Downloads & Free Reading Options - Results

This study investigates how game strategies (Predict-Observe-Explain vs. traditional) interact with students' prior math performance to influence cognitive load, motivation, and mathematical reasoning outcomes in digital game-based learning. Using a quasi-experimental design with 173 fifth-grade students, linear mixed-effects models revealed a significant interaction: higher-performing students showed greater reasoning gains in POE games despite higher cognitive load, while lower-performing students benefited more from traditional games due to reduced cognitive demands and better-supported motivation. Results emphasize tailoring game-based learning to cognitive readiness and optimizing scaffolding for inclusive design.

It is still unclear on what and how tool and learner characteristics influence tool use and consequently performance in computer-based learning environments (CBLEs). This study examines the relationships between tool-related characteristics (tool presentation: non-/embedded tool and instructional cues: non-/explained tool functionality) and learner-related characteristics (self-efficacy and goal orientation) as well as their effects on tool use (quantity and quality) and performance in a CBLE. One hundred and forty students, without statically difference in prior knowledge, were randomly assigned to the four conditions (Embedded and non-embedded with explained tool functionality and embedded and non-embedded with non-explained tool functionality) to study a hypertext. Results reveal that embedding tools influenced positively quantity of tool use and negatively quality of tool use. Partial effects of explained tool functionality were found. There were significant interactions of goal orientation (mastery avoidance) and condition on quality of tool use. Performance approach influenced quality of tool positively and self-efficacy influenced negatively quantity of tool use. Only quantity of tool use affected performance. The implications of these results for future research on tool use in CBLE's are discussed. (Contains 3 tables and 8 figures.)

This work is in support of a multi-year project to evaluate people-centered aspects of computer-based training (CBT) or technology assisted instruction. The core component of the work will be developing a longitudinal database that will include individual difference measures of non-cognitive characteristics (e.g., personality, temperament, interests) of students, student engagement and satisfaction with instruction, student outcomes, and early job satisfaction and performance. This report is a selected literature review to provide a starting point from which to view the interaction between individual differences in non-cognitive characteristics, instructional methods, and learning outcomes to inform decisions on which non-cognitive characteristics to use in the research. Two-hundred and thirty-five research and theoretical articles, book chapters, and reports were reviewed. Dozens of non-cognitive characteristics and their association with learning outcomes were scrutinized, with a special emphasis on those associated with computer-based training. Eight categories of individual difference constructs were found to be important predictors of outcomes. motivation to learn; intrinsic motivation; metacognitive abilities; goal orientation; personal control beliefs; personality measures (e.g., conscientiousness, flexibility, emotional control); organizational commitment and perceptions of fairness; and attitudes towards training. While the results were not predetermined, following the review of critical non-cognitive characteristics, the recommendation is to utilize a Navy-developed battery of personality traits referred to as the Navy Computer Adaptive Personality Scales (NCAPS) because it's 19 traits cover the characteristics that should be measured, the battery is owned by the Navy and is not proprietary, has proven to be fake resistant, has been evaluated many Navy personnel, and it is web-enabled which reduces barriers to administration.

This thesis evaluated the first iteration of Navy Postgraduate School (NPS) on-line course Space Systems - Technology and Applications'. The thesis goal was to evaluate: course design; course effectiveness; student expectations and motivation before and affective reactions after the course; and typical patterns of on-line learning behavior according to Long's theory. Ultimately, the thesis intended to derive lessons learned and recommendations for future distributed learning (DL) courses and future research in this area There were thirty regular, resident NPS students in the sample. Data was collected from nine sources: four on-line questionnaires, Student Opinion Forms, on-line tracking:, final grades, interviews with students, and interviews with the instructor. Even though the course first course taught by this instructor, it was successful. Student expectations, motivation, affective and utility reactions were predominantly positive. For students, the most important advantage of the course was time flexibility and convenience. Students and the instructor agreed about central role of interactivity/feedback in a DL, which should be included into models of training effectiveness evaluation. Long-Dziuban's protocol results were surprising, but solid conclusions cannot be made without focused research. Results provided several recommendations about the course design, pedagogical improvements, instructional counseling ,and future research.

This paper describes the history of education reform in Washington State since the early 1980s. Specifically, the paper traces the educational system's transition to one based on performance and student learning. During the 1980s, education reform measures in Washington were targeted at segments of the system, rather than at the whole system. In 1992, the state legislature passed Engrossed Substitute House Bill (ESHB) 5953, which created a Commission on Student Learning, mandated a local control provision, and reviewed teacher-preparations and certification requirements. In 1993, the legislature passed ESHB 1209, which placed four student learning goals in statute and created performance standards. Subsequent developments included the establishment of Subject Advisory Councils, citizen/educator groups who help to develop the essential academic learning requirements, and Student Learning Improvement Grants that provide additional time and resources for site-based planning and staff development. In 1994, Washington's application for federal funding under the Goals 2000 Act was accepted. (LMI)

This paper discusses the use of portfolios, performance assessments, and standards-based learning. The National Council for Accreditation of Teacher Education and other professional groups call for performance assessments that are as authentic as possible, and authenticity, in this context, means resembling real-world challenges. A portfolio is simply a collection of these authentic tasks organized to demonstrate student competence in various standards. Portfolios can be used in a variety of ways, but should be guided by design principles whether in paper and pencil or electronic formats. Portfolio design begins with preplanning that includes definition of contents. Development of scoring guides and effective rubrics is essential so that the portfolio can be constructed to meet these rubrics and then evaluated through them. (Contains 12 references.) (SLD)

In this paper we study the problem of content-based image retrieval. In this problem, the most popular performance measure is the top precision measure, and the most important component of a retrieval system is the similarity function used to compare a query image against a database image. However, up to now, there is no existing similarity learning method proposed to optimize the top precision measure. To fill this gap, in this paper, we propose a novel similarity learning method to maximize the top precision measure. We model this problem as a minimization problem with an objective function as the combination of the losses of the relevant images ranked behind the top-ranked irrelevant image, and the squared Frobenius norm of the similarity function parameter. This minimization problem is solved as a quadratic programming problem. The experiments over two benchmark data sets show the advantages of the proposed method over other similarity learning methods when the top precision is used as the performance measure.

Heterogeneous computing, which combines devices with different architectures, is rising in popularity, and promises increased performance combined with reduced energy consumption. OpenCL has been proposed as a standard for programing such systems, and offers functional portability. It does, however, suffer from poor performance portability, code tuned for one device must be re-tuned to achieve good performance on another device. In this paper, we use machine learning-based auto-tuning to address this problem. Benchmarks are run on a random subset of the entire tuning parameter configuration space, and the results are used to build an artificial neural network based model. The model can then be used to find interesting parts of the parameter space for further search. We evaluate our method with different benchmarks, on several devices, including an Intel i7 3770 CPU, an Nvidia K40 GPU and an AMD Radeon HD 7970 GPU. Our model achieves a mean relative error as low as 6.1%, and is able to find configurations as little as 1.3% worse than the global minimum.

Heterogeneous computing, which combines devices with different architectures, is rising in popularity, and promises increased performance combined with reduced energy consumption. OpenCL has been proposed as a standard for programing such systems, and offers functional portability. It does, however, suffer from poor performance portability, code tuned for one device must be re-tuned to achieve good performance on another device. In this paper, we use machine learning-based auto-tuning to address this problem. Benchmarks are run on a random subset of the entire tuning parameter configuration space, and the results are used to build an artificial neural network based model. The model can then be used to find interesting parts of the parameter space for further search. We evaluate our method with different benchmarks, on several devices, including an Intel i7 3770 CPU, an Nvidia K40 GPU and an AMD Radeon HD 7970 GPU. Our model achieves a mean relative error as low as 6.1%, and is able to find configurations as little as 1.3% worse than the global minimum.

In this paper, we investigate the problem of optimization multivariate performance measures, and propose a novel algorithm for it. Different from traditional machine learning methods which optimize simple loss functions to learn prediction function, the problem studied in this paper is how to learn effective hyper-predictor for a tuple of data points, so that a complex loss function corresponding to a multivariate performance measure can be minimized. We propose to present the tuple of data points to a tuple of sparse codes via a dictionary, and then apply a linear function to compare a sparse code against a give candidate class label. To learn the dictionary, sparse codes, and parameter of the linear function, we propose a joint optimization problem. In this problem, the both the reconstruction error and sparsity of sparse code, and the upper bound of the complex loss function are minimized. Moreover, the upper bound of the loss function is approximated by the sparse codes and the linear function parameter. To optimize this problem, we develop an iterative algorithm based on descent gradient methods to learn the sparse codes and hyper-predictor parameter alternately. Experiment results on some benchmark data sets show the advantage of the proposed methods over other state-of-the-art algorithms.

Model-based reasoning about physical systems has several well-known advantages over heuristic expert systems. These include correctness of conclusions, explanations of conclusions, ease of modifiability and ease of transfer of expertise to new physical systems. On the other hand, reasoning from a model can be slow. This thesis explores ways to augment a model-based diagnostic program with a learning component, so that it speeds up as it solves problems. Several learning components are proposed, each exploiting a different kind of similarity between diagnostic examples. Through analysis and experiments, we explore the effect each learning component has on the performance of a model-based diagnostic program. We also analyze more abstractly the performance effects of Explanation-Based Generalization, a technology that is used in several of the proposed learning components.

Technology-based systems for education, training, and performance-aiding (including decision-aiding) may pose the ultimate test for validating approaches to integrate humans with automated systems. These systems need to model students and users. The models they generate, as well as the interactions based on them, must adapt to the evolving knowledge and skills of individual students and users. Evaluation findings suggest that such adaptations are feasible, worthwhile, and cost effective. Data drawn from many evaluations of technology-based education and training indicate overall that these systems can reduce costs by one-third and that they can also either reduce the time to achieve instructional objectives by one-third or increase achievement (holding time constant) by one-third. The likely impact on military readiness and effectiveness suggested by these findings is significant. Evaluations of technology-based performance aiding systems have suggested similar results of increased personnel effectiveness and cost-savings. They suggest a need to determine and readjust the balance between resources allocated to training and resources allocated to performance-aiding systems. Development of sharable, reusable objects and capabilities for assembling these objects on demand and in real time will substantially increase accessibility and will reduce costs of education, training, and performance-aiding while making them asynchronously and continuously available regardless of distance and time. Specifications and capabilities for such objects are the goals of much current research and development (R&D).

Adaptive e-Learning systems are able to adjust to a user's learning needs, usually by user modeling or tracking progress. Such learner-adaptive behavior has rapidly become a hot topic for e-Learning, furthered in part by the recent rapid increase in the use of MOOCs (Massive Open Online Courses). A lack of general, individual, and situational data about student populations currently hampers the infusion of effective adaptive behavior into existing e-Learning platforms. This contribution presents original research on using differences in individual learning styles. Factors related to performance, motivation, satisfaction, and previous knowledge were targeted and used to assess the effectiveness of the approach. We discuss alternative bases for adaptation (e.g. cognitive styles), style distributions in student populations, and conclude with repercussions for adaptive behavior in HCI in general. [For the full proceedings, see ED562095.]

Sixteen scholarly papers and twenty abstracts comprise the content of the eighteenth annual Brick & Click Libraries Conference, held annually at Northwest Missouri State University in Maryville, Missouri. The proceedings, authored by academic librarians and presented at the conference, portray the contemporary and future face of librarianship. The 2018 paper and abstract titles include: (1) Committee on Diversity & Inclusion: Cultivating an Inclusive Library Environment (Orolando Duffus, Andrea Malone, Margaret Dunn, Lisa Cruces, Matthew Moore, Annie Wu, and Frederick Young); (2) Checking Out the LGBT+ (Kayla Reed); (3) Tailoring Library Instruction to Adult Students: Applying the Science and Methods of Andragogy for Modern Instructional and Reference Services (Eric Deatherage and Jason Smith); (4) Library-Faculty Collaboration for OER Promotion and Implementation (Paula Martin); (5) The Facts of Fiction: Research for Creative Writers (Addison Lucchi); (6) Location and the Collection Connection (Kayla Reed and Amber Carr); (7) Gay for No Pay: How to Maintain an LGBTQ+ Collection with No Budget (Rachel Wexelbaum); (8) A Step Up: Piloting Integrated Information Literacy Instruction Throughout a Discipline (Nathan Elwood and Robyn Hartman); (9) Not Just a Collection: The Emergence and Evolution of Our Contemporary Collection (Hong Li and Kayla Reed); (10) Flipster: How One Community College Library Supports Faculty and Student Academic Needs with Flipster Digital Magazines (Stephen Ambra); (11) Three Ring Circus: A Model for Understanding and Teaching Students about Bias (Virginia Cairns); (12) Demystifying DH: How to Get Started with Digital Humanities (Sherri Brown and Forstot Burke); (13) Academic Libraries Embracing Technology with a Purpose (Lavoris Martin); (14) (A)ffective Management: A People First Management Approach (Ryan Weir); (15) Plugged & Unplugged Active Learning Strategies for One Shots (Judy Bastin, Justina Mollach, Leslie Pierson, Ruth Harries, and Teresa Mayginnes); (16) Giving a Booster Shot to Your One Shot: Incorporating Engaging Activities into Library Instruction (Kelly Leahy, Gwen Wilson, and Angela Beatie); (17) Adventures with Omeka.net: Metadata, Workflows, and Exhibit-based Storytelling at UNO Libraries (Yumi Ohira, Angela Kroeger, and Lori Schwartz); (18) Online Badge Classes for High School Students (Angela Paul); (19) Fake News: The Fun, the Fear, and the Future of Resource Evaluation (Lindsay Brownfield); (20) Making Outreach the Library's Mission (April K. Miller); (21) Active Learning for Metaliteracies: Digital Modules from the New Literacies Alliance (Rachel R. Vukas, Prasanna Vaduvathiriyan, and Brenda Linares); (22) Calculating Return on Investment in Libraries (Nicholas Wyant); (23) Crossing Borders: Expanding Digitization Efforts Across Library Departments (Jay Trask, Jane Monson, and Jessica Hayden); (24) From Silos to Collaboration (Joyce Meldrem); (25) Key Performance Indicator Tracking Using Google Forms (Joshua Lambert); (26) Bridging the Gap: Providing Equal Access of Library Resources and Services to Distance Learners (Nancy Crabtree, Xiaocan (Lucy) Wang, Bob Black); (27) Coming to the Plains: Latino/a Stories in Nebraska (Laurinda Weisse, Michelle Warren, and Jacob Rosdail); (28) Five Keys to #SocialMediaSuccess in Academic Libraries (Hannah E. Christian and Alison Hanner); (29) Easy Information Literacy Assessments for Small Academic Libraries (Julie Pinnell); (30) Traversing the Path: A Library Director's Guide to the Higher Learning Commission's Open Pathway for Accreditation (Sandy Moore); (31) Drawing Magic: Visualizing the Internet to Introduce Information Literacy (Kelly Leahy); (32) Chatspeak for Librarians: Best Practices for Chat Reference (Tanner D. Lewey); (33) The Creative Learning Spiral: A Python Learner in the Library (Greta Valentine); (34) The Poet's Papers: Literary Research in the Small College Archives (Martha A. Tanner); (35) Giving Students an Edge: Enhancing Resumes with a Digital Information Research Certificate (Rachel R. Vukas); and (36) Where Did You Get That eBook? Comparison of Student/Faculty Use of eBooks, Library Space, and Citation Management Programs (Alice B. Ruleman). (Individual papers contain references.) [For the 2017 proceedings, see ED578189.]

This article summarizes my experiences in developing an educational initiative known as the Career Fair Project within the curriculum of the Legal Administration/Law Clerk program at Durham College over the past ten years. A performance-based, student-centered project, the Career Fair empowers students to work collaboratively with one another in order to gain employment within their chosen field. In this article, I discuss the rationale and structure of the course, as well as the outcomes of the project, and methods of evaluation. I also discuss numerous problems that I have encountered along the way, and theorize potential solutions in hopes that this information might be helpful to other educators who wish to develop similar performance-based projects within a college environment.

Previously, we reported qualitative findings showing that students who experienced a problem- or enquiry-based course (EBL) in a first-year seminar program had greater confidence in their academic abilities, were more engaged, and were better prepared for upper-year courses. In the current paper, we provide quantitative data to substantiate the students' qualitative conclusions. We present results to show that these students do perform at a significantly higher level compared with members of the control group who did not experience an EBL course. Using survey data, we show that the EBL students shift the way they access information compared with peers: they preferentially use more sophisticated resources for research. At the same time, students report greater engagement in the community, and student engagement is known to contribute to increased academic performance. (Contains 3 figures and 3 tables.)

Previously, we reported qualitative findings showing that students who experienced a problem- or enquiry-based course (EBL) in a first-year seminar program had greater confidence in their academic abilities, were more engaged, and were better prepared for upper-year courses. In the current paper, we provide quantitative data to substantiate the students' qualitative conclusions. We present results to show that these students do perform at a significantly higher level compared with members of the control group who did not experience an EBL course. Using survey data, we show that the EBL students shift the way they access information compared with peers: they preferentially use more sophisticated resources for research. At the same time, students report greater engagement in the community, and student engagement is known to contribute to increased academic performance. (Contains 3 figures and 3 tables.)

The study was designed to describe the basis of a potential mathematics learning theory founded on the relationship between language and thinking, to relate the subjects' overt verbalization and performance after they had been taught a mathematical structure, to relate overt verbalization and the number of discovered rules, and to examine the strategies used. Forty girls, 11-12 years old, were randomly assigned to one of four instructional situations: (1) subjects talk aloud while doing mathematical activities and then are silent, (2) subjects are silent while doing the activities and afterwards answer questions and explain findings, (3) subjects verbalize both during and after the activities, and (4) subjects do not verbalize either during or after the activities. A machine wired to embody the Klein-Four Group structure was used as a manipulative aid. Among the results found were: (1) subjects performed better, retained more and discovered more rules when they were silent while doing the activities and afterwards answered questions and explained findings; (2) there was significant interaction between learning and questioning; (3) the subjects' overt verbalization during questioning did not accelerate learning; and (4) the nature of the rule and the number of discovered rules influenced performance. (Author/DT)

Brain tumor is an uncontrolled growth of abnormal cell in the brain. Early diagnosis of brain tumor has a crucial step in this type of cancer, which is fatal. Magnetic resonance imaging (MRI) is one of the examination tools to examine brain anatomy in clinical practice. The high resolution and clear separation of the tissue enable medical experts to identify brain tumor. The earlier of brain tumor is detected, the wider of treatment options. However, manually analysed of brain anatomy on MRI images are time-consuming. Computer-aided diagnosis with automated way is helpful solution to help management with unreliable degrees of automation to trace various tissue boundaries. This study proposes convolutional neural network (CNN) with its excellences to automated features extraction in convolution layer. The popular architectures of CNN, i.e., visual geometry group16 (VGG16), residual network-50 (resNet-50), inceptionV3, mobileNet, and efficientNetB7 in medical image processing are compared to brain tumor classification task. As the results, VGG16 outperformed other architectures of CNN in this study. VGG16 yields 100% accuracy, precision, sensitivity, specificity, and F1-score for testing set data. The results show the excellent performance in classifying brain tumor and no tumor from MRI images that demonstrate the efficiency of system suggested.

With the rapid growth of digital technology communications are overwhelmed by network data traffic. The demand for the internet is growing every day in today's cyber world, raising concerns about network security. Big Data are a term that describes a vast volume of complicated data that is critical for evaluating network patterns and determining what has occurred in the network. Therefore, detecting attacks in a large network is challenging. Intrusion detection system (IDS) is a promising cybersecurity research field. In this paper, we proposed an efficient classification scheme for IDS, which is divided into two procedures, on the CSE-CIC-IDS-2018 dataset, data pre-processing techniques including under-sampling, feature selection, and classifier algorithms were used to assess and decide the best performing model to classify invaders. We have implemented and compared seven classifier machine learning algorithms with various criteria. This work explored the application of the random forest (RF) for feature selection in conjunction with machine learning (ML) techniques including linear regression (LR), k-Nearest Neighbor (k-NN), classification and regression trees (CART), Bayes, RF, multi layer perceptron (MLP), and XGBoost in order to implement IDSS. The experimental results show that the MLP algorithm in the most successful with best performance with evaluation matrix.

Today the U.S. higher education system is facing a crisis regarding its perceived quality. One model for improving quality is competency-based education, in which an institution clearly defines the specific competencies expected of its graduates. A key challenge is how to help more people, particularly adults, succeed at the post¬secondary level and earn degrees. This paper examines the current state of competency-based postsecondary education in the U.S., profiling the various types of competency-based, or competency-focused, models that currently exist, the extent to which these programs assess for student competencies or learning outcomes, and the extent to which these programs operate outside of a credit-based system. These programs can help inform other institutions interested in developing a stronger focus on competencies, whether by demonstrating the possibilities of high-quality programs or by facilitating the recognition of learning acquired outside of traditional instruction. Included in the report are 11 appendices: (A-1) University of Maryland University College Computer and Information Science Program Outcomes Guide; (A-2) University of Maryland University College Institutional Level Learning Outcomes: Student Learning Expectations; (A-3) Alverno College Social Science Example of Relationships among Outcomes; (A-4) Brandman University's Degree Qualifications; (A-5) Marylhurst University's Liberal Arts Core Outcomes: (A-6) Westminster College Consumers and Markets Sequence; (A-7) Southern New Hampshire University Year One Modules, Themes, and Integrating Experiences; (A-8) Southern New Hampshire University Year One Competency Reinforcement Plan; (A-9) Western Governors University Domain for Bachelor of Science in Accounting; (A-10) Western Governors University Subdomains and Competencies for Accounting Domain;, and (A-11) DePaul University School for New Learning's Master of Arts in Applied Professional Studies (MAAPS). [See the follow-up report: "2013 CAEL Forum & News: Competency-Based Education" in ERIC at ED547415.]

There is a new conversation taking place in public education on creating systemic incentives through school finance to encourage schools to innovate and be rewarded for positive student outcomes and performance. What if education funding was not based on seat-time, but on rewarding student performance? Performance-based funding is a term that captures this new concept. Performance-based funding means that funding is tied to an outcome-a policy outcome. In higher education, institutions seek outcomes tied to degree completion. In K-12 education, the performance-based funding outcomes have historically been tied to course completion (as the outcome). As students successfully develop competencies and complete courses, they would trigger payment and performance-based funding incentives. It is important to protect quality and ensure student performance outcomes are validated through independent assessments and/or end-of-course exams. As interest in online courses grows, many states are looking for guidance regarding costs, models and performance-based funding policies. The research in this report is focused on online learning to focus on the issues of equity and adequacy of funding in the context of performance-based funding models. Appended are: (1) K-12 Education Funding; (2) Examining Costs of Online Learning; (3) Outcomes-Based Measures of Student Performance for Online Schools; (4) Individuals Participating in Research Meetings and Phone Interviews (5) Definitions; and (6) References and Resources.

There is a new conversation taking place in public education on creating systemic incentives through school finance to encourage schools to innovate and be rewarded for positive student outcomes and performance. What if education funding was not based on seat-time, but on rewarding student performance? Performance-based funding is a term that captures this new concept. Performance-based funding means that funding is tied to an outcome-a policy outcome. In higher education, institutions seek outcomes tied to degree completion. In K-12 education, the performance-based funding outcomes have historically been tied to course completion (as the outcome). As students successfully develop competencies and complete courses, they would trigger payment and performance-based funding incentives. It is important to protect quality and ensure student performance outcomes are validated through independent assessments and/or end-of-course exams. As interest in online courses grows, many states are looking for guidance regarding costs, models and performance-based funding policies. The research in this report is focused on online learning to focus on the issues of equity and adequacy of funding in the context of performance-based funding models. Appended are: (1) K-12 Education Funding; (2) Examining Costs of Online Learning; (3) Outcomes-Based Measures of Student Performance for Online Schools; (4) Individuals Participating in Research Meetings and Phone Interviews (5) Definitions; and (6) References and Resources.

The enhanced penetration of the renewable energy sources (RES) is dependent on microgrid (MG) to a power system is impact stability of the system due to a variation in dynamic properties of the MG from a traditional generator. As a result, analyzing the new issues with dynamic stability and controlling the operation of the power system in the connection of rising MG penetration becomes critical. This paper contains a MG system with renewable energy assisted, superconducting magnetic energy storage (SMES) storage and an extreme learning machine (ELM) based proportional integral derivative (PID) controller. The effect of renewable-based MG penetration on a dynamic stability and control of the multi machine multi area system under varied operating situations is comprehensively investigated in this study. Non-linear time-domain simulations and several performance indicators are used to evaluate the controller's ability with the different MG penetration percentages under various disturbances and operational conditions.

The development of a voluminous database aimed at performance evaluation of the vehicle detection and tracking algorithms is described here. The vehicle database has been created using many recorded traffic videos and annotated automatically by applying some convolutional neural network (CNN) – based object detectors. It has been split into training, validation and testing datasets and then successfully used to train, validate and test deep learning-based vehicle detectors. Some multiple vehicle detection and tracking simulations are also described. A transfer learning-based vehicle classification solution using this database and those detection and counting results is also provided here.

This article explores the theoretical foundations of cybersecurity self-efficacy, metacognition—particularly Judgment of Performance (JoP)—and learning achievement in the context of cybersecurity education. The paper outlines the role of self-efficacy in shaping individual motivation, behavioral regulation, and performance in high-stakes digital environments. It integrates findings from various educational and operational settings, highlighting how hands-on experiences such as Capture the Flag (CTF) competitions, gamified learning platforms, and AI-supported tutoring systems can enhance learners’ confidence and cognitive regulation. Special attention is given to the interaction between metacognitive skills and JoP accuracy, which are shown to support self-regulated learning, strategic decision-making, and resilience under pressure. The manuscript also discusses emerging tools like learning analytics and immersive environments that support real-time feedback and reflective practices. This theoretical framework forms the basis for empirical studies that assess student feedback and task performance in immersive VR/3D cybersecurity challenges, collected during cyber training events in 2025. These insights are intended to support the development of more effective educational interventions and tools for evaluating learner progress and experience in cybersecurity education. This study was funded by Digital Europe Programme: Project ATHENA (grant number 101127970). This study was funded by Digital Europe Programme: Project CHESS ( grant number 101087529 This study was funded by Digital Europe Programme: Project: 101158617 — EE SIC VIII — DIGITAL-2023-DEPLOY-04

This paper reports on two research studies. The first study investigates teachers monitoring strategies during computer-assisted composition instruction. The findings reveal four principles underlying a strategic, proximal instruction process. The four principles are collaborative assessment, guided practice, instructional branching, and learner self-monitoring skills development. The computer is described as a cognitive tool supporting and facilitating teachers' active involvement in students' writing process, which represents a change in the traditional process-writing pedagogy that focuses on the analysis of students' writing products. The second study reports on a related issue concerning the design of electronic performance-based assessment systems. The findings are a set of guidelines instructional designers and educators can use in planning for the use of electronic assessment systems. Both studies point to the need for technological solutions to provide efficient, valid, and reliable information to teachers and learners in order to enhance the instructional processes that accompany active, engaged learning experiences. Two figures illustrate: a common interaction pattern between teacher, students, and peers in a computerized learning environment; and the proximal instruction model. A table diagramming the guidelines for management of performance-based assessment and an appendix outlining the guidelines formed in the second study are also included. (Contains 49 references.) (Author/DLS)

Social and emotional (SE) skills are interpersonal, self-regulatory, and task-related behaviors that are important for adaption to and successful performance in educational and workplace settings (Casillas et al., 2015). Students can improve these skills through social and emotional learning (SEL) interventions or programming designed to teach SE skill development to students in school settings (e.g., Mahoney et al., 2018). While teaching SEL shows promising results and continues to gain traction in the K-12 education space, little consensus exists on what label should be used to describe this skill set (e.g., Berg et al., 2017). A recent study with parents suggests that, although support for social and emotional learning (SEL)-related skills is relatively strong, the label "social-emotional learning" itself is unfavorable (Tyner, 2021). This conclusion was based on a rank order item which presented parents with a list of 12 undefined, SEL-related program titles and asked them to select three school programs they would most and least want their child enrolled in. To test the assumption that respondents perceived all program labels as teaching SEL, the authors conducted a survey with parents and caregivers of ACT® test takers. Results from this follow-up study demonstrate that the program label "life skills" represents a far broader range of skills than those we would typically conceptualize as SE skills. While skills such as financial management, completing taxes, and taking care of homes and automobiles are undoubtedly important, they are fundamentally different from SE skills such as persistence, empathy, and resilience. [For the 2021 study "How to Sell SEL: Parents and the Politics of Social-Emotional Learning," see ED616280.]

Social and emotional (SE) skills are interpersonal, self-regulatory, and task-related behaviors that are important for adaption to and successful performance in educational and workplace settings (Casillas et al., 2015). Students can improve these skills through social and emotional learning (SEL) interventions or programming designed to teach SE skill development to students in school settings (e.g., Mahoney et al., 2018). While teaching SEL shows promising results and continues to gain traction in the K-12 education space, little consensus exists on what label should be used to describe this skill set (e.g., Berg et al., 2017). A recent study with parents suggests that, although support for social and emotional learning (SEL)-related skills is relatively strong, the label "social-emotional learning" itself is unfavorable (Tyner, 2021). This conclusion was based on a rank order item which presented parents with a list of 12 undefined, SEL-related program titles and asked them to select three school programs they would most and least want their child enrolled in. To test the assumption that respondents perceived all program labels as teaching SEL, the authors conducted a survey with parents and caregivers of ACT® test takers. Results from this follow-up study demonstrate that the program label "life skills" represents a far broader range of skills than those we would typically conceptualize as SE skills. While skills such as financial management, completing taxes, and taking care of homes and automobiles are undoubtedly important, they are fundamentally different from SE skills such as persistence, empathy, and resilience. [For the 2021 study "How to Sell SEL: Parents and the Politics of Social-Emotional Learning," see ED616280.]

To make Alternative Learning System (ALS) more flexible and inclusive, the researchers aimed to explore the possibilities of utilizing mobile learning as a delivery mode and tool in increasing the delivery of instruction and enhancing learning experiences of its clienteles of the Accreditation and Equivalency (A&E) Program. This study is divided into three (3) folds composed of the following: (a) first phase -- design and development of the Mobile-based Enhancement Learning Material (MELMA); (b) second phase -- implementation and experiment of the developed learning material; and, (c) learning experience assessment. The researchers employed sequential mixed method research design and involved the use of both quantitative (content, technical and aesthetic acceptability level checklist; pre- and post-test; significance assessment questionnaire) and qualitative (document analysis; in-depth interviews) instruments. Three groups of respondents were also purposively selected composed of the following: (a) first phase -- fifty (50) product validators; (b) second phase -- fifteen (15) A&E learners; and, (c) third phase -- ten (10) A&E learner and teacher-respondents. Using the Analysis, Design, Development, Implementation and Evaluation (ADDIE) Model, the analyzed data went through quantitative (mean, t-test, analysis of variance) and qualitative (consensual qualitative research) means. The first phase yielded acceptable and excellent ratings for the developed mobile learning material. The second phase showed a significant difference on the pre- and post-test results of the participating learners. It helped them increase their learning performance with the aid of the developed material. The third phase also revealed that there is a positive experience both from the teachers and learners because of the flexible learning and differentiated instruction designs. [Paper presented at International Research Conference on Higher Education (4th, 2017)]

To make Alternative Learning System (ALS) more flexible and inclusive, the researchers aimed to explore the possibilities of utilizing mobile learning as a delivery mode and tool in increasing the delivery of instruction and enhancing learning experiences of its clienteles of the Accreditation and Equivalency (A&E) Program. This study is divided into three (3) folds composed of the following: (a) first phase -- design and development of the Mobile-based Enhancement Learning Material (MELMA); (b) second phase -- implementation and experiment of the developed learning material; and, (c) learning experience assessment. The researchers employed sequential mixed method research design and involved the use of both quantitative (content, technical and aesthetic acceptability level checklist; pre- and post-test; significance assessment questionnaire) and qualitative (document analysis; in-depth interviews) instruments. Three groups of respondents were also purposively selected composed of the following: (a) first phase -- fifty (50) product validators; (b) second phase -- fifteen (15) A&E learners; and, (c) third phase -- ten (10) A&E learner and teacher-respondents. Using the Analysis, Design, Development, Implementation and Evaluation (ADDIE) Model, the analyzed data went through quantitative (mean, t-test, analysis of variance) and qualitative (consensual qualitative research) means. The first phase yielded acceptable and excellent ratings for the developed mobile learning material. The second phase showed a significant difference on the pre- and post-test results of the participating learners. It helped them increase their learning performance with the aid of the developed material. The third phase also revealed that there is a positive experience both from the teachers and learners because of the flexible learning and differentiated instruction designs. [Paper presented at International Research Conference on Higher Education (4th, 2017)]

Breast cancer is the one common cause of death in both developed worlds and the most death-causing disease diagnosed among women. Early recognition of this condition can help to minimize death rates. The breast problem statement, in brief, is not reliable for accuracy recognition. They have a high degree of classification accuracy as well as diagnostic capabilities. The most common classifications are normal, benign cancer, and malignant cancer. Machine learning (ML) techniques are now widely used in the classification of breast cancer. In this paper, some machine learning technics have been investigated to diagnose breast cancer (BC) on magnetic resonance imaging (MRI) images using multi-step processes. The first step has been to take the MRI image as an input image and have been pre-processing an image, then use feature extraction by using (scaleinvariant feature transform (SIFT), histogram of oriented gradient (HOG), local binary patterns (LBP), bag of words (BoW), and edge-oriented histogram (EOH)). Next step we implement the classifying algorithms (KNN, decision tree (DT), naïve Bayes, ANN, SVM, RF, AdaBoost), have been used to detect and classify the normal or breast cancer region for this purpose datasets like ACRIN-Contralateral-Breast-MRI, In and breast cancer MRI dataset) has been collected our breast cancer MRI images from Erbil and Sulaymaniyah hospital the results was 91.9%, the result of ACRIN was 97% and the results Breast Cancer was 92.3%.

This document, which is part of a series of training materials explaining the benefits and use of performance-focused learning, examines the goal-based learning approach to education and training as a means of accelerating performance change. Each page of the document consists of a graphic and bulleted list of information pertaining to one of the following topics: changing performance; performance change curves; shifting the focus of education; knowledge and performance; comparison of learning environments; structure in a lecture-based environment; structure in a goal-based environment; developing a web of support for goal-based learning; typical support for goal-based learning; forms of goal-based learning; what goal-based learning is not; the paradox of goal-based learning as learning that is both flexible and consistent; developing goal-based learning; key lessons learned from goal-based learning; organizational readiness for goal-based learning; and other readiness factors. Contains 11 references. (MN)

In the context of the lockdown area of COVID-19, animated and gag cartoon-based intervention has a significant role in promoting students' learning performance at elementary levels. ICT-based intervention on students with ADHD mostly uses in UK, USA, Greece, Iran, and Poland. Still, recent researches are conducting in India,especially in the Northeastern region, to assess its effectiveness. Based on the literature and research questions,the current study aimed to evaluate the effect of animated and gag cartoon-based e-contents intervention on experimental groups' learning performance with those in the comparison group in Silchar town, Assam. 90 students with ADHD were assigned to experimental groups and a comparison group. The experimental group I (n=30, age ranged 10.5-11.5 SD= 11) was exposed to animated cartoon based e-contents instruction, and experimental group II (n=30, age ranged 10.5-11.5 SD= 11) was assigned to gag cartoon based e-contents in Environmental studies. Simultaneously, the traditional approach was used in the comparison group (n=30, age ranged 10.5-11.5 SD= 11). The quasi-experimental design was used to examine the effect of animated and gag cartoon-based e-contents intervention on experimental groups' learning performance with those in the comparison group. It resulted in animated cartoon-based e-contents and gag cartoon-based e-content intervention that significantly affected participants' performance over the comparison group.

This study aims to summarise and critically appraise the literature examining the effect of PCT interventions on the learning and/or performance of sports skills.

The first stage of a 3-part research project designed to assess the appropriateness of using cooperative learning groups in communication courses with a significant oral performance component, a study assessed student perceptions about the appropriateness of using cooperative learning groups in a communication class with more than half of the final grade based on performance in oral performances. Students in five sections of the basic public speaking course taught by three instructors participated. Instructors for each section were given instructions about how to structure cooperative learning groups. A total of 49 useable surveys were collected at the end of the semester, a response rate of 81.6%. Results indicated that students (1) judged the cooperative learning methods used to be effective; (2) felt more prepared as a result of the required practice sessions structured into the cooperative learning group; (3) were actively involved in providing feedback to members of their learning group, and that overall the feedback was judged to be valuable; (4) found the cooperative learning groups to be useful in helping them control their public speaking anxiety; and (5) felt that their competence as communicators generally and public speakers specifically improved as a result of interaction in the cooperative learning groups. (Contains 26 references, 11 notes, and 6 tables of data.) (RS)

In January 2010, the General Assembly passed the Complete College Tennessee Act (CCTA), a comprehensive reform agenda seeking to transform public higher education through changes in academic, fiscal, and administrative policies at the state and institutional levels. While the higher education landscape has been shaped by the CCTA, it is also evolving with the adoption of the "Drive to 55," and the FOCUS Act, which alters the governance structure of higher education by giving six universities, previously under the Tennessee Board of Regents, independent governing boards. This year, Tennessee Higher Education Commission (THEC) released an update to the 2015-2025 Master Plan entitled "Enabling the Competitive Edge: Tennessee Higher Education in the New Economy." The update on Tennessee's higher education landscape and progress toward the "Drive to 55" emphasizes three domains: student success, family prosperity, and the future workforce. This Fact Book includes the following performance categories and illustrative indicators: (1) Student preparation, such as admission rates, freshman class profiles, and learning support placement and success rates, by subject area; (2) Student participation, such as college-going rates, overall enrollment, and enrollment by critical student subpopulations; (3) Student progression, such as end-of-term enrollment counts, freshman-to-sophomore retention rates, the number of students passing credit hour benchmarks under the higher education funding formula and lottery scholarship renewal rates; (4) Student success and completion, such as student transfer activity and subsequent academic performance, graduation rates, time to degree, credentials awarded, and credentials awarded per one hundred (100) full-time equivalent enrolled students; (5) Workforce participation, such as labor market supply and demand, employer satisfaction survey results, job placement rates, and licensure passage rates; (6) Academic trends, such as student engagement survey results, changes to the academic program inventory, low-producing academic programs, the number and percentage of accredited programs, and the percentage of lower division instructional courses taught by full-time faculty, part-time faculty, and graduate assistants; (7) Financing trends, such as state appropriation levels and net tuition revenues, state and total subsidies per student, and degree costs; and (8) Affordability trends, such as in-state and out-of-state tuition rates, net costs of attendance, and need-based and merit-based student financial aid. [For "Tennessee Higher Education Fact Book: 2018-19," see ED599417.]

To attain highly efficient instructional conditions, it is important to adapt instruction to the individual trainee. This so-called personalization of training by dynamic/automatic task selection is the focus of this paper. Recently, cognitive load measures have been proposed as a useful addition to conventional performance measures like speed and accuracy. The combination of conventional performance measures and cognitive load measures can be used to obtain information about the mental efficiency of instructional conditions. The paper argues that the dynamic/automatic selection of learning tasks on the basis of mental efficiency will have a significant influence on the optimization of the learning process. First, a review of various task selection approaches is given. A distinction is made between static and dynamic task selection approaches. In both procedures the training is based on the trainee's prior knowledge. However, the selection of a set of learning tasks can either be chosen by the teacher/trainer prior to the start of the training (static procedures) or can be adjusted during the training (dynamic procedures). It has been proposed that when teaching student complex cognitive skills, part-task training can have a higher learning efficiency and reduced training costs than whole-task training (Wightman & Lintern, 1985). The whole-task and part-task approaches are used as subcategories for static and dynamic selection, yielding four approaches. Then, a new approach using mental efficiency in a dynamic whole-task procedure is presented. (Contains 31 references.) (Author/AEF)

School districts altered instructional delivery from solely in-person to a combined virtual/in-person model in the 2020-2021 school year in response to the COVID-19 pandemic. The Houston Independent School District (HISD) offered in-person and virtual instruction to early learners at school-based programs (SBP), early childhood centers (ECC), and charter/magnet schools. The analyses were conducted across the three campus types (SBP, ECC, and charter/magnet). Descriptive analyses were performed using enrollment trends, teachers' and parents' perceptions of HISDs virtual prekindergarten program compared to charter and magnet schools, and student performance. Students' academic growth was measured using beginning-of-year and end-of-year academic performance on the CIRCLE language and literacy and mathematics assessments in the 2020-2021 school year. The analysis showed a decrease of 28.4 percentage points in the number of prekindergarten students in the district. Of those enrolled in the prekindergarten program, 50.1 percent were learning in person, 28.7 percent learned virtually, and 21.2 percent alternated between virtual and in-person. Teachers with prior experience had higher mean ratings for using technology for virtual instruction and supporting parents with accessing virtual learning tools. A higher percentage of teachers at ECCs reported using technology in the classrooms before the implementation of the virtual learning model (96.2%) compared to teachers at SBPs (87.8%) and charter/magnet schools (81.4%). Parents' and teachers' perceptions of students' academic progress in the 2020-2021 school year varied by campus type. A lower proportion of teachers at ECCs reported a high confidence level that their students would make academic progress (41.2%) compared to ECCs (52.2%) and charter/magnet schools (63.6%). However, a higher proportion of parents whose child attended an ECC reported their child's school performance was much better/better than expected compared to parents of prekindergartners at SBPs (66.7%) and charter/magnet schools (50.5%). Prekindergarten students who learned in person showed a higher increase in academic performance on CIRCLE language and literacy and mathematics assessments compared to virtual learners. [Cover title: "Descriptive Analysis of Teachers' and Parents' Perceptions of Virtual Prekindergarten Program and Learning Gains among Prekindergarten Children in HISD, 2020-2021."]

The concept of classification of organisms in Biology seems to pose a problem to Secondary School students in Kenya. Though, the topic is important for understanding of the basic elements of the subject. The Examinations Council in Kenya has identified teacher centred pedagogical techniques as one of the main causes for this. Project based learning technique has been successfully used as an instructional technique in countries like Japan and United States of America to improve students' academic achievement and positive attitude development among students in science practicals in Biology. The study therefore investigated the influence of project based learning technique on achievement of students in classification of organisms among Secondary Schools in Kenya with particular focus on academic achievement and attitude development. Quasi experimental design was employed. Stratified random sampling technique was used to select eight classes from four secondary schools with a total of 360 students. Students' Attitude Questionnaires and Classification Achievement Test were used as instruments. Analysis of Variance and t-test were employed in data analysis. The study revealed that project based learning technique enabled students to improve in academic achievement as well as developing positive attitude towards classification of organisms. The study recommends to the Ministry of Education and its agencies to adopt project based learning technique among other student centred instructional techniques for teaching of practical oriented concepts like Classification of Organisms.

Developing an engineering design is resource-intensive and time consuming, particularly for the floats of a floatplane design, due to its complexity and limited testing facilities. Intelligent-based computational design (IBCD) techniques, which integrate computational design techniques and machine learning (ML) algorithms, offer a solution to reduce required testing by providing predictions. This paper proposes a deep learning (DL) based IBCD method for modeling floats' lift-to-drag coefficient ratio (CL/CD), where DL is one of the most powerful ML. The proposed method consists of two phases: hyper-parameter optimization and DL model training and evaluation. A genetic algorithm (GA) is employed in the first phase to explore complex hyper-parameter combinations efficiently. Evaluation of the predicted CL/CD of the floats using the DL model resulted in a satisfactory R-squared of 0.9329 and the lowest mean squared error (MSE) of 0,001536. These results demonstrate the ability of DL model to predict the float's performance accurately and can facilitate further design optimization. Thus, the proposed method can offer a time-efficient and cost effective solution for predicting float performance, aiding in optimizing floatplane designs and enhancing their functionalities.

Developing an engineering design is resource-intensive and time consuming, particularly for the floats of a floatplane design, due to its complexity and limited testing facilities. Intelligent-based computational design (IBCD) techniques, which integrate computational design techniques and machine learning (ML) algorithms, offer a solution to reduce required testing by providing predictions. This paper proposes a deep learning (DL) based IBCD method for modeling floats' lift-to-drag coefficient ratio (CL/CD), where DL is one of the most powerful ML. The proposed method consists of two phases: hyper-parameter optimization and DL model training and evaluation. A genetic algorithm (GA) is employed in the first phase to explore complex hyper-parameter combinations efficiently. Evaluation of the predicted CL/CD of the floats using the DL model resulted in a satisfactory R-squared of 0.9329 and the lowest mean squared error (MSE) of 0,001536. These results demonstrate the ability of DL model to predict the float's performance accurately and can facilitate further design optimization. Thus, the proposed method can offer a time-efficient and cost effective solution for predicting float performance, aiding in optimizing floatplane designs and enhancing their functionalities.

The evolution of the internet of things as a promising and modern technology has facilitated daily life. Its emergence was accompanied by challenges represented by its frequent exposure to attacks and its being a target for intruders who exploit the gaps in this technology in terms of the nature of its heterogeneous data and its large quantity. This made the study of cyber security an urgent necessity to monitor infrastructures It has network flaw detection and intrusion detection that helps protect the network by detecting attacks early and preventing them. As a result of advances in machine learning techniques, especially deep learning and its ability to selflearning and feature extraction with high accuracy, the research exploits deep learning to analyze the real data set of CSE-CIC-IDS2018 network traffic, which includes normal behavior and attacks, and evaluate our deep model long short-term memory (LSTM), That achieves accuracy of detection up to 99%. 

K-means is an iterative algorithm used with clustering task. It has more characteristics such as simplicity. In the same time, it suffers from some of drawbacks, sensitivity to initial centroid values that may produce bad results, they are based on the initial centroids of clusters that would be selected randomly. More suggestions have been given in order to overcome this problem. Ensemble learning is a method used in clustering; multiple runs are executed that produce different results for the same data set. Then the final results are driven. According to this hypothesis, more ensemble learning techniques have been suggested to deal with the clustering problem. One of these techniques is "Three ways method". However, in this paper, three ways method as an ensemble technique would be suggested to be merged with kmean algorithm in order to improve its performance and reduce the impact of initial centroids on results. Then it was compared with traditional k-means results through practical work that was executed using popular data set. The evaluation of the hypothesis was done through computing related metrics.

Performance-based learning is one of the keys to cracking open the assumptions that undergird the current educational codes, structures, and practices. By finally moving beyond the traditions of a time-based system, greater customized educational services can flourish, preparing more and more students for college and careers. This proposed policy framework, designed to expedite state policy development in performance-based learning, may be applied to all next generation learning. Building upon the 2011 Competency-Based Learning Summit convened by the International Association for K-12 Online Learning (iNACOL) and the Council of Chief State School Officers (CCSSO), this discussion explores how state policy can loosen the regulatory environment that is handcuffing administrators and educators who are ready to move toward student-centered, competency-based models of learning. Transitioning to a competency-based system requires deep analysis and wide-reaching creativity. Thus, chief state school officers will want to work collaboratively, drawing on insights and innovations from other states in order to expedite the process of constructing a set of policies that promote innovation and breakthrough strategies, rather than the traditional compliance model. (Contains 8 resources, 7 online resources and 3 footnotes.) [Funding for this paper was provided by the Stupski Foundation. For related report, "It's Not a Matter of Time: Highlights from the 2011 Competency-Based Summit," see ED537332.]

The aim of this work is to adapt and test, in a Brazilian public school, the ACE model proposed by Borkulo for evaluating student performance as a teaching-learning process based on computational modeling systems. The ACE model is based on different types of reasoning involving three dimensions. In addition to adapting the model and introducing innovative methodological procedures and instruments for collecting and analyzing data, our main results showed that the ACE model is superior than written tests for discriminating students on the top and bottom of the scale of scientific reasoning abilities, while both instruments are equivalent for evaluating students in the middle of the scale.

Instructional Scaffolding is a learning process designed to promote a deeper level of learning and asynchronous learning involves learning via e-mails, blogs etc. The study was conducted in the urban area of Assam, India aimed to find out the effects of Online Technology Based Scaffolding (OTBS)on the asynchronous learning performance in relations to self-efficacy and persistence of school students. Non Equivalent Quasi Experimental Design used to conduct the experiment where pre test and post test used to assess the learning performance over traditional approach. Class VIII students (N = 40) of 'School 1' was counted as the non scaffolding group and 35 students of class VIII of 'School 2' was assigned for the Scaffolding group. It was resulted that 1) the effects of online technology based scaffolding was statistically significant over traditional approach, 2) misinformed factor of self-efficacy was comparatively lower in scaffolding group than the traditional group, and 3) the mean score of current purpose perusing factor of persistence found higher among the participants of the scaffolding group than the participants of the traditional group because of the influence of scaffolding learning.

The adoption and impact of curriculum innovation are most commonly researched in terms of roles that teachers and school administrators are expected to play. A study in New Zealand focused on the responses of students. This integrated studies project developed alternative approaches to the traditional junior secondary school curriculum. It included extensive use of computers and a enquiry-based approach that explored Maori and European cultures while integrating mathematics, history, and language arts skills. Data were collected from 3 cohorts of students each over 3 years to provide complete data on 415 students. Results show that students who elected to be involved in the integrated studies program had more positive attitudes toward computer use than those in the traditional school program, and they also reported more enjoyment of out-of-class activities. Students in the integrated studies program were significantly less alienated than their counterparts in the regular grade-9 programs, and academic achievement was also higher for English, mathematics, and science. Results demonstrate a lag of over 3 years between the implementation of the curriculum elements and their complete use, and that student attitudes toward the program developed favorably over time. This suggests the importance of support and encouragement for teachers as they work to implement the innovations. (Contains 1 table and 11 references.) (SLD)

COVID-19 has impacted education worldwide. In this troublesome situation, it is hard enough for an institution to predict a student’s performance. Students’ performance prediction has always been a complex task and this pandemic situation has led this task to be more complex. The main focus of this work is to come up with a machine learning model based on a classical machine learning technique to predict the change in students’ performance due to COVID-19. Initially, some relevant features are selected, based on which the data are collected from students of some private universities in Bangladesh. After the entire data set is formed, we preprocessed the dataset to remove redundancy and noise. These preprocessed data are used for testing and training using the proposed model. The model is extensively evaluated in this way using three separate classical machine learning techniques, namely linear regression, k-nearest neighbors (k-NN), and decision tree. Finally, the results of the entire experiment follow, demonstrating the power of the machine learning model in such an application. It is observed that the proposed model with linear regression exhibits the best performance with an R2 error of 0.07% and an accuracy of 99.84%.

COVID-19 has impacted education worldwide. In this troublesome situation, it is hard enough for an institution to predict a student’s performance. Students’ performance prediction has always been a complex task and this pandemic situation has led this task to be more complex. The main focus of this work is to come up with a machine learning model based on a classical machine learning technique to predict the change in students’ performance due to COVID-19. Initially, some relevant features are selected, based on which the data are collected from students of some private universities in Bangladesh. After the entire data set is formed, we preprocessed the dataset to remove redundancy and noise. These preprocessed data are used for testing and training using the proposed model. The model is extensively evaluated in this way using three separate classical machine learning techniques, namely linear regression, k-nearest neighbors (k-NN), and decision tree. Finally, the results of the entire experiment follow, demonstrating the power of the machine learning model in such an application. It is observed that the proposed model with linear regression exhibits the best performance with an R2 error of 0.07% and an accuracy of 99.84%.