Online Learning Mechanism Research Articles

Deep Learning-Based Real-Time Data Quality Assessment and Anomaly Detection for Large-Scale Distributed Data Streams

Time delay and data quality degradation pose significant challenges in large-scale distributed data streams processing. This paper proposes a deep learning-based realtime data quality assessment and anomaly detection method for distributed streaming data environments. The proposed approach integrates quality-aware feature extraction with adaptive deep neural networks to enable real-time quality monitoring and anomaly detection. A multi-dimensional quality assessment framework is developed, incorporating temporal-spatial correlations and stream characteristics for comprehensive quality evaluation. The system implements a distributed architecture with parallel processing capabilities, enabling scalable operations across multiple nodes while maintaining low-latency responses. A novel online learning mechanism is introduced to adapt model parameters dynamically, ensuring robust performance under evolving data patterns. Experimental evaluation conducted on three large-scale datasets, including industrial IoT sensors (2.5TB), network traffic (1.8TB), and financial transactions (3.2TB), demonstrates superior performance compared to traditional methods. The system achieves 97.8% detection accuracy while maintaining processing latency below 10ms, with linear scalability up to 128 nodes. Results show consistent performance improvement across different operational scenarios, with 95% precision in anomaly detection and throughput exceeding 1.2 million events per second.

Learning Lithium-Ion Battery Health in Real-Time

Health-conscious decision-making by battery management systems (BMS) is crucial for the safe operation of lithium-ion batteries. Traditional battery models often struggle to accurately predict the dynamic behavior of lithium-ion batteries over time due to the degradation of their health. The Electrochemical-Thermal-Aging (ETA) model, which integrates single-particle, thermal, and aging models, is comprehensive but computationally intensive and challenging to implement on-board a BMS. Recent advancements in physics-informed neural networks (PINNs) provide a promising solution by approximating the solution to the ETA dynamics. Additionally, PINNs offer a significant advantage by avoiding the need for discretization when solving partial differential equations (PDEs) based on ETA models. However, PINNs face difficulties in predicting the future behavior of lithium-ion cells when confronted with new, unseen data resulting from accelerated degradation, internal faults, or abusive conditions.Our recent work addresses this challenge by developing a novel approach that enhances PINNs with online learning capabilities. By embedding the PINN with the ability to learn continuously, we enable it to adapt to health degradation, maintaining its predictive accuracy over the battery's lifespan. We leverage dynamical system theory to develop the online training algorithm to update the weights. The deviation between the model's predicted output and the measured values is used to adjust the weights of the PINN's final layer online, ensuring that the model's predictions are continuously refined based on the latest data. An essential aspect of our approach is providing analytical guarantees for the convergence of the neural network's weight estimation error and the model output error. Using Lyapunov theory for dynamical systems, we ensure that the adjustments made during online learning lead to a stable and convergent solution. This theoretical foundation is crucial for validating the reliability and robustness of the proposed online learning mechanism. To demonstrate the effectiveness of our online learning scheme, we present simulation results that highlight the performance improvements achieved. These simulations show that the enhanced PINN can accurately track the battery's health status over time, adapting to changes in internal parameters and external conditions. The results also illustrate the computational efficiency of the proposed method, making it feasible for on-board implementation in real-world BMS applications.In summary, the poster will present our novel approach to embedding PINNs with online learning capabilities, which represents a significant advancement in battery management systems. By enabling real-time adaptation to battery health degradation, our method ensures more accurate predictions, safer operation, and a prolonged battery lifespan. The integration of dynamical system theory, online learning mechanisms, and theoretical guarantees provides a robust and reliable solution for next-generation BMS applications. Through this work, we aim to pave the way for more intelligent, adaptable, and health-conscious battery management systems.

A Literature Review on Digital Creativity in Higher Education—Toward a Conceptual Model

The present study relies on the assumption that in today’s knowledge society and the use of online knowledge, learning, and communication mechanisms, pre-university and university teachers, students of all ages, those preparing for teaching careers or other careers, parents, managers, and administrators have become increasingly interested in acquiring proficiency, capitalization, and stimulation of digital creativity. As such, research must provide, primarily for teachers and those preparing for this career, systematic results on digital creativity. This study aims to identify the current state-of-the-art in scientific knowledge on the comprehensive topic of digital creativity: conceptual frameworks and theoretical highlights, elaborating a model of digital creativity, and identifying and formulating future research opportunities. The investigation relies on the key concept of digital creativity and aims to cover two aspects: 1. basic meanings of the concept of digital creativity, specifying its applicability in the educational process at pre-university and university levels, and 2. identifying a set of concepts that are complementary to digital creativity. The research is a conceptual literature review, implying the use of literature review and bibliographic analysis methods. The study relies on a comprehensive literature review of papers selected from the three international databases WEB of SCIENCE, Science Direct, and Scopus from 2020 to 2024. The literature search enabled us to identify an initial sample of 1848 studies in the three databases and finally select 31 relevant articles. The analysis of the relevant papers allowed us to formulate conclusions related to the two investigated aspects: 1. The concept of digital creativity has two types of definitions—explicit and implicit; 2. The concept of digital creativity is correlated with other complementary concepts: digital competence, mobile networking creativity, multi-team digital creativity, the use of artificial intelligence, and digital innovation. Our findings support us in proposing ways to effectively use and stimulate digital creativity in the educational process and build 21st-century skills and competences for students of all ages and at all levels of education.

Online learning experiences of female students living with albinism amidst a global pandemic: evidence from South Africa and Zimbabwe

Our paper explores the online learning experiences of two female students living with albinism in South Africa and Zimbabwe during the COVID-19 era. Using a two-case qualitative research design, we conducted in-depth face to face interviews with the purposefully selected sample. The intention was to bring to the fore, the unique online experiences of female students living with albinism. Data were thematically analyzed. Literature suggests that the unique health and physiological characteristics of students living with albinism tend to present unique but surmountable challenges. Our study established that the gender and physiology of the students tended to complicate or enhance their desire to benefit from online learning during the COVID-19 era. The findings make a case for academic inclusion in universities’ academic projects. Institutions May consider instituting pedagogically inclusive online learning mechanisms that are not only gender sensitive, but are in sync with, not only the COVID-19 realities, but similar calamities.

Principle of Educational Management in Hybrid Learning

Hybrid learning occasioned by the introduction of the traditional face to face instruction with online learning mechanisms, has developed as a prominent educational approach in response to the developing landscape of teaching and process of learning. The professional management of hybrid learning ecosystems is important for ensuring best students’ engagement, the outcomes of learning, and the general success. The study investigates the principles of educational management of hybrid learning environments highlighting basic consideration and methods for the management. Based on the established principles of educational management, such as effective communication, support system, assessment and curriculum patterns, this study investigates how these principles apply in the realm of hybrid learning. In addition to that, the study explores the contribution of technology in helping management processes and improving student experiences in hybrid environments. Understanding and adopting the tenants of educational management in hybrid learning setups, educators can come up with dynamic, inclusive, and effective learning experiences that satisfies the diverse demands of the students and promote academic excellence

Shifting to Open and Distance Learning (ODL) During Pandemic Era: Learning Experiences of the Undergraduate Students at UiTM Cawangan Kelantan

The outbreak of COVID-19 has urged the education providers to transform their method of teaching and learning into Open and Distance Learning (ODL). The sudden implementation of these online learning may cause some negative experiences for students as there are several issues that may impede them during the learning process. Hence, the aim of this study is to assess the university students’ learning experiences on ODL implementation during pandemic era using UiTM Cawangan Kelantan as a case study. A total of 2217 students participated in the study to share their experiences about ODL learning experiences. The data were analysed using descriptive statistics techniques to check on the students’ perceptions, the challenges of ODL faced by them and their most preferred online platform for ODL. The findings reveal that majority of the students gave neutral response on the effectiveness of ODL implementation. As overall, most of the students preferred live chat such as WhatsApp and Telegram as a platform for the learning process. Some negative feedback were given by the students such as not effective for the students with limited facilities, problem with internet coverage and difficult to adapt and perform well through ODL implementation. The findings of the study can be used by the higher education providers to improvise the online learning mechanism in future.

A novel online learning-based linear quadratic regulator for vanadium redox flow battery in DC microgrids

This paper proposes a novel learning-based linear quadratic regulator (LQR) to overcome the long-lasting problem of model dependency in the existing vanadium redox flow battery (VRB) control approaches. Compared to the conventional model-dependent control methods, such as PI control and model predictive control (MPC), the proposed method automatically updates the optimal control policy through the online learning mechanism without any knowledge of the VRB system dynamics. The ability of the proposed method to handle uncertainties is verified by simulations under various scenarios.

HyperDE: An Adaptive Hyper-Heuristic for Global Optimization

In this paper, a novel global optimization approach in the form of an adaptive hyper-heuristic, namely HyperDE, is proposed. As the naming suggests, the method is based on the Differential Evolution (DE) heuristic, which is a well-established optimization approach inspired by the theory of evolution. Additionally, two other similar approaches are introduced for comparison and validation, HyperSSA and HyperBES, based on Sparrow Search Algorithm (SSA) and Bald Eagle Search (BES), respectively. The method consists of a genetic algorithm that is adopted as a high-level online learning mechanism, in order to adjust the hyper-parameters and facilitate the collaboration of a homogeneous set of low-level heuristics with the intent of maximizing the performance of the search for global optima. Comparison with the heuristics that the proposed methodologies are based on, along with other state-of-the-art methods, is favorable.

The Influence of Online Learning Behavior on Learning Performance

Online education is a significant part of information education. It is an effective way to uncover online learning mechanisms and improve the quality of online teaching by exploiting the behavioral data of online learning platforms for learning performance prediction and analysis. In this paper, we focus on the learner’s learning behavior in an online teaching scenario and explore the predictive effectiveness and impact mechanism of each behavioral feature by building a predictive model based on a machine learning algorithm. Experimental results show that three behavioral characteristics, namely the number of visits to course materials, lecture review time, and assignment, intensively influence learning performance. By comparing various machine learning algorithms, it is found that the random forest algorithm has better prediction results.

An Intelligence-Defined Networking Architecture With Importance-Based Network Resource Control

As network complexity increases because of the diversification of applications and services based on Internet of Autonomous Things (IoAT), it is difficult for humans to design the optimal control rule for software-defined networking (SDN) controllers. Intelligence-defined networking, called IDN, is proposed to overcome this limitation through machine learning (ML) algorithms. Since the existing IDN approaches are mostly designed to optimize only the network Quality of Services (QoS), including throughput, jitter, and latency, the controllers do not consider the importance of data in the applications and the services. This causes the controller to allocate insufficient resources to the crucial data flow, which leads critical problems, such as self-driving car accidents. To prevent this problem, we propose an importance-based IDN architecture that enables network controllers to manage network traffic with an importance levels (ILs) of data flows. First, we devise an importance estimation scheme to set the IL for the flows. Second, a dynamic resource allocation model of the controllers is developed by means of deep learning algorithms in order to make the optimal network resource. Additionally, an online learning mechanism based on weighted auto-labeling is adopted to continue enhancing the adaptability of the resource allocation model on runtime as the network conditions change. The evaluation results of the proposed architecture under various autonomous things scenarios show that the loss rate for data flows of higher importance is reduced by one-quarter compared to the case of a network controller without IL and that the bandwidth waste ratio is reduced by 10% compared to the rule-based model.

UMN Sims: Development of Student Life Simulation Game to Introduce Onsite Campus Activities using React.js

After two years of the implementation of the online learning mechanism due to the Covid-19 pandemic, Universitas Multimedia Nusantara, a private university in Tangerang, Indonesia, has again implemented an onsite learning mechanism. Based on the initial survey conducted, it was found that many students were not aware of various academic and non-academic activities that could be carried out on campus and its surroundings. Therefore, in this paper, we try to make a simulation game to introduce students to various activities that can be done on campus and its surroundings. Simulation games are developed mainly using React.js with the Axioos library to perform HTTP requests to access web services. We evaluated this game using the Hedonic-Motivation System Adoption Model or HMSAM to determine the level of player satisfaction. We conducted experiments on high school students, freshmen and sophomores and found that high school students' HMSAM scores were higher, assuming high school students had a curiosity about life as university students. Behavioral intention to use and immersion obtained using HMSAM were 80.25% and 80.31%, respectively, this means that these games strongly agree that respondents are interested in playing UMNSims and playing it immersively. Keywords: HMSAM, ReactJS, Simulation Game, university, web-based

QoS-aware congestion control with online learning

In emerging applications such as industrial control and autonomous driving, end-to-end deterministic quality of service (QoS) transmission guarantee has become an urgent problem to be solved. Internet congestion control algorithms are essential to the performance of applications. However, existing congestion control schemes follow the best-effort principle of data transmission without the perception of application QoS requirements. To enable data delivery within application QoS constraints, we leverage an online learning mechanism to design Crimson, a novel congestion control algorithm in which each sender continuously observes the gap between current performance and pre-defined QoS. Crimson can change rates adaptively that satisfy application QoS requirements as a result. Across many emulation environments and real-world experiments, our proposed scheme can efficiently balance the different trade-offs between throughput, delay and loss rate. Crimson also achieves consistent performance over a wide range of QoS constraints under diverse network scenarios.

Online time-series forecasting using spiking reservoir

IoT-based automated systems require efficient online time-series analysis and forecasting and there is a growing requirement to enable such processing at the low-cost constrained edge devices. Classical approaches such as Online Autoregressive Integrated Moving Average (Online ARIMA), Seasonal ARIMA (SARIMA) etc. and Artificial Neural Network (ANN) based techniques including Long-short Term Memory (LSTM) do not cater to this niche requirement due to their memory and computation power requirements. Neuromorphic computing and bio-plausible spiking neural networks, being both data and energy efficient, may offer a better solution. In this work, a novel spiking reservoir based network is proposed for online time series forecasting that relies on temporal spike encoding with a feedback driven online learning mechanism. The proposed network is capable of avoiding rapidly fading memory problem. The prediction accuracy of the network (tested on nine time-series datasets) outperforms conventional methods like SARIMA, Online ARIMA, Stacked LSTM, achieving up to 8% higher R2 score while using negligible buffer memory.

A META-SYNTHESIS STUDY ON ADAPTATION OF ACADEMIC AND RESEARCH MANAGEMENT DURING THE COVID-19 PANDEMIC IN PUBLIC UNIVERSITIES

The ability of a tertiary institution to pursue academic studies and research during the pandemic and now endemic covid-19 is a measure of the strength of the university or college in facing the biggest test of this decade. Thus, various measures and approaches have been taken by the university management, especially in ensuring the momentum of study operations and even research at the university does not stop just like that. However, the biggest challenge seen and felt by all parties including students as the most important stakeholders is the online teaching and learning mechanism. Similarly, research planning and the ability of researchers are seen as very critical, especially in the success of data collection activities with the community and matters related to the study location. This meta-synthesis study potrayed the continuous efforts of the university in adapting and at the same time maintaining the quality of knowledge and skills taught are so significant and become the reason for the development of academic activities in crisis situations especially covid-19 pandemic. Studies from 2020 to 2021 were referred specifically to review teaching and learning situations as well as research during pandemics and endemics show that academics and researchers have an optimistic and dynamic attitude in adapting their respective teaching and research approaches.

Reform and Intelligent Innovation Path of College Football Teaching and Training Based on Mixed Teaching Mode

Computers and network information technology are changing our production and lifestyles and at the same time profoundly affecting the development of education. Higher education informatization focuses on the integration of information technology and higher education. It is necessary to comprehensively deepen education reform and innovate teaching methods and promote the overall improvement of education. This article aims to explore the innovative path of teaching reform based on college football training based on the current situation of college football teaching. Using literature research and questionnaire surveys, physical education teaching focuses on the unification of theory and practice. Both are indispensable. Mixed teaching is more in line with the teaching laws of physical education disciplines. It inherits the advantages of e-learning (i.e., the Internet online learning mechanism). Free from the constraints of a single online learning mode, this article designs a football training reform and innovative teaching experiment, using smart sensor technology to detect the body posture control of students participating in football teaching and training, and using questionnaires to track the nutritional adjustment of the students participating in the experiment. Through the football training scores and physical fitness index data of students in the teaching experiment, the application effect of football teaching and training reform in this research is analyzed. Experiments show that with a 95% confidence level, the football teaching and training reform program of this research can improve the football training effect of students. Colleges and universities currently pay more attention to football catching and offensive practice, especially the offensive practice accounts for 68.2%. Therefore, it has important practical value and theoretical significance to study the mixed teaching effect of using MOOC resources in football teaching. The use of blended teaching can cultivate students’ autonomous learning ability, make students transform from passive learning to active learning, stimulate students’ learning motivation, and enable students to have a strong interest in football.

Learning from home: the impact of COVID-19 on teaching and learning

PurposeWith the impact of COVID-19, the educational system across the world had to be reviewed and readapted. Both learners and tutors were forced to adopt the online teaching and learning mechanism. Learners had to cope with the drastic teaching mode. In all of these, the student’s level of satisfaction remains pivotal. Teaching and learning remain successful if the students are satisfied and engaged. Therefore, this study aims to identify and assess factors that influenced students’ level of learning from home satisfaction.Design/methodology/approachThis study followed a mixed-method approach. Online focus groups were arranged to devise indicators for factors like the accessibility of lecturers, support from university and conducive home environment. A questionnaire was designed and disseminated through an online survey. A response rate of (N = 169) was received. The proposed framework was tested in two phases: confirmatory factor analysis and partial least square structural equation modeling.FindingsThe findings revealed that education/life balance and learners’ commitment have a positive and significant relationship with learning from home satisfaction. Accessibility of lecturers and a conducive home environment positively influenced education and life balance. Learners’ commitment was influenced by the accessibility of lecturers, education and life balance and support from the university. Support from university was positively influenced by the accessibility of lecturers and a conducive home environment.Originality/valueVarious studies focused on the quality of online teaching and learning, and very few studies paid attention to the day-to-day lives of learners at the tertiary level. This study has borrowed organizational factors and adapted them to the students’ lives with two theoretical foundations which will enable a better understanding of the students.

EFEKTIVITAS PEMBELAJARAN DARING DI PROGRAM STUDI PENDIDIKAN BAHASA DAN SASTRA INDONESIA MENGGUNAKAN APLIKASI MOODLE (ELEARNING FKIP UNPAS)

This research was motivated by learning affected by the Covid-19 pandemic at the Indonesian Language and Literature Education Study Program (PBSI), Faculty of Teacher Training and Education (FKIP), Pasundan University (Unpas) Bandung using the Moodle application (e-learning FKIP Unpas). This study aims to determine the mechanism for implementing the online learning process carried out at PBSI FKIP Unpas, to determine student satisfaction in carrying out online learning using the Moodle application, and to identify the effectiveness of online learning implemented at PBSI FKIP Unpas. The research method used is descriptive research with a qualitative approach that focuses on ongoing events and the results will be analyzed. The results of this study, namely the online learning mechanism applied at PBSI FKIP Unpas, were compiled and implemented systematically, starting from student registration as course participants to the actual learning process. After analyzing the survey results, most of the students were satisfied with the online learning mechanism of PBSI FKIP Unpas. Respondents assessed that online learning using the Moodle application (e-learning FKIP Unpas) was very effective at 23%, effective at 54%, and less effective at (16%). However, there were a small number of students who rated it as ineffective, namely 7%, while those who rated it very ineffective were 0%. Based on this, online learning using the Moodle application (E-Learning FKIP Unpas), is effectively applied in the Indonesian Language and Literature Education Study Program FKIP Unpas.

Optimal control method of HVAC based on multi-agent deep reinforcement learning

In HVAC control problems, model-based optimal control methods have been extensively studied and verified by many researchers, but they highly depend on the accuracy of the model, a large amount of historical data, and the deployment of different sensors. With respect to the above problems, this paper proposed a Multi-Agent deep reinforcement learning method for the building Cooling Water System Control (MA-CWSC) to optimize the load distribution, cooling tower fan frequency, and cooling water pump frequency of different types of chillers, which provide a model-free and online learning mechanism. In the learning process, unlike the traditional reinforcement learning methods, the proposed control method uses five agents to control different controllable parts by parallel learning, which can greatly reduce the action space and speed up the convergence rate. In order to verify the effectiveness of the propose-d control method, based on the actual building cooling water system parameters and related historical data, we construct an experimental building cooling water system model, and moreover, we test the MA-CWSC method, model-based control method (optimal control method), baseline method, and single-agent deep reinforcement learning (deep Q-network). The experimental results show that the energy saving performance of the proposed MA-CWSC method is significantly better than the rule-based control method (11.1% improvement), and is very close to that of the model-based control method (only 0.5% difference). In addition, the MA-CWSC method has a faster learning rate compared to the deep Q-network (DQN) control method.

Design of Intelligent Solar PV Power Generation Forecasting Mechanism Combined with Weather Information under Lack of Real-Time Power Generation Data

In order to reduce the cost of data transmission, the meter data management system (MDMS) of the power operator usually delays time to obtain the power generation information of a solar photovoltaic (PV) power generation system. Although this approach solves the problem of data transmission cost, it brings more challenges to the solar PV power generation forecast. Because power operators usually need real-time solar PV power generation as a basis for the power dispatch, but considering the cost of communication, they cannot always provide corresponding historical power generation data in real time. In this study, an intelligent solar PV power generation forecasting mechanism combined with weather information is designed to cope with the issue of the absence of real-time power generation data. Firstly, the Pearson correlation coefficient analysis is used to find major factors with a high correlation in relation to solar PV power generation to reduce the computational burden of data fitting via a deep neural network (DNN). Then, the data preprocessing, including the standardization and the anti-standardization, is adopted for data-fitting or real-time solar PV power generation data to take as the input data of a long short-term memory neural network (LSTM). The salient features of the proposed DNN-LSTM model are: (1) only the information of present solar PV power generation is required to forecast the one at the next instant, and (2) an on-line learning mechanism is helpful to adjust the trained model to adapt different solar power plant or environmental conditions. In addition, the effectiveness of the trained model is verified by six actual solar power plants in Taiwan, and the superiority of the proposed DNN-LSTM model is compared with other forecasting models. Experimental verifications show that the proposed forecasting model can achieve a high accuracy of over 97%.

Establishing Cognitive Strategies to Support Online English Language Learners’ Critical Thinking

Shifting to online-based teaching and learning has become the most obvious way to increase pedagogy sustainability in nations with the current implementation of an online learning environment. Research has shown that online learning and teaching in English language subjects influence the cognitive strategy environment. This research aims to determine and explore the cognitive strategy implemented in the online learning environment to support the critical thinking of English language learners. This study is a quantitative descriptive study accomplished online utilising survey methodologies. The sample selection method is simple random sampling. Primary data was collected in this study by distributing questionnaires to 115 respondents via the internet. Data was gathered via the completion of questions provided to all respondents in the survey. In addition, the data were evaluated in order to be characterised and described. The data analysis findings of English language lecturers’ surveys indicate that the cognitive strategy used during the online teaching situation was prosperous and profligate. The cognitive strategies necessitated implementing the online English language learners’ critical thinking included procedural, metacognitive, conceptual, and strategic. This research establishes that the cognitive strategy implementation of online learning is commonly advantageous during remote teaching. The survey indicated that just 73% of English lecturers considered cognitive strategy online platforms as beneficial as traditional meetings for supporting critical thinking. Positive responses were obtained from the questionnaire. In addition, the statement is relevant to their experience and background in online teaching, which indicates a good attitude toward the cognitive strategy that works well in their online learning mechanism. The research suggests that online pedagogy and English lecturers’ must be developed to pave the way for a potential future authentic online strategy. The research may illustrate the difficulties associated with online education and potential areas for improvement..