• The Journal of Information Technology
• /
• v.8 no.1
• /
• pp.119-124
• /
• 2005

Learning procedure in the neural network is updating of weights between neurons. Unadequate initial learning coefficient causes excessive iterations of learning process or incorrect learning results and degrades learning efficiency. In this paper, adaptive learning algorithm is proposed to increase the efficient in the learning algorithms of Kohonens Self-Organization Neural networks. The algorithm updates the weights adaptively when learning procedure runs. To prove the efficiency the algorithm is experimented to clustering of the random weight. The result shows improved learning rate about 42~55% ; less iteration counts with correct answer.

• Proceedings of the Safety Management and Science Conference
• /
• 2008.04a
• /
• pp.185-196
• /
• 2008

The objective of this study is to evaluate the mechanical behaviors and structural integrity of the weldment of high strength steel by using an acoustic emission (AE) techniques. Simple tension and AE tests were conducted against the 3 kind of welding test specimens. In order to analysis the effectiveness of weldability, joinability and structural integrity, we used K-means clustering method as a unsupervised learning pattern recognition algorithm for obtained multivariate AE main data sets, such as AE counts, energy, amplitude, hits, risetime, duration, counts to peak and rms signals. Through the experimental results, the effectiveness of the proposed method is discussed.

• Journal of the Korea Safety Management & Science
• /
• v.10 no.4
• /
• pp.127-134
• /
• 2008

The objective of this study is to evaluate the mechanical behaviors and structural integrity of the weldment of high strength steel by using an acoustic emission (AE) techniques. Monotonic simple tension and AE tests were conducted against the 3 kinds of welded specimen. In order to analysis the effectiveness of weldability, joinability and structural integrity, we used K-means clustering method as a unsupervised learning pattern recognition algorithm for obtained multi-variate AE main data sets, such as AE counts, energy, amplitude, hits, risetime, duration, counts to peak and rms signals. Through the experimental results, the effectiveness of the proposed method is discussed.

• International Journal of Advanced Culture Technology
• /
• v.9 no.1
• /
• pp.190-195
• /
• 2021

Currently, one of the fields leading the 4th industrial revolution is the image recognition field of artificial intelligence, which is showing good results in many fields. In this paper, using is a YOLO V2 model, which is one of the image recognition models, we intend to classify and select into three types according to the characteristics of fruits. To this end, it was designed to proceed the number of iterations of learning 9000 counts based on 640 mandarin image data of 3 classes. For model evaluation, normal, rotten, and unripe mandarin oranges were used based on images. We as a result of the experiment, the accuracy of the learning model was different depending on the number of learning. Normal mandarin oranges showed the highest at 60.5% in 9000 repetition learning, and unripe mandarin oranges also showed the highest at 61.8% in 9000 repetition learning. Lastly, rotten tangerines showed the highest accuracy at 86.0% in 7000 iterations. It will be very helpful if the results of this study are used for fruit farms in rural areas where labor is scarce.

• Journal of Physiology & Pathology in Korean Medicine
• /
• v.22 no.3
• /
• pp.624-629
• /
• 2008

Effects of Kyungohkgo Ga Nokyong(Yongohkgo) on growth development and learning ability were investigated growth and intellectual impairment rat with insufficient nutrition diet. We divided male Sprague-Dawley rats into 4 groups. They were Normal group, Growth deficiency rat with insufficient nutrition diet group, Growth deficiency rat with 0.1% Yongohkgo group and 0.2% Yongohkgo group. They were administered for 5 weeks. We measured body weight, and morris water maze test in escape distance, escape time and escape speed, serum growth hormone, insulin-like growth factor and thyroid stimulating hormone, RBC, concentration of Hb and PCV ratio, total WBC and its composition, the values of GOT and GPT activities. The results are as follows that Yongohkgo 0.1%, 0.2% groups were showed significantly different than control groups in body weight and the counts of RBC. In the morris water maze test, in escape distance and escape time, in concentration of Hb and PCV ratio, 0.2% Yongohkgo group were significantly different than control groups. Serum growth hormone, insulin- like growth factor and thyroid stimulating hormone showed a tendency to increase in Yongohkgo groups. The counts of total WBC and its composition, GOT, GPT activities showed no significantly different in all treatment groups. These results suggested that Yongohkgo have an effect of promoting growth and learning ability of rats and might be effect to treat various kinds of growth and learning ability delay in children.

• Transactions of the Korean Society of Machine Tool Engineers
• /
• v.18 no.2
• /
• pp.170-177
• /
• 2009

The objective of this study is to evaluate an acoustic emission (AE) source characterization and fracture behavior of the SM45C steel by using back-propagation neural network (BPN). In previous research Ref. [8] about k-nearest neighbor classifier (k-NNC) continuity, we used K-means clustering method as an unsupervised learning method for obtaining multi-variate AE main data sets, such as AE counts, energy, amplitude, risetime, duration and counts to peak. Similarly, we applied k-NNC and BPN as a supervised learning method for obtaining multi-variate AE working data sets. According to the error of convergence for determinant criterion Wilk's ${\lambda}$, heuristic criteria D&B(Rij) and Tou values are discussed. As a result, in k-NNC before fracture signal is detected or when fracture signal is detected, showed that produce some empty classes in BPN. And we confirmed that could save trouble in AE signal processing if suitable error of convergence or acceptable encoding error give to BPN.

• Nuclear Engineering and Technology
• /
• v.54 no.12
• /
• pp.4684-4692
• /
• 2022

Radionuclide identification is an important part of the nuclear material identification system. The development of artificial intelligence and machine learning has made nuclide identification rapid and automatic. However, many methods directly use existing deep learning models to analyze the gamma-ray spectrum, which lacks interpretability for researchers. This study proposes an explainable radionuclide identification algorithm based on the convolutional neural network and class activation mapping. This method shows the area of interest of the neural network on the gamma-ray spectrum by generating a class activation map. We analyzed the class activation map of the gamma-ray spectrum of different types, different gross counts, and different signal-to-noise ratios. The results show that the convolutional neural network attempted to learn the relationship between the input gamma-ray spectrum and the nuclide type, and could identify the nuclide based on the photoelectric peak and Compton edge. Furthermore, the results explain why the neural network could identify gamma-ray spectra with low counts and low signal-to-noise ratios. Thus, the findings improve researchers' confidence in the ability of neural networks to identify nuclides and promote the application of artificial intelligence methods in the field of nuclide identification.

• English Language & Literature Teaching
• /
• v.17 no.4
• /
• pp.229-250
• /
• 2011

Despite the social turn in literacy studies, few empirical studies have investigated the practical applications and learning experiences of multimodal composition pedagogy. Using a qualitative research approach, this study examines undergraduates' experiences in producing multimodal texts. Findings report that students' experiences in a multimodal composition class epitomize enjoyable learning. Students enjoyed their learning process because (a) the multimodal literacy curriculum filled the pedagogical gap between the conventional school-sponsored alphabetic literacy pedagogy and widespread out-of-school multimodal literacy practices and (b) the usefulness of the curriculum helped students enhance their intrinsic motivation to learn and compose. By questioning fundamental assumptions about what counts as knowledge in the current ecology of literacies, the authors argue for a dynamic view of literacy into practice.

• 대한공업교육학회지
• /
• v.30 no.1
• /
• pp.106-119
• /
• 2005

Many problems with the offline class, which is the traditional education type in corporations or universities, were indicated and people hoped that e-learning, which is web-based instruction, would solve these problems. However, e-learning also has weak points in that it should be self-paced and media-based in many ways. Therefore, when considering the good and weak points of offline classes and e-learning, blended learning seems to be necessary. Until now, blended learning has usually been used in corporations, and there have been almost no studies on the effectiveness or management of blended learning in universities. Thus, in this study, I would like to design blended classes, manage them at the level of university classes, and verify the effectiveness of blended classes, by comparing academic achievement, student participation, and student satisfaction. The subject students who signed up for Computer & Technology at C University in 2005 were divided into three study groups: offline class, online class, and blended class. The offline class was taught using the traditional class teaching method. For the online class and the blended class, multimedia contents were developed and a different LMS was used. The results of 13 weeks of teaching are as follows. For the academic achievement in the offline, online and blended classes, there was no statistically significant difference (f=2.387, p=.096). But when comparing the average achievement, the average of the blended class was higher than that of the other classes, so that it can be said that the blended class has positive effects on academic achievement. Second, when comparing the learners' participation in the online class and the blended class, the total posts were 85 and 138 respectively, which shows a considerable difference. The hit counts for each post in the online class and the blended class are 10 and 20, respectively. Moreover, the login counts for subjects are 3 in the online class and 4 in the blended class. In the questionnaire for the students' academic satisfaction in the online class and the blended class, all of the 15 items showed higher satisfaction in the blended class. Considering all these results, if adequate media are properly combined, the blended class is better than either the pure online class or the pure offline class.

• ETRI Journal
• /
• v.42 no.2
• /
• pp.217-229
• /
• 2020

Long-term electroencephalography (EEG) monitoring is time-consuming, and requires experts to interpret EEG signals to detect seizures in patients. In this paper, we propose a novel automated method called adaptive slope of wavelet coefficient counts over various thresholds (ASCOT) to classify patient episodes as seizure waveforms. ASCOT involves extracting the feature matrix by calculating the mean slope of wavelet coefficient counts over various thresholds in each frequency subband. We validated our method using our own database and a public database to avoid overtuning. The experimental results show that the proposed method achieved a reliable and promising accuracy in both our own database (98.93%) and the public database (99.78%). Finally, we evaluated the performance of the method considering various window sizes. In conclusion, the proposed method achieved a reliable seizure detection performance with a short-term window size. Therefore, our method can be utilized to interpret long-term EEG results and detect momentary seizure waveforms in diagnostic systems.