Search

Laboratory work is essential in most applied science and engineering specific coursers. The continuous advances in Information and Communication Technologies be able the use of online laboratories (OLs). Their use is increased due to the advantages they offer compared to conventional laboratories, as unfortunately verified during Covid-19 pandemic. OLs and Learning Management Systems (LMSs) are key technologies in the learning process field. Normally, the integration of OLs with LMSs is carried out using proprietary or ad-hoc solutions. Furthermore, there are standards that can be used for this integration, such as: LTI, IEEE P92741.1 (xAPI), SCORM and IEEE1876. This work elaborates a survey which has been provided to the experts of online laboratories for them to complete. Gathered information allowed authors to diagnose the level of knowledge and use of the main standards for creating and integrating OLs. In fact, the analysis of this information confirms that there is a lack of criteria for selecting one standard against the others and how use it in the online laboratory development. This work gives guidelines for selecting the most appropriate standard according to desired characteristics of the online laboratory under development.

Increasing the effectiveness of programming education has emerged as an important goal in teaching programming languages in the last decade. Automatic evaluation of the correctness of the student’s source code saves teachers time and effort and allows a more comprehensive focus on the preparation of assignments with integrated feedback. The study aims to present an approach that will enable effective testing of students’ source codes within object-oriented programming courses while minimising the demands on teachers when preparing the assignment. This approach also supports variability in testing and preventing student cheating. Based on the principles of different types of testing (black-box, white-box, grey-box), an integrated solution for source code verification was designed and verified. The basic idea is to use a reference class, which is assumed to be part of every assignment, as the correct solution. This reference class is compared to the student solution using the grey-box method. Due to their identical interface (defined by assignment), comparing instance states and method outputs is a matter of basic programming language mechanisms. A significant advantage is that a random generation of test cases can be used in such a case, while the rules for their generation can be determined using simple formulas. The proposed procedure was implemented and gradually improved over 4 years on groups of bachelor students of applied informatics with a high level of acceptance.

Project moveSMART uses a web-based platform to integrate physical education with computer science (CS) and computational thinking (CT). This article describes a series of tutorials to introduce elementary students to CS/CT by making connections to physical activity and grade-level curricula in other subjects. [ABSTRACT FROM AUTHOR]

With the COVID-19 pandemic, online university education assumed greater importance. We propose a practical e-learning model with suggestions to augment online education of all forms to support effective technical education. We also share our experiences of implementations of some of these suggestions in online teaching. [ABSTRACT FROM AUTHOR]

In this article, behaviour of students in the e-learning environment is analyzed. The novel pipeline is proposed based on video facial processing. At first, face detection, tracking and clustering techniques are applied to extract the sequences of faces of each student. Next, a single efficient neural network is used to extract emotional features in each frame. This network is pre-trained on face identification and fine-tuned for facial expression recognition on static images from AffectNet using a specially developed robust optimization technique. It is shown that the resulting facial features can be used for fast simultaneous prediction of students’ engagement levels (from disengaged to highly engaged), individual emotions (happy, sad, etc.,) and group-level affect (positive, neutral or negative). This model can be used for real-time video processing even on a mobile device of each student without the need for sending their facial video to the remote server or teacher's PC. In addition, the possibility to prepare a summary of a lesson is demonstrated by saving short clips of different emotions and engagement of all students. The experimental study on the datasets from EmotiW (Emotion Recognition in the Wild) challenges showed that the proposed network significantly outperforms existing single models. [ABSTRACT FROM AUTHOR]

The widespread adoption of digital tools for learning enables extensive data collection about employee learning activities. This article analyzes how corporations use data to inform and support their learning needs. This article presents a qualitative analysis of what managers and executives overseeing learning and development initiatives in their respective organization's value in learning analytics (LA) at the workplace. The interviews were analyzed through a framework that combined lenses of value-sensitive design and value creation to uncover the values underpinning current applications of LA in the corporate sector. We found that corporations vary in their strategic LA orientations when it comes to organizational learning and development. However, although aspirational narratives of executives are diverse, their activities are not. The implications of the rich qualitative findings suggest that learning sciences can support organizational learning by creating diagnostic tools to identify organizational orientations for LA. Computer scientists can also support decision-making and research on corporate LA by creating tools that can help map organizational aspirations and activities in relation to the various data sources, stemming from different organizational units. We conclude by outlining these recommendations. [ABSTRACT FROM AUTHOR]

Educational Data Virtual Lab (EDVL) is an open-source platform for data exploration and analysis that combines the power of a coding environment, the convenience of an interactive visualization engine, and the infrastructure needed to handle the complete data lifecycle. Based on the building blocks of the FIWARE European platform and Apache Zeppelin, this tool allows domain experts to become acquainted with data science methods using the data available within their own organization, ensuring that the skills they acquire are relevant to their field and driven by their own professional goals. We used EDVL in a pilot study in which we carried out a focus group within a multinational company to gain insight into potential users' perceptions of EDVL, both from the educational and operational points of view. The results of our evaluation suggest that EDVL holds a great potential to train the workforce in data science skills and to enable collaboration among professionals with different levels of expertise. [ABSTRACT FROM AUTHOR]

One of the main challenges of massive open online courses (MOOCs) is the effective facilitation of learners in the course forum. The more learners participating in the forum, the more difficult it is for instructors to provide timely support. The effective intervention of teaching assistants (TAs) can play a crucial role in mitigating this issue; however, questions arise regarding the instructional approaches that TAs follow and whether they can effectively promote learning in a MOOC environment. In this article, we study the instructional approaches that TAs followed in two MOOCs of different subject matters, using mixed-methods. The goal was to evaluate the pedagogies that TAs adopted in facilitating learners, following a broadly accepted framework on MOOC instructional quality. Content analysis and interviews with the TAs provided insights into their intervention strategies, while linguistic and social network analysis enhanced the findings. The study revealed that the TAs most often were rushing to provide direct answers rather than guiding learners in finding themselves an answer. Problem-solving approaches and collaboration were not adequately promoted. The findings of this article provide design implications for the development of supportive tools that could automatically assess facilitators’ instructional approaches in the forum. Thus, timely feedback could be provided to instructors so as to refine the instructional design of their online courses and to provide guided support to their learner facilitators while courses are running. [ABSTRACT FROM AUTHOR]

This article introduces Xiao-Shih, the first intelligent question answering bot on Chinese-based massive open online courses (MOOCs). Question answering is critical for solving individual problems. However, instructors on MOOCs must respond to many questions, and learners must wait a long time for answers. To address this issue, Xiao-Shih integrates many novel natural language processing and machine learning approaches to achieve state-of-the-art performance. Furthermore, Xiao-Shih has a built-in self-enriched mechanism for expanding the knowledge base through open community-based question answering. This article proposes a novel approach, known as spreading question similarity (SQS), which iterates similar keywords on our keyword networks to find duplicate questions. Compared with BERT, an advanced neural language model, the results showed that SQS outperforms BERT on recall and accuracy above a prediction probability threshold of 0.8. After training, Xiao-Shih achieved a perfect correct rate. Furthermore, Xiao-Shih outperforms Jill Watson 1.0, which is a noted question answering bot, on answer rate with the self-enriched mechanism. [ABSTRACT FROM AUTHOR]

In the last years, educational community is asking for a change by introducing social tools in formal learning contexts. Besides, although meaningful research on the effects of social learning can be found in the literature, all authors agree that further research is needed. Social Network Analysis (SNA) has been proven to be a useful tool for both researchers and teachers. However, both standard SNA tools and the most popular social network sites are external to the Learning Management Systems (LMS) anda, therefore, they are hard to integrate by teachers into their educational designs. This study shows MSocial, a novel tool integrated into the Moodle LMS, that allows students interact in their social networks without losing their presence in the LMS. MSocial monitors student activity in social networks (both inside and outside of Moodle), calculates SNA metrics and shows the results promptly in the Moodle course site. Therefore, teachers can easily understand, visualize, and analyze the social participation and interaction of students and incorporate these results to improve the learning process.

Existing literature has provided strong evidence that educational escape rooms are engaging and effective learning activities when they are properly conducted in the classroom. However, no prior research has determined whether the positive effects of these novel educational activities on students’ perceptions and learning persist when conducted remotely. This article performs, for the first time, a comparative study of the effectiveness of face-to-face and remote educational escape rooms. For this purpose, two versions of the same educational escape room were conducted: one in-class and one remotely. Both experiences were evaluated by means of three different instruments: (1) a pre-test and a post-test for measuring learning gains, (2) a questionnaire for assessing students’ perceptions, and (3) a web platform for recording student interaction data during the activities. The results obtained suggest that, although remote educational escape rooms for learning programming can be as engaging as their face-to-face counterparts, their learning effectiveness is somewhat lower.

In this article, we investigate a novel reinforcement-learning (RL)-based scheme to address the optimal leader–follower consensus issue for constrained-input continuous-time multiagent systems. First, as for input-constrained problems, the fundamental smooth assumption is not satisfied for the value function. To deal with this problem, we employ the method of vanishing viscosity solutions to relax this smooth assumption to a continuity assumption for the value function, which broadens the scope of RL applications. Second, the control cost functions take a more general form which guarantees the continuity of the optimal control policy instead of the specific integrand form used in previous input-constrained RL-based schemes. Based on these results, we introduce a novel identifier–critic–actor structure to extend the conventional critic–actor RL framework into a distributed model-free one, where the learning of the identifier, critic, and actor is online and simultaneous. We provide the simulation examples to validate the effectiveness of the proposed scheme. [ABSTRACT FROM AUTHOR]

Traditionally, engineering degree programs are reasonably good at transmitting technical knowledge to their students; regardless of the particular conditions of each university, lecturers usually have extensive experience in doing so. However, employers are demanding soft skills in addition to the abilities our engineers master, and strategies used to incorporate them within our compact and information-packed programs are required. This article presents an experience in promoting skills for communicating and interacting with ordinary people in technical issues, performed among electrical engineering students from the radio-frequency (RF) branch. Our proposal is to use role-playing techniques to train students to communicate with people who are afraid of electromagnetic (EM) exposure effects, which has immediate implications during 5G network deployment. Moreover, the use of forum facilities in e-learning platforms allows such techniques to be integrated within an online course. [ABSTRACT FROM AUTHOR]

This article investigated the impact of immersive virtual reality (VR) content, using 360-degree videos, in undergraduate education. To improve the delivery and reality of 360-degree VR content, we filmed the video in the third person so that the viewers could feel like they were in the environment where the lecture was conducted. To verify the educational effects, 33 university students participated in our experiment. We conducted pretest learning, using 360-degree videos, and posttest learning via conventional 2-D videos for statistical analysis. A paired t-test was used to compare the means of the pretest and the posttest. In addition, learning via 360-degree videos was assessed for its effectiveness through questionnaires consisting of five measurement elements—engagement, immersion, motivation, cognitive benefits, and perceived learning effectiveness—and comparing them to the existing 2-D video method, based on e-learning. From the results, we confirmed that the teaching material delivered through 360-degree VR content allows students to be more focused, immersed, and interested than 2-D learning modes. Furthermore, the high scores of cognitive and perceived learning elements imply that VR-based 360-degree educational content can encourage more active participation than traditional lectures and can improve the ability to analyze and organize study lessons. [ABSTRACT FROM AUTHOR]

Making students become intrinsically motivated to participate in daily class activities is an open challenge that has been addressed in different ways. In this article, we evaluate the impact of an educational innovation project, named TrivialCV, in terms of student motivation, engagement, and learning outcomes. We analyze the impact of two types of activities: multiplayer team-based competitions created for delivery during a live class session and single-player competitions created to be played asynchronously from home. We deployed these activities in two different computer engineering courses (programming fundamentals and operating systems) and used questionnaires, grades, and activity tracking in the virtual campus to evaluate their impact. After the analysis, we can assert that the use of TrivialCV activities was useful for providing additional motivation to the students and improving their engagement with the courses in which they were deployed. TrivialCV activities were very well received by the students, who considered them fun, engaging, and useful. [ABSTRACT FROM AUTHOR]

The pandemic is forcing educational institutions around the world to rapidly transition from face-to-face instruction to e-learning. However, e-learning has some multifaceted problems that are waiting to be solved and optimized. Therefore, this article aims to discuss how we can optimize online learning systems to benefit traditional education systems in the long run. Based on the findings, this article suggests policy makers, educators, and system developers not only develop continuous training for teachers and learners on e-learning to improve their skills and acceptance of online education systems but also create a more comprehensive technical environment to accelerate adoption and increase the intention to continue using e-learning systems. In addition, high-quality online education designs that include interaction mechanisms and adaptations to students’ perceptions and attitudes toward online education are needed, and the design of online education systems should be varied according to the different characteristics of educational subjects. The proposed insights in this article are applicable to a wide set of projects for online education. Moreover, this article calls on scholars from around the world to further explore the critical factors for adapting online education and optimizing online education environments to propose better solutions for future crises similar to the COVID-19 pandemic. [ABSTRACT FROM AUTHOR]

Sustainability science has been gaining traction within academic research and teaching over the past few decades, and it is now entering a phase of greater maturity as a discipline, with links to sectors beyond the academy including energy, finance, advocacy, development, design, construction, agriculture, and healthcare. Concurrently, the growth of the internet has spurred innovation in platform development. This convergence has generated technology-enabled opportunities for training sustainability scholars and practitioners grounded in active learning pedagogies. Case studies are one active learning strategy becoming established within the fields of environment and sustainability. However, the traditional case study form and mode of delivery, which excel at a retroactive analysis of strategic decision-making and outcomes, can create challenges for teaching about issues involving emergent technologies and rapidly changing scientific and social dimensions. This article describes one novel platform, called Gala, for teaching and learning about sustainability science through case studies that address these challenges. Gala closely integrates text and multimedia components, is easily updated, and embraces content consumers as content creators and vice versa. We also report here the results of one pilot study that investigated differences between graduate and undergraduate student interactions with different case components. Results indicate that both groups heavily used the case narrative, but varied in their use of multimedia components. Such insights into patterns of case use are informing decisions about improvements to Gala and directions for future case production, teaching, and assessment, in order to create better tools for teaching and learning about sustainability. [ABSTRACT FROM AUTHOR]

Education and training are among the fields taking advantage of serious games (SGs). In this article, we present Gea 2: A New Earth, a digital SG developed as an immersive 3-D virtual learning environment, integrating several educational resources, including multimedia learning material, communication tools, and intelligent tutoring support. The game aims to complement traditional classroom activities in science, technology, engineering, and mathematics (STEM), for high school students and teachers. It incorporates an intelligent pedagogical agent that can converse in natural language with the student and provides unsolicited hints during gameplay. This article presents the game and its evaluation based on experiments involving about 100 participants. We think that the results presented here add to the research on game-based learning for STEM, by proposing a complex game system, where artificial intelligence techniques are integrated to support students’ learning, and by confirming that the game experience can be attractive for the learners, also in very constrained classroom environments, such as those we operated in. [ABSTRACT FROM AUTHOR]

The papers in this special section focus on designing technologies to support professional and workplace learning in situated practices. In an era of global, organizational, and technological change, all of which are transforming the world of work, professional and workplace learning are critical for employability and organizational competitiveness. A range of fundamental transformations is changing how people work. Digital technologies are replacing human labor and, at the same time, are accelerating the expansion of job roles and work practices. Work is becoming increasingly specialized, which means that professionals in collaborative and networked ways across discipline and organization boundaries. In parallel, labor is increasingly decentralized, making decisionmaking more distributed and raising the need for remote communication and collaboration. Subsequently, work is becoming more independent from time and place, as people connect, collaborate, and work via digital technologies. These changes come with a need for substantial and continuous workplace learning, and with the need for changes in how workplace learning happens. Of course, digital technologies are already used to provide learning and training in workplaces. However, most of these learning technologies have been developed for formal education (e.g., K- 12 and higher education) rather than in workplace contexts. There is a need to understand and evidence workplace learning needs and to further develop technologies that can support and scale workplace learning. [ABSTRACT FROM AUTHOR]

Due to the sudden transition towards online learning caused by COVID-19, we realized the urgency of deepening our understanding of the distributed collaborative design process. We conducted critical incident interviews to identify the challenges that design students and professional designers encountered and how they overcame them in their work. Based on the insights from this first study, we carried out an online survey to corroborate our insights. We found that the shift towards online meetings impacts the collaborative design process. Most notably it causes reduced co-exploration in the design process, especially on and with physical objects. This leads us to the conclusion that the current strategies are not adequate for facilitating designers who collaborate remotely. Following our insights, we provide suggestions for enhancing the supporting tools for the distributed collaborative design process. [ABSTRACT FROM AUTHOR]

The aim of this research is to develop a prototype of a multi-agent systems (MAS), which manages the resources and activities in a smart classroom (SaCI). We developed three specific objectives. First, we designed the architecture of the MAS that manages the resources and activities in a virtual learning environment (VLE), for which we used the related works on MAS in a VLE. Second, we implement the architecture of the MAS, the prototype proposed has been called GREAC (Gestion de Recursos y Actividades), which considers the learning style (LS) of the student to manage the resources and activities in an adaptive way, and uses the methodology Prometheus, Moodle 3.6 (VLE), and the framework Jade, for the development of the prototype of the MAS. Third, we evaluated the prototype GREAC in a SaCI in an academic context. In the evaluation participated 54 students from the Universidad Nacional de Loja, of the Carrera de Ingenieria en Sistemas. As a result of the evaluation by the students, we obtained a level of satisfaction 92.6 %. Finally, this work concludes that the implementation of a MAS in a VLE allows the dynamic management of resources and activities, taking into account the students LS, allowing an adaptive and personalized management. [ABSTRACT FROM AUTHOR]

COVID-19 has accelerated and scaled out remote education beyond our wildest imagination. How can we retain benefits, eliminate downsides, and build education better for the long-term future? [ABSTRACT FROM AUTHOR]

Accurately grading open-ended assignmentsin large or massive open online courses is nontrivial. Peer review is a promising solution but can be unreliable due to few reviewers and an unevaluated review form. To date, no work has leveraged sentiment analysis in the peer-review process to inform or validate grades or utilized aspect extraction to craft a review form from what students actually communicated. This article utilizes, rather than discards, student data from review form comments to deliver better information to the instructor. In this article, we detail the process by which we create our domain-dependent lexicon and aspect-informed review form as well as our entire sentiment analysis algorithm, which provides a fine-grained sentiment score from text alone. We end by analyzing validity and discussing conclusions from our corpus of over 6800 peer reviews from nine courses to understand the viability of sentiment in the classroom for increasing the information from and reliability of grading open-ended assignments in large courses. [ABSTRACT FROM AUTHOR]

With the proliferation of mobile devices and online video services, mobile learning has grown both in popularity and complexity. New challenges include the multitude of mobile devices with different characteristics and limitations, as well as the exponential growth in educational multimedia content. Streaming multimedia content to mobile devices is a resource intensive task that requires significant resources such as network bandwidth, with demand expected to increase beyond future networks capacity as users adopt new technologies such as UHD, AR, VR, 3D and 360-degree video. While a number of adaptive m-learning systems have been previously proposed, none of these have thoroughly addressed the adaptation of educational multimedia content. This article proposes the novel DQAMLearn framework that aims to support mobile learner’s seamless access to educational multimedia content from a variety of mobile devices with different characteristics. Moreover, as mobile users are increasingly becoming quality-aware, the framework integrates novel mechanisms for decreasing the video quality in a controlled way, with the aim to support a good learner quality of experience (QoE) even in resource constrained situations. A comprehensive subjective study was conducted to evaluate the proposed framework. The results showed that the framework enables both high learning achievement from educational multimedia clips, with 12% and 83% correct response rates for pre and post-test questionnaires, respectively, and high learner QoE with a mean video quality rating of 79.19 on a 0–100 acceptability scale. [ABSTRACT FROM AUTHOR]

This article describes the implementation of a remote access experiment, called Web-lab 1553B (Digital Data Bus Lab), for courses in the Avionics Systems Program. Its main objective is to present a study about the use of two rotation model subgroups: 1) rotation laboratory and 2) station rotation. In the “rotation laboratory” subgroup, the teacher does not need to leave the classroom, once the access (rotation) to the lab is done remotely and not physically. On the other hand, in the “station rotation” subgroup, rotation is performed by changing the experiment functions, and thus students not having to physically change stations. This article shows some preliminary results, which demonstrate that the application of the aforementioned methods not only increases the teaching–learning process effectiveness (quantitative results) but also improves student satisfaction (qualitative results) and motivation. Two comparative tests were carried out to quantitatively verify the impact on the student’s accomplishment: Test 1, following the theory class, and Test 2, following the demo class and lab practice, using the Web-lab 1553B. Results show an average increase of 27.69% in the student grades. Considering qualitative assessments, students’ rates for the Web-lab 1553B method showed that 81.25% of the questions achieved rate “5” (top level), demonstrating that Web-lab meets the desired needs. These assessments serve as data sources for the validation of the hybrid teaching models, concerning classes and laboratory practices, as proposed, showing a high potential for practical implementation in different levels of education institutions. [ABSTRACT FROM AUTHOR]

The rapid, global spread of COVID-19 has led to an unprecedented rise in enrollments in online learning experiences among learners of all ages. In this article, we explore the impact of the global pandemic on a massive open online course, Problem Solving Using Computational Thinking, with a particular focus on the topics learners chose for their final projects. The Computational Thinking MOOC was designed using a project-based learning approach and aims to provide learners with an introduction to the "big ideas" of computational thinking using a range of case studies that encompass topics such as airport surveillance, epidemiology, and human trafficking. Beyond observing a sharp increase in enrollment and engagement at the time the pandemic began, we discuss ways in which the course's project-based pedagogy allowed learners to bring their present experiences and concerns together with the course's subject matter in order to meet the learning objectives for the course. Many learners chose to address aspects of the pandemic in the course's final project and applied ideas about computational thinking to peer-graded assignments that conveyed an individualized sense of importance and urgency. We assert that this approach, along with the inclusion of a timely epidemiology case study, enabled learners to more deeply internalize the role that computational thinking can play in their own lives and in society as a whole. [ABSTRACT FROM AUTHOR]

In e-learning recommender systems, interpersonal information between learners is very scarce, which makes it difficult to apply collaborative filtering (CF) techniques to achieve recommendations. In this study, we propose a hybrid filtering recommendation approach ($SI-IFL$ S I - I F L ) combining learner influence model (LIM), self-organization based (SOB) recommendation strategy, and sequential pattern mining (SPM) together for recommending learning objects (LOs) to learners. The method works as follows: (i) LIM is applied to acquire the interpersonal information by computing the influence that a learner exerts on others. LIM consists of learner similarity, knowledge credibility, and learner aggregation, meanwhile, LIM is independent of ratings. Furthermore, to address the uncertainty and fuzzy natures of learners, intuitionistic fuzzy logic (IFL) is applied to optimize the LIM. (ii) A SOB recommendation strategy is applied to recommend the optimal learner cliques for active learners by simulating the influence propagation among learners. Influence propagation means that a learner can move towards active learners, and such behaviors can stimulate the moving behaviors of his/her neighbors. This SOB recommendation approach achieves a stable structure based on distributed and bottom-up behaviors of individuals. (iii) SPM is applied to decide the final learning objects (LOs) and navigational paths based on the recommended learner cliques. The experimental results demonstrate that $SI-IFL$ S I - I F L can provide personalized and diversified recommendations, and it shows promising efficiency and adaptability in e-learning scenarios. [ABSTRACT FROM AUTHOR]

Recently, ultrasound has been increasingly used in emergency departments (EDs) due to its promising noninvasive and portable characteristics. Traditional ultrasound training is a complex process that requires knowledge gain, development of psychomotor skills, and visual perception. However, it usually takes a long time for a novice sonographer to finish an ultrasound training program, and the training may target on limited applications. The e-learning, information, and communication technology on digital devices to support learning may overcome the limitations of the traditional ultrasound training. In this article, we design and implement a mobile e-learning platform for ultrasound training, namely MEUS, to address the abovementioned issues. The MEUS provides an interactive learning environment for teachers and learners. Teachers and learners can access MEUS and receive real-time feedback anywhere and anytime through a mobile device or desktop. We implement a multistage learning process. To evaluate the learning performance of using the platform, we add the MEUS platform as a part of the emergency ultrasound training course with the Post Graduate Year (PGY) program in the ED at the National Taiwan University Hospital. Our study shows that the MEUS platform can significantly increase the learning efficacy of the PGY students. [ABSTRACT FROM AUTHOR]

A smart campus provides students who are geographically scattered with online tools to get access to learning resources and laboratories. Although these remote laboratories have the potential and capabilities to implement different learning experiments, most of them are configured in a static fashion, being able to serve only one experiment for a given period of time. This lack of adaptability and flexibility causes long waiting queues of students in certain overloaded remote laboratories, while others are underused. To overcome this limitation, a smart campus can incorporate new techniques and paradigms such as Network function virtualization (NFV) and Software-defined networking (SDN), which are rapidly modifying current cloud services and applications to provide autonomous and adaptive solutions. In this context, the main contribution of this paper is an SDN/NFV-based architecture with autonomic capabilities to adapt to the remote laboratories configuration according to the end-user demand. The proposed architecture will be able to optimize computing resources to ensure the users’ quality of service in a smart campus with remote laboratories. A use case with a remote laboratory, based on the control of a servo motor, shows how our solution can dynamically change the lab service under specific circumstances. Finally, experiments using diverse configurations and service descriptions show the performance and suitability of the proposal. [ABSTRACT FROM AUTHOR]

Rapid development of open educational resources (OER) is often motivated by a new educational paradigm. This paradigm tends to address current challenges such as reusing, goal-oriented remixing, revising, and redistributing OER. This paper proposes a system to automatically remix, which is a step towards automatic course generation. This system uses an agent-based modeling (ABM) approach to profile OER and simulate the process of selecting and linking the appropriate ones. Each resource profile provides us with metadata that shows its domain and is completed using an appropriate ontology before agents’ interactions. ABM enables cooperation between OER to satisfy requirements of the input syllabus. Using the interaction ability, the most compatible agents, which cover the concepts of the input syllabus, can be linked. The efficiency of the proposed method is evaluated through applying the implemented system on two datasets: ARIADNE and SlideShare. In our experimental results on OER related to the domain of e-learning, we report on the system precision and user satisfaction. Our approach is generic; it can be used with any OER repository and any content management system with a free reuse and remix license. [ABSTRACT FROM AUTHOR]

This paper studies students engagement in e-learning environments in which students work independently and solve problems without external supervision. We propose a new method to infer engagement patterns of users in such self-directed environments. We view engagement as a continuous process in time, measured along chosen axes that are derived from student data in the system using unsupervised learning (Principal Component Analysis). We construct a trajectory of user activity by projecting the user's scores along the selected PCs at regular time intervals. This approach is applied to a popular e-learning software for K12 math education that is used by thousands of students worldwide. We identify cohorts of users according to the way their trajectory changes over time (e.g., monotone up, monotone down, and constant). Each of the cohorts exhibits distinct behavioral dynamics and differed substantially in the amount of time users spent in the e-learning system. Specifically, one cohort included students that dropped out of the system after choosing very difficult problems that they were not able to complete, while another cohort included students users that chose more diverse problems and stayed longer in the system. In future work, these results can be used by teachers or intelligent tutors to track students’ engagement in the system and decide whether and how to intervene. [ABSTRACT FROM AUTHOR]

We present a system that automatically generates deictic gestures for animated pedagogical agents (APAs). The system takes audio and text as input, which define what the APA has to say, and generates animated gestures based on a set of rules. The automatically generated gestures point to the exact locations of elements on a whiteboard nearby the APA, which are calculated by searching for keywords mentioned in the speech. We conducted a study with 100 subjects, in which we compared lecture videos containing gestures automatically-scripted by the system to videos of the same lecture containing manually-scripted gestures. The study results show that the manually-scripted and automatically-scripted lectures had comparable number of gestures, and that the gestures were timed equally well. [ABSTRACT FROM AUTHOR]

With the wide spread large-scale e-learning environments such as MOOCs, peer assessment has been popularly used to measure the learner ability. When the number of learners increases, peer assessment is often conducted by dividing learners into multiple groups to reduce the learner's assessment workload. However, in such cases, the peer assessment accuracy depends on the method of forming groups. To resolve that difficulty, this study proposes a group formation method to maximize peer assessment accuracy using item response theory and integer programming. Experimental results, however, have demonstrated that the method does not present sufficiently higher accuracy than a random group formation method does. Therefore, this study further proposes an external rater assignment method that assigns a few outside-group raters to each learner after groups are formed using the proposed group formation method. Through results of simulation and actual data experiments, this study demonstrates that the proposed external rater assignment can substantially improve peer assessment accuracy. [ABSTRACT FROM AUTHOR]

The use of recommendation systems has been extensively applied to different fields from the suggestion of multimedia content to other areas such as education. Thereof, the integration of affect-related information becomes a key factor to enhance the user experience and, when it comes to learning, it can be translated into the maximization of knowledge acquisition. The existence of affect-enhanced datasets is imperative when training a recommender learning system. This article introduces a new dataset gathered under realistic conditions, which includes affect-related information of 33 learners from different knowledge background interacting with an e-learning platform. Matrix factorization techniques were evaluated in order to prove the wealthiness of the dataset and to demonstrate that it could potentially replace the adaptation mechanism of the introduced platform and maximizing the learner's knowledge acquisition. [ABSTRACT FROM AUTHOR]

Motivated by the success of student-centered language-learning tools and data-driven platforms commonly found in popular commercial and entertainment platforms, we present a new adaptive educational web platform for engineering students. The presented adaptive education platform (ADEPT) focuses on personalizing learning in large college classes by enabling proactive and continuous student engagement. In this paper, we present the principles, implementation strategies, and initial results obtained by working with an early version of ADEPT in a required sophomore-level circuits course at Purdue University. Initial results underline the potential of this web tool to identify the challenging concepts for students as well as prepare instructors to modify the way concepts are presented to the students. In addition, ADEPT helps to reveal the student engagement habits and studying patterns that may not be easily identifiable through other means, such as self-reporting.

Besides technical skills, industry increasingly demands professional skills. Teaching these at university is challenging and still rare. In computer science, new technologies and relevant tools emerge quickly. Keeping curricula up to date is challenging. We present a teaching methodology that teaches students the art of teaching. Students learn how to communicate technical content efficiently in lecture, self-learning, and teamwork settings. Students work with constructive alignment and active listening. In addition, they learn how to assess self-learning training materials systematically, and how to give constructive feedback. Finally, they access a technical topic from a new dimension by practically teaching it. During the course, the students train the described skills under guidance by developing new exercises. The article presents our course unit that teaches important professional skills and produces a continuous stream of new teaching material, which is integrated into class. The unit enables teaching relevant and new technologies early. The presented methodology has been continuously applied at the Technical University of Munich since 2013. Through 2019, more than 150 students have participated. More than 60 new exercises were created. A structured evaluation shows the power of the approach. [ABSTRACT FROM AUTHOR]

The advent and increased use of new technologies, such as innovative mulsemedia and multi-modal content distribution mechanisms, have brought new challenges and diverse opportunities for technology enhanced learning (TEL). NEWTON is a Horizon 2020 European project that revolutionizes the educational process through innovative TEL methodologies and tools, integrated in a pan-European STEM-related learning network platform. This article focuses on one of these novel TEL methodologies (i.e., mulsemedia) and presents how NEWTON enables mulsemedia- enhanced teaching and learning of STEM subjects, with a particular focus on telecommunication and networking related modules. The article also discusses the very promising results of NEWTON case studies carried out with engineering students across two different universities in Spain and Ireland, respectively. The case studies focused on analyzing the impact on the learning process of the mulsemedia-enhanced teaching in the context of telecommunication and networking modules. The main conclusion of the article is that mulsemedia-enhanced education significantly increases students' learning experience and improves their knowledge gain. [ABSTRACT FROM AUTHOR]

Contribution: A redesign of the Moodle platform to adapt digital educational content [learning objects (LOs)] to the specific needs of students with disabilities. The approach, extendable to a range of disciplines, was empirically tested with blind and deaf engineering students. Background: Previous studies identified difficulties that blind and deaf students face in accessing digital content for learning. General guidelines and specific tools are available to help educators adapt digital content and existing learning platforms for access by students with varying abilities/capacities. Such tools are usually for a specific disability rather than a range of capacities, and few provide empirical evidence of effectiveness. Intended Outcomes: The engineering-related digital content adapted using the techniques described in this paper should enable blind and/or deaf students to use an oscilloscope, understand communication channels, and distinguish the different types of telecommunication networks. Application Design: The Moodle learning platform was adapted using existing e-learning accessibility standards so that digital LOs could be automatically presented in formats accessible to blind and/or deaf students. This model is extendable for other types of disabilities, helping educators adapt existing content for access by students with differing capacities. The teacher adds content (in non-auditory and non-visual formats to describe content otherwise inaccessible to deaf or blind students) and students upload reusable profiles/metadata describing their specific accessibility needs to connect to suitably adjusted content. Findings: Learning improvement with the adjusted platform was evaluated via multiple choice pre- and post-tests. Students’ learning performance improved significantly across all groups: blind (45%), deaf (46.25%) and deaf-blind (87.5%). [ABSTRACT FROM AUTHOR]

Contribution: An original application of individual rotation to blended learning, which mixes e-learning, discussion groups (seminars), practical laboratory work, and self-motivated tasks called “mini-projects.” Background: In examining the changing practices toward students’ transferable skills in higher education, current teaching needs to devote much more attention to using multi-modular teaching methods to foster students’ key transferable skills, such as logical, analytical, and creative thinking. Intended Outcomes: Rotational blended learning is intended to maximize students’ engagement and improve educational outcomes during the learning process. Application Design: A rotated form of existing teaching methods—e-learning, seminars, and group projects—was proposed. A quasi-experimental design, involving classroom observation, student surveys, and overall results, was used with two cohorts of computer system engineering students, one a controlled cohort taught using traditional techniques and the other an experimental cohort taught using a novel rotational blended system. Findings: The influence of blended learning on the subsequent development of critical transferable skills was demonstrated. Results suggest that rotational blended learning is an ideal way to address these challenges, since it allows computer engineering students to reassess and enhance the core skills and competencies they need to acquire in their learning experiences. [ABSTRACT FROM AUTHOR]

Question classification is a key point in many applications, such as Question Answering (QA, e.g., Yahoo! Answers), Information Retrieval (IR, e.g., Google search engine), and E-learning systems (e.g., Bloom's tax. classifiers). This paper aims to carry out a systematic review of the literature on automatic question classifiers and the technology directly involved. Automatic classifiers are responsible for labeling a certain evaluation item using a type of categorization as a selection criterion. The analysis of 80 primary studies previously selected revealed that SVM is the main algorithm of the Machine Learning used, while BOW and TF-IDF are the main techniques for feature extraction and selection, respectively. According to the analysis, the taxonomies proposed by Li and Roth and Bloom were the most used ones for the classification criteria, and Accuracy/Precision/Recall/F1-score were proven to be the most used metrics. In the future, the objective is to perform a meta-analysis with the studies that authorize the availability of their data. [ABSTRACT FROM AUTHOR]

Automatic recommendation has become an increasingly relevant problem to industries, which allows users to discover new items that match their tastes and enables the system to target items to the right users. In this paper, we propose a deep learning (DL) based collaborative filtering framework, namely, deep matrix factorization (DMF), which can integrate any kind of side information effectively and handily. In DMF, two feature transforming functions are built to directly generate latent factors of users and items from various input information. As for the implicit feedback that is commonly used as input of recommendation algorithms, implicit feedback embedding (IFE) is proposed. IFE converts the high-dimensional and sparse implicit feedback information into a low-dimensional real-valued vector retaining primary features. Using IFE could reduce the scale of model parameters conspicuously and increase model training efficiency. Experimental results on five public databases indicate that the proposed method performs better than the state-of-the-art DL-based recommendation algorithms on both accuracy and training efficiency in terms of quantitative assessments. [ABSTRACT FROM AUTHOR]

Global corporations are characterized by a large number of employees and geographically dispersed offices. Moreover, the competitiveness in the global market requires them to invest in their human resources to be able to remain a step ahead of competition. Implementing large scale classical education in such environments is challenging and costly. Mobile e-learning (m-learning) allows users to tailor their professional training and education to their needs and time constraints. However, in self-paced education, it is very hard to keep user retention and engagement. To achieve the latter, we have designed and developed an m-learning platform for corporate environments based on the triggering persuasive technology principle that try to incite users in regularly using the platform. We have evaluated the application in-the-wild in corporate environments of differently sized companies with 300 users. Users were subjected to three different conditions: no triggering, simple regular triggering, and adaptive triggering. The results show that the use of adaptive triggering in m-learning increases user engagement as well as course completion rates more than simple regular triggering and no triggering. [ABSTRACT FROM AUTHOR]

This paper presents the work done to support student dropout risk prevention in a real online e-learning environment: A Spanish distance university with thousands of undergraduate students. The main goal is to prevent students from abandoning the university by means of retention actions focused on the most at-risk students, trying to maximize the effectiveness of institutional efforts in this direction. With this purpose, we generated predictive models based on the C5.0 algorithm using data from more than 11,000 students collected along five years. Then, we developed SPA (Sistema de Predicciæn de Abandono, dropout prediction system in Spanish), an early warning system that uses these models to generate static early dropout-risk predictions and dynamic periodically updated ones. It also supports the recording of the resulting retention-oriented interventions for further analysis. SPA is in production since 2017 and is currently in its fourth semester of continuous use. It has calculated more than 117,000 risk scores to predict the dropout risk of more than 5700 students. About 13,000 retention actions have been recorded. The white-box predictive models used in production provided reasonably good results, very close to those obtained in the laboratory. On the way from research to production, we faced several challenges that needed to be effectively addressed in order to be successful. In this paper, we share the challenges faced and the lessons learnt during this process. We hope this helps those who wish to cross the road from predictive modeling with potential value to the exploitation of complete dropout prevention systems that provide sustained value in real production scenarios. [ABSTRACT FROM AUTHOR]

The systems used in education have evolved bringing new teaching approaches and impacting in traditional learning methods. Also, they present new scenarios to students aiming to improve activities and practices in online systems. Along with this evolution the growing availability of contents has promoting an overload of data. Nowadays, many online educational systems have too much information, content and activities available to students making choices difficult. Educational systems supported by recommender systems allow to offer contents toward the needs and profiles of the students. This work presents a student profile model based on students' preferences and their interactions in online educational systems, with the purpose of allowing the recommendation of contents. To demonstrate the feasibility and evaluation of the proposed model two study scenarios were developed with information from an educational institution. In these scenarios recommendations were carried out using collaborative filtering and content based approaches. From the evaluation of these scenarios, the results demonstrate the viability of the model in producing recommendations that aim to aid the students' learning process when interacting with online educational systems. [ABSTRACT FROM AUTHOR]

Contribution: The main contribution is to provide practitioners and researchers with an insight in efficiently and effectively employing multi-attribute decision making (MADM) methods in e-learning website selection problems. Background: Advances in information systems and the Internet have resulted in e-learning websites becoming an important interface for learners. The selection of a particular website directly affects the end users’ performance, with the objectives and expected outcomes of e-learning being largely dependent on the quality of the teaching processes and the effectiveness of online access. Evaluation and selection of these websites has thus become a decisive issue for learners and developers. Intended Outcomes: This paper is expected to provide decision support to leaders in education and educational organizations in effectively analyzing and managing their e-learning system. A knowledge-based system may be developed to develop and assess e-learning websites in a meaningful way. Application Design: A hybrid MADM method, fuzzy complex proportional assessment (COPRAS) is used to evaluate, rank, and select e-learning websites of “C” programming language. The fuzzy COPRAS method is validated by comparison with the technique for order preference by similarity to ideal solution, on an existing selection problem of 12 websites available in the open literature. Findings: Fuzzy COPRAS provides a comprehensive ranking of e-learning websites based on fitness degree values. The case study results rated tutorialspoint.com as the best of the e-learning websites considered, and howstuffworks.com was rated lowest. This paper shows that fuzzy COPRAS could be useful for e-learning practitioners, developers, and the researchers. [ABSTRACT FROM AUTHOR]

A social learning network (SLN) emerges when users exchange information on educational topics with structured interactions. The recent proliferation of massively scaled online (human) learning, such as massive open online courses (MOOCs), has presented a plethora of research challenges surrounding SLN. In this paper, we ask: how efficient are these networks? We propose a method in which the SLN efficiency is determined by comparing user benefit in the observed network to a benchmark of maximum utility achievable through optimization. Our method defines the optimal SLN through utility maximization subject to a set of constraints that can be inferred from the network, and given multiple solutions searches for the one closest to the observed network so as to require the least amount of change to user behavior in practice. Through evaluation on four MOOC discussion forum data sets and optimizing over millions of variables, we find that the SLN efficiency can be rather low (from 76% to 90% depending on the specific parameters and data set), which indicates that much can be gained through optimization. We find that the gains in global utility (i.e., average across users) can be obtained without making the distribution of local utilities (i.e., utility of individual users) less fair. We also propose an algorithm for realizing the optimal network through curated news feeds in online SLN. [ABSTRACT FROM AUTHOR]