Search results for: opposition based learning

Authors: Henriette van Rensburg, Betty Adcock

Abstract:

Concern about the lack of knowledge of quality teaching and teacher retention in rural and remote areas of Australia, has caused academics to improve pre-service teachers’ understanding of this problem. The participants in this study were forty students enrolled in an undergraduate educational course (EDO3341 Teaching in rural and remote communities) at the University of Southern Queensland in Toowoomba in 2012. This study involved an innovative case-based teaching approach in order to broaden their generally under-informed understanding of teaching in a rural and remote area. Three themes have been identified through analysing students’ critical reflections: learning expertise, case-based learning support and authentic learning. The outcomes identified the changes in pre-service teachers’ understanding after they have deepened their knowledge of the realities of teaching in rural and remote areas.

Keywords: rural and remote education, case based teaching, innovative education approach, higher education

Procedia PDF Downloads 491

Authors: Violeta Damjanovic-Behrendt

Abstract:

This paper presents an approach for optimal cyber security decisions to protect instances of a federated Internet of Things (IoT) platform in the cloud. The presented solution implements the repeated Stackelberg Security Game (SSG) and a model called Stochastic Human behaviour model with AttRactiveness and Probability weighting (SHARP). SHARP employs the Subjective Utility Quantal Response (SUQR) for formulating a subjective utility function, which is based on the evaluations of alternative solutions during decision-making. We augment the repeated SSG (including SHARP and SUQR) with a reinforced learning algorithm called Naïve Q-Learning. Naïve Q-Learning belongs to the category of active and model-free Machine Learning (ML) techniques in which the agent (either the defender or the attacker) attempts to find an optimal security solution. In this way, we combine GT and ML algorithms for discovering optimal cyber security policies. The proposed security optimization components will be validated in a collaborative cloud platform that is based on the Industrial Internet Reference Architecture (IIRA) and its recently published security model.

Keywords: security, internet of things, cloud computing, stackelberg game, machine learning, naive q-learning

Procedia PDF Downloads 354

Authors: Mohammed Belalia Douma

Abstract:

Interactive learning is considered as the most effective strategy of learning, it is an educational philosophy based on the learner's contribution and involvement mainly in classroom and how he interacts toward his small society “classroom”, and the level of his collaboration into challenge, discovering, games, participation, all these can be provided through the interactive learning, which aims to activate the learner's role in the operation of learning, which focuses on research and experimentation, and the learner's self-reliance in obtaining information, acquiring skills, and forming values and attitudes. Whereas not based on memorization only, but rather on developing thinking and the ability to solve problems, on teamwork and collaborative learning. With the exchange or roles - teacher to student- , when the student will be more active and performing operations more than the student under the interactive learning method; we might face a several issues dealing with class controlling management, noise, and stability of learning… etc. This research paper is observing the application of the interactive learning on reality “classroom” and answers several assumptions and analyzes the issues coming up of these strategies mainly: noise, class control…etc The research sample was about 150 student of the 3rd year primary school in “Chlef” district, Algeria, level: beginners in the range of age 08 to 10 years old . We provided a questionnaire of confidential fifteen questions and also analyzing the attitudes of learners during three months. it have witnessed as teachers a variety of strategies dealing with applying the interactive learning but with a different issues; time management, noise, uncontrolled classes, overcrowded classes. Finally, it summed up that although the active education is an inevitably effective method of teaching, however, there are drawbacks to this, in addition to the fact that not all theoretical strategies can be applied and we conclude with solutions of this case study.

Keywords: interactive learning, student, learners, strategies.

Procedia PDF Downloads 59

Authors: Jolanta Jonak

Abstract:

Traditional classrooms are challenging for many students, but especially for students that learn differently due to cognitive makeup, learning preferences, or disability. These students often require different teaching approaches and learning opportunities to benefit from learning. Teachers frequently divert to using one teaching approach, the one that matches their own learning style. For instance, teachers that are auditory learners, likely default to providing auditory learning opportunities. However, if a student is a visual learner, he/she may not fully benefit from that teaching style. Based on research, students and their parents’ feedback, large numbers of students are not provided the type of education and types of supports they need in order to be successful in an academic environment. This eventually leads to not learning at an appropriate rate and ultimately leading to skill deficiencies and deficits. Providing varied learning approaches promote high academic and social-emotional growth of all students and it will prevent inaccurate Special Education referrals. Varied learning opportunities can be delivered for all students by providing Differentiated Instruction (DI). This type of instruction allows each student to learn in the most optimal way regardless of learning preferences and cognitive learning profiles. Using Differentiated Instruction will lead to a high level of student engagement and learning. In addition, experiencing success in the classroom, will contribute to increased social emotional wellbeing. Being cognizant of how teaching approaches impact student's learning, school staff can avoid inaccurate perceptions about the students’ learning abilities, unnecessary referrals for special education evaluations, and inaccurate decisions about the presence of a disability. This presentation will illustrate learning differences due to various factors, how to recognize them, and how to address them through Differentiated Instruction.

Keywords: special education, disability, differences, differentiated instruction, social emotional wellbeing

Procedia PDF Downloads 49

Authors: Haozhe Xiang

Abstract:

With the rise of the Internet of Things era, intelligent products are gradually integrating into people's lives. Pedestrian trajectory prediction has become a key issue, which is crucial for the motion path planning of intelligent agents such as autonomous vehicles, robots, and drones. In the current technological context, deep learning technology is becoming increasingly sophisticated and gradually replacing traditional models. The pedestrian trajectory prediction algorithm combining neural networks and attention mechanisms has significantly improved prediction accuracy. Based on in-depth research on deep learning and pedestrian trajectory prediction algorithms, this article focuses on physical environment modeling and learning of historical trajectory time dependence. At the same time, social interaction between pedestrians and scene interaction between pedestrians and the environment were handled. An improved pedestrian trajectory prediction algorithm is proposed by analyzing the existing model architecture. With the help of these improvements, acceptable predicted trajectories were successfully obtained. Experiments on public datasets have demonstrated the algorithm's effectiveness and achieved acceptable results.

Keywords: deep learning, graph convolutional network, attention mechanism, LSTM

Procedia PDF Downloads 70

Authors: Rodrigo Aguiar, Adelino Ferreira

Abstract:

Road traffic accidents are the leading cause of unnatural death and injuries worldwide, representing a significant problem of road safety. In this context, the use of artificial intelligence with advanced machine learning techniques has gained prominence as a promising approach to predict traffic accidents. This article investigates the application of machine learning algorithms to develop traffic accident frequency prediction models. Models are evaluated based on performance metrics, making it possible to do a comparative analysis with traditional prediction approaches. The results suggest that machine learning can provide a powerful tool for accident prediction, which will contribute to making more informed decisions regarding road safety.

Keywords: machine learning, artificial intelligence, frequency of accidents, road safety

Procedia PDF Downloads 89

Authors: Nurhak Altın, Ayhan Kucukmanisa, Oguzhan Urhan

Abstract:

It is important that highways are in good condition for traffic safety. Road crashes (road cracks, erosion of lane markings, etc.) can cause accidents by affecting driving. Image processing based methods for detecting road cracks are available in the literature. In this paper, a deep learning based road crack detection approach is proposed. YOLO (You Look Only Once) is adopted as core component of the road crack detection approach presented. The YOLO network structure, which is developed for object detection, is trained with road crack images as a new class that is not previously used in YOLO. The performance of the proposed method is compared using different training methods: using randomly generated weights and training their own pre-trained weights (transfer learning). A similar training approach is applied to the simplified version of the YOLO network model (tiny yolo) and the results of the performance are examined. The developed system is able to process 8 fps on NVIDIA Jetson TX1 development kit.

Keywords: deep learning, embedded platform, real-time processing, road crack detection

Procedia PDF Downloads 338

Authors: Bunki Enid Pitsoane

Abstract:

This study was influenced by the radical change in the tutorial system of UNISA, immigrating from face to face to E-tutoring. The study was undertaken to investigate the perceptions, experiences, and views of E-tutors in relation to active learning. The study is aimed at capturing the views and experiences of E-tutors as they are deemed to implement active learning within their E-tutoring. The problem was traced from Developmental and behaviorist’s theorists perspective and factors related to perception, experience, and views of E-tutors on active learning. The research is aligned with the views of constructivism which put more emphasis on situated learning, chaos, and digital factors. The basis of the theory is that learning is developmental, situational and context-sensitive and also digital. The theorists further purports that the tutor’s conception of teaching and learning influence their tutoring style. In order to support or reject the findings of the literature study, qualitative research in the form of interviews and document analysis were conducted. The sample of the study constituted of 10 E-tutors who are involved in tutoring modules from the College of Education. The identified E-tutors were randomly selected based on their availability. The data concerning E-tutors perception and experience was analysed and interpreted. The results of the empirical study indicated that some tutors are struggling to implement active learning because they are digital immigrants or they lack in digital knowledge which affect productivity in their teaching.

Keywords: E-Tutoring, active learning, perceptions, views

Procedia PDF Downloads 222

Authors: Ratima Damkham, Natasha Dejdumrong, Priyakorn Pusawiro

Abstract:

This paper explored the solution in game design to help game designers in the educational game designing using digital educational game model canvas (DEGMC) and digital educational game form (DEGF) based on Outcome-based Education program. DEGMC and DEGF can help designers develop an overview of the game while designing and planning their own game. The way to clearly assess players’ ability from learning outcomes and support their game learning design is by using the tools. Designers can balance educational content and entertainment in designing a game by using the strategies of the Business Model Canvas and design the gameplay and players’ ability assessment from learning outcomes they need by referring to the Constructive Alignment. Furthermore, they can use their design plan in this research to write their Game Design Document (GDD). The success of the research was evaluated by four experts’ perspectives in the education and computer field. From the experiments, the canvas and form helped the game designers model their game according to the learning outcomes and analysis of their own game elements. This method can be a path to research an educational game design in the future.

Keywords: constructive alignment, constructivist theory, educational game, outcome-based education

Procedia PDF Downloads 354

Authors: Ágoston Nagy

Abstract:

This paper introduces a method of using unsupervised machine learning to visualize the feature space of a dataset in 2D, in order to find most characteristic segments in the set. After dimension reduction, users can select clusters by manual drawing. Selected clusters are recorded into a data model that is used for later predictions, based on realtime data. Predictions are made with supervised learning, using Gesture Recognition Toolkit. The paper introduces two example applications: a semantic audio organizer for analyzing incoming sounds, and a gesture database organizer where gestural data (recorded by a Leap motion) is visualized for further manipulation.

Keywords: gesture recognition, machine learning, real-time interaction, visualization

Procedia PDF Downloads 353

Authors: Mst. Tuhin Akter, Sharun Akter Khushbu, S. M. Shaqib

Abstract:

A real-time system architecture is highly effective for monitoring and detecting various damaged products or fruits that may deteriorate over time or become infected with diseases. Deep learning models have proven to be effective in building such architectures. However, building a deep learning model from scratch is a time-consuming and costly process. A more efficient solution is to utilize deep neural network (DNN) based transfer learning models in the real-time monitoring architecture. This study focuses on using a novel strawberry dataset to develop effective transfer learning models for the proposed real-time monitoring system architecture, specifically for evaluating and detecting strawberry freshness. Several state-of-the-art transfer learning models were employed, and the best performing model was found to be Xception, demonstrating higher performance across evaluation metrics such as accuracy, recall, precision, and F1-score.

Keywords: strawberry freshness evaluation, deep neural network, transfer learning, image augmentation

Procedia PDF Downloads 90

Authors: Ng Siew Keow, Cheah Seeh Lee

Abstract:

Students’ strong engagement in learning increases their motivation and satisfaction to learn, be resilient to combat academic stress. Engagement in learning is even crucial in the open distance learning (ODL) setting, where the adult students are learning remotely, lessons and learning materials are mostly delivered via online platforms. This study aimed to explore the relationship between learning engagement and perceived academic stress levels of adult students who enrolled in ODL learning mode. In this descriptive correlation study during the 2021-2022 academic years, 101 adult students from Wawasan Open University, Malaysia (WOU) were recruited through convenient sampling. The adult students’ online learning engagement levels and perceived academic stress levels were identified through the self-report Online Student Engagement Scale (OSE) and the Perception of Academic Stress Scale (PASS). The Pearson correlation coefficient test revealed a significant positive relationship between online student engagement and perceived academic stress (r= 0.316, p<0.01). The higher scores on PASS indicated lower levels of perceived academic stress. The findings of the study supported the assumption of the importance of engagement in learning in promoting psychological well-being as well as sustainability in online learning in the open distance learning context.

Keywords: student engagement, academic stress, open distance learning, online learning

Procedia PDF Downloads 161

Authors: I. Pelizzoni, C. Cavallini, I. Salvaderi, F. Cavallini

Abstract:

We present the improvement in writing skills obtained by 94 participants (aged between six and 10 years) with special educational needs through a writing enhancement program based on fluency principles. The study was planned and conducted with a single-subject experimental plan for each of the participants, in order to confirm the results in the literature. These results were obtained using precision teaching (PT) methodology to increase the number of written graphemes per minute in the pre- and post-test, by curriculum based measurement (CBM). Results indicated an increase in the number of written graphemes for all participants. The average overall duration of the intervention is 144 minutes in five months of treatment. These considerations have been analyzed taking account of the complexity of the implementation of measurement systems in real operational contexts (an Italian learning center) and important aspects of replicability and cost-effectiveness of such interventions.

Keywords: curriculum based measurement, precision teaching, writing skill, Italian learning center

Procedia PDF Downloads 128

Authors: K. Z. Tang, K. Ameek, K. Kuang

Abstract:

Rapid product development cycles and changing economic conditions compel businesses to find new ways to stay relevant and effective. One of the ways which many companies have adopted is to spur innovations within the various team-based units in the organization. It would be relevant and important to ensure our graduates are ready to excel in such evolving conditions within their professional eco-systems. However, it is not easy to understand the interplays of nurturing team innovation and improving students’ learning, in the context of engineering education. In this study, we seek to understand team innovation and explore ways to improve students’ performance and learning, via motivation and mentorship. Learning goals from a group of students are collected during a carefully designed two-week long summer programme to provide insights on the main themes, within the context of learning and working in a team.

Keywords: team innovation, mentorship, motivation, learning

Procedia PDF Downloads 282

Authors: Yasith Nayana, Janani Manamperuma, Lalindi Amarasinghe, Sasanka Kodithuwakku

Abstract:

Educational games are popular among current e-learning systems. The approach to education through interactive media is expected to motivate students and encourage participation and engagement. ‘Kalayathra’ is an adaptive, player centered e-learning 3D game. The game identifies the player’s personality and adapt the gaming environment according to the player’s preference. Our platform measures the student’s performance and support learning through player assessment. Player experience is a good measure of the level of fun and education presented to players. To assess the level of playability we introduce an educational playability model. ‘Kalayathra’ is developed according to the GCE O/L syllabus and teaching guide in Sri Lankan education system. The game is capable of guiding players into the environment and aid them in tasks and activities depending on how much the player requires help.

Keywords: e-learning, games, adaptive, personality, gamification, player experience

Procedia PDF Downloads 433

Authors: Jamaludin Jamaludin, Suparji Suparji, Lilik Anifah, I. Gusti Putu Asto Buditjahjanto, Eppy Yundra

Abstract:

Learning media is an important and main component in the learning process. By using currently available media, cadets still have difficulty understanding how the piston engine works, so they are not able to apply these concepts appropriately. This study aims to examine the development of interactive media for piston engine courses in order to improve student learning outcomes. The research method used is a literature study of several articles, journals and proceedings of interactive media development results from 2010-2020. The results showed that the development of interactive media is needed to support the learning process and influence the cognitive abilities of students. With this interactive media, learning outcomes can be improved and the learning process can be effective.

Keywords: interactive media, learning outcomes, learning process, literature study

Procedia PDF Downloads 151

Authors: Yitao Lei, Xingxiang Zhai, Burra Venkata Durga Kumar

Abstract:

As the quantity and complexity of computing in large-scale software systems increase, distributed system computing becomes increasingly important. The distributed system realizes high-performance computing by collaboration between different computing resources. If there are no efficient resource scheduling resources, the abuse of distributed computing may cause resource waste and high costs. However, resource scheduling is usually an NP-hard problem, so we cannot find a general solution. However, some optimization algorithms exist like genetic algorithm, ant colony optimization, etc. The large scale of distributed systems makes this traditional optimization algorithm challenging to work with. Heuristic and machine learning algorithms are usually applied in this situation to ease the computing load. As a result, we do a review of traditional resource scheduling optimization algorithms and try to introduce a deep reinforcement learning method that utilizes the perceptual ability of neural networks and the decision-making ability of reinforcement learning. Using the machine learning method, we try to find important factors that influence the performance of distributed system computing and help the distributed system do an efficient computing resource scheduling. This paper surveys the application of deep reinforcement learning on distributed system computing resource scheduling proposes a deep reinforcement learning method that uses a recurrent neural network to optimize the resource scheduling, and proposes the challenges and improvement directions for DRL-based resource scheduling algorithms.

Keywords: resource scheduling, deep reinforcement learning, distributed system, artificial intelligence

Procedia PDF Downloads 111

Authors: Leena Ahmad Khaleel Alfarani, Maggie McPherson, Neil Morris

Abstract:

Over the last decade, m-learning has been widely accepted and utilized by many western universities. However, Saudi universities face many challenges in utilizing such technology, a central one being to encourage teachers to use such technology. Although there are several factors that affect faculty members’ participation in the adoption of m-learning, this paper focuses merely on two factors, the usefulness and ease of using m-learning. A sample of 279 faculty members in one Saudi university has responded to the online survey. The results of the study have revealed that there is a statistically significant relationship (at the 0.05 level) between both usefulness and ease of using m-learning factors and the intention of teachers to use m-learning currently and in the future.

Keywords: mobile learning, diffusion of innovation theory, technology acceptance, faculty adoption

Procedia PDF Downloads 545

Authors: Panita Wannapiroon, Noppadon Phumeechanya, Sitthichai Laisema

Abstract:

This study is the research and development which is intended to: 1) design the ubiquitous cloud learning management system and: 2) assess the suitability of the design of the ubiquitous cloud learning management system. Its methods are divided into 2 phases. Phase 1 is the design of the ubiquitous cloud learning management system, phase 2 is the assessment of the suitability of the design the samples used in this study are work done by 25 professionals in the field of Ubiquitous cloud learning management systems and information and communication technology in education selected using the purposive sampling method. Data analyzed by arithmetic mean and standard deviation. The results showed that the ubiquitous cloud learning management system consists of 2 main components which are: 1) the ubiquitous cloud learning management system server (u-Cloud LMS Server) including: cloud repository, cloud information resources, social cloud network, cloud context awareness, cloud communication, cloud collaborative tools, and: 2) the mobile client. The result of the system suitability assessment from the professionals is in the highest range.

Keywords: learning management system, cloud computing, ubiquitous learning, ubiquitous learning management system

Procedia PDF Downloads 520

Authors: Ibtissem Daoudi, Raoudha Chebil, Wided Lejouad Chaari

Abstract:

Serious games constitute actually a recent and attractive way supposed to replace the classical boring courses. However, the choice of the adapted serious game to a specific learning environment remains a challenging task that makes teachers unwilling to adopt this concept. To fill this gap, we present, in this paper, a multi-agent-based simulator allowing to predict the impact of a serious game integration in a learning environment given several game and players characteristics. As results, the presented tool gives intensities of several emotional aspects characterizing learners reactions to the serious game adoption. The presented simulator is tested to predict the effect of basing a coding course on the serious game ”CodeCombat”. The obtained results are compared with feedbacks of using the same serious game in a real learning process.

Keywords: emotion, learning process, multi-agent simulation, serious games

Procedia PDF Downloads 398

Authors: A. Kavya Sri, K. Sai Vineela, R. Vanitha, S. Rohith

Abstract:

Large pieces of ice that break from the glaciers are known as icebergs. The threat that icebergs pose to navigation, production of offshore oil and gas services, and underwater pipelines makes their detection crucial. In this project, an automated iceberg tracking method using deep learning techniques and satellite images of icebergs is to be developed. With a temporal resolution of 12 days and a spatial resolution of 20 m, Sentinel-1 (SAR) images can be used to track iceberg drift over the Southern Ocean. In contrast to multispectral images, SAR images are used for analysis in meteorological conditions. This project develops a web-based graphical user interface to detect and track icebergs using sentinel-1 images. To track the movement of the icebergs by using temporal images based on their latitude and longitude values and by comparing the center and area of all detected icebergs. Testing the accuracy is done by precision and recall measures.

Keywords: synthetic aperture radar (SAR), icebergs, deep learning, spatial resolution, temporal resolution

Procedia PDF Downloads 91

Authors: Ying Zhou, Jian-Hua Wang

Abstract:

Based upon the self-determination theory and self-regulated learning theory, this study examined the predictiveness of goal orientation and self-regulated learning strategies on academic achievement of adult students in distance learning. The results show a positive relation between goal orientation and the use of self-regulated strategies, and academic achievements. A significant and positive indirect relation of mastery goal orientation through self-regulated learning strategies was also found. In addition, results pointed to a positive indirect impact of performance-approach goal orientation on academic achievement. The effort regulation strategy fully mediated this relation. The theoretical and instructional implications are discussed. Interventions can be made to motivate students’ mastery or performance approach goal orientation and help them manage their time or efforts.

Keywords: goal orientation, self-regulated strategies, achievement, adult distance students

Procedia PDF Downloads 274

Authors: Zolzaya Badmaavanchig, Shoko Miyamoto

Abstract:

In recent years, universal design for learning (hereinafter referred to as "UDL"), which aims to establish an inclusive education system and to make all children, regardless of their disabilities, experts in learning, has been attracting attention, and there have been some attempts to incorporate it into regular classrooms where children with developmental disabilities and those who show such tendencies are enrolled. The purpose of this study was to examine the effectiveness and challenges of implementing UDL in Japanese elementary schools based on the previous literature. As a method, we first searched for articles on UDL for learning and UDL in the classroom from 2010 to 2022. In addition, we selected practice studies that targeted children with special educational support needs and the classroom as a whole. In response to the extracted literature, this bridge examined the following five perspectives: (1) subjects and grades in which UDL was practiced, (2) methods to grasp the actual conditions of the children, (3) consideration for children with special needs during class, (4) form of class, and (5) effects of the practice. Based on the results, we would like to present issues related to future UDL efforts in Japanese elementary schools.

Keywords: universal design for learning, regular elementary school class, children with special education needs, special educational support

Procedia PDF Downloads 62

Authors: Piyanut Sujit

Abstract:

The purposes of this study were to: 1) create knowledge and develop web based instruction on the puppet show, 2) evaluate the effectiveness of the web based instruction on the puppet show by using the criteria of 80/80, and 3) compare and analyze the achievement of the students before and after learning with web based instruction on the puppet show. The population of this study included 53 students in the Program of Library and Information Sciences who registered in the subject of Reading and Reading Promotion in semester 1/2011, Suansunandha Rajabhat University. The research instruments consisted of web based instruction on the puppet show, specialist evaluation form, achievement test, and tests during the lesson. The research statistics included arithmetic mean, variable means, standard deviation, and t-test in SPSS for Windows. The results revealed that the effectiveness of the developed web based instruction was 84.67/80.47 which was higher than the set criteria at 80/80. The student achievement before and after learning showed statistically significant difference at 0.05 as in the hypothesis.

Keywords: puppet, puppet show, web based instruction, library and information sciences

Procedia PDF Downloads 367

Authors: Cristina Galván-Fernández

Abstract:

Students differentiate the different contexts of their lives as well as employment, hobbies or studies. In higher education is needed to transfer the experiential knowledge to theory and viceversa. However, is difficult to achieve than students use their personal experiences and social readings for get the learning evidences. In an experience with 178 education students from Chile and Spain we have used an e-portfolio system and a methodology for 4 years with the aims of help them to: 1) self-regulate their learning process and 2) use social networks and professional experiences for make the learning evidences. These two objectives have been controlled by interviews to the same students in different moments and two questionnaires. The results of this study show that students recognize the ownership of their learning and progress in planning and reflection of their own learning.

Keywords: competences, e-portfolio, higher education, self-regulation

Procedia PDF Downloads 299

Authors: Susan Diamond

Abstract:

Machine Learning workloads have traditionally been run in high-performance computing (HPC) environments, where users log in to dedicated machines and utilize the attached GPUs to run training jobs on huge datasets. Training of large neural network models is very resource intensive, and even after exploiting parallelism and accelerators such as GPUs, a single training job can still take days. Consequently, the cost of hardware is a barrier to entry. Even when upfront cost is not a concern, the lead time to set up such an HPC environment takes months from acquiring hardware to set up the hardware with the right set of firmware, software installed and configured. Furthermore, scalability is hard to achieve in a rigid traditional lab environment. Therefore, it is slow to react to the dynamic change in the artificial intelligent industry. Watson Deep Learning as a service, a cloud-based deep learning platform that mitigates the long lead time and high upfront investment in hardware. It enables robust and scalable sharing of resources among the teams in an organization. It is designed for on-demand cloud environments. Providing a similar user experience in a multi-tenant cloud environment comes with its own unique challenges regarding fault tolerance, performance, and security. Watson Deep Learning as a service tackles these challenges and present a deep learning stack for the cloud environments in a secure, scalable and fault-tolerant manner. It supports a wide range of deep-learning frameworks such as Tensorflow, PyTorch, Caffe, Torch, Theano, and MXNet etc. These frameworks reduce the effort and skillset required to design, train, and use deep learning models. Deep Learning as a service is used at IBM by AI researchers in areas including machine translation, computer vision, and healthcare. 

Keywords: deep learning, machine learning, cognitive computing, model training

Procedia PDF Downloads 208

Authors: Jacqueline Mary Jepson

Abstract:

This article provides a case study centred on the development of 13 video episodes which have been created to enhance student engagement with a post graduate online course in Project Management. The student group was unique as their online course needed to provide for asynchronistic learning and an adult learning pedagogy. In addition, students had come from a wide range professional backgrounds, with some having no Project Management experience, while others had 20 years or more. Students had to gain an understanding of an advanced body of knowledge and the course needed to achieve the academic requirements to qualify individuals to apply their learning in a range of contexts for professional practice and scholarship. To achieve this, a 13 episode case study was developed along with supportive learning materials based on the relocation of a zoo. This unique project provided a learning environment where the project could evolve over each video episode demonstrating the application of Project Management methodology which was then tied into the learning outcomes for the course and the assessment tasks. Discussion forums provided a way for students to converse and demonstrate their own understanding of content and how Project Management methodology can be applied.

Keywords: project management, adult learning, video case study, asynchronistic education

Procedia PDF Downloads 338

Authors: Ksenia Meshkova

Abstract:

With increasing amount of multimedia data, it is very important to automate and speed up the process of obtaining meta. This process means not just recognition of some object or its movement, but recognition of the entire scene versus separate frames and having timeline segmentation as a final result. Labeling datasets is time consuming, besides, attributing characteristics to particular scenes is clearly difficult due to their nature. In this article, we will consider autoencoders application to unsupervised scene recognition and clusterization based on interpretable features. Further, we will focus on particular types of auto encoders that relevant to our study. We will take a look at the specificity of deep learning related to information theory and rate-distortion theory and describe the solutions empowering poor interpretability of deep learning in media content processing. As a conclusion, we will present the results of the work of custom framework, based on autoencoders, capable of scene recognition as was deeply studied above, with highlights generation resulted out of this recognition. We will not describe in detail the mathematical description of neural networks work but will clarify the necessary concepts and pay attention to important nuances.

Keywords: neural networks, computer vision, representation learning, autoencoders

Procedia PDF Downloads 127

Authors: Ali Berkan Ural

Abstract:

This paper presents the analysis, evaluation, and pre-diagnosis of early stage breast based diagnostic problems (breast cancer, nodulesorlumps) by Computer Aided Diagnosing (CAD) system from mammogram radiological images. According to the statistics, the time factor is crucial to discover the disease in the patient (especially in women) as possible as early and fast. In the study, a new algorithm is developed using advanced image processing and deep learning method to detect and classify the problem at earlystagewithmoreaccuracy. This system first works with image processing methods (Image acquisition, Noiseremoval, Region Growing Segmentation, Morphological Operations, Breast BorderExtraction, Advanced Segmentation, ObtainingRegion Of Interests (ROIs), etc.) and segments the area of interest of the breast and then analyzes these partly obtained area for cancer detection/lumps in order to diagnosis the disease. After segmentation, with using the Spectrogramimages, 5 different deep learning based methods (specified Convolutional Neural Network (CNN) basedAlexNet, ResNet50, VGG16, DenseNet, Xception) are applied to classify the breast based problems.

Keywords: computer aided diagnosis, breast cancer, region growing, segmentation, deep learning

Procedia PDF Downloads 95

Authors: Rui Mao, Heming Ji, Xiaoyu Wang

Abstract:

Keystroke dynamics are widely used in identity recognition. It has the advantage that the individual typing rhythm is difficult to imitate. It also supports continuous authentication through the keyboard without extra devices. The existing keystroke dynamics authentication methods based on machine learning have a drawback in supporting relatively complex scenarios with massive data. There are drawbacks to both feature extraction and model optimization in these methods. To overcome the above weakness, an authentication model of keystroke dynamics based on deep learning is proposed. The model uses feature vectors formed by keystroke content and keystroke time. It ensures efficient continuous authentication by cooperating attention mechanisms with the combination of CNN and Bi-LSTM. The model has been tested with Open Data Buffalo dataset, and the result shows that the FRR is 3.09%, FAR is 3.03%, and EER is 4.23%. This proves that the model is efficient and accurate on continuous authentication.

Keywords: keystroke dynamics, identity authentication, deep learning, CNN, LSTM

Procedia PDF Downloads 155