Learning Performance Research Articles

Performance of machine learning versus the national early warning score for predicting patient deterioration risk: a single-site study of emergency admissions

ObjectivesIncreasing operational pressures on emergency departments (ED) make it imperative to quickly and accurately identify patients requiring urgent clinical intervention. The widespread adoption of electronic health records (EHR) makes rich...

LearningViz: a dashboard for visualizing, analyzing and closing learning performance gaps—a case study approach

The availability of large-scale learning data presents unprecedented opportunities for investigating student learning processes. However, it is challenging for instructors to fully make sense of this data and effectively support their teaching practices. This study introduces LearningViz, an interactive learning analytics dashboard to help instructors identify, analyze, and close performance gaps among students in their classes. In this dashboard, we incorporated three modules to enhance human and computer interactions for better supporting the teaching practices: the Student Overall Performance Analysis Module, which provides a comprehensive understanding of students’ learning in the course; the Student Group Performance Analysis Module, which examines performance gaps across different groups and identifies factors contributing to these gaps; and the Final Exam Item Analysis Module, which evaluates the quality of exam questions and identifies strategies for closing performance gaps. The overall design of the platform follows a user-centered approach, integrating data analysis with various visualization strategies in a unified platform. A case study is then conducted to highlight the effectiveness of LearningViz in supporting instructors analyzing students’ learning patterns and associated factors impacting learning performance. We further conduct a usability test with several domain experts, to evaluate the usefulness and effectiveness of this platform in supporting the teaching practices. Our findings underscore the platform's ability to support instructors in detecting performance gaps among students, investigating influential factors, evaluating assessment quality and implementing targeted instructional strategies for closing performance gaps.

Mobile learning in the classroom – Should students bring mobile devices for learning, or should these be provided by schools?

AbstractIn recent years, the importance of mobile devices has increased for education in general and more specifically for science and mathematics education. In the classroom, approaches for teaching with mobile devices include using student-owned devices (“bring your own device”; BYOD approach) or using school-owned devices from central pools (POOL approach). While many studies point out features of mobile learning and BYOD that are conducive to learning, a research gap can be identified in the analysis of effects of mobile device access concepts on teaching–learning processes. Thus, this study aimed to empirically compare BYOD and POOL approaches in terms of learning performance and cognitive performance (subject knowledge development, cognitive load, concentration performance). Furthermore, the analyses included specific characteristics and preconditions (gender, socioeconomic status, fear of missing out, problematic smartphone use). A quasi-experimental study (two groups) was conducted in year 8 and 9 physics classes (N = 339 students) in which smartphones are used for different purposes. The present data show no group differences between the BYOD and the POOL approach in the group of learners with respect to subject knowledge development, cognitive load, and concentration performance. However, individual findings in subsamples indicate that the POOL approach may be beneficial for certain learners (e.g., learners with low fear of missing out or learners tending toward problematic smartphone use). For school practice, these results indicate that organizational, economic, and ecological aspects appear to be the main factors in deciding about the mobile device access concept.

Architecture-Aware Augmentation: A Hybrid Deep Learning and Machine Learning Approach for Enhanced Parkinson’s Disease Detection

Parkinson’s Disease (PD) is a progressive neurodegenerative disorder affecting millions worldwide. Early detection is crucial for improving patient outcomes. Spiral drawing analysis has emerged as a non-invasive tool to detect early motor impairments associated with PD. This study examines the performance of hybrid deep learning and machine learning models in detecting PD using spiral drawings, with a focus on the impact of data augmentation techniques. We compare the accuracy of Vision Transformer (ViT) with K-Nearest Neighbors (KNN), Convolutional Neural Networks (CNN) with Support Vector Machines (SVM), and Residual Neural Networks (ResNet-50) with Logistic Regression, evaluating their performance on both augmented and non-augmented data. Our findings reveal that ViT with KNN, initially achieving 96.77% accuracy on unaugmented data, experienced a notable decline across all augmentation techniques, suggesting it relies heavily on global patterns in spiral drawings. In contrast, ResNet-50 with Logistic Regression showed consistent improvement with data augmentation, reaching 93.55% accuracy when rotation and flipping techniques were applied. These results highlight that hybrid models respond differently to augmentation, and careful selection of augmentation strategies is necessary for optimizing model performance. Our study provides important insights into the development of reliable diagnostic tools for early PD detection, emphasizing the need for appropriate augmentation techniques in medical image analysis.

The effect of problem-based learning on improving problem-solving, self-directed learning, and critical thinking ability for the pharmacy students: A randomized controlled trial and meta-analysis.

This study aimed to comprehensively evaluate the effect of PBL on problem-solving, self-directed learning, and critical thinking ability of pharmaceutical students through a randomized controlled trial (RCT) and meta-analysis of RCTs. In 2021, 57 third-year pharmacy students from China Pharmaceutical University were randomly divided into a PBL group and a lecture-based learning (LBL) group. Mean scores were compared between the two groups for problem-solving, self-directed learning, communication skills, critical thinking, and final exam grades. Students' feedback on the implementation of PBL was also collected. A meta-analysis was subsequently performed. Two authors independently conducted a comprehensive search of two databases (PubMed and CNKI). Eligible studies with effective data were included and the valuable data were extracted for analysis. Quality of involved studies was assessed by the Cochrane Collaboration's tool. All analyses of statistics were conducted using the 'metafor' package in R software. The PBL group had significantly higher mean scores for problem-solving (8.43±1.56) and self-directed learning (7.39±1.19) than the LBL group (7.02±1.72 and 6.41±1.28, respectively). The PBL group also showed better communication skills (8.86±1.47) than the LBL group (7.68±1.89). The mean level of critical thinking was significantly higher in the PBL group than the LBL group (p = 0.02). The PBL group also had better final exam grades (79.86±1.38) compared to the LBL group (68.1±1.76). Student feedback on PBL implementation was positive. The outcome of subsequent meta-analysis including 8 eligible studies involved 1819 participants showed that the use of PBL significantly improved problem-solving ability (SMD = 1.12, 95% CI = 0.25-1.99) and PBL was also associated with better performance in self-directed learning (SMD = 1.55, 95% CI = 0.64-2.45). However, there was no significant difference in the final exam score in the PBL group compared to the LBL control group (SMD = 0.23, 95% CI = -0.08-0.53). This study found that PBL is an effective teaching method for pharmacy students.

Predicting hepatitis C infection with machine learning algorithms: a prospective study

<p>Globally, chronic hepatitis C virus (HCV) infection affects millions of people and leads to a high number of deaths annually. In 2019, the World Health Organization (WHO) recorded around 290,000 deaths related to HCV, a virus transmitted mainly through blood that causes liver damage. The virus has infected more than 169 million people worldwide. This study aims to compare the performance of machine learning (ML) models for HCV detection. ML models such as logistic regression (LR), random forest (RF), decision tree (DT), and catBoost classifier (CATBC) were used. To carry out this task, a dataset of 615 patient records, and 14 variables were used. This research process was carried out in multiple phases, encompassing model understanding, data analysis and cleaning, ML model training, and subsequent model evaluation. The results revealed that the gradient boosting (GB) model stood out by achieving the best performance and highest accuracy, achieving a rate of 94% in HCV detection, this demonstrates outstanding performance compared to the other models such as LR, RF, k-nearest neighbor (KNN), DT, and CATBC, which obtained accuracy rates of 89%, 93%, 85%, 93%, 93%, and 92%, respectively. It can be concluded that the GB model stands out as the best algorithm for this task.</p>

Exploring the performance and interpretability of hybrid hydrologic model coupling physical mechanisms and deep learning

Recently, differentiable modeling techniques have emerged as a promising approach to bidirectionally integrating neural networks and hydrologic models, achieving performance levels close to deep learning models while preserving the ability to output physical states and fluxes. However, there remains a lack of systematic exploration into the performance and physical interpretability of hybrid models that use neural networks to replace the runoff generation and routing processes in regionalized modeling. This research developed 12 regionalized hybrid models based on a differentiable parameter learning (DPL) framework, utilizing the Hydrologiska Byråns Vattenbalansavdelning (HBV) model as the foundational backbone. These hybrid models incorporate neural networks to replace the various physical processes within the runoff generation and routing modules. The publicly available CAMELS dataset is employed to evaluate the performance and interpretability of these hybrid models. The results show that while the median Nash-Sutcliffe efficiency (NSE) and Kling-Gupta efficiency (KGE) coefficients for all hybrid models are lower than those of the purely data-driven regionalized long short-term memory neural network (LSTM) model (median NSE: 0.742, median KGE: 0.762), the best-performing hybrid model (median NSE: 0.731, median KGE: 0.761) approaches the LSTM model and has better physical interpretability. Embedding neural networks does not inherently guarantee improved performance and may, in some cases, even result in reduced performance. The degree of performance enhancement is not significantly correlated with the number of embedded neural networks. Compared to replacing the runoff generation process, substituting the routing process with neural networks yields more substantial performance improvements and enables the learning of different routing patterns based on the catchment’s static attributes. This study underscores the importance of reasonably balancing the location, complexity, and quantity of embedded neural networks to achieve a trade-off between model performance and interpretability in hybrid modeling. These insights contribute to advancing regionalized hybrid modeling development.

Self-Supervised 3D Scene Flow Estimation and Motion Prediction Using Local Rigidity Prior.

In this article, we investigate self-supervised 3D scene flow estimation and class-agnostic motion prediction on point clouds. A realistic scene can be well modeled as a collection of rigidly moving parts, therefore its scene flow can be represented as a combination of rigid motion of these individual parts. Building upon this observation, we propose to generate pseudo scene flow labels for self-supervised learning through piecewise rigid motion estimation, in which the source point cloud is decomposed into local regions and each region is treated as rigid. By rigidly aligning each region with its potential counterpart in the target point cloud, we obtain a region-specific rigid transformation to generate its pseudo flow labels. To mitigate the impact of potential outliers on label generation, when solving the rigid registration for each region, we alternately perform three steps: establishing point correspondences, measuring the confidence for the correspondences, and updating the rigid transformation based on the correspondences and their confidence. As a result, confident correspondences will dominate label generation, and a validity mask will be derived for the generated pseudo labels. By using the pseudo labels together with their validity mask for supervision, models can be trained in a self-supervised manner. Extensive experiments on FlyingThings3D and KITTI datasets demonstrate that our method achieves new state-of-the-art performance in self-supervised scene flow learning, without any ground truth scene flow for supervision, even performing better than some supervised counterparts. Additionally, our method is further extended to class-agnostic motion prediction and significantly outperforms previous state-of-the-art self-supervised methods on nuScenes dataset.

Investigation of Chinese undergraduate EFL learners’ online communication strategies

With the outbreak of the Covid-19 pandemic, online learning has gained popularity throughout the world in recent years. How EFL learners communicate with their teachers and classmates online has sparked great interest with a view to enhancing their performance in online learning. This study aimed to explore the online communication strategies of Chinese undergraduate EFL learners and how they vary according to gender and English proficiency. The quantitative method was adopted in this study. The participants were 120 undergraduate EFL learners from a public university in China. The questionnaire on online communication strategies, consisting of 30 items, was developed. Cronbach alpha and factor analysis were conducted to evaluate the reliability and validity of the questionnaire. The descriptive statistics and independent T-tests were used for data analysis. The study reached the conclusion that the most commonly used online communication strategies are Reduction strategies, followed by Focus on Form, Social-cultural, Paralinguistic, Compensatory, and Interactional strategies. Male and female learners differed significantly in the use of reduction strategies, focus on form strategies, social-cultural strategies, and paralinguistic strategies. Furthermore, there is a significant difference in the use of all types of online communication strategies between good learners and poor learners, with good learners applying more online communication strategies than poor learners. The study indicates that teachers should raise students' awareness of online communication strategies that are conducive to online communication and learning. Training on online communication strategies should be provided with a view to enhancing students' communication competence as well as English proficiency.

Personalized learning model based on machine learning algorithms

<p>Machine learning algorithms have been widely applied in the field of personalized learning within educational information technology. By leveraging big data analysis and data mining techniques, machine learning can help identify patterns and trends in students' learning behaviors, preferences, and performance. This information can then be used to tailor educational resources and experiences to meet the individual needs and unique characteristics of each learner. Machine learning has made great progress and achievements in the teaching process of universities, but there are also some shortcomings. Such as data dependence, over-fitting and under-fitting, explanatory problems, need a lot of computing resources, data bias, sensitive to outliers, cannot solve all problems, and the challenge of data privacy, through the analysis of machine learning algorithm model, efforts to find ways to expand the dimension of personalized learning classroom, meet the students in learning objectives, learning content, learning methods of the special characteristics and unique needs, to guide students to actively explore and research, obtain innovation and appropriate learning results.</p>

Deterministic Gradient-Descent Learning of Linear Regressions: Adaptive Algorithms, Convergence Analysis and Noise Compensation.

Weight learning forms a basis for the machine learning and numerous algorithms have been adopted up to date. Most of the algorithms were either developed in the stochastic framework or aimed at minimization of loss or regret functions. Asymptotic convergence of weight learning, vital for good output prediction, was seldom guaranteed for online applications. Since linear regression is the most fundamental component in machine learning, we focus on this model in this paper. Aiming at online applications, a deterministic analysis method is developed based on LaSalle's invariance principle. Convergence conditions are derived for both the first-order and the second-order learning algorithms, without resorting to any stochastic argument. Moreover, the deterministic approach makes it easy to analyze the noise influence. Specifically, adaptive hyperparameters are derived in this framework and their tuning rules disclosed for the compensation of measurement noise. Comparison with four most popular algorithms validates that this approach has a higher learning capability and is quite promising in enhancing the weight learning performance.

Decision Support Systems for Lifelong Learning: Leveraging Information Systems to Enhance Learning Quality in Higher Education

This study investigates the impact of technology adoption—specifically AI Tools, Decision Support Systems (DSS), and Learning Management Systems (LMS)—on higher education. As these technologies reshape educational paradigms, understanding their effects on learning performance, satisfaction, and the adoption and usage of these tools is critical. The research aims to empirically examine the relationships between technology adoption, self-efficacy, and key educational outcomes. It explores the direct effects of AI Tools, DSS, and LMS on learning performance and satisfaction, as well as the role of self-efficacy as a mediator. Utilizing a quantitative approach, the study collected data from 356 students via a distributed questionnaire. Variables measured include technology adoption, self-efficacy, learning performance, satisfaction, and adoption and usage of educational tools. Data analysis was conducted using SPSS and Origin, incorporating regression, mediation, and moderation analyses. The study found significant positive effects of technology adoption on learning performance (β = 0.45, p < 0.01), satisfaction (β = 0.40, p < 0.01), and adoption and usage (β = 0.50, p < 0.01). Self-efficacy significantly mediated these relationships, indicating that higher confidence in using technology enhances its benefits. This research extends Bandura's social cognitive theory by empirically validating the mediating role of self-efficacy in technology adoption within educational contexts. The findings provide actionable insights for educators and policymakers, suggesting that boosting students' confidence in using technology can amplify its positive effects on learning outcomes.

Learning from interactive video: the influence of self-explanations, navigation, and cognitive load

AbstractRecent research has shown that enhancing instructional videos with questions, such as self-explanation prompts, and thus shifting the process from receptive to constructive learning, is beneficial to learning. However, the inclusion of questions is often confounded with the implementation of learner pacing through navigation features. Furthermore, previous studies have often not controlled for learning time. To address these shortcomings, an experiment (N = 128) was conducted. Participants watched an instructional video about cloud formation and lightning, with learning time controlled. In a 2 × 2 between-subjects design, navigation features (learner pacing vs. system pacing) and self-explanation prompts (prompts vs. no prompts) were manipulated. The results showed no effects of navigation features and self-explanation prompts on learning performance. While navigation features did not affect cognitive load, self-explanation prompts increased both intrinsic and extraneous cognitive load. Overall, the quality of responses to prompts was low but positively related to comprehension. The results are discussed in terms of the Interactive-Constructive-Active-Passive framework and Cognitive Load Theory. They highlight the importance of boundary conditions when investigating the effects of interactive features in instructional videos.

Eye Movements of Learners with Different Cognitive Styles Watching Digital Learning Videos

In response to the arrival of the post-COVID-19- era, digital learning has become crucial in education. The use of multimedia digital teaching materials in education is becoming increasingly common. Each individual has preferences regarding methods for absorbing information and, in turn, has a distinct learning model. Differences in learners’ cognitive styles may be related to their learning performance and their eye movements when browsing information. Therefore, this study explored differences in the eye movements of learners with different cognitive styles (text-based and image-based) while reading graphic information. First, we used the Style of Processing scale to identify learners’ cognitive styles. Next, a pretest was administered, and the participants were invited to watch educational videos. Finally, a posttest was conducted. We collected eye movement data while the participants watched the videos, and the collected data were used for subsequent analysis. Examining the use of digital learning materials for different subjects indicated that the participants with image-based cognitive styles focused more on graphic reading than did those with text-based cognitive styles.

Mindfulness in Surgical Training (MiST): A Modified Mindfulness Curriculum for Surgical Residents

Residents experience numerous work-related and personal stressors that make it difficult to focus in the operating room, negatively impacting learning and surgical performance. Mindfulness-based cognitive therapy decreases anxiety and improves memory and learning. This study aimed to create a feasible and desirable modified mindfulness curriculum for surgical residents. This was a prospective cohort study using multiple methods design to assess a 12-week modified mindfulness curriculum tailored to busy surgical trainees involving a 30 min group session weekly and 15 minutes home practice daily. The main outcomes were program feasibility and desirability. Focus groups explored how mindfulness techniques were used in the operating room. Secondary outcomes were measured in a pre- and post- intervention design assessing surgical performance, anxiety, confidence and burnout using validated assessment scales. Outcome measures were collected at baseline, immediately following the course and at 3 months postintervention. Academic obstetrics and gynecology residency program. Obstetrics and gynecology residents in postgraduate years 2-5 at the University of Toronto were invited to participate in Mindfulness in Surgical Training. Twelve (20%) out of 61 eligible residents enrolled in the program and 8 (67%) completed the course. There was a statistically significant decrease in anxiety (p < 0.001) and increase in surgical confidence (p = 0.007) following the mindfulness curriculum using validated survey tools. There was no change in burnout or surgical performance evaluations. Thematic analysis identified that mindfulness tools were beneficial and regularly utilized by participants in the operating room with sustained use 3 months post intervention. The biggest barrier to participation in the mindfulness curriculum was time. Participants felt the residency program should support ongoing mindfulness training to promote a positive culture shift. A modified mindfulness curriculum designed for surgical trainees is feasible, desirable, reduces anxiety and increases surgical confidence.

Relationship Between Continuous Improvement and Organizational Performance in A European Bank: The Mediating Role of Organizational Learning

This study attempted to determine the relationship between continuous improvement and organizational performance, with organizational learning functioning as the mediating variable, among Data and Analytics professionals at a European Bank located in Manila. The descriptive-correlational research design was used to determine if there are existing relationship between continuous improvement and organizational performance as its independent and dependent variables, respectively. The components of continuous improvement were management empowerment and leadership and employee participation. The following were the parameters used for assessing organizational performance: cost-effectiveness, business objectives, customer satisfaction, employee satisfaction, and innovation. The survey containing Continuous Improvement Questionnaire, Dimensions of Learning Organization Questionnaire (DLOQ) and Organizational Performance Questionnaire were distributed among data and analytics professionals with the goal of gathering data. The study’s findings indicated a significant relationship between continuous improvement and organizational performance. Furthermore, there was a significant mediation influence with organizational learning; however, on a partial level, as continuous improvement positively influences organizational performance. The study concluded with an understanding of an implication and the development of recommendations for the organization and future researchers that could be useful in the study of continuous improvement, organizational learning, and organizational performance.

Meta-Learning to Rank for Sparsely Supervised Queries

Supervisory signals are a critical resource for training learning to rank models. In many real-world search and retrieval scenarios, these signals may not be readily available or could be costly to obtain for some queries. The examples include domains where labeling requires professional expertise, applications with strong privacy constraints, and user engagement information that are too scarce. We refer to these scenarios as sparsely supervised queries which pose significant challenges to traditional learning to rank models. In this work, we address sparsely supervised queries by proposing a novel meta-learning to rank framework which leverages fast learning and adaption capability of meta-learning. The proposed approach accounts for the fact that different queries have different optimal parameters for their rankers, in contrast to traditional learning to rank models which only learn a global ranking model applied to all the queries. In consequence, the proposed method would yield significant advantages especially when new queries are of different characteristics with the training queries. Moreover, the proposed meta-learning to rank framework is generic and flexible. We conduct a set of comprehensive experiments on both public datasets and a real-world e-commerce dataset. The results demonstrate that the proposed meta-learning approach can significantly enhance the performance of learning to rank models with sparsely labeled queries.

Effects of multilevel metacognition on group performance and regulation in collaborative learning

Effects of multilevel metacognition on group performance and regulation in collaborative learning

Learning performance and physiological feedback-based evaluation for human–robot collaboration

The development of Industry 4.0 has resulted in tremendous transformations in the manufacturing sector to supplement the human workforce through collaboration with robots. This emphasis on a human-centered approach is a vital aspect in promoting resilience within manufacturing operations. In response, humans need to adjust to new working conditions, including sharing areas with no apparent separations and with simultaneous actions that might affect performance. At the same time, wearable technologies and applications with the potential to gather detailed and accurate human physiological data are growing rapidly. These data lead to a better understanding of evaluating human performance while considering multiple factors in human–robot collaboration. This study uses an approach for assessing human performance in human–robot collaboration. The assessment scenario necessitates understanding of how humans perceive collaborative work based on several indicators, such as perceptions of workload, performance, and physiological feedback. The participants were evaluated for around 120 min. The results showed that human performance improved as the number of repetitions increased, and the learning performance value was 92%. Other physiological indicators also exhibited decreasing values as the human performance tended to increase. The findings can help the industry to evaluate human performance based on workload, performance, and physiological feedback information. The implication of this assessment can serve as a foundation for enhancing resilience by refining work systems that are adaptable to humans without compromising performance.

Random image masking and in-batch feature mixing for self-supervised learning

Contrastive learning has already had a significant impact in the field of self-supervised learning. The contrast of positive and negative samples is critical for contrastive learning. Recently, a revolutionary breakthrough has revealed that it is possible to learn meaningful representations without the need of negative samples. Nevertheless, using two randomly augmented views of the same instance as positive samples inevitably leads to its limitations. The design of positive samples becomes critically important and often necessitates domain-specific expertise. These methods invariably emphasize maximizing the similarity between two views, with a focus on global information invariance: different views of the same image yield approximately similar representations after encoding. Hence, we propose a novel methodology aimed at generating more intricate positive instances at both the image and feature level, termed random image masking and in-batch feature mixing. The former introduces local information loss through random masking of images, compelling the model to learn generalizable representations that focus on local information. The latter generates virtual positive samples by mixing samples within the batch in feature space, breaking free from the limitations of traditional data augmentation. We validate the superiority of our proposed method through experiments on several public datasets, the proposed method significantly enhances the self-supervised learning performance for downstream tasks, particularly in classification and object detection tasks.