subset is then fed into the LDA model for stroke classification. The hyperparameters of the LDA model are found using a grid search strategy. However, the conventional accuracy metric does not truly reflect the performance of ML models. Therefore, we employed several evaluation metrics to validate the efficiency of the proposed model. Consequently, we evaluated the proposed model’s accuracy, sensitivity, specificity, area under the curve (AUC), and receiver operator characteristic (ROC). The experimental results show that the proposed model achieves a sensitivity and specificity of 98.51% and 97.56%, respectively, with an accuracy of 99.24% and a balanced accuracy of 98.00%.

Synaptic weight dynamics underlying memory consolidation: Implications for learning rules, circuit organization, and circuit function

Systems consolidation is a common feature of learning and memory systems, in which a long-term memory initially stored in one brain region becomes persistently stored in another region. We studied the dynamics of systems consolidation in simple circuit architectures with two sites of plasticity, one in an early-learning and one in a late-learning brain area. We show that the synaptic dynamics of the circuit during consolidation of an analog memory can be understood as a temporal integration process, by which transient changes in activity driven by plasticity in the early-learning area are accumulated into persistent synaptic changes at the late-learning site. This simple principle naturally leads to a speed-accuracy tradeoff in systems consolidation and provides insight into how the circuit mitigates the stability-plasticity dilemma of storing new memories while preserving core features of older ones. Furthermore, it imposes two constraints on the circuit. First, the plasticity rule at the late-learning site must stably support a continuum of possible outputs for a given input. We show that this is readily achieved by heterosynaptic but not standard Hebbian rules. Second, to turn off the consolidation process and prevent erroneous changes at the late-learning site, neural activity in the early-learning area must be reset to its baseline activity. We provide two biologically plausible implementations for this reset that propose functional roles in stabilizing consolidation for core elements of the cerebellar circuit.

Cluster partition-fuzzy broad learning-based fast detection and localization framework for false data injection attack in smart distribution networks

The distributed renewable energy generations, as accessible and easily targets for attackers, introduce an extra false data injection attack (FDIA) threat in the smart distribution networks. Scattered attack points and complex attack features hinder the elimination of potential threats. In this context, an FDIA fast detection and pinpoint localization framework is proposed. This framework identifies abnormal signals and attacked nodes from the unique topology structure and status contiguity of smart distribution networks, namely, spatial-temporal correlations of power grids, by using a cluster partition-fuzzy broad learning system (CP-FBLS). Unlike most existing FDIA detection methods, which are dedicated to high accuracy but neglect the urgent need for rapid detection in smart distribution networks, the proposed CP-FBLS framework maintains the fast computational nature of a fuzzy broad learning system (FBLS), while avoiding the accuracy degradation caused by high-dimension of data in large-scale smart distribution networks. Moreover, the multi-layer structure of the proposed framework recognizes the location of FDIA, bridging the research gap of attack localization. To comprehensively evaluate the proposed strategy, datasets containing various FDIA types are constructed. Numerical simulations based on the above datasets in modified IEEE 34-bus and 123-bus distribution systems are implemented. The results of the case studies showed that the proposed method can achieve 98.43 % accuracy with 0.34 ms detection time, realizing rapid detection and localization of various FDIAs with satisfactory accuracy.

Physics-supported Bayesian machine learning for chatter prediction with process damping in milling

Chatter stability of milling operations is a complicated phenomenon causing serious productivity issues in the manufacturing industry, yet a shop-floor implementable solution is lacking. This paper follows a physics-supported Bayesian machine learning approach and incorporates the potential effect of process damping on the stability of the process. Using a likelihood function based on the Nyquist stability criterion, the learning system monitors the actual stability state of the process during arbitrary cuts and refines the underlying model parameter uncertainties in the structural dynamics, cutting force coefficients, as well as the process damping. The framework can operate with limited training data and display the remaining uncertainties in stability predictions to the machine operator. Experimental case studies show the effectiveness of the proposed method and highlight the importance of considering process damping for certain endmills.

Peningkatan Hasil Belajar Siswa dengan Sikap Kritis Materi Sifat-Sifat Gereja Melalui Model Problem Based Learning (PBL) pada Siswa Kelas XI Fase F SMK Negeri 1 Lahomi Kabupaten Nias Barat

Study This aiming For increase results Study student eye lesson Catholic Religious Education and Character​ through the Problem Based Learning model . Research action class This implemented as many as 2 cycles with four stages namely : planning, implementation, observation, reflection. Subject study This is student Class XI of State Vocational School 1 Lahomi as many as 6 students. Research This use technique analysis descriptive qualitative . Indicator action study This is a learning system finished namely achievement KKM value ≥75. Success Study measured if every student has reach value ≥75 then it is said succeed finished And in a way classical if as many as 80% of students has reach value ≥75 then it is said finished in a way classical. Problems in study This is lack of activity student For Study especially on moment do questions practice , difficulty student understand material the lesson in progress discussed , students not enough brave disclose opinion And Not yet capable think critical. For minimize problem said, researchers carry out study action class with objective For know improvement results Study student on eye lesson Catholic Religious Education and Character​ with material main Properties Church on class XI phase F SMK 1 Lahomi Regency West Nias Year 2024/2025 lessons through problem based learning model . Based on test results learning gained​ on end cycle I with average results Study student of 16.67 with percentage completeness which is 33.34%. Meanwhile test results Study on cycle II experienced significant improvement​ with average results obtained Study of 98.33 with percentage completeness namely 98.33% with category proficient . With thus, then can concluded that improvement results Study student with attitude critical material properties Church through the problem based learning (PBL) model student class XI phase F SMK 1 Lahomi regency West Nias Year 2024/2025 learning results change Study students who can proven with achievement Criteria Minimum Completion Criteria (KKM) that has been achieved set .

Accurate measurement of key structures in CBD patients using deep learning

The prevalence of common bile duct (CBD) stones presents a widespread and intricate clinical challenge that necessitates precise measurements for effective clinical management. In response, we developed a practical deep learning system designed to precisely measure key structures within endoscopic retrograde cholangiopancreatography (ERCP) images in CBD patients. This system excels in detecting key structures in CBD patients and provides comprehensive information about the quantity, size, location, and dimensions of the common bile duct. While the size, location, and dimensions of the CBD are derived from the scaling ratio of the estimated duodenoscope diameter, the quantity of stones is determined independently of this conversion between pixels and millimeters. A rigorous evaluation of the system’s performance was conducted utilizing diverse metrics such as the Dice coefficient, accuracy, area under the receiver operating characteristic (AUROC) curve, and a comprehensive confusion matrix. A dataset of 418 images from 403 patients was divided into three sets: 314 images (75%) from 299 patients (74%) for training, 52 images (12.5%) from 52 patients (13%) for validation, and 52 images (12.5%) from 52 patients (13%) for testing. The proposed system demonstrated remarkable results in key structure detection and measurement, with Dice coefficients and overall AUROCs of 95.6% and 95.0% for the duodenoscope, 94.6% and 94.0% for the bile duct, and 94.1% and 92.0% for stone identification, respectively. The system’s timely, precise measurements of CBD structures can significantly enhance clinical decision-making, improve patient outcomes, and streamline diagnostic and management processes.

Challenges of employing unformalized skills: Recognizing and leveraging informal expertise for national development in Tanzania

This paper explores the challenges and implications of informal skills acquisition in Tanzania and its impact on national development. Through two case studies—one focusing on a driving class and the other on an electrical technician—it highlights the tension between formal certification and practical experience. The findings reveal that informal skilled laborers, while highly proficient, face significant barriers to employment, market competition, and project tendering due to the absence of formal recognition and certification. Despite their lack of formal qualifications, these workers contribute meaningfully to their families, communities, and the national economy. The study underscores the critical gap in valuing informal expertise relative to formal qualifications, and it advocates for the re-evaluation of the certification system to better recognize and leverage informal skills. The paper argues that integrating informal skills into the formal economy is crucial for inclusive national development. It calls for a Recognition of Prior Learning (RPL) system to validate experiential learning and create alternative, affordable certification pathways. This approach would not only level the playing field for informal workers but also enhance their ability to compete in national projects alongside formally educated professionals. By acknowledging and supporting informal expertise, Tanzania can build a more robust, versatile workforce that is better equipped to meet the demands of various industries, ultimately driving sustainable national development.

Integrating Building Information Modelling and Artificial Intelligence in Construction Projects: A Review of Challenges and Mitigation Strategies

Artificial intelligence (AI), including machine learning and decision support systems, can deploy complex algorithms to learn sufficiently from the large corpus of building information modelling (BIM) data. An integrated BIM-AI system can leverage the insights to make smart and informed decisions. Hence, the integration of BIM-AI offers vast opportunities to extend the possibilities of innovations in the design and construction of projects. However, this synergy suffers unprecedented challenges. This study conducted a systematic literature review of the challenges and constraints to BIM-AI integration in the construction industry and categorise them into different taxonomies. It used 64 articles, retrieved from the Scopus database using the PRISMA protocol, that were published between 2015 and July 2024. The findings revealed thirty-nine (39) challenges clustered into six taxonomies: technical, knowledge, data, organisational, managerial, and financial. The mean index score analysis revealed financial (µ = 30.50) challenges are the most significant, followed by organisational (µ = 23.86), and technical (µ = 22.29) challenges. Using Pareto analysis, the study highlighted the twenty (20) most important BIM-AI integration challenges. The study further developed strategic mitigation maps containing strategies and targeted interventions to address the identified challenges to the BIM-AI integration. The findings provide insights into the competing issues stifling BIM-AI integration in construction and provide targeted interventions to improve synergy.

A-179 Improving specimen labeling errors in the pediatric emergency department at a tertiary care hospital

Abstract Background Specimen labeling errors, such as unlabeled, mislabeled, or sample-requisition mismatches, pose risks to patient care. Correcting these errors, particularly for pediatric patients, is crucial to avoid unnecessary stress to the family and time spent on follow-up. In 2018, the BCCH Emergency Department (ED) reported the highest labeling error rates among all departments at our tertiary care hospital, accounting for approximately 30% of all errors. Methods To address this issue, a quality improvement project following the Model for Improvement framework was initiated in 2019 at the ED. An interdisciplinary team approach was adopted initially, and a positive patient identification system (PPID) was implemented later in February 2022. Patient Safety Learning System (PSLS) data from January 2019 to November 2023 were analyzed to evaluate the impact of these interventions. Results 1763 (22.6%) of the 7802 PSLS events recorded during this period were sample labeling errors. Ward collections accounted for the majority of labeling errors compared to laboratory collections (20.3% vs 2.4%). In the ED, most of the labeling errors occurred with unspecified samples (25%), swabs (24%), urine (23%), and blood culture (19%). The interdisciplinary team intervention initially reduced labeling errors, but its impact was inconsistent. However, the implementation of PPID in 2022 led to a significant and consistent decrease in labeling errors in the ED (4.51 errors/month vs 1.14 errors/month; p = 0.01) (Figure. 1). Conclusions While the implementation of PPID has notably improved labeling errors in the ED, complete elimination remains challenging. In addition, more specific PSLS filing categories are needed to capture sources of labeling error from unspecified samples. Continued efforts are necessary to achieve the goal of zero labeling errors.

Humanistic place leadership, cultural enhancement, and change in industrial districts

Drawing inspiration from the works of Fiorenza Belussi on multiple paths of development in local productive systems, the essay contributes to the ongoing debate on factors facilitating such transformations, focusing primarily on industrial districts (IDs) as open learning systems. With specific reference to the Italian context, this paper recalls the delicate balance between socio-cultural and economic elements within IDs transitioning towards new paths of development, crucial for facing the current technological and societal challenges. e essay proposes to look at the possible diffusion of humanistic management (HM) and cultural enhancement of economic strategies, as new triggering factors against lock-in conditions and tendencies toward degradation during phases of ID transformation. HM prioritizes the well-being of individuals and ethical decision-making, aligning with a broader societal purpose. Cultural enhancement, rooted in a common heritage, supports place identity, and suggests bases for multiple specializations. e convergence of HM and cultural enhancement within IDs may offer to renewed place-leaderships levers for strategies of transition pointing to sustainable paths that align with broader societal goals in an era of disrupting challenges.

Multi-label semantic segmentation of magnetic resonance images of the prostate gland

Prostate segmentation is a substantial factor in the diagnostic pathway of suspicious prostate lesions. Medical doctors are assisted by computer-aided detection and diagnosis systems systems and methods derived from artificial intelligence deep learning-based systems have to be trained on existing data. Especially freely available labeled prostate magnetic resonance imaging data is rare. With regard to this problem, we show a method to combine two existing small datasets to form a bigger one. We present a data processing pipeline consisting of a cascaded network architecture that is able to perform multi-label semantic segmentation, hence, capable of classifying each pixel in a T2\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\ extrm{T}_{2}$$\\end{document}-weighted image not only to one class but to a subset of the prostate zones and classes of interest. This delivers richer information such as overlapping zones that are key to medical radiological examination. Additionally, we describe how to integrate expert knowledge in our deep learning system. To increase data variety for training and evaluation we use image augmentation on our two datasets—a freely available dataset and our new open-source dataset. To combine the datasets we denoise the contourings in our dataset by using an effective yet simple algorithm based on standard computer vision methods only. The performance of the presented methodology is compared and evaluated using the dice score metric and 5-fold cross-validation on all datasets. Although we trained on tiny datasets our method achieves excellent segmentation quality and is even able to detect prostate cancer. Our method to combine the two datasets reduces segmentation errors and increases data variety. The proposed architecture significantly improves performance by including expert knowledge via feature-map concatenation. On the initiative for collaborative computer vision benchmarking dataset we achieve on average dice scores of approximately 91% for the whole prostate gland, 67% for the peripheral zone and 75% for the prostate central gland. We find that image augmentation except contrast limited adaptive histogram equalisation did not have much influence on the segmentation quality. Derived and enhanced from existing methods we present an approach that is able to deliver multi-label semantic segmentation results for prostate magnetic resonance imaging. This approach is simple and could be applied to other applications of deep learning as well. It improves the segmentation results by a large margin. Once tweaked to the data, our denoising and combination algorithm delivers robust and accurate results even on data with segmentation errors.

Compulsory Training in Post-pandemic Times - Presence, Digital, Hybrid & Co. Using the Example of the University Hospital of Würzburg

With the end of the pandemic in April 2023 the whole learning environment had run through some major changes. The act of normal teaching was not that easy anymore during the pandemic. The size of a class had to be cut down to a size that fits the hygienic guidelines. Therefore, methods like blended learning and virtual classrooms came to life stronger than before. The methods were not new but never before needed in such a strong matter. Due to the experience with these methods of education, teaching and learning stayed different. Topics like virtual reality and even augmented reality stayed and are to some point part of the educational system nowadays. Due to the pandemic, it may seem that these new kinds of methods are worse compared to the ones we are used to but several studies can prove the opposite.The University Hospital Würzburg (UKW) has adapted to these kinds of changes. Every specialist training was formed to fit the necessary changes during the pandemic. It was not easy to keep up good practical education with the correct distance measurement, open windows for fresh air every 45 minutes or an FFP2 mask for protection. The rest of the close to 190 trainings held in the rooms of the academy of the UKW had to be cancelled or cut down to a group size allowed. Most of the known digital teaching and learning system that were installed at that time remained and are in use.

A bibliometric review of flipped classroom approaches for enhanced students’ language learning

The widespread use of contemporary flipped classrooms for teaching, especially in language learning, provides new insights. Flipped classrooms are two important aspects of language learning that produce impactful outcomes and understanding. It is essential that these processes be used to strengthen the system of language learning. We used PRISMA statement templates for bibliometric analysis. The analysis is based on publications from 2014 to 2023. We examined and analyzed 309 databases from Scopus to gain insights into flipped classrooms and language learning. Concurrently, distinct inclusion and exclusion standards were applied for the screening process. Through a bibliometric analysis, we have determined the distribution of publications on the impact of flipped classrooms on language learning between 2014 and 2023. Additionally, we have identified the most significant journals and authors, the prominent countries engaged, the primary research keywords, and the key subject areas explored. This study illustrates trends and suggestions for flipped classroom and language learning. Suggestions are also made to improve students’ academic performance by making full use of the effects of flipped classroom on language learning.