Optimal Online Learning Research Articles

PEMANFAATAN MEDIA SOSIAL DALAM OPTIMALISASI PEMBELAJARAN DARING DI DESA GUNUNGPRING

Social media as a brave learning media is something that is still rarely done by the people of Gunungpring Village. Therefore, this time the assistance was carried out in an effort to optimize online learning by using social media. This empowerment assistance uses the ABCD (Asset Base Community Development) approach using several stages, namely discovery, dream, define, design and destiny. Because in this new normal era, parents are minimizing expenses, then the author provides a solution by using social media which is free but has many advantages, namely telegram. With the assistance program in the form of a telegram for learning media, it is expected that children use their original cellphones to playing game, but now children are taught to use mobile phones to learn so that this potential can make children more effective in carrying out learning at home.

Website Development As An Online Classroom For Online Classes In Distance Learning

The spread of COVID-19 outbreaks makes learning activities in Indonesia become online learning. Online learning is followed by new problems, those are limited internet data, bad internet networks, and the use of many applications. Online learning in Indonesia must be immediately optimized because the COVID-19 outbreak is not over yet and there is a discourse that online learning will be applied permanently. The teacher can optimize online learning by building a website as an online classroom. If teachers build their websites, they can adjust the features on the website based on the online learning problems and the subject matter needs. Based on a questionnaire distributed, 83% of student respondents agreed if the teacher made a website. Besides, most of teacher respondents think the website can optimize online learning. The idea of the website as an online classroom is expected to contribute ideas and alternatives for online learning programs in Indonesia.

Development of Web-Based TOEFL Learning Media (E-TOEFL) Using Moodle at Universitas Bung Hatta

This research aims to produce a website-based TOEFL learning system (E-TOEFL) used for online TOEFL learning at Universitas Bung Hatta. The research method used in this research is a research and development (R&D) method. The steps taken in this study are the first stage that is to analyze the need to develop an E-TOEFL system-based learning management system (LMS), namely with observation and interview. Then carried out the design of this LMS-based E-TOEFL system. The next step is system testing by testing access speeds, link structures, and testing with a wide variety of browsers. The last step is the system implementation phase. In the implementation of E-TOEFL at Universitas Bung Hatta, some students have not participated in this E-TOEFL because TOEFL does design to be self-learning. So that in the future, it is necessary to apply blended learning to increase the enthusiasm of students in learning Toefl. More development also needs to be done to optimize online learning activities. One of the steps that need to do is to socialize continuously related to TOEFL’s learning and simulation to students to make students more motivated to use this E-TOEFL. Then there is also the dissemination of E-TOEFL learning policies to the academic students of Universitas Bung Hatta.

Student’s Responses on Learning in the Early COVID-19 Pandemic

The coronavirus disease (COVID-19) pandemic forced many universities to apply online learning. The purpose of this study was to break down the online learning process in the early pandemic as well as effective and optimal online learning. The design of this research is descriptive qualitative research. The data were collected through observation, questionnaires, interviews, and documentation. Interestingly, this study found online learning using the WhatsApp Group to be the most effective in the early COVID-19 pandemic. WhatsApp is easy, simple, and does not require a large data quota package. Through WhatsApp accounts, learning took place optimally because students and lecturers could communicate and share PowerPoint files, Microsoft Word files, JPGs, Voice Notes, Videos, and other learning resource links. The study recommends that other researchers uncover the solution to obstacles experienced by students in online learning and the development of other media to implement effective online lectures.

Inclusive considerations for optimal online learning in times of disasters and crises

Emergencies can create situations where traditional face-to-face courses need to quickly be transferred to be online. When transitioning, it is easy to focus on simply moving content into an online learning management system quickly and training instructors and students in the basics of how to use the platform in a pared-down learning structure. However, this article argues that approaching emergency course redesign with accessibility in mind at the start of the transition will ensure that more students, including students with disabilities, succeed in the online learning environment. This also helps ensure the course is designed for optimal student learning outcomes.

UTILIZATION OF EDMODO AS PART OF OPTIMIZATION LEARNING IN NETWORK

Edmodo is a form of learning application used by academics in order to carry out online learning. This research was conducted with the aim of informing more widely about Edmodo and describing how to use it to be able to optimize online learning. The research method used is a qualitative research method with a type of literature study. The results of the research conducted are that Edmodo has provided complete facilities in the context of implementing virtual learning. Intended as virtual learning because the learning process is carried out in the network. Educators (teachers / lecturers) have breadth in the implementation of classroom management because students can be made classes as classes in conventional learning. Educators can optimize the dissemination of information as well as in applications that have provided many features to make tests with various types of tests offered, educators can also provide assignments with predetermined collection times so as to optimize classroom management, educators can provide a direct assessment of the assignments students, giving awards with various types of awards. The advantage of using Edmodo is that there is no limit on space and time in learning because at any time it can take part in learning, it does not depend on the presence of educators in front of the class because wherever educators can still provide optimal learning.

Optimizing the Use of E-Worksheet and PHET Simulation in Science Learning during the Covid-19 Pandemic

This study examines the use of e-worksheets and PHET Simulation as virtual laboratories in science. This study aims to determine whether the use of PHET Simulation with e-worksheet guidance can optimize online learning to minimize learning loss due to the Covid 19 pandemic. E-worksheets are used as scaffolding media that are used to help guide students to do online practicums. The research was conducted in two stages. The first stage is conducting a direct trial of learning using PHET Simulation and e-worksheets through Google Classroom. For the second stage, the activities include collecting data using an accessibility questionnaire using a google form. The data were then analyzed by usability test covering aspects of usefulness, ease of use, ease of learning, and satisfaction. The usability test results show that the percentage of respondents to the questions given is dominated by the Agree (S) and Strongly Agree (SS) statements on all usability test indicators. This shows that the use of e-worksheets and PHET Simulation as a virtual laboratory fulfils the element of usability.

Study Smart Not Hard

Learners’ concentration is an essential factor for learning and acquisition. The duration of concentration varies from one individual to another. Some learners have a long duration of concentration; whereas, others have a short one. Leaving the learner in front of a screen for a random duration is a strategy that does not optimize online learning. In this perspective, we have implemented an approach based on three intelligent agents customizing and adapting the learning time according to the learner's concentration time. The first agent is called "Detector Agent" (DA).The task of this agent is to measure the learner's concentration time and to detect the factors that may influence it. The second agent is named "Scheduler Agent" (SA). Its function is to cut the time of a session in proportion to the concentration time measured by DA and to program relaxing breaks in order to regenerate the degree of concentration. And finally, the third agent is called "Rectifier Agent" (RA). The latter is responsible for rectifying the factors that negatively influence the concentration of the learners. These agents continuously communicate between each other in order to ensure an efficient treatment. The experienced results show that the approach contributes effectively in the acquisition and the performance of learners. Success rates have risen sharply and the learners express a growing satisfaction.

Moderation of an online discussion on communication skills – A GSMC FAIMER experience

This article was migrated. The article was not marked as recommended. Discussion forums provide the ability for asynchronous discussion to occur over a period of time. The ability to learn asynchronously gives flexibility in learning and is one of the primary benefits of online learning. The combination of the asynchronous and interactive nature of the online learning environment encourages participants to engage in higher-order learning and reflection. To foster an optimal online learning community three components are important: cognitive presence, social presence and teacher presence in order to ensure that effective group discussions take place that can share meaning, identify areas of doubt, and attempt to reach a consensus and understanding of the subject being discussed. Moderating an online discussion is an art of its own and requires meticulous planning and execution strategies. Maintaining participant interest throughout the session can be quite challenging. This article is to share our experience in moderating a session as a part of the FAIMER programme in GSMC regional institution in Mumbai, India.

Social gaming is inSIDE: Impact of anonymity and group identity on performance in a team game-based learning environment

The present experiment aimed to determine how quiz performance in a team game-based learning environment can be predicted from the Social Identity model of Deindividuation Effects (SIDE). According to this model, anonymity influences social behavior by accentuating the salience of group identity and reducing interpersonal differences, leading to greater group identification and motivation to work for one's own group. As these effects could lead to higher cognitive performance, the goal of the present research was to extend predictions based on the SIDE model on performance in online game-based learning environments. After measuring their prior computing knowledge, 343 Master Degree students were placed in virtual teams on a trivial criterion to perform a series of online quizzes about computing and the Internet. An anonymous (or individuated) username was attributed to each team member to connect to the online learning environment, and information about comparison between teams was used to manipulate the degree of salience of group identity (high versus low). As predicted by the SIDE model, anonymity boosted performance when group identity was salient, but only for students with low prior knowledge. Unexpectedly, it was also found that anonymity boosted the performance of students with high prior knowledge when group identity was not salient. A similar pattern was found for perceived mastery of computing and the Internet. Theoretical and practical implications of the SIDE model are discussed, and specifically its application to social gaming to optimize online learning.

Self-regulation of motivation when learning online: the importance of who, why and how

Successful online students must learn and maintain motivation to learn. The Self-regulation of Motivation (SRM) model (Sansone and Thoman 2005) suggests two kinds of motivation are essential: Goals-defined (i.e., value and expectancy of learning), and experience-defined (i.e., whether interesting). The Regulating Motivation and Performance Online (RMAPO) project examines implications using online HTML lessons. Initial project results suggested that adding usefulness information (enhancing goals-defined motivation) predicted higher engagement levels (enhancing experience), which in turn predicted motivation (interest) and performance (HTML quiz) outcomes. The present paper examined whether individual interest in computers moderated these results. When provided the utility value information, students with higher (relative to lower) individual interest tended to display higher engagement levels, especially when usefulness was framed in terms of personal versus organizational applications. In contrast, higher engagement levels continued to positively predict outcomes regardless of individual interest. We discuss implications for designing optimal online learning environments.

Preference learning on an OSGi based home gateway

The goal of ubiquitous computing is to create intelligent environment. To make the environment adapt rationally according to the desire of users, the system should be able to guess users' interest, by learning users' preferences. Users' preferences are sometimes conflicting and needs to be resolved. When many users are involved in a ubiquitous environment, the decisions of one user can be affected by the desires of others. This makes learning and prediction of user preferences difficult. In this paper we prove that learning and prediction of user preference is NP-Hard. So, we propose Bayesian RN-Metanetwork, a multilevel Bayesian network to model user preference and priority. This is a semi optimal online learning approach. By using game theory we prove that the method we use will certainly converge after a while. We also provide implementation details of the metanetwork on an OSGi based home gateway.

A locally recurrent fuzzy neural network with application to the wind speed prediction using spatial correlation

In this paper, a locally feedback dynamic fuzzy neural network (LF-DFNN) for modeling of temporal processes is suggested. The model is composed of dynamic TSK-type fuzzy rules where the consequent sub-models are implemented by recurrent neural networks with internal feedback paths and dynamic neuron synapses. The LF-DFNN exhibits some interesting features, such as enhanced representation power, local modeling characteristics, model parsimony, and stable learning. Training of the LF-DFNN models is achieved using an optimal on-line learning scheme, the decoupled recursive prediction error algorithm (DRPE). The method has reduced computational demands and is derived through decomposition of the weight vector to several mutually exclusive weight groups. The partial derivatives required for the implementation of the training algorithm are calculated using the adjoint model approach, adapted to the fuzzy network's architecture exercised here. The paper deals with the wind speed prediction in wind farms, using spatial information from remote measurement stations. The LF-DFNN networks are used as advanced forecast models, providing multi-step ahead wind speed estimates from 15 min to 3 h ahead. Extensive simulation results demonstrate that our models exhibit superior performance compared to other network types suggested in the literature. Furthermore, it is shown that DRPE outperforms three gradient descent algorithms, in training of the recurrent forecast models.

Long-Term Wind Speed and Power Forecasting Using Local Recurrent Neural Network Models

This paper deals with the problem of long-term wind speed and power forecasting based on meteorological information. Hourly forecasts up to 72-h ahead are produced for a wind park on the Greek island of Crete. As inputs our models use the numerical forecasts of wind speed and direction provided by atmospheric modeling system SKIRON for four nearby positions up to 30 km away from the wind turbine cluster. Three types of local recurrent neural networks are employed as forecasting models, namely, the infinite impulse response multilayer perceptron (IIR-MLP), the local activation feedback multilayer network (LAF-MLN), and the diagonal recurrent neural network (RNN). These networks contain internal feedback paths, with the neuron connections implemented by means of IIR synaptic filters. Two novel and optimal on-line learning schemes are suggested for the update of the recurrent network's weights based on the recursive prediction error algorithm. The methods assure continuous stability of the network during the learning phase and exhibit improved performance compared to the conventional dynamic back propagation. Extensive experimentation is carried out where the three recurrent networks are additionally compared to two static models, a finite-impulse response NN (FIR-NN) and a conventional static-MLP network. Simulation results demonstrate that the recurrent models, trained by the suggested methods, outperform the static ones while they exhibit significant improvement over the persistent method.

An adaptive learning algorithm for a wavelet neural network

Abstract: An optimal online learning algorithm of a wavelet neural network is proposed. The algorithm provides not only the tuning of synaptic weights in real time, but also the tuning of dilation and translation factors of daughter wavelets. The algorithm has both tracking and smoothing properties, so the wavelet networks trained with this algorithm can be efficiently used for prediction, filtering, compression and classification of various non-stationary noisy signals.

Locally recurrent neural networks for long-term wind speed and power prediction

The paper deals with a real-world application, the long-term wind speed and power forecasting in a wind farm using locally recurrent multilayer networks as forecast models. To cope with the complexity of the process and to improve the performance of the models, a class of optimal on-line learning algorithms is employed for training the locally recurrent networks based on the recursive prediction error (RPE) algorithm. A global RPE algorithm is devised and three local learning algorithms are suggested by partitioning the GRPE into a set of sub-problems at the neuron level to reduce computational complexity and storage requirements. Experimental results on the wind prediction problem demonstrate that the proposed algorithms exhibit enhanced performance, in terms of convergence speed and the accuracy of the attained solutions, compared to conventional gradient-based methods. Furthermore, it is shown that the suggested recurrent forecast models outperform the atmospheric and time-series models.

Reparametrization-covariant theory for on-line learning of probability distributions.

We discuss the on-line learning of probability distributions in a reparametrization covariant formulation. Reparametrization covariance plays an essential role not only to respect an intrinsic property of "information" but also for pattern recognition problems. We can obtain an optimal on-line learning algorithm with reparametrization invariance, where the conformal gauge connects a covariant formulation with a noncovariant one in a natural way.

Field Theoretical Analysis of On-Line Learning of Probability Distributions

On-line learning of probability distributions is analyzed from the field theoretical point of view. We can obtain an optimal on-line learning algorithm, since a renormalization group enables us to control the number of degrees of freedom of a system according to the number of examples. We do not learn parameters of a model, but probability distributions themselves. Therefore, the algorithm requires no a priori knowledge of a model.

Optimal online learning: a Bayesian approach

A recently proposed Bayesian approach to online learning is applied to learning a rule defined as a noisy single layer perceptron. In the Bayesian online approach, the exact posterior distribution is approximated by a simple parametric posterior that is updated online as new examples are incorporated to the dataset. In the case of binary weights, the approximate posterior is chosen to be a biased binary distribution. The resulting online algorithm is shown to outperform several other online approaches to this problem.

Globally optimal on-line learning rules for multi-layer neural networks

We present a method for determining the globally optimal on-line learning rule for a soft committee machine under a statistical mechanics framework. This rule maximizes the total reduction in generalization error over the whole learning process. A simple example demonstrates that the locally optimal rule, which maximizes the rate of decrease in generalization error, may perform poorly in comparison.