• Journal of Korean Home Economics Education Association
• /
• v.30 no.4
• /
• pp.165-185
• /
• 2018

The purpose of the study was to show the development process of the total instructional alignment based on the 2015 revised technology home economics curriculum, and to present the detailed aspects of this task. The results of this study were as follows: First, the clarity phase of the performance established the annual evaluation plan to ensure the consistency of the achievement standard, the learning objectives and the assessment objectives through an analysis of the curriculum. Second, the development phase of the performance tasks and the grading criteria analyzed the contents area and determined the performance assessment task. The performance assessment consisted of a portfolio and project activity types, and also comprised of knowledge markets, as well as e-NIE tasks, including individual tasks and collective tasks. Third, the performance assessment and the results phase were conducted after planning the class operation according to the instructors' directions and arrangement of the schedule into block classes. Afterwards, the performance assessment tasks were revised by the teaching-learning community and class reflection on practical application. As these results demonstrate, developing performance tasks that are centered on the learner's development processes can contribute to teachers' expertise and improvement of instruction.

• KIPS Transactions on Computer and Communication Systems
• /
• v.10 no.3
• /
• pp.71-80
• /
• 2021

The selection of an appropriate neural network algorithm is an important step for accurate data prediction in machine learning. Many algorithms based on basic artificial neural networks have been devised to efficiently predict future data. These networks include deep neural networks (DNNs), recurrent neural networks (RNNs), long short-term memory (LSTM) networks, and gated recurrent unit (GRU) neural networks. Developers face difficulties when choosing among these networks because sufficient information on their performance is unavailable. To alleviate this difficulty, we evaluated the performance of each algorithm by comparing their errors and processing times. Each neural network model was trained using a tax dataset, and the trained model was used for data prediction to compare accuracies among the various algorithms. Furthermore, the effects of activation functions and various optimizers on the performance of the models were analyzed The experimental results show that the GRU and LSTM algorithms yields the lowest prediction error with an average RMSE of 0.12 and an average R2 score of 0.78 and 0.75 respectively, and the basic DNN model achieves the lowest processing time but highest average RMSE of 0.163. Furthermore, the Adam optimizer yields the best performance (with DNN, GRU, and LSTM) in terms of error and the worst performance in terms of processing time. The findings of this study are thus expected to be useful for scientists and developers.

• Journal of Information Science Theory and Practice
• /
• v.7 no.3
• /
• pp.6-20
• /
• 2019

Learning analytics, or educational data mining, is an emerging field that applies data mining methods and tools for the exploitation of data coming from educational environments. Learning management systems, like Moodle, offer large amounts of data concerning students' activity, performance, behavior, and interaction with their peers and their tutors. The analysis of these data can be elaborated to make decisions that will assist stakeholders (students, faculty, and administration) to elevate the learning process in higher education. In this work, the power of Excel is exploited to analyze data in Moodle, utilizing an e-learning course developed for enhancing the information computer technology skills of school teachers in primary and secondary education in Greece. Moodle log files are appropriately manipulated in order to trace daily and weekly activity of the learners concerning distribution of access to resources, forum participation, and quizzes and assignments submission. Learners' activity was visualized for every hour of the day and for every day of the week. The visualization of access to every activity or resource during the course is also obtained. In this fashion teachers can schedule online synchronous lectures or discussions more effectively in order to maximize the learners' participation. Results depict the interest of learners for each structural component, their dedication to the course, their participation in the fora, and how it affects the submission of quizzes and assignments. Instructional designers may take advice and redesign the course according to the popularity of the educational material and learners' dedication. Moreover, the final grade of the learners is predicted according to their previous grades using multiple linear regression and sensitivity analysis. These outcomes can be suitably exploited in order for instructors to improve the design of their courses, faculty to alter their educational methodology, and administration to make decisions that will improve the educational services provided.

• Journal of the Korea Computer Graphics Society
• /
• v.27 no.5
• /
• pp.45-54
• /
• 2021

Recently, deep learning-based denoising approaches have been actively studied. In particular, with the advances of blind denoising techniques, it become possible to train a deep learning-based denoising model only with noisy images in an image domain where it is impossible to obtain a clean image. We no longer require pairs of a clean image and a noisy image to obtain a restored clean image from the observation. However, it is difficult to recover the target using a deep learning-based denoising model trained by only noisy images if the distribution of the noisy image is far from the distribution of the clean image. To address this limitation, unpaired image denoising approaches have recently been studied that can learn the denoising model from unpaired data of the noisy image and the clean image. ISCL showed comparable performance close to that of supervised learning-based models based on pairs of clean and noisy images. In this study, we propose suitable normalization techniques for each purpose of architectures (e.g., generator, discriminator, and extractor) of ISCL. We demonstrate that the proposed method outperforms state-of-the-art unpaired image denoising approaches including ISCL.

• The Journal of the Acoustical Society of Korea
• /
• v.21 no.4E
• /
• pp.146-155
• /
• 2002

We present a new technique for achieving source separation when given only a single charmel recording. The main idea is based on exploiting the inherent time structure of sound sources by learning a priori sets of time-domain basis functions that encode the sources in a statistically efficient manner. We derive a learning algorithm using a maximum likelihood approach given the observed single charmel data and sets of basis functions. For each time point we infer the source parameters and their contribution factors. This inference is possible due to the prior knowledge of the basis functions and the associated coefficient densities. A flexible model for density estimation allows accurate modeling of the observation, and our experimental results exhibit a high level of separation performance for simulated mixtures as well as real environment recordings employing mixtures of two different sources. We show separation results of two music signals as well as the separation of two voice signals.

• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10b
• /
• pp.61-64
• /
• 1992

In this paper, a visual pattern recognition system is proposed, which can recognize both a pattern and its location. This system, referred to as the expanded neocognitron, has the following capabilities: (1) A higher performance in extraction of features, and (2) A new capability for recognizing the locations of patterns. This system adopts the learning and recognizing mechanism of the neocognitron. First, the ability to classify pattern is enhanced by improving the mechanisms of feature extraction and learning algorithm. Second, the function of detecting the location of each pattern is realized by developing an architecture which does not reduce structure, i.e., the unit density is constant all the way from the input stage to the output stage.

• Annual Conference of KIPS
• /
• 2023.05a
• /
• pp.539-541
• /
• 2023

COVID-19 is a viral pandemic disease that spreads widely all around the world. The only way to identify COVID-19 patients at an early stage is to stop the spread of the virus. Different approaches are used to diagnose, such as RT-PCR, Chest X-rays, and CT images. However, these are time-consuming and require a specialized lab. Therefore, there is a need to develop a time-efficient diagnosis method to detect COVID-19 patients. The proposed machine learning (ML) approach predicts the presence of coronavirus based on clinical symptoms. The clinical dataset is collected from the Israeli Ministry of Health. We used different ML classifiers (i.e., XGB, DT, RF, and NB) to diagnose COVID-19. Later, classifiers are optimized with the Bayesian hyperparameter optimization approach to improve the performance. The optimized RF outperformed the others and achieved an accuracy of 97.62% on the testing data that help the early diagnosis of COVID-19 patients.

• Structural Engineering and Mechanics
• /
• v.92 no.4
• /
• pp.341-348
• /
• 2024

This paper revisits our recent work on rapid and accurate design of reinforced concrete (R/C) columns and bridge piers using Artificial Neural Networks (ANNs). Both rectangular and circular, solid and hollow sections are treated. The new functions for rectangular sections now accommodate a much greater aspect ratio, making them suitable for all sections typically used for bridge piers, without sacrificing performance. For the first time, to the best of our knowledge, new design functions for T-beams and singly-reinforced rectangular beams are also derived. The error estimation is presented in detail using extremely extensive test sets, while auxiliary ANNs are employed to screen out improper data input. All design functions are sufficiently accurate, unconditionally stable, and orders of magnitude faster than any iterative section analysis procedure. The forward feed of the final ANNs has been translated into optimized code in all popular programming languages, which can be easily used without the need of specialized software, even on a spreadsheet.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.3
• /
• pp.133-140
• /
• 2022

We investigate the performance of deep learning-based Korean language models on a task of predicting the score range of Korean essays written by foreign students. We construct a data set containing a total of 304 essays, which include essays discussing the criteria for choosing a job ('job'), conditions of a happy life ('happ'), relationship between money and happiness ('econ'), and definition of success ('succ'). These essays were labeled according to four letter grades (A, B, C, and D), and a total of eleven essay score range prediction experiments were conducted (i.e., five for predicting the score range of 'job' essays, five for predicting the score range of 'happiness' essays, and one for predicting the score range of mixed topic essays). Three deep learning-based Korean language models, KoBERT, KcBERT, and KR-BERT, were fine-tuned using various training data. Moreover, two traditional probabilistic machine learning classifiers, naive Bayes and logistic regression, were also evaluated. Experiment results show that deep learning-based Korean language models performed better than the two traditional classifiers, with KR-BERT performing the best with 55.83% overall average prediction accuracy. A close second was KcBERT (55.77%) followed by KoBERT (54.91%). The performances of naive Bayes and logistic regression classifiers were 52.52% and 50.28% respectively. Due to the scarcity of training data and the imbalance in class distribution, the overall prediction performance was not high for all classifiers. Moreover, the classifiers' vocabulary did not explicitly capture the error features that were helpful in correctly grading the Korean essay. By overcoming these two limitations, we expect the score range prediction performance to improve.

• Journal of Institute of Control, Robotics and Systems
• /
• v.7 no.6
• /
• pp.509-517
• /
• 2001

To deal with non-linearities and time-varying characteristics of hydraulic systems, in this paper, the neuro-fuzzy controller has been introduced. This controller does not require and accurate mathematical model for the nonlinear factor. In order to solve general fuzzy inference problems, the input membership function and fuzzy reasoning rules are used for determining the controller parameters. These parameters are determined by using the learning algorithm. The control performance of the neuro-fuzzy controller is evaluated through a series of experiments for the various types of inputs while applying disturbances to the hydraulic system. The performance of this controller was compared with those of PID and PD controllers. From these results, We observe be said that the position tracking performance of neuro-fuzzy is better those of PID and PD controllers.