Active Learning Scenario Research Articles

PENGEMBANGAN SKENARIO PEMBELAJARAN AKTIF BERBASIS MIKiR UNTUK SISWA KELAS V PADA PEMBELAJARAN MATEMATIKA DI SEKOLAH DASAR

Learning scenario is the important thing in teaching process. Designing of good learning scenario is one of effort to increase learning quality. The aims of this study is to see the validity and practicality of the developed MIKiR-based active learning scenario. This study was carried out at the Integrated Islamic Elementary School (SDIT) Nurul Ilmi with the number of study subjects are 40 students from two fifth grades of Mathematics. The method used to develop the MIKiR-based active learning scenario is ADDIE (Analysis, Design, Development, Implementation, and Evaluation). The study instruments used in this study were: validation sheets to measure product feasibility, and teacher response questionnaires to see the practicality of the product. The results obtained are as follows, for the level of validity of the MIKiR-based active learning scenario, the average score is 3.67 with a very valid category consisting of three aspects, namely content, material, form, presentation, language, and legibility. Furthermore, the level of practicality obtained a very practical category with an average value of 84.76%. This can be seen from four aspects, namely the attractiveness of the device to student interest, ease of use and content, increased student activity, and time efficiency.

Uncertainty quantification in multivariable regression for material property prediction with Bayesian neural networks

With the increased use of data-driven approaches and machine learning-based methods in material science, the importance of reliable uncertainty quantification (UQ) of the predicted variables for informed decision-making cannot be overstated. UQ in material property prediction poses unique challenges, including multi-scale and multi-physics nature of materials, intricate interactions between numerous factors, limited availability of large curated datasets, etc. In this work, we introduce a physics-informed Bayesian Neural Networks (BNNs) approach for UQ, which integrates knowledge from governing laws in materials to guide the models toward physically consistent predictions. To evaluate the approach, we present case studies for predicting the creep rupture life of steel alloys. Experimental validation with three datasets of creep tests demonstrates that this method produces point predictions and uncertainty estimations that are competitive or exceed the performance of conventional UQ methods such as Gaussian Process Regression. Additionally, we evaluate the suitability of employing UQ in an active learning scenario and report competitive performance. The most promising framework for creep life prediction is BNNs based on Markov Chain Monte Carlo approximation of the posterior distribution of network parameters, as it provided more reliable results in comparison to BNNs based on variational inference approximation or related NNs with probabilistic outputs.

Beyond the Passive Absorption of Information: Engaging Students in the Critical Reading of Scientific Articles

AbstractThere is a consensus within the science education community that primary scientific literature is a legitimate and desirable educational resource. Moreover, critical reading of scientific articles is widely recognized as a key aspect of scientific literacy. However, university science courses rarely provide students with explicit opportunities to cultivate their critical reading skills. Much of the reason for this is that instructors tend to hold a passive learning view of reading in which students are expected to absorb information from scientific articles. The purpose of this study was to provide research evidence that an active learning scenario (ALS) combining (1) argumentation, (2) peer critique (also referred to as peer assessment), and (3) the Task-Oriented Reading Instruction framework (Ritchey & List, College Teaching, 70(3), 280–295, 2022) could be a concrete and realistic possibility for engaging students in the critical reading of scientific papers. The data analyzed in this study were the written critiques of scientific research articles and written peer feedback produced by sixty-one university students (38 females and 23 males, 19–25 years old). The results indicate that the ALS effectively offered students explicit opportunities to become more active and more critical readers of scientific articles, producing arguments, anticipating counterarguments, and constructing rebuttals. Implications related to critical reading instruction in science education and supporting students’ development of critical reading skills are discussed.

Generalisable Dialogue-based Approach for Active Learning of Activities of Daily Living

While Human Activity Recognition systems may benefit from Active Learning by allowing users to self-annotate their Activities of Daily Living (ADLs), many proposed methods for collecting such annotations are for short-term data collection campaigns for specific datasets. We present a reusable dialogue-based approach to user interaction for active learning in activity recognition systems, which utilises semantic similarity measures and a dataset of natural language descriptions of common activities (which we make publicly available). Our approach involves system-initiated dialogue, including follow-up questions to reduce ambiguity in user responses where appropriate. We apply this approach to two active learning scenarios: (i) using an existing CASAS dataset, demonstrating long-term usage; and (ii) using an online activity recognition system, which tackles the issue of online segmentation and labelling. We demonstrate our work in context, in which a natural language interface provides knowledge that can help interpret other multi-modal sensor data. We provide results highlighting the potential of our dialogue- and semantic similarity-based approach. We evaluate our work: (i) quantitatively, as an efficient way to seek users’ input for active learning of ADLs; and (ii) qualitatively, through a user study in which users were asked to compare our approach and an established method. Results show the potential of our approach as a hands-free interface for annotation of sensor data as part of an active learning system. We provide insights into the challenges of active learning for activity recognition under real-world conditions and identify potential ways to address them.

How to share the leadership competence among the team members in active learning scenarios: Before, during and after COVID-19 pandemic

Teamwork is one of the most demanded generic competencies by international organizations, and higher education institutions train and assess that competence to prepare students for working life. Leadership is a crucial part of teamwork development, and previous research has shown that shared leadership tasks between team members present more advantages than the traditional concept of a formal leader. Shared leadership seems to be the best option in the academic context due to the university students’ characteristics. This paper aims to prove that students can identify, distinguish and exercise shared leadership actions based on the needs that arise during the development of teamwork and that derive from the teamwork method applied rather than by the training modality that is followed (face-to-face – online). The achievement of the aim has been possible through a qualitative study of the teamwork development of 40 teams of new university entrance (237 students) with the Comprehensive Teamwork Competency Formation Model. The research has been carried out during three consecutive academic courses, with different training modalities for each course, forced by the COVID-19 pandemic (face-to-face for the pre-COVID-19 course, online for the COVID-19 course and face-to-face during the post-COVID-19 course). The shared leadership tasks and responsibilities, defined by students, were categorized in the same way independently of the training modality, which validates the proposed ontology. Also, the three academic courses studied the evolution of the primary shared leadership responsibilities by category. Besides, it is concluded that the primary responsibilities for each category remained unchanged during the three academic years but that some other categories were affected to some extent by the exceptionality caused by COVID-19. The ontology validated here constitutes a recommendation for future teams working with an evidence-based methodology.

A study on learning analytics of using mobile augmented reality application to enhance cultural competence for design cultural creation in higher education

AbstractBackgroundTaiwan's higher education institutions prioritize interdisciplinary knowledge and cultural competence in cultural design, emphasizing the value of immersion in the local environment to develop cultural competence. However, challenges arise from the disappearance of traditional local lifestyles and limitations of traditional outdoor lectures such as group size and learning environment. Mobile augmented reality technology serves as a personal learning tool to overcome these challenges, enabling learners to uncover hidden knowledge in scenic locations and improving learning efficiency.ObjectiveThe study utilized learning analytics to investigate the impact of using mobile augmented reality on enhancing learners' cultural competence.MethodsThe study used both qualitative and quasi‐experimental quantitative research to evaluate the effects of mobile augmented reality on cultural competence. A total of 78 third‐year university students participated in the study, with 37 of them using mobile augmented reality to learn about culture. The researchers utilized both the Mann–Whitney U test and interviews to compare the effectiveness of mobile augmented reality with that of a ‘real’ tour guide in field courses.ResultsThe experimental results indicate that learners were able to continuously seek and find information about local culture through field exploration with mobile augmented reality, resulting in similar effects to those of attending a live lecture.ConclusionsIt showed that incorporating mobile AR into field visits can create gamified and active learning scenarios that enhance immersion and drive active learning in local culture. The results in greater cultural knowledge and understanding, leading to the development of cultural competence in students.

Active Bayesian Deep Learning With Vector Sensor for Passive Sonar Sensing of the Ocean

Multiple studies have explored and reported on using passive sonar scalar pressure fields for both biological and man-made target classification. Here, inclusion of the vector field is considered and evaluated as an input feature in the classification of ship noises. We find that the inclusion of vector field information can significantly improve the understanding of the classification model performance. Based on an analysis of the epistemic and aleatoric uncertainties of probabilistic classification models, using vector field information as additional inputs leads to an <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$F_{1}$</tex-math></inline-formula> score of 0.8 in classification, comparable to scalar pressure field inputs only, but with 39% lower epistemic uncertainty in favor of utilizing the vector field component. Moreover, we verify that a similar conclusion applies to the active learning scenario, where we demonstrate the ability to utilize vector field information and epistemic uncertainty estimates to train a model that achieves an <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$F_{1}$</tex-math></inline-formula> score of 0.8 while using only 23% of the overall data. The ability to decompose uncertainty into an aleatoric and epistemic component leads to additional model explainability.

Transferring knowledge between topics in systematic reviews

In the medical domain, a systematic review (SR) is a well-structured process aimed to review all available literature on a research question. This is however a laborious task, both in terms of money and time. As such, the automation of a SR with the aid of technology has received interest in several research communities, among which the Information Retrieval community. In this work, we experiment on the possibility of leveraging previously conducted systematic reviews to train a classifier/ranker which is later applied to a new SR. We also investigate on the possibility of pre-training Deep Learning models and eventually tuning them in an Active Learning process. Our results show that the pre-training of these models deliver a good zero-shot (i.e., with no fine-tuning) ranking, achieving an improvement of 79% for the MAP metric, with respect to a standard classifier trained on few in-domain documents. However, the pre-trained deep learning algorithms fail to deliver consistent results when continuously trained in an Active Learning scenario: our analysis shows that using smaller sized models and employing adapter modules might enable an effective active learning training.

Active multitask learning with uncertainty-weighted loss for coronary calcium scoring.

The coronary artery calcification (CAC) score is an independent marker for the risk of cardiovascular events. Automatic methods for quantifying CAC could reduce workload and assist radiologists in clinical decision-making. However, large annotated datasets are needed for training to achieve very good model performance, which is an expensive process and requires expert knowledge. The number of training data required can be reduced in an active learning scenario, which requires only the most informative samples to be labeled. Multitask learning techniques can improve model performance by joint learning of multiple related tasks and extraction of shared informativefeatures. We propose an uncertainty-weighted multitask learning model for coronary calcium scoring in electrocardiogram-gated (ECG-gated), noncontrast-enhanced cardiac calcium scoring CT. The model was trained to solve the two tasks of coronary artery region segmentation (weak labels) and coronary artery calcification segmentation (strong labels) simultaneously in an active learning scenario to improve model performance and reduce the number of samples needed for training. We compared our model with a single-task U-Net and a sequential-task model as well as other state-of-the-art methods. The model was evaluated on 1275 individual patients in three different datasets (DISCHARGE, CADMAN, orCaScore), and the relationship between model performance and various influencing factors (image noise, metal artifacts, motion artifacts, image quality) wasanalyzed. Joint learning of multiclass coronary artery region segmentation and binary coronary calcium segmentation improved calcium scoring performance. Since shared information can be learned from both tasks for complementary purposes, the model reached optimal performance with only 12% of the training data and one-third of the labeling time in an active learning scenario. We identified image noise as one of the most important factors influencing model performance along with anatomical abnormalities and metalartifacts. Our multitask learning approach with uncertainty-weighted loss improves calcium scoring performance by joint learning of shared features and reduces labeling costs when trained in an active learningscenario.

Examining Students’ Perceptions towards Video-Based and Video-Assisted Active Learning Scenarios in Journalism and Communication Courses

Audiovisual content is highly employed in modern education curricula. This study aimed to identify students’ perceptions regarding the use of video in the courses they attend. Quantitative research was conducted among students of journalism and communication studies. Four courses were used for this study that employ video-based and video-assisted teaching strategies for active learning. Students used videos in different educational settings for flipped learning, blended learning, and autonomous self-paced learning. Their perceptions were examined through an extended form of the Technology Acceptance Model, encompassing more parameters (such as self-efficacy, perceived enjoyment, satisfaction, attitude, and intention of use) to identify causal relationships. In addition, the roles of technical parameters, such as internet connection, audio quality, and video quality, along with the duration of a video, were explored. The findings reveal students’ positive attitudes towards videos, and all factors were significantly related to the intention of use. In addition, the technical aspects of the video did not constitute a problem in the educational process, most likely due to contemporary quality in internet connections and audiovisual productions along with playback equipment. The duration of the videos was also not presented as an issue as there are flexible ways to embed videos in the learning process. Overall, videos can be used in multiple active learning scenarios to enhance students’ motivation and engagement and provide a joyful, collaborative, and hospitable learning environment. Students are familiar with digital tools and technologies and seem to enjoy audiovisual material in the learning process.

An Ensemble-Based Fuzzy Rough Active Learning Approach for Broken Rotor Bar Detection in Nonstationary Environment

Most existing fault detection methods rely on batch learning, which assumes that all data is available and can be accessed over and over again. Such an approach is impossible in the face of nonstationary environments. This paper proposes an online combination of an Ensemble framework, Fuzzy Rough Active Learning, and drift detection, namely EFRAL, in which the structure of the ensemble automatically grows according to the data distribution. Since obtaining a real industrial data label is costly and time-consuming in practice, this paper incorporates an active learning scenario based on fuzzy rough sets to reduce operator labeling effort. This method gives weight to training samples based on the fuzzy rough sets theory and the samples are selected based on their weight. The proposed algorithm is applied to detect a broken rotor bar fault of induction motors to show its feasibility in industrial environments. The attained results show the efficiency of the proposed method. Furthermore, the efficiency of EFRAL is evaluated through data stream problems.

Investigation of Combining Logitboost(M5P) under Active Learning Classification Tasks

Active learning is the category of partially supervised algorithms that is differentiated by its strategy to combine both the predictive ability of a base learner and the human knowledge so as to exploit adequately the existence of unlabeled data. Its ambition is to compose powerful learning algorithms which otherwise would be based only on insufficient labelled samples. Since the latter kind of information could raise important monetization costs and time obstacles, the human contribution should be seriously restricted compared with the former. For this reason, we investigate the use of the Logitboost wrapper classifier, a popular variant of ensemble algorithms which adopts the technique of boosting along with a regression base learner based on Model trees into 3 different active learning query strategies. We study its efficiency against 10 separate learners under a well-described active learning framework over 91 datasets which have been split to binary and multi-class problems. We also included one typical Logitboost variant with a separate internal regressor for discriminating the benefits of adopting a more accurate regression tree than one-node trees, while we examined the efficacy of one hyperparameter of the proposed algorithm. Since the application of the boosting technique may provide overall less biased predictions, we assume that the proposed algorithm, named as Logitboost(M5P), could provide both accurate and robust decisions under active learning scenarios that would be beneficial on real-life weakly supervised classification tasks. Its smoother weighting stage over the misclassified cases during training as well as the accurate behavior of M5P are the main factors that lead towards this performance. Proper statistical comparisons over the metric of classification accuracy verify our assumptions, while adoption of M5P instead of weak decision trees was proven to be more competitive for the majority of the examined problems. We present our results through appropriate summarization approaches and explanatory visualizations, commenting our results per case.

Optimalisasi Pendekatan MIKiR Sebagai Solusi Pembelajaran Abad 21 Bagi Guru SD Muhammadiyah Kota Medan

The study aims to answer the learning challenges on the 21st century. Teachers are required to improve their abilities, especially in managing student-oriented learning. Teacher creativity is the basis for designing meaningful learning. Participants of this activity were teachers of SD Muhammadiyah 29 Sei Mencirim and SD Muhammadiyah 38 Medan Krio. The target of this activity is that teachers are able to design active learning scenarios that can be applied in class. The methods for implementing this activity are: 1) Initial meeting with school principals, 2) Training activities, 3) Evaluation, and 4) Follow-up plans. Based on the results of the activity, the teachers have been able to design active learning scenarios and 21st century skills. The implementer hopes that participants will familiarize themselves with implementing the results of activities in supporting the learning being carried out and can pass them on to other teachers.

Teaching business ethics in a digital world

PurposeWhile teaching of business ethics has been increasing in business schools worldwide, universities still face increasing pressure to do more to proactively defend and help avoid unethical business practices and scandals calling for more responsible education. This study aims to examine teaching business ethics in light of recent technological advances (i.e. teaching via the use of digital devices) and well-established pedagogical practices.Design/methodology/approachThis study uses a 2 × 2 experimental design examining the effect of active (vs passive) and presence (vs absence) of digital devices in student learning of 192 US students.FindingsThe findings suggest that the active learning scenario, the usage of laptops and phones helped students get higher results in the test compared to active learning with no digital devices or passive learning with digital devices.Originality/valueActive learning practices such as group discussions and peer assessment or the flipped classroom approach make a difference for business ethics teaching where students need to develop inquiry and interest for the subject and engage in ethical dilemmas and real-life examples. Further, students in the active learning scenario performed better in knowledge tests when they were asked to use their digital devices.

What more than a hundred project groups reveal about teaching visualization

The growing number of students can be a challenge for teaching visualization lectures, supervision, evaluation, and grading. Moreover, designing visualization courses by matching the different experiences and skills of the students is a major goal in order to find a common solvable task for all of them. Particularly, the given task is important to follow a common project goal, to collaborate in small project groups, but also to further experience, learn, or extend programming skills. In this article, we survey our experiences from teaching 116 student project groups of 6 bachelor courses on information visualization with varying topics. Moreover, two teaching strategies were tried: 2 courses were held without lectures and assignments but with weekly scrum sessions (further denoted by TS1) and 4 courses were guided by weekly lectures and assignments (further denoted by TS2). A total number of 687 students took part in all of these 6 courses. Managing the ever growing number of students in computer and data science is a big challenge in these days, i.e., the students typically apply a design-based active learning scenario while being supported by weekly lectures, assignments, or scrum sessions. As a major outcome, we identified a regular supervision either by lectures and assignments or by regular scrum sessions as important due to the fact that the students were relatively unexperienced bachelor students with a wide range of programming skills, but nearly no visualization background. In this article, we explain different subsequent stages to successfully handle the upcoming problems and describe how much supervision was involved in the development of the visualization project. The project task description is given in a way that it has a minimal number of requirements but can be extended in many directions while most of the decisions are up to the students like programming languages, visualization approaches, or interaction techniques. Finally, we discuss the benefits and drawbacks of both teaching strategies.Graphic abstract

An interdisciplinary digital learning and research factory: The Smart Production Lab

To accelerate digital transformation processes, more efforts are needed both in applied research and development as well as in practical implementation. For this reason, numerous learning and research factories have been build up during the last decade. The aim of this paper is to present an interdisciplinary concept model for a learning and research factory aiming to increase significantly the digital competences of concerned stakeholders. The proposed model has three dimensions: (i) content, (ii) function and (iii) methodology and is realized through a specific structure (Lab-in-Lab). Based on this concept model, the Smart Production Lab (Smart Lab) has been designed and implemented by the Institute of Industrial Management at FH JOANNEUM University of applied sciences, Austria. Essentially the Smart Lab is about the digital transformation of connecting businesses – horizontally along the entire value chain and vertically with focus on data generation, integration, and usage. The development of use cases in the two maturity levels – state-of-the-art and research – allows the transfer of knowledge of digital transformation processes for companies as well as active learning scenarios for the qualification of future industrial engineers. As an innovative experimental environment, the Smart Lab addresses companies, researchers, students, and – with the integrated fabrication laboratory (FabLab) – the interested public in general.

Online Tool Condition Monitoring Based on Parsimonious Ensemble.

Existing methodologies for tool condition monitoring (TCM) still rely on batch approaches which cannot cope with a fast sampling rate of a metal cutting process. Furthermore, they require a retraining process to be completed from scratch when dealing with a new set of machining parameters. This paper presents an online TCM approach based on Parsimonious Ensemble+ (pENsemble+). The unique feature of pENsemble+ lies in its highly flexible principle where both the ensemble structure and base-classifier structure can automatically grow and shrink on the fly based on the characteristics of data streams. Moreover, the online feature selection scenario is integrated to actively sample relevant input attributes. This paper presents advancement of a newly developed ensemble learning algorithm, pENsemble, where the online active learning scenario is incorporated to reduce the operator's labeling effort. The ensemble merging scenario is proposed which allows reduction of ensemble complexity while retaining its diversity. Experimental studies utilizing two real-world manufacturing data streams: 1) metal turning and 2) 3-D-printing processes and comparisons with well-known algorithms were carried out. Furthermore, the efficacy of pENsemble+ was examined using benchmark concept drift data streams. It has been found that pENsemble+ incurs low structural complexity and results in a significant reduction of the operator's labeling effort.

Inverse reinforcement learning via nonparametric spatio-temporal subgoal modeling

Advances in the field of inverse reinforcement learning (IRL) have led to sophisticated inference frameworks that relax the original modeling assumption of observing an agent behavior that reflects only a single intention. Instead of learning a global behavioral model, recent IRL methods divide the demonstration data into parts, to account for the fact that different trajectories may correspond to different intentions, e.g., because they were generated by different domain experts. In this work, we go one step further: using the intuitive concept of subgoals, we build upon the premise that even a single trajectory can be explained more efficiently locally within a certain context than globally, enabling a more compact representation of the observed behavior. Based on this assumption, we build an implicit intentional model of the agent's goals to forecast its behavior in unobserved situations. The result is an integrated Bayesian prediction framework that significantly outperforms existing IRL solutions and provides smooth policy estimates consistent with the expert's plan. Most notably, our framework naturally handles situations where the intentions of the agent change over time and classical IRL algorithms fail. In addition, due to its probabilistic nature, the model can be straightforwardly applied in active learning scenarios to guide the demonstration process of the expert.

Sparse coding based classifier ensembles in supervised and active learning scenarios for data classification

Sparse coding and dictionary learning has recently gained great interest in signal, image and audio processing applications through representing each problem instance by a sparse set of atoms. This also allows us to obtain different representations of feature sets in machine learning problems. Thus, different feature views for classifier ensembles can be obtained using sparse coding. On the other hand, nowadays unlabelled data is abundant and active learning methods with single and classifier ensembles received great interest. In this study, Random Subspace Dictionary Learning (RDL) and Bagging Dictionary Learning (BDL) algorithms are examined by learning ensembles of dictionaries through feature/instance subspaces. Besides, ensembles of dictionaries are evaluated under active learning framework as promising models and they are named as Active Random Subspace Dictionary Learning (ARDL) and Active Bagging Dictionary Learning (ABDL) algorithms. Active learning methods are compared with their Support Vector Machines counterparts. The experiments on eleven datasets from UCI and OpenML repositories has shown that selecting instance and feature subspaces for dictionary learning model increases the number of correctly classified instances for the most of the data sets while SVM has superiority over all of the applied models. Furthermore, using an active learner generally increases the chance of improved classification performance as the number of iterations is increased.

Scaffolding type-2 classifier for incremental learning under concept drifts

The proposal of a meta-cognitive learning machine that embodies the three pillars of human learning: what-to-learn, how-to-learn, and when-to-learn, has enriched the landscape of evolving systems. The majority of meta-cognitive learning machines in the literature have not, however, characterized a plug-and-play working principle, and thus require supplementary learning modules to be pre-or post-processed. In addition, they still rely on the type-1 neuron, which has problems of uncertainty. This paper proposes the Scaffolding Type-2 Classifier (ST2Class). ST2Class is a novel meta-cognitive scaffolding classifier that operates completely in local and incremental learning modes. It is built upon a multivariable interval type-2 Fuzzy Neural Network (FNN) which is driven by multivariate Gaussian function in the hidden layer and the non-linear wavelet polynomial in the output layer. The what-to-learn module is created by virtue of a novel active learning scenario termed the uncertainty measure; the how-to-learn module is based on the renowned Schema and Scaffolding theories; and the when-to-learn module uses a standard sample reserved strategy. The viability of ST2Class is numerically benchmarked against state-of-the-art classifiers in 12 data streams, and is statistically validated by thorough statistical tests, in which it achieves high accuracy while retaining low complexity.