• Proceedings of the Korea Water Resources Association Conference
• /
• 2007.05a
• /
• pp.58-66
• /
• 2007

This paper introduces the new generic dynamic neuro-fuzzy local modeling system (DNFLMS) that is based on a dynamic Takagi-Sugeno (TS) type fuzzy inference system for complex dynamic hydrological modeling tasks. The proposed DNFLMS applies a local generalization principle and an one-pass training procedure by using the evolving clustering method to create and update fuzzy local models dynamically and the extended Kalman filtering learning algorithm to optimize the parameters of the consequence part of fuzzy local models. The proposed DNFLMS is applied to develop the inference model to forecast the flow of Waikoropupu Springs, located in the Takaka Valley, South Island, New Zealand, and the influence of the operation of the 32 Megawatts Cobb hydropower station on springs flow. It is demonstrated that the proposed DNFLMS is superior in terms of model accuracy, model complexity, and computational efficiency when compared with a multi-layer perceptron trained with the back propagation learning algorithm and well-known adaptive neural-fuzzy inference system, both of which adopt global generalization.

• Journal of the Semiconductor & Display Technology
• /
• v.18 no.2
• /
• pp.1-5
• /
• 2019

As the number of IOT devices is growing rapidly, various 'see-thru connection' techniques have been reported for efficient communication with them. In this paper, we propose a deep learning based IOT device recognition system for interaction with these devices. The overall system consists of a TensorFlow based deep learning server and two Android apps for data collection and recognition purposes. As the basic neural network model, we adopted Google's inception-v3, and modified the output stage to classify 20 types of IOT devices. After creating a data set consisting of 1000 images of 20 categories, we trained our deep learning network using a transfer learning technology. As a result of the experiment, we achieve 94.5% top-1 accuracy and 98.1% top-2 accuracy.

• Journal of The Korean Association of Information Education
• /
• v.20 no.5
• /
• pp.475-486
• /
• 2016

As software transforms the structure of industry, it becomes a key measure in determining market competitiveness. Therefore, various educational efforts have been attempted in Korea to cultivate software professionals to secure software competitiveness. While previous studies had focused mainly on the cognitive effectiveness of software education, the authors tried to focus on affective perspectives. The authors, therefore, aimed to analyze the predictive power of the recognition of software importance and learning flow on affective outcomes, such as efficacy of computational thinking skills, and attitude toward, and satisfaction with, software education. The data were collected from 103 sixth grade students who participated in a software education. Results show that software importance and learning flow had significant predictive power on affective outcomes; Learning flow mediated the relationship between software importance and affective outcomes. This study provides practical implications for improving affective outcomes in the design and implementation of software education.

• Journal of the Korea Convergence Society
• /
• v.11 no.1
• /
• pp.349-357
• /
• 2020

This study is a descriptive study designed to identify the relationships among positive psychological capital, physical health status, depression, interpersonal relationship and learning flow. The subjects were 181 nursing students and the data collection was from May 8 to June 20, 2019. Data analysis methods were descriptive statistics, t-tests, ANOVA, Pearson's correlation coefficients, and stepwise multiple regression, using the SPSS 22.0 program. Positive psychological capital showed statistical differences according to age, grade, motive for major choice, major satisfaction and subjective health status. Positive psychological capital was correlated with depression(r=-.454, p<.001), interpersonal relationship(r=.611, p<.001) and learning flow(r=.452, p<.001). The factors affecting learning flow were positive psychological capital(β=.414, p<.001), major satisfaction(β=.177, p=.014), and grade(β=-.150, p=.026), which explained 24.4% of the variance. Therefore, it is necessary to develop and apply educational programs that can promote positive psychological capital in nursing students.

• Journal of The Korean Association of Information Education
• /
• v.25 no.3
• /
• pp.557-569
• /
• 2021

In preparation for the future society, the educational curriculum is changing according to the trend of the times, and with the advent of the era of the 4th Industrial Revolution, the purpose of the new 2015 revised curriculum was suggested to foster the convergence creativity of students. The purpose of software education is to promote creativity and further develop problem-solving skills in connection with real life. In addition, flow in learning leads to outstanding educational achievement. However, in elementary school computer education, there is still a lack of development of a convergence class model for students to easily immerse themselves and promote creative problem-solving skills. Therefore, in this study, we designed convergence computer education using Novel Engineering, which is a convergence class model suitable for these educational conditions and applied it to classes. Further, to measure the effect on the improvement of learning flow and creative problem-solving ability. the Novel Engineering-based computer class was applied to the experimental group for 6th graders, and the general computer class was applied to the control group. As a result of the pre-post test between groups, it was found that computer classes using Novel Engineering had a positive effect on learning flow and creative problem-solving ability.

• Nuclear Engineering and Technology
• /
• v.52 no.4
• /
• pp.784-790
• /
• 2020

The accelerated corrosion by Flow Accelerated Corrosion (FAC) has caused unexpected rupture of piping, hindering the safety of nuclear power plants (NPPs) and sometimes causing personal injury. For the safety, it may be necessary to select some pipes in terms of condition monitoring and to measure the change in thickness of pipes in real time. Direct current potential drop (DCPD) method has advantages in on-line monitoring of pipe wall thinning. However, it has a disadvantage in that it is difficult to quantify thinning due to various thinning shapes and thus there is a limitation in application. The machine learning approach has advantages in that it can be easily applied because the machine can learn the signals of various thinning shapes and can identify the thinning using these. In this paper, finite element analysis (FEA) was performed by applying direct current to a carbon steel pipe and measuring the potential drop. The fundamental machine learning was carried out and the piping thinning model was developed. In this process, the features of DCPD to thinning were proposed.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.8
• /
• pp.2948-2963
• /
• 2015

In the field of computer vision, visual surveillance systems have recently become an important research topic. Growth in this area is being driven by both the increase in the availability of inexpensive computing devices and image sensors as well as the general inefficiency of manual surveillance and monitoring. In particular, the ultimate goal for many visual surveillance systems is to provide automatic activity recognition for events at a given site. A higher level of understanding of these activities requires certain lower-level computer vision tasks to be performed. So in this paper, we propose an intelligent activity recognition model that uses a structure learning method and a classification method. The structure learning method is provided as a K2-learning algorithm that generates Bayesian networks of causal relationships between sensors for a given activity. The statistical characteristics of the sensor values and the topological characteristics of the generated graphs are learned for each activity, and then a neural network is designed to classify the current activity according to the features extracted from the multiple sensor values that have been collected. Finally, the proposed method is implemented and tested by using PETS2013 benchmark data.

• International Journal of Contents
• /
• v.13 no.4
• /
• pp.16-22
• /
• 2017

The purpose of this study was to investigate the daily life experiences of medical students and to explore gender differences in these experiences using the Experience Sampling Method (ESM) as the method. The instrument, the Experience Sampling Form (ESF), consisted of questions on the external and internal experiences of the respondents. Data were collected from 2,035 ESFs by 91 students (male=52, female=39) at three medical schools for one week. The data was analyzed using the statistical tests of the t-test and ${\chi}^2$ test. Activity places were significantly different by gender (${\chi}^2=16.576$, p=.001). Males spent more time in learning places such as schools, libraries, etc., whereas females spent their time in personal places, including their homes, dormitories, etc. Males undertook more learning activities than did females, and females undertook more social/leisure activities and basic life activities than did male students (${\chi}^2=18.753$, p=.001). They were in a learning place and performing learning activities. There were significant perceptual differences between males and females about their flow levels, competency levels, and difficulty levels, based on the activity type. These results can help us to understand the daily lives of medical students and can be useful in developing counseling programs and educational activities for students.

• Journal of the Semiconductor & Display Technology
• /
• v.18 no.2
• /
• pp.48-52
• /
• 2019

Virtual metrology, which is one of APC techniques, is a method to predict characteristics of manufactured films using machine learning with saving time and resources. As the photoresist is no longer a mask material for use in high aspect ratios as the CD is reduced, hard mask is introduced to solve such problems. Among many types of hard mask materials, amorphous carbon layer(ACL) is widely investigated due to its advantages of high etch selectivity than conventional photoresist, high optical transmittance, easy deposition process, and removability by oxygen plasma. In this study, VM using different machine learning algorithms is applied to predict the thickness of ACL and trained models are evaluated which model shows best prediction performance. ACL specimens are deposited by plasma enhanced chemical vapor deposition(PECVD) with four different process parameters(Pressure, RF power, $C_3H_6$ gas flow, $N_2$ gas flow). Gradient boosting regression(GBR) algorithm, random forest regression(RFR) algorithm, and neural network(NN) are selected for modeling. The model using gradient boosting algorithm shows most proper performance with higher R-squared value. A model for predicting the thickness of the ACL film within the abovementioned conditions has been successfully constructed.

• Structural Engineering and Mechanics
• /
• v.86 no.2
• /
• pp.181-195
• /
• 2023

To predict the rheological behaviours along with the compressive strength of self-compacting concrete that incorporates environmentally friendly ingredients as cement substitutes, a comparative evaluation of machine learning methods is conducted. To model four parameters, slump flow diameter, L-box ratio, V-funnel time, as well as compressive strength at 28 days-a complete mix design dataset from available pieces of literature is gathered and used to construct the suggested machine learning standards, SVM, MARS, and Mp5-MT. Six input variables-the amount of binder, the percentage of SCMs, the proportion of water to the binder, the amount of fine and coarse aggregates, and the amount of superplasticizer are grouped in a particular pattern. For optimizing the hyper-parameters of the MARS model with the lowest possible prediction error, a gravitational search algorithm (GSA) is required. In terms of the correlation coefficient for modelling slump flow diameter, L-box ratio, V-funnel duration, and compressive strength, the prediction results showed that MARS combined with GSA could improve the accuracy of the solo MARS model with 1.35%, 11.1%, 2.3%, as well as 1.07%. By contrast, Mp5-MT often demonstrates greater identification capability and more accurate prediction in comparison to MARS-GSA, and it may be regarded as an efficient approach to forecasting the rheological behaviors and compressive strength of SCC in infrastructure practice.