Search results for: opposition based learning

Authors: Tina Stavredes

Abstract:

Increasingly borderless learning opportunities including online learning are expanding. Singapore University of Social Science (SUSS) conducted research in February of 2017 to determine the level of consumer interest in undertaking a completely online distance learning degree program across three countries in the Asian Pacific region. The target audience was potential bachelor degree and post-degree students from Malaysia, Indonesia, and Vietnam. The results gathered were used to assess the market size and ascertain the business potential of online degree programs in Malaysia, Indonesia and Vietnam. Secondly, the results were used to determine the most receptive markets to prioritise entry and identify the most receptive student segments. In order to achieve the key outcomes, the key points of understanding were as follows: -Motivations for higher education & factors that influence the choice of institution, -Interest in online learning, -Interest in online learning from a Singapore university relative to other foreign institutions, -Key drivers and barriers of interest in online learning. An online survey was conducted from from 7th Feb 2017 to 27th Feb 2017 amongst n=600 respondents aged 21yo-45yo, who have a basic command of English, A-level qualifications and above, and who have an intent to further their education in the next 12 months. Key findings from the study regarding enrolling in an online program include the need for a marriage between intrinsic and extrinsic motivation factors and the flexibility and support offered in an online program. Overall, there was a high interest for online learning. Survey participants stated they are intrinsically motivated to learn because of their interest in the program of study and the need for extrinsic rewards including opportunities for employment or salary increment in their current job. Seven out of ten survey participants reported they are motivated to further their education and expand their knowledge to become more employable. Eight in ten claims that the feasibility of furthering their education depends on cost and maintaining a work-life balance. The top 2 programs of interest are business and information and communication technology. They describe their choice of university as a marriage of both motivational and feasibility factors including cost, choice, quality of support facilities, and the reputation of the institution. Survey participants reported flexibility as important and stated that appropriate support assures and grows their intent to enrol in an online program. Respondents also reported the importance of being able to work while studying as the main perceived advantage of online learning. Factors related to the choice of an online university emphasized the quality of support services. Despite concerns, overall there was a high interest for online learning. One in two expressed strong intent to enrol in an online programme of study. However, unfamiliarity with online learning is a concern including the concern with the lack of face-to-face interactions. Overall, the findings demonstrated an interest in online learning. A main driver was the ability to earn a recognised degree while still being able to be with the family and the ability to achieve a ‘better’ early career growth.

Keywords: distance education, student motivations, online learning, online student needs

Procedia PDF Downloads 110

Authors: Leigh Ann Wilson, Melanie Borrego

Abstract:

The purpose of institutional learning outcomes (ILOs) is to assess what students across the university know and what they do not. The issue is gathering this information in a systematic and usable way. This presentation will explain how one institution has engineered this process for both student success and maximum faculty curriculum and course design input. At Brandman University, there are three levels of learning outcomes: course, program, and institutional. Institutional Learning Outcomes (ILOs) are mapped to specific courses. Faculty course developers write the signature assignments (SAs) in alignment with the Institutional Learning Outcomes for each course. These SAs use a specific rubric that is applied consistently by every section and every instructor. Each year, the 12-member General Education Team (GET), as a part of their work, conducts the calibration and assessment of the university-wide SAs and the related rubrics for one or two of the five ILOs. GET members, who are senior faculty and administrators who represent each of the university's schools, lead the calibration meetings. Specifically, calibration is a process designed to ensure the accuracy and reliability of evaluating signature assignments by working with peer faculty to interpret rubrics and compare scoring. These calibration meetings include the full time and adjunct faculty members who teach the course to ensure consensus on the application of the rubric. Each calibration session is chaired by a GET representative as well as the course custodian/contact where the ILO signature assignment resides. The overall calibration process GET follows includes multiple steps, such as: contacting and inviting relevant faculty members to participate; organizing and hosting calibration sessions; and reviewing and discussing at least 10 samples of student work from class sections during the previous academic year, for each applicable signature assignment. Conversely, the commitment for calibration teams consist of attending two virtual meetings lasting up to three hours in duration. The first meeting focuses on interpreting the rubric, and the second meeting involves comparing scores for sample work and sharing feedback about the rubric and assignment. Next, participants are expected to follow all directions provided and participate actively, and respond to scheduling requests and other emails within 72 hours. The virtual meetings are recorded for future institutional use. Adjunct faculty are paid a small stipend after participating in both calibration meetings. Full time faculty can use this work on their annual faculty report for "internal service" credit.

Keywords: assessment, assurance of learning, course design, institutional learning outcomes, rubrics, signature assignments

Procedia PDF Downloads 266

Authors: Ja-Keoung Koo, Kensuke Nakamura, Hyohun Kim, Dongwha Shin, Yeonseok Kim, Ji-Su Ahn, Byung-Woo Hong

Abstract:

The machine learning techniques based on a convolutional neural network (CNN) have been actively developed and successfully applied to a variety of image analysis tasks including reconstruction, noise reduction, resolution enhancement, segmentation, motion estimation, object recognition. The classical visual information processing that ranges from low level tasks to high level ones has been widely developed in the deep learning framework. It is generally considered as a challenging problem to derive visual interpretation from high dimensional imagery data. A CNN is a class of feed-forward artificial neural network that usually consists of deep layers the connections of which are established by a series of non-linear operations. The CNN architecture is known to be shift invariant due to its shared weights and translation invariance characteristics. However, it is often computationally intractable to optimize the network in particular with a large number of convolution layers due to a large number of unknowns to be optimized with respect to the training set that is generally required to be large enough to effectively generalize the model under consideration. It is also necessary to limit the size of convolution kernels due to the computational expense despite of the recent development of effective parallel processing machinery, which leads to the use of the constantly small size of the convolution kernels throughout the deep CNN architecture. However, it is often desired to consider different scales in the analysis of visual features at different layers in the network. Thus, we propose a CNN model where different sizes of the convolution kernels are applied at each layer based on the random projection. We apply random filters with varying sizes and associate the filter responses with scalar weights that correspond to the standard deviation of the random filters. We are allowed to use large number of random filters with the cost of one scalar unknown for each filter. The computational cost in the back-propagation procedure does not increase with the larger size of the filters even though the additional computational cost is required in the computation of convolution in the feed-forward procedure. The use of random kernels with varying sizes allows to effectively analyze image features at multiple scales leading to a better generalization. The robustness and effectiveness of the proposed CNN based on random kernels are demonstrated by numerical experiments where the quantitative comparison of the well-known CNN architectures and our models that simply replace the convolution kernels with the random filters is performed. The experimental results indicate that our model achieves better performance with less number of unknown weights. The proposed algorithm has a high potential in the application of a variety of visual tasks based on the CNN framework. Acknowledgement—This work was supported by the MISP (Ministry of Science and ICT), Korea, under the National Program for Excellence in SW (20170001000011001) supervised by IITP, and NRF-2014R1A2A1A11051941, NRF2017R1A2B4006023.

Keywords: deep learning, convolutional neural network, random kernel, random projection, dimensionality reduction, object recognition

Procedia PDF Downloads 272

Authors: Ghazal Faraj, Andras Micsik

Abstract:

The CIDOC Conceptual Reference Model (CRM) is an extensible ontology that provides integrated access to heterogeneous and digital datasets. The CIDOC-CRM offers a “semantic glue” intended to promote accessibility to several diverse and dispersed sources of cultural heritage data. That is achieved by providing a formal structure for the implicit and explicit concepts and their relationships in the cultural heritage field. The COURAGE (“Cultural Opposition – Understanding the CultuRal HeritAGE of Dissent in the Former Socialist Countries”) project aimed to explore methods about socialist-era cultural resistance during 1950-1990 and planned to serve as a basis for further narratives and digital humanities (DH) research. This project highlights the diversity of flourished alternative cultural scenes in Eastern Europe before 1989. Moreover, the dataset of COURAGE is an online RDF-based registry that consists of historical people, organizations, collections, and featured items. For increasing the inter-links between different datasets and retrieving more relevant data from various data silos, a shared federated ontology for reconciled data is needed. As a first step towards these goals, a full understanding of the CIDOC CRM ontology (target ontology), as well as the COURAGE dataset, was required to start the work. Subsequently, the queries toward the ontology were determined, and a table of equivalent properties from COURAGE and CIDOC CRM was created. The structural diagrams that clarify the mapping process and construct queries are on progress to map person, organization, and collection entities to the ontology. Through mapping the COURAGE dataset to CIDOC-CRM ontology, the dataset will have a common ontological foundation with several other datasets. Therefore, the expected results are: 1) retrieving more detailed data about existing entities, 2) retrieving new entities’ data, 3) aligning COURAGE dataset to a standard vocabulary, 4) running distributed SPARQL queries over several CIDOC-CRM datasets and testing the potentials of distributed query answering using SPARQL. The next plan is to map CIDOC-CRM to other upper-level ontologies or large datasets (e.g., DBpedia, Wikidata), and address similar questions on a wide variety of knowledge bases.

Keywords: CIDOC CRM, cultural heritage data, COURAGE dataset, ontology alignment

Procedia PDF Downloads 131

Authors: M. Roberts Masillamani

Abstract:

There is a phenomenal difference, as regard to the teaching methodology adopted at the urban and the rural area colleges. However, bright and talented student may be from rural back ground even. But there is huge dearth of the digitization in the rural areas and lesser developed countries. Today’s students need new skills to compete and successful in the future. Education should be combination of practical, intellectual, and social skills. What does this mean for rural classrooms and how can it be achieved. Rural colleges are not able to hire the best resources, since the best teacher’s aim is to move towards the city. If city is provided everywhere, then there will be no rural area. This is possible by forming university clusters (UC). The University cluster is a group of renowned and accredited universities coming together to bridge this dearth. The UC will deliver the live lectures and allow the students’ from remote areas to actively participate in the classroom. This paper tries to present a plan of action of providing a better live classroom teaching and learning system from the city to the rural and the lesser developed countries. This paper titled “University Clusters using ICT for teaching and learning” provides a true concept of opening live digital classroom windows for rural colleges, where resources are not available, thus reducing the digital divide. This is different from pod casting a lecture or distance learning and eLearning. The live lecture can be streamed through digital equipment to another classroom. The rural students can collaborate with their peers and critiques, be assessed, collect information, acquire different techniques in assessment and learning process. This system will benefit rural students and teachers and develop socio economic status. This will also will increase the degree of confidence of the Rural students and teachers. Thus bringing about the concept of ‘Train the Trainee’ in reality. An educational university cloud for each cluster will be built remote infrastructure facilities (RIF) for the above program. The users may be informed, about the available lecture schedules, through the RIF service. RIF with an educational cloud can be set by the universities under one cluster. This paper talks a little more about University clusters and the methodology to be adopted as well as some extended features like, tutorial classes, library grids, remote laboratory login, research and development.

Keywords: lesser developed countries, digital divide, digital learning, education, e-learning, ICT, library grids, live classroom windows, RIF, rural, university clusters and urban

Procedia PDF Downloads 456

Authors: Martins Y. Otache, John J. Musa, Abayomi I. Kuti, Mustapha Mohammed

Abstract:

The performance of an artificial neural network (ANN) is contingent on a host of factors, for instance, the network optimisation scheme. In view of this, the study examined the general implications of the ANN training optimisation algorithm on its forecast performance. To this end, the Bayesian regularisation (Br), Levenberg-Marquardt (LM), and the adaptive learning gradient descent: GDM (with momentum) algorithms were employed under different ANN structural configurations: (1) single-hidden layer, and (2) double-hidden layer feedforward back propagation network. Results obtained revealed generally that the gradient descent with momentum (GDM) optimisation algorithm, with its adaptive learning capability, used a relatively shorter time in both training and validation phases as compared to the Levenberg- Marquardt (LM) and Bayesian Regularisation (Br) algorithms though learning may not be consummated; i.e., in all instances considering also the prediction of extreme flow conditions for 1-day and 5-day ahead, respectively especially using the ANN model. In specific statistical terms on the average, model performance efficiency using the coefficient of efficiency (CE) statistic were Br: 98%, 94%; LM: 98 %, 95 %, and GDM: 96 %, 96% respectively for training and validation phases. However, on the basis of relative error distribution statistics (MAE, MAPE, and MSRE), GDM performed better than the others overall. Based on the findings, it is imperative to state that the adoption of ANN for real-time forecasting should employ training algorithms that do not have computational overhead like the case of LM that requires the computation of the Hessian matrix, protracted time, and sensitivity to initial conditions; to this end, Br and other forms of the gradient descent with momentum should be adopted considering overall time expenditure and quality of the forecast as well as mitigation of network overfitting. On the whole, it is recommended that evaluation should consider implications of (i) data quality and quantity and (ii) transfer functions on the overall network forecast performance.

Keywords: streamflow, neural network, optimisation, algorithm

Procedia PDF Downloads 139

Authors: Ivan Kuzio

Abstract:

The GIZ/TTC project is the first of its kind in the Middle East, which allows the development of a teaching training programme to degree level based on modern methodologies. The graduates from this college are part of the Saudization programme and will, over the next four years be part of and eventually run the new Colleges of Excellence. The new Colleges of Excellence are being developed to create a local vocationally trained workforce and will run initially alongside the current Colleges of Technology.

Keywords: blended learning, pedagogy, training, key competencies, social skills, cognitive development

Procedia PDF Downloads 295

Authors: Saziye Darendeli

Abstract:

Motivation has a significant impact on a second/foreign language learning process, so it plays a vital role while achieving the learning goal. As it is defined by Simon (1967, p. 29), motivation is “a goal terminating mechanism, permitting goals to be processed serially.”AccordingtoSimon, if a learning goal is activated and enough attention is given, the learner starts learning. In connection with this view, the more attention is given on a subject, and the more activation takes place on it, the quicker learning will occur. Moreover, today almost every teacher is familiar with the term “distance education” regardless of their student's age group. As it is stated by Visser (2002), when compared to the traditional classrooms, in distance education, the rate and success of language learningdecreasesandone of the most essential reasons is that motivating students in distance education contexts, in which interaction is lower, is much more challenging than face-to-face training especially with young learners(Lim& Kim, 2003). Besides, there are limited numbers of studies conducted on motivational strategies for young learners in distance education contexts since we have been experiencing full time the online schooling process recently, yet online teaching seems to be permanent in our lives with the new technological era. Therefore, there appears to be a need for various strategies to motivate young learners in distance education, and the current study aims to find out the strategies that young learners’ teachers use to increase their students’ motivation level in distance education. To achieve this aim, a qualitative research approach and a phenomenological method with an interpretive design will be used. The participants, who are teachers of young learners, will be interviewed using a structured interview format consisting of 7 questions. As the participants are young learners’teacherswhohavebeenexperiencingteaching online, exploring thestrategiesthattheyusetoincreasetheirstudents’ motivationlevelwillprovidesomesuggestionsaboutthemotivationalstrategiesforfuture online classes. Also, in this paper, I will move beyond the traditional classrooms that have face-to-face lessons and discuss the effective motivational strategies for young learners in distance education.

Keywords: motivation, distance education, young learners, strategies

Procedia PDF Downloads 172

Authors: Amin Naboureh, Ainong Li, Jinhu Bian, Guangbin Lei, Hamid Ebrahimy

Abstract:

Large-scale land cover mapping has become a new challenge in land change and remote sensing field because of involving a big volume of data. Moreover, selecting the right classification method, especially when there are different types of landscapes in the study area is quite difficult. This paper is an attempt to compare the performance of different machine learning (ML) algorithms for generating a land cover map of the China-Central Asia–West Asia Corridor that is considered as one of the main parts of the Belt and Road Initiative project (BRI). The cloud-based Google Earth Engine (GEE) platform was used for generating a land cover map for the study area from Landsat-8 images (2017) by applying three frequently used ML algorithms including random forest (RF), support vector machine (SVM), and artificial neural network (ANN). The selected ML algorithms (RF, SVM, and ANN) were trained and tested using reference data obtained from MODIS yearly land cover product and very high-resolution satellite images. The finding of the study illustrated that among three frequently used ML algorithms, RF with 91% overall accuracy had the best result in producing a land cover map for the China-Central Asia–West Asia Corridor whereas ANN showed the worst result with 85% overall accuracy. The great performance of the GEE in applying different ML algorithms and handling huge volume of remotely sensed data in the present study showed that it could also help the researchers to generate reliable long-term land cover change maps. The finding of this research has great importance for decision-makers and BRI’s authorities in strategic land use planning.

Keywords: land cover, google earth engine, machine learning, remote sensing

Procedia PDF Downloads 104

Authors: Chassel T. Paras, Tryxzy Q. Dela Cruz, Ma. Carmela Lousie V. Goingco, Pauline L. Tolentino, Carmela S. Dizon

Abstract:

In implementing inclusive education, parental involvement is measured to be an irreplaceable contributing factor. Parental involvement is described as an indispensable aspect of the teaching-learning process and has a remarkable effect on the student's academic performance. However, there are still differences in the viewpoints, expectations, and needs of both parents and teachers that are not yet fully conveyed in their relationship; hence, the perceptions of SNED teachers are essential in their collaboration with parents. This qualitative study explored how SNED teachers perceive the roles of parental involvement in the teaching-learning process of students with special needs. To answer this question, one-on-one face-to-face semi-structured interviews with three SNED teachers in a selected public school in Angeles City, Philippines, that offer special needs education services were conducted. The gathered data are then analyzed using Interpretative Phenomenological Analysis (IPA). The results revealed four superordinate themes, which include: (1) roles of parental involvement, (2) parental involvement opportunities, (3) barriers to parental involvement, and (4) parent-teacher collaboration practices. These results indicate that SNED teachers are aware of the roles and importance of parental involvement; however, despite parent-teacher collaboration, there are still barriers that impede parental involvement. Also, SNED teachers acknowledge the big roles of parents as they serve as main figures in the teaching-learning process of their children with special needs. Lastly, these results can be used as input in developing a school-facilitated parenting involvement framework that encompasses the contribution of SNED teachers in planning, developing, and evaluating parental involvement programs, which future researchers can also use in their studies

Keywords: parental involvement, special needs education, teaching-learning process, teachers’ perceptions, special needs education teachers, interpretative phenomenological analysis

Procedia PDF Downloads 88

Authors: Yun-Ju Chiu, Feng-Yi Chen

Abstract:

Kinematics is the beginning to learn mechanics in physics course. The concept of acceleration plays an important role in learning kinematics. Teachers usually instruct the conception through the formulas and graphs of kinematics and the well-known law F = ma. However, over the past few decades, a lot of researchers reveal numerous students’ difficulties in learning acceleration. One of these difficulties is that students frequently confuse acceleration with velocity and force. Why is the concept of acceleration so difficult to learn? The aim of this study is to understand the conceptual evolution of acceleration through the historical textual analysis. Text analysis and one-to-one interviews with high school students and teachers are used in this study. This study finds the history of science constructed from textbooks is usually quite different from the real evolution of history. For example, most teachers and students believe that the best-known law F = ma was written down by Newton. The expression of the second law is not F = ma in Newton’s best-known book Principia in 1687. Even after more than one hundred years, a famous Cambridge textbook titled An Elementary Treatise on Mechanics by Whewell of Trinity College did not express this law as F = ma. At that time of Whewell, the early mid-nineteenth century Britain, the concept of acceleration was not only ambiguous but also confused with the concept of force. The process of learning the concept of acceleration is analogous to its conceptual development in history. The study from the perspective of historical textual analysis will promote the understanding of the concept learning difficulties, the development of professional physics teaching, and the improvement of the context of physics textbooks.

Keywords: acceleration, textbooks, mechanics, misconception, history of science

Procedia PDF Downloads 239

Authors: Ambali Alade Odebowale, Andargachew Mekonnen Berhe, Haroldo T. Hattori, Andrey E. Miroshnichenko

Abstract:

Solar thermophotovoltaic systems (STPV) have emerged as a promising solution to overcome the Shockley-Queisser limit, a significant impediment in the direct conversion of solar radiation into electricity using conventional solar cells. The STPV system comprises essential components such as an optical concentrator, selective emitter, and a thermophotovoltaic (TPV) cell. The pivotal element in achieving high efficiency in an STPV system lies in the design of a spectrally selective emitter or absorber. Traditional methods for designing and optimizing selective emitters are often time-consuming and may not yield highly selective emitters, posing a challenge to the overall system performance. In recent years, the application of machine learning techniques in various scientific disciplines has demonstrated significant advantages. This paper proposes a novel nanostructure composed of four-layered materials (SiC/W/SiO2/W) to function as a selective emitter in the energy conversion process of an STPV system. Unlike conventional approaches widely adopted by researchers, this study employs a machine learning-based approach for the design and optimization of the selective emitter. Specifically, a random forest algorithm (RFA) is employed for the design of the selective emitter, while the optimization process is executed using genetic algorithms. This innovative methodology holds promise in addressing the challenges posed by traditional methods, offering a more efficient and streamlined approach to selective emitter design. The utilization of a machine learning approach brings several advantages to the design and optimization of a selective emitter within the STPV system. Machine learning algorithms, such as the random forest algorithm, have the capability to analyze complex datasets and identify intricate patterns that may not be apparent through traditional methods. This allows for a more comprehensive exploration of the design space, potentially leading to highly efficient emitter configurations. Moreover, the application of genetic algorithms in the optimization process enhances the adaptability and efficiency of the overall system. Genetic algorithms mimic the principles of natural selection, enabling the exploration of a diverse range of emitter configurations and facilitating the identification of optimal solutions. This not only accelerates the design and optimization process but also increases the likelihood of discovering configurations that exhibit superior performance compared to traditional methods. In conclusion, the integration of machine learning techniques in the design and optimization of a selective emitter for solar thermophotovoltaic systems represents a groundbreaking approach. This innovative methodology not only addresses the limitations of traditional methods but also holds the potential to significantly improve the overall performance of STPV systems, paving the way for enhanced solar energy conversion efficiency.

Keywords: emitter, genetic algorithm, radiation, random forest, thermophotovoltaic

Procedia PDF Downloads 44

Authors: Mohammad Bakhshandeh

Abstract:

Structural Health Monitoring (SHM) is a crucial and necessary practice that plays a vital role in ensuring the safety and integrity of critical structures, and in particular, bridges. The continuous monitoring of bridges for signs of damage or degradation through Bridge Health Monitoring (BHM) enables early detection of potential problems, allowing for prompt corrective action to be taken before significant damage occurs. Although all monitoring techniques aim to provide accurate and decisive information regarding the remaining useful life, safety, integrity, and serviceability of bridges, understanding the development and propagation of damage is vital for maintaining uninterrupted bridge operation. Over the years, extensive research has been conducted on BHM methods, and experts in the field have increasingly adopted new methodologies. In this article, we provide a comprehensive exploration of the various BHM approaches, including sensor-based, non-destructive testing (NDT), model-based, and artificial intelligence (AI)-based methods. We also discuss the challenges associated with BHM, including sensor placement and data acquisition, data analysis and interpretation, cost and complexity, and environmental effects, through an extensive review of relevant literature and research studies. Additionally, we examine potential solutions to these challenges and propose future research ideas to address critical gaps in BHM.

Keywords: structural health monitoring (SHM), bridge health monitoring (BHM), sensor-based methods, machine-learning algorithms, and model-based techniques, sensor placement, data acquisition, data analysis

Procedia PDF Downloads 72

Authors: Elham Azimzadeh, Hamidollah Hassanlouei, Hadi Nobari, Georgian Badicu, Jorge Pérez-Gómez, Luca Paolo Ardigò

Abstract:

Children with cerebral palsy (CP) have weak physical abilities and their limitations may have an effect on performing everyday motor activities. One of the most important and common debilitating factors in CP is the malfunction in the upper extremities to perform motor skills and there is strong evidence that task-specific training may lead to improve general upper limb function among this population. However, augmented feedback enhances the acquisition and learning of a motor task. Practice conditions may alter the difficulty, e.g., the reduced frequency of PG could be more challenging for this population to learn a motor task. So, the purpose of this study was to investigate the effect of physical guidance (PG) on learning a tracking task in children with cerebral palsy (CP). Twenty-five independently ambulant children with spastic hemiplegic CP aged 7-15 years were assigned randomly to five groups. After the pre-test, experimental groups participated in an intervention for eight sessions, 12 trials during each session. The 0% PG group received no PG; the 25% PG group received PG for three trials; the 50% PG group received PG for six trials; the 75% PG group received PG for nine trials; and the 100% PG group, received PG for all 12 trials. PG consisted of placing the experimenter's hand around the children's hand, guiding them to stay on track and complete the task. Learning was inferred by acquisition and delayed retention tests. The tests involved two blocks of 12 trials of the tracking task without any PG being performed by all participants. They were asked to make the movement as accurate as possible (i.e., fewer errors) and the number of total touches (errors) in 24 trials was calculated as the scores of the tests. The results showed that the higher frequency of PG led to more accurate performance during the practice phase. However, the group that received 75% PG had significantly better performance compared to the other groups in the retention phase. It is concluded that the optimal frequency of PG played a critical role in learning a tracking task in children with CP and likely this population may benefit from an optimal level of PG to get the appropriate amount of information confirming the challenge point framework (CPF), which state that too much or too little information will retard learning a motor skill. Therefore, an optimum level of PG may help these children to identify appropriate patterns of motor skill using extrinsic information they receive through PG and improve learning by activating the intrinsic feedback mechanisms.

Keywords: cerebral palsy, challenge point framework, motor learning, physical guidance, tracking task

Procedia PDF Downloads 57

Authors: Scott Hollier, Ruchi Permvattana

Abstract:

The purpose of this abstract is to share the findings of a recently completed disability-related Internet of Things (IoT) project undertaken at Curtin University in Australia. The project focused on identifying how IoT could support people with disabilities with their educational outcomes. To achieve this, the research consisted of an analysis of current literature and interviews conducted with students with vision, hearing, mobility and print disabilities. While the research acknowledged the ability to collect data with IoT is now a fairly common occurrence, its benefits and applicability still need to be grounded back into real-world applications. Furthermore, it is important to consider if there are sections of our society that may benefit from these developments and if those benefits are being fully realised in a rush by large companies to achieve IoT dominance for their particular product or digital ecosystem. In this context, it is important to consider a group which, to our knowledge, has had little specific mainstream focus in the IoT area –people with disabilities. For people with disabilities, the ability for every device to interact with us and with each other has the potential to yield significant benefits. In terms of engagement, the arrival of smart appliances is already offering benefits such as the ability for a person in a wheelchair to give verbal commands to an IoT-enabled washing machine if the buttons are out of reach, or for a blind person to receive a notification on a smartphone when dinner has finished cooking in an IoT-enabled microwave. With clear benefits of IoT being identified for people with disabilities, it is important to also identify what implications there are for education. With higher education being a critical pathway for many people with disabilities in finding employment, the question as to whether such technologies can support the educational outcomes of people with disabilities was what ultimately led to this research project. This research will discuss several significant findings that have emerged from the research in relation to how consumer-based IoT can be used in the classroom to support the learning needs of students with disabilities, how industrial-based IoT sensors and actuators can be used to monitor and improve the real-time learning outcomes for the delivery of lectures and student engagement, and a proposed method for students to gain more control over their learning environment. The findings shared in this presentation are likely to have significant implications for the use of IoT in the classroom through the implementation of affordable and accessible IoT solutions and will provide guidance as to how policies can be developed as the implications of both benefits and risks continue to be considered by educators.

Keywords: disability, higher education, internet of things, students

Procedia PDF Downloads 99

Authors: Chongya Song, Kang Yen, Alexander Pons, Jin Liu

Abstract:

In the field of machine learning, the ensemble has been employed as a common methodology to improve the performance upon multiple base classifiers. However, the true predictions are often canceled out by the false ones during consensus due to a phenomenon called “curse of correlation” which is represented as the strong interferences among the predictions produced by the base classifiers. In addition, the existing practices are still not able to effectively mitigate the problem of imbalanced classification. Based on the analysis on our experiment results, we conclude that the two problems are caused by some inherent deficiencies in the approach of consensus. Therefore, we create an enhanced ensemble algorithm which adopts a designed rank-based chain-mode consensus to overcome the two problems. In order to evaluate the proposed ensemble algorithm, we employ a well-known benchmark data set NSL-KDD (the improved version of dataset KDDCup99 produced by University of New Brunswick) to make comparisons between the proposed and 8 common ensemble algorithms. Particularly, each compared ensemble classifier uses the same 22 base classifiers, so that the differences in terms of the improvements toward the accuracy and reliability upon the base classifiers can be truly revealed. As a result, the proposed rank-based chain-mode consensus is proved to be a more effective ensemble solution than the traditional consensus approach, which outperforms the 8 ensemble algorithms by 20% on almost all compared metrices which include accuracy, precision, recall, F1-score and area under receiver operating characteristic curve.

Keywords: consensus, curse of correlation, imbalance classification, rank-based chain-mode ensemble

Procedia PDF Downloads 120

Authors: Xabier Olaz Moratinos

Abstract:

Drone navigation is critical, particularly during the initial phases, such as the initial ascension, where pilots may fail due to strong external interferences that could potentially lead to a crash. In this ongoing work, a drone has been successfully trained to perform an ascent of up to 6 meters at speeds with external disturbances pushing it up to 24 mph, with the DQN algorithm managing external forces affecting the system. It has been demonstrated that the system can control its height, position, and stability in all three axes (roll, pitch, and yaw) throughout the process. The learning process is carried out in the Gazebo simulator, which emulates interferences, while ROS is used to communicate with the agent.

Keywords: machine learning, DQN, Gazebo, navigation

Procedia PDF Downloads 59

Authors: Oumaima Khlifati, Khadija Baba

Abstract:

Pavement distress is the main factor responsible for the deterioration of road structure durability, damage vehicles, and driver comfort. Transportation agencies spend a high proportion of their funds on pavement monitoring and maintenance. The auscultation of pavement distress was based on the manual survey, which was extremely time consuming, labor intensive, and required domain expertise. Therefore, the automatic distress detection is needed to reduce the cost of manual inspection and avoid more serious damage by implementing the appropriate remediation actions at the right time. Inspired by recent deep learning applications, this paper proposes an algorithm for automatic road distress detection and classification using on the Deep Convolutional Neural Network (DCNN). In this study, the types of pavement distress are classified as transverse or longitudinal cracking, alligator, pothole, and intact pavement. The dataset used in this work is composed of public asphalt pavement images. In order to learn the structure of the different type of distress, the DCNN models are trained and tested as a multi-label classification task. In addition, to get the highest accuracy for our model, we adjust the structural optimization hyper parameters such as the number of convolutions and max pooling, filers, size of filters, loss functions, activation functions, and optimizer and fine-tuning hyper parameters that conclude batch size and learning rate. The optimization of the model is executed by checking all feasible combinations and selecting the best performing one. The model, after being optimized, performance metrics is calculated, which describe the training and validation accuracies, precision, recall, and F1 score.

Keywords: distress pavement, hyperparameters, automatic classification, deep learning

Procedia PDF Downloads 68

Authors: Marie Alexandre, Jean Bernatchez

Abstract:

This paper focuses on the complexity of the face-to-face-to-distance learning transition process. Our research action aims to support the process of transition from classroom to distance learning for teachers in higher education with regard to pedagogical practices that can meet the various needs of students using digital learning environments. In Quebec and elsewhere in the world, the advent of digital education is helping to transform teaching, which is significantly changing the role of teachers. While distance education implies a dissociation of teaching and learning to a variable degree in space and time, distance education (DE) is becoming more and increasingly becoming a preferred option for maintaining the delivery of certain programs and providing access to programs and to provide access to quality activities throughout Quebec. Given the impact of teaching practices on educational success, this paper reports on the results of three research objectives: 1) To document teachers' knowledge of teaching in distance education through the design, experimentation and production of a repertoire of the determinants of pedagogical practices in response to students' needs. 2) Explain, according to a gendered logic, the adequacy between the pedagogical practices implemented in distance learning and the response to the profiles and needs expressed by students using digital learning environments; 3) Produce a model of a support approach during the process of transition from classroom to distance learning at the college level. A mixed methodology, i.e., a quantitative component (questionnaire survey) and a qualitative component (explanatory interviews and living lab) was used in cycles that were part of an ongoing validation process. The intervention includes the establishment of a professional collaboration group, webinars training webinars for the participating teachers on the didactic issue of knowledge-teaching in FAD, the didactic use of technologies, and the differentiated socialization models of educational success in college education. All of the tools developed will be used by partners in the target environment as well as by all teacher educators, students in initial teacher training, practicing teachers, and the general public. The results show that access to training leading to qualifications and commitment to educational success reflects the existing links between the people in the educational community. The relational stakes of being present in distance education take on multiple configurations and different dimensions of learning testify to needs and realities that are sometimes distinct depending on the life cycle. This project will be of interest to partners in the targeted field as well as to all teacher trainers, students in initial teacher training, practicing college teachers, and to university professors. The entire educational community will benefit from digital resources in education. The scientific knowledge resulting from this action research will benefit researchers in the fields of pedagogy, didactics, teacher training and pedagogy in higher education in a digital context.

Keywords: action research, didactics, digital learning environment, distance learning, higher education, pedagogy technological, pedagogical content knowledge

Procedia PDF Downloads 68

Authors: Xiuying Wang, Changliang Li, Bo Xu

Abstract:

A fixed-length feature vector is required for many machine learning algorithms in NLP field. Word embeddings have been very successful at learning lexical information. However, they cannot capture the compositional meaning of sentences, which prevents them from a deeper understanding of language. In this paper, we introduce a novel hierarchical tree long short-term memory (HTLSTM) model that learns vector representations for sentences of arbitrary syntactic type and length. We propose to split one sentence into three hierarchies: short phrase, long phrase and full sentence level. The HTLSTM model gives our algorithm the potential to fully consider the hierarchical information and long-term dependencies of language. We design the experiments on both English and Chinese corpus to evaluate our model on sentiment analysis task. And the results show that our model outperforms several existing state of the art approaches significantly.

Keywords: deep learning, hierarchical tree long short-term memory, sentence representation, sentiment analysis

Procedia PDF Downloads 338

Authors: Malavika Sundararajan

Abstract:

This study introduces the use of mindfulness techniques in the classroom to make individuals aware of how the creative thinking process works, resulting in more constructive learning and application. Case observation method was utilized within a classroom setting in a graduate class in the Business School. It entailed, briefing the student participants about the use of a template called the dots and depths map, and having them complete it for themselves, compare it to their team members and reflect on the outputs. Finally, they were debriefed about the use of the template and its value to their learning and creative application process. The major finding is the increase in awareness levels of the participants following the use of the template, leading to a subsequent pursuit of diverse knowledge and acquisition of relevant information and not jumping to solutions directly, which increased their overall creative outputs for the given assignment. The significant value of this study is that it can be applied to any classroom on any subject as a powerful mindfulness tool which increases creative problem solving through constructive knowledge building.

Keywords: connecting dots, mindful awareness, constructive knowledge building, learning creatively

Procedia PDF Downloads 129

Authors: Panagiota A. Kotsoni, George S. Ypsilandis

Abstract:

Learning an L2 is a demanding process for all students and in particular for those with learning disabilities (LD) who demonstrate an inability to catch up with their classmates’ progress in a given period of time. This area of study, i.e. examining children with learning disabilities in L2 has not (yet) attracted the growing interest that is registered in L1 and thus remains comparatively neglected. It is this scientific field that this study wishes to contribute to. The longitudinal purpose of this study is to locate effective Supportive Feedback Strategies (SFS) and add to the quality of learning in second language vocabulary in both typically developing (TD) and LD children. Specifically, this study aims at investigating and comparing the performance of TD with LD children on two different types of SFSs related to vocabulary short and long-term retention. In this study two different SFSs have been examined to a total of ten (10) unknown vocabulary items. Both strategies provided morphosyntactic clarifications upon new contextualized vocabulary items. The traditional SFS (direct) provided the information only in one hypertext page with a selection on the relevant item. The experimental SFS (engaging) provided the exact same split information in three successive hypertext pages in the form of a hybrid dialogue asking from the subjects to move on to the next page by selecting the relevant link. It was hypothesized that this way the subjects would engage in their own learning process by actively asking for more information which would further lead to their better retention. The participants were fifty-two (52) foreign language learners (33 TD and 19 LD) aged from 9 to 12, attending an English language school at the level of A1 (CEFR). The design of the study followed a typical pre-post-post test procedure after an hour and after a week. The results indicated statistically significant group differences with TD children performing significantly better than the LD group in both short and long-term memory measurements and in both SFSs. As regards the effectiveness of one SFS over another the initial hypothesis was not supported by the evidence as the traditional SFS was more effective compared to the experimental one in both TD and LD children. This difference proved to be statistically significant only in the long-term memory measurement and only in the TD group. It may be concluded that the human brain seems to adapt to different SFS although it shows a small preference when information is provided in a direct manner.

Keywords: learning disabilities, memory, second/foreign language acquisition, supportive feedback

Procedia PDF Downloads 271

Authors: Misrak Assefa

Abstract:

The Amharic language is the second most widely spoken Semitic language in the world. There are several new overloaded on the web. Searching some useful documents from the web on a specific topic, which is written in the Amharic language, is a challenging task. Hence, document categorization is required for managing and filtering important information. In the classification of Amharic text news, there is still a gap in the domain of information that needs to be launch. This study attempts to design an automatic Amharic news classification using a supervised learning mechanism on four un-touch classes. To achieve this research, 4,182 news articles were used. Naive Bayes (NB) and Decision tree (j48) algorithms were used to classify the given Amharic dataset. In this paper, k-fold cross-validation is used to estimate the accuracy of the classifier. As a result, it shows those algorithms can be applicable in Amharic news categorization. The best average accuracy result is achieved by j48 decision tree and naïve Bayes is 95.2345 %, and 94.6245 % respectively using three categories. This research indicated that a typical decision tree algorithm is more applicable to Amharic news categorization.

Keywords: text categorization, supervised machine learning, naive Bayes, decision tree

Procedia PDF Downloads 174

Authors: Marco Frieslaar, Bing Chu, Eric Rogers

Abstract:

Stroke is a devastating illness that is the second biggest cause of death in the world (after heart disease). Where it does not kill, it leaves survivors with debilitating sensory and physical impairments that not only seriously harm their quality of life, but also cause a high incidence of severe depression. It is widely accepted that early intervention is essential for recovery, but current rehabilitation techniques largely favor hospital-based therapies which have restricted access, expensive and specialist equipment and tend to side-step the emotional challenges. In addition, there is insufficient funding available to provide the long-term assistance that is required. As a consequence, recovery rates are poor. The relatively unexplored solution is to develop therapies that can be harnessed in the home and are formulated from technologies that already exist in everyday life. This would empower individuals to take control of their own improvement and provide choice in terms of when and where they feel best able to undertake their own healing. This research seeks to identify how effective post-stroke, rehabilitation therapy can be applied to upper limb mobility, within the physical context of a home rather than a hospital. This is being achieved through the design and construction of an automation scheme, based on iterative learning control and the Riener muscle model, that has the ability to adapt to the user and react to their level of fatigue and provide tangible physical recovery. It utilizes a SMART Phone and laptop to construct an iterative learning control (ILC) system, that monitors upper arm movement in three dimensions, as a series of exercises are undertaken. The equipment generates functional electrical stimulation to assist in muscle activation and thus improve directional accuracy. In addition, it monitors speed, accuracy, areas of motion weakness and similar parameters to create a performance index that can be compared over time and extrapolated to establish an independent and objective assessment scheme, plus an approximate estimation of predicted final outcome. To further extend its assessment capabilities, nerve conduction velocity readings are taken by the software, between the shoulder and hand muscles. This is utilized to measure the speed of response of neuron signal transfer along the arm and over time, an online indication of regeneration levels can be obtained. This will prove whether or not sufficient training intensity is being achieved even before perceivable movement dexterity is observed. The device also provides the option to connect to other users, via the internet, so that the patient can avoid feelings of isolation and can undertake movement exercises together with others in a similar position. This should create benefits not only for the encouragement of rehabilitation participation, but also an emotional support network potential. It is intended that this approach will extend the availability of stroke recovery options, enable ease of access at a low cost, reduce susceptibility to depression and through these endeavors, enhance the overall recovery success rate.

Keywords: home-based therapy, iterative learning control, Riener muscle model, SMART phone, stroke rehabilitation

Procedia PDF Downloads 250

Authors: Minghui Zhang, Sirine Fkaier, Sabri Fernana, Svetlana Kiseleva, Denis Kiselev

Abstract:

The real-time identification of bacteria and fungi is difficult because they emit much weaker signals than pollen. In 2020, Plair developed Rapid-E+, which extends abilities of Rapid-E to detect smaller bioaerosols such as bacteria and fungal spores with diameters down to 0.3 µm, while keeping the similar or even better capability for measurements of large bioaerosols like pollen. Rapid-E+ enables simultaneous measurements of (1) time-resolved, polarization and angle dependent Mie scattering patterns, (2) fluorescence spectra resolved in 16 channels, and (3) fluorescence lifetime of individual particles. Moreover, (4) it provides 2D Mie scattering images which give the full information on particle morphology. The parameters of every single bioaerosol aspired into the instrument are subsequently analysed by machine learning. Firstly, pure species of microbes, e.g., Bacillus subtilis (a species of bacteria), and Penicillium chrysogenum (a species of fungal spores), were aerosolized in a bioaerosol chamber for Rapid-E+ training. Afterwards, we tested microbes under different concentrations. We used several steps of data analysis to classify and identify microbes. All single particles were analysed by the parameters of light scattering and fluorescence in the following steps. (1) They were treated with a smart filter block to get rid of non-microbes. (2) By classification algorithm, we verified the filtered particles were microbes based on the calibration data. (3) The probability threshold (defined by the user) step provides the probability of being microbes ranging from 0 to 100%. We demonstrate how Rapid-E+ identified simultaneously microbes based on the results of Bacillus subtilis (bacteria) and Penicillium chrysogenum (fungal spores). By using machine learning, Rapid-E+ achieved identification precision of 99% against the background. The further classification suggests the precision of 87% and 89% for Bacillus subtilis and Penicillium chrysogenum, respectively. The developed algorithm was subsequently used to evaluate the performance of microbe classification and quantification in real-time. The bacteria and fungi were aerosolized again in the chamber with different concentrations. Rapid-E+ can classify different types of microbes and then quantify them in real-time. Rapid-E+ enables classifying different types of microbes and quantifying them in real-time. Rapid-E+ can identify pollen down to species with similar or even better performance than the previous version (Rapid-E). Therefore, Rapid-E+ is an all-in-one instrument which classifies and quantifies not only pollen, but also bacteria and fungi. Based on the machine learning platform, the user can further develop proprietary algorithms for specific microbes (e.g., virus aerosols) and other aerosols (e.g., combustion-related particles that contain polycyclic aromatic hydrocarbons).

Keywords: bioaerosols, laser-induced fluorescence, Mie-scattering, microorganisms

Procedia PDF Downloads 75

Authors: Kanu Priya Mohan

Abstract:

Innovation in education is one of the essential pathways to achieve both educational, and development goals in today’s dynamically changing world. Over the last decade, coaching and mentoring have been applied in the field of education as positive intervention techniques for fostering teaching and learning reforms in the developed countries. The context of this research was Thailand’s educational reform process, wherein a project on coaching and mentoring (C&M) was launched in 2014. The C&M project endeavored to support the professional development of the school teachers in the various provinces of Thailand, and to also enable them to apply C&M for teaching innovative instructional techniques. This research aimed to empirically investigate the learning outcomes for the master trainers, who trained for coaching and mentoring as the first step in the process to train the school teachers. A mixed method study was used for evaluating the learning outcomes of training in terms of cognitive- behavioral-affective dimensions. In the first part of the research a quantitative research design was incorporated to evaluate the effects of learner characteristics and instructional techniques, on the learning outcomes. In the second phase, a qualitative method of in-depth interviews was used to find details about the training outcomes, as well as the perceived barriers and enablers of the training process. Sample size constraints were there, yet these exploratory results, integrated from both methods indicated the significance of evaluating training outcomes from the three dimensions, and the perceived role of other factors in the training. Findings are discussed in terms of their implications for the training of C&M, and also their impact in fostering positive education through innovative educational techniques in the developing countries.

Keywords: cognitive-behavioral-affective learning outcomes, mixed method research, teachers in Thailand, training evaluation

Procedia PDF Downloads 260

Authors: Syed Romat Ali Shah, Bilal Shoaib, Saleem Akhtar, Munib Ahmad, Shahan Sadiqui

Abstract:

Classification is a data mining technique, which is recognizedon Machine Learning (ML) algorithm. It is used to classifythe individual articlein a knownofinformation into a set of predefinemodules or group. Web mining is also a portion of that sympathetic of data mining methods. The main purpose of this paper to analysis and compare the performance of Naïve Bayse Algorithm, Decision Tree, K-Nearest Neighbor (KNN), Artificial Neural Network (ANN)and Support Vector Machine (SVM). This paper consists of different ML algorithm and their advantages and disadvantages and also define research issues.

Keywords: Data Mining, Web Mining, classification, ML Algorithms

Procedia PDF Downloads 271

Authors: R. Ramesh, K. K. Shivaraman

Abstract:

The main aim of this paper is to handle the energy requirement in an efficient manner by merging the advanced digital communication and control technologies for smart grid applications. In order to reduce user home load during peak load hours, utility applies several incentives such as real-time pricing, time of use, demand response for residential customer through smart meter. However, this method provides inconvenience in the sense that user needs to respond manually to prices that vary in real time. To overcome these inconvenience, this paper proposes a convolutional neural network (CNN) with k-means clustering machine learning model which have ability to forecast energy requirement in short term, i.e., hour of the day or day of the week. By integrating our proposed technique with home energy management based on Bluetooth low energy provides predicted value to user for scheduling appliance in advanced. This paper describes detail about CNN configuration and k-means clustering algorithm for short-term energy forecasting.

Keywords: convolutional neural network, fuzzy logic, k-means clustering approach, smart home energy management

Procedia PDF Downloads 289

Authors: Nik Siti Hanifah Nik Ahmad

Abstract:

This research discuss about teaching cupping therapy or hijama by using an online 3D Virtual Learning Environment. The experimental platform was using of flash and Second Life as 2D and 3D comparison. 81 samples have been used in three experiments with 21 in the first and 30 in each second and third. The design of the presentation was tested in five categories such as effectiveness, ease of use, efficacy, aesthetic and users’ satisfaction. The results from three experiments had shown promising outcome for usage of the technique to be implement in teaching Cupping Therapy as well as other alternative or conventional medicine knowledge especially for training.

Keywords: medical and health, cupping therapy or hijama, second life, online 3D VLE, virtual worlds

Procedia PDF Downloads 405

Authors: Sapfo A. Mortaki

Abstract:

This article examines the presence of Greek immigrant artists (painters and sculptors) in Paris during 1970-2010. The aim is to highlight their presence in the French capital through archival research in the daily and periodical press as well as present the impact of their artistic activity on the French intellectual life and society. At the same time, their contribution to the development of cultural life in Greece becomes apparent. The integration of those migrant artists into an environment of cultural coexistence and the understanding of the social phenomenon of their migration, in the context of postmodernity, are being investigated. The cultural relations between the two countries are studied in the context of support mechanisms, such as the Greek community, cultural institutions, museums and galleries. The recognition of the Greek artists by the French society and the social dimension in the context of their activity in Paris, are discussed in terms of the assimilation theory. Since the 1970s, and especially since the fall of the dictatorship in Greece, in opposition to the prior situation, artists' contacts with their homeland have been significantly enhanced, with most of them now travelling to Paris, while others work in parallel in both countries. As a result, not only do the stages of the development of their work through their pursuits become visible, but, most importantly, the artistic world becomes informed about the multifaceted expression of art through the succession of various contemporary currents. Thus, the participation of Greek artists in the international cultural landscape is demonstrated.

Keywords: artistic migration, cultural impact, Greek artists, postmodernity, theory of assimilation

Procedia PDF Downloads 294