• Journal of Engineering Education Research
• /
• v.15 no.4
• /
• pp.109-117
• /
• 2012

The purpose of this study is to give the preliminary data and suggestion for introducing and spreading e-learning engineering education through analyzing learning behaviors, learning motivations, and satisfaction of e-learning engineering students. Especially, this comparative study analyzes each research domain according to majors and grades, thereby suggesting more specific and practical results. 2,745 students registered in 38 subjects of e-learning in 2 Universities were analyzed for this study. The study result shows that engineering students are attending around 2 e-learning subjects with a duration of about 30 minutes once a week. The main of learning motivation for e-learning was not easy test level and feasibility of acquiring credit but advantages of e-learning such as freedom of time and space, learning by repetition. The satisfaction scores of e-learning were lower compared to the aspects of system and contents Based on these results, first, an active spread of e-learning to engineering education is necessary because the demand from the engineering students is high enough and they have desirable learning behavior and learning motivation for it. Second, the characteristics of grades need to be taken into consideration on operation of e-learning. Third, a successful e-learning process needs more meticulous and active operation.

• ETRI Journal
• /
• v.15 no.2
• /
• pp.35-51
• /
• 1993

This paper presents function approximation based on nonparametric estimation. As an estimation model of function approximation, a three layered network composed of input, hidden and output layers is considered. The input and output layers have linear activation units while the hidden layer has nonlinear activation units or kernel functions which have the characteristics of bounds and locality. Using this type of network, a many-to-one function is synthesized over the domain of the input space by a number of kernel functions. In this network, we have to estimate the necessary number of kernel functions as well as the parameters associated with kernel functions. For this purpose, a new method of parameter estimation in which linear learning rule is applied between hidden and output layers while nonlinear (piecewise-linear) learning rule is applied between input and hidden layers, is considered. The linear learning rule updates the output weights between hidden and output layers based on the Linear Minimization of Mean Square Error (LMMSE) sense in the space of kernel functions while the nonlinear learning rule updates the parameters of kernel functions based on the gradient of the actual output of network with respect to the parameters (especially, the shape) of kernel functions. This approach of parameter adaptation provides near optimal values of the parameters associated with kernel functions in the sense of minimizing mean square error. As a result, the suggested nonparametric estimation provides an efficient way of function approximation from the view point of the number of kernel functions as well as learning speed.

• The Transactions of the Korea Information Processing Society
• /
• v.5 no.11
• /
• pp.2897-2907
• /
• 1998

The production-rule generation from training examples is a hard problem that has large space and many local optimal solutions. Many learning methods are proposed for production-rule generation and genetic algorithms is an alternative learning method. However, traditional genetic algorithms has been known to have an obstacle in converging at the global solution area and show poor efficiency of production-rules generated. In this paper, we propose a production-rule generating method which uses genetic algorithm learning. By analyzing optimal sub-solutions captured by genetic algorithm learning, our method takes advantage of its schema structure and thus generates relatively small rule set.

• Journal of Korean Elementary Science Education
• /
• v.32 no.2
• /
• pp.139-158
• /
• 2013

This study investigated elementary students' (grade 4~6) learning progressions for water cycling drawn from iterative assessments using ordered multiple-choice (OMC) items. An assessment system, which consisted of construct map, item design, outcome space, and measurement model, was employed in this study to examine children's learning progressions. At the first stage of the assessment system, a construct map was designed on which children's conceptual understandings from naive to most sophisticated were represented. At the item design stage, 8 OMC items were drawn from the construct map. Each item option of the OMC items was scored from 0 to 3 according to its level of understanding at the stage of outcome space. As a measurement model, Rasch model, a branch of item response theory, was applied to interpreting the outcomes of the OMC items. This cycle of assessment system was furtherly implemented iteratively in order to elaborate on the first version of water cycling learning progression. In conclusion, children's understanding of water cycling could be described in two aspects: water distribution and water movement. We identified children's conjectural developmental pathways about water cycling existed from superficial and naive accounts to more complex and abstract accounts.

• Journal of Korea Society of Industrial Information Systems
• /
• v.12 no.5
• /
• pp.71-85
• /
• 2007

In recent years, the need of the education for the Internet ethics is emphasized as the number of offenses committed by juveniles is increasing. So many researches on the methodologies about educating the Internet ethics have been being performed But, those are not competent to make young people inspire the sense of ethics in the cyber-space because they get trained by the cramming system of education with a textbook in their schools. In this paper, we developed a web-based learning model and an learning system including various multimedia materials that induce the learners for ethics to study actively. And then we applied our system to school education for the Internet ethics, and verified the effectiveness of our learning model and system.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.49 no.10
• /
• pp.883-892
• /
• 2021

This paper provides a process for developing deep learning-based aerial object detection models that can run in realtime on onboard. To improve object detection performance, we pre-process and augment the training data in the training stage. In addition, we perform transfer learning and apply a weighted cross-entropy method to reduce the variations of detection performance for each class. To improve the inference speed, we have generated inference acceleration engines with quantization. Then, we analyze the real-time performance and detection performance on custom aerial image dataset to verify generalization.

• Journal of Korea Multimedia Society
• /
• v.24 no.2
• /
• pp.178-185
• /
• 2021

Three-dimensional high-resolution magnetic resonance imaging (MRI) provides fine-level anatomical information for disease diagnosis. However, there is a limitation in obtaining high resolution due to the long scan time for wide spatial coverage. Therefore, in order to obtain a clear high-resolution(HR) image in a wide spatial coverage, a super-resolution technology that converts a low-resolution(LR) MRI image into a high-resolution is required. In this paper, we propose a super-resolution technique through filter learning based on information on the surrounding gradient information in 3D space from 3D MRI images. In the learning step, the gradient features of each voxel are computed through eigen-decomposition from 3D patch. Based on these features, we get the learned filters that minimize the difference of intensity between pairs of LR and HR images for similar features. In test step, the gradient feature of the patch is obtained for each voxel, and the filter is applied by selecting a filter corresponding to the feature closest to it. As a result of learning 100 T1 brain MRI images of HCP which is publicly opened, we showed that the performance improved by up to about 11% compared to the traditional interpolation method.

• International Journal of Computer Science & Network Security
• /
• v.23 no.2
• /
• pp.193-198
• /
• 2023

The article focuses on the areas of education activities in using techniques for teaching and learning with information and communication technologies (ICTs), researching and analyzing the available ICTs, gearing the technologies to the specific psychological and pedagogical conditions, independently building and modeling ICTs, enlarging and developing their use in the learning environment. The visualization of scientific research has been determined to be part of the educational support for building students' ICT competence during teaching and learning and is essential to the methodology culture. There have been specified main tasks for pedagogy technologies (PTs) to develop the skills of adaptability to the global digital space in students, their effective database operation and using the data bases as necessary elements for learning and as part of professional training for research. We provided rationalization for implementing the latest ICTs into the Ukrainian universities' curricula, as well as creating modern methods for using the technologies in the learning / teaching process and scientific activities.

• The KIPS Transactions:PartB
• /
• v.10B no.6
• /
• pp.587-592
• /
• 2003

Many real world control problems have continuous states and actions. When the state space is continuous, the reinforcement learning problems involve very large state space and suffer from memory and time for learning all individual state-action values. These problems need function approximators that reason action about new state from previously experienced states. We introduce Fuzzy Q-Map that is a function approximators for 1 - step Q-learning and is based on fuzzy clustering. Fuzzy Q-Map groups similar states and chooses an action and refers Q value according to membership degree. The centroid and Q value of winner cluster is updated using membership degree and TD(Temporal Difference) error. We applied Fuzzy Q-Map to the mountain car problem and acquired accelerated learning speed.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.4
• /
• pp.406-411
• /
• 2005

In this paper, we present the area-based Q-learning to search a target object using multiple robot. To search the target in Markovian space, the robots should recognize their surrounding at where they are located and generate some rules to act upon by themselves. Under area-based Q-learning, a robot, first of all, obtains 6-distances from itself to environment by infrared sensor which are hexagonally allocated around itself. Second, it calculates 6-areas with those distances then take an action, i.e., turn and move toward where the widest space will be guaranteed. After the action is taken, the value of Q will be updated by relative formula at the state. We set up an experimental environment with five small mobile robots, obstacles, and a target object, and tried to search for a target object while navigating in a unknown hallway where some obstacles were placed. In the end of this paper, we presents the results of three algorithms - a random search, area-based action making (ABAM), and hexagonal area-based Q-teaming.