• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.1
• /
• pp.413-435
• /
• 2018

Although the accuracy of handwritten character recognition based on deep networks has been shown to be superior to that of the traditional method, the use of an overly deep network significantly increases time consumption during parameter training. For this reason, this paper took the training time and recognition accuracy into consideration and proposed a novel handwritten character recognition algorithm with newly designed network structure, which is based on an extended nonlinear kernel residual network. This network is a non-extremely deep network, and its main design is as follows:(1) Design of an unsupervised apriori algorithm for intra-class clustering, making the subsequent network training more pertinent; (2) presentation of an intermediate convolution model with a pre-processed width level of 2;(3) presentation of a composite residual structure that designs a multi-level quick link; and (4) addition of a Dropout layer after the parameter optimization. The algorithm shows superior results on MNIST and SVHN dataset, which are two character benchmark recognition datasets, and achieves better recognition accuracy and higher recognition efficiency than other deep structures with the same number of layers.

• Journal of The Korean Association of Information Education
• /
• v.23 no.6
• /
• pp.583-590
• /
• 2019

The purpose of this study is to develop an educational program for learning deep learning concepts for elementary school students. The model of education program was developed the deep-learning teaching method based on CT element-oriented teaching and learning model. The subject of the developed program is the artificial intelligence image recognition CNN algorithm, and we have developed 9 educational programs. We applied the program over two weeks to sixth graders. Expert validity analysis showed that the minimum CVR value was more than .56. The fitness level of learner level and the level of teacher guidance were less than .80, and the fitness of learning environment and media above .96 was high. The students' satisfaction analysis showed that students gave a positive evaluation of the average of 4.0 or higher on the understanding, benefit, interest, and learning materials of artificial intelligence learning.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.3
• /
• pp.237-246
• /
• 2019

In order to improve the accuracy of the rating prediction of the recommendation model, not only user-item rating data are used but also consider auxiliary information of item such as comments, tags, or descriptions. The traditional approaches use a word-level model of the bag-of-words for the auxiliary information. This model, however, cannot utilize the auxiliary information effectively, which leads to shallow understanding of auxiliary information. Convolution neural network (CNN) can capture and extract feature vector from auxiliary information effectively. Thus, this paper proposes character-level deep-Convolution Neural Network based matrix factorization (Char-DCNN-MF) that integrates deep CNN into matrix factorization for a novel recommendation model. Char-DCNN-MF can deeper understand auxiliary information and further enhance recommendation performance. Experiments are performed on three different real data sets, and the results show that Char-DCNN-MF performs significantly better than other comparative models.

• Economic and Environmental Geology
• /
• v.55 no.6
• /
• pp.737-760
• /
• 2022

For the geological disposal of high-level radioactive wastes (HLW), an understanding of deep subsurface environment is essential through geological, hydrogeological, geochemical, and geotechnical investigations. Although South Korea plans the geological disposal of HLW, only a few studies have been conducted for characterizing the geochemistry of deep subsurface environment. To guide the hydrogeochemical research for selecting suitable repository sites, this study overviewed the status and trends in hydrogeochemical characterization of deep groundwater for the deep geological disposal of HLW in developed countries. As a result of examining the selection process of geological disposal sites in 8 countries including USA, Canada, Finland, Sweden, France, Japan, Germany, and Switzerland, the following geochemical parameters were needed for the geochemical characterization of deep subsurface environment: major and minor elements and isotopes (e.g., ³⁴S and ¹⁸O of SO₄^2-, ¹³C and ¹⁴C of DIC, ²H and ¹⁸O of water) of both groundwater and pore water (in aquitard), fracture-filling minerals, organic materials, colloids, and oxidation-reduction indicators (e.g., Eh, Fe²⁺/Fe³⁺, H₂S/SO₄^2-, NH₄⁺/NO₃^-). A suitable repository was selected based on the integrated interpretation of these geochemical data from deep subsurface. In South Korea, hydrochemical types and evolutionary patterns of deep groundwater were identified using artificial neural networks (e.g., Self-Organizing Map), and the impact of shallow groundwater mixing was evaluated based on multivariate statistics (e.g., M3 modeling). The relationship between fracture-filling minerals and groundwater chemistry also has been investigated through a reaction-path modeling. However, these previous studies in South Korea had been conducted without some important geochemical data including isotopes, oxidationreduction indicators and DOC, mainly due to the lack of available data. Therefore, a detailed geochemical investigation is required over the country to collect these hydrochemical data to select a geological disposal site based on scientific evidence.

• Korean Journal of Materials Research
• /
• v.9 no.5
• /
• pp.460-466
• /
• 1999

The deep level electron traps in AP-MOCVD GaAs/undoped Al\ulcornerGa\ulcornerAs/n-type GaAs heterostructures have been investigated by means of Deep Level Transient Spectroscopy DLTS). In terms of the experimental procedure, GaAs/undoped Al\ulcornerGa\ulcornerAs/n-type GaAs heterostructures were deposited on 2" undoped semi-insulating GaAs wafers by the AP-MOCVD method at $650^{\circ}C$ with TMGa, AsH3, TMAl, and SiH4 gases. The n-type GaAs conduction layers were doped with Si to the target concentration of about 2$\times$10\ulcornercm\ulcorner. The Al content was targeted to x=0.5 and the thicknesses of Al\ulcornerGa\ulcornerAs layers were targeted from 0 to 40 nm. In order to investigate the electrical characteristics, an array of Schottky diodes was built on the heterostructures by the lift-off process and Al thermal evaporation. Among the key results of this experiment, the deep level electron traps at 0.742~0.777 eV and 0.359~0.680 eV were observed in the heterostructures; however, only a 0.787 eV level was detected in n-type GaAs samples without the Al\ulcornerGa\ulcornerAs overlayer. It may be concluded that the 0.787 eV level is an EL2 level and that the 0.742~0.777 eV levels are related to EL2 and residual oxygen impurities which are usually found in MOCVD GaAs and Al\ulcornerGa\ulcornerAs materials grown at $630~660^{\circ}C$. The 0.359~0.680 eV levels may be due to the defects related with the al-O complex and residual Si impurities which are also usually known to exist in the MOCVD materials. Particularly, as the Si doping concentration in the n-type GaAs layer increased, the electron trap concentrations in the heterostructure materials and the magnitude of the C-V hysteresis in the Schottky diodes also increased, indicating that all are intimately related.ated.

• Journal of Korea Water Resources Association
• /
• v.51 no.12
• /
• pp.1207-1216
• /
• 2018

Discharge or water level predictions at tidally affected river reaches are currently still a great challenge in hydrological practices. This research aims to predict water level of the tide dominated site, Jamsu bridge in the Han River downstream. Physics-based hydrodynamic approaches are sometimes not applicable for water level prediction in such a tidal river due to uncertainty sources like rainfall forecasting data. In this study, TensorFlow deep learning framework was used to build a deep neural network based LSTM model and its applications. The LSTM model was trained based on 3 data sets having 10-min temporal resolution: Paldang dam release, Jamsu bridge water level, predicted tidal level for 6 years (2011~2016) and then predict the water level time series given the six lead times: 1, 3, 6, 9, 12, 24 hours. The optimal hyper-parameters of LSTM model were set up as follows: 6 hidden layers number, 0.01 learning rate, 3000 iterations. In addition, we changed the key parameter of LSTM model, sequence length, ranging from 1 to 6 hours to test its affect to prediction results. The LSTM model with the 1 hr sequence length led to the best performing prediction results for the all cases. In particular, it resulted in very accurate prediction: RMSE (0.065 cm) and NSE (0.99) for the 1 hr lead time prediction case. However, as the lead time became longer, the RMSE increased from 0.08 m (1 hr lead time) to 0.28 m (24 hrs lead time) and the NSE decreased from 0.99 (1 hr lead time) to 0.74 (24 hrs lead time), respectively.

• The Korean Journal of Food And Nutrition
• /
• v.22 no.2
• /
• pp.261-271
• /
• 2009

Deep sea water exists at depths of over 200m under the sea. As no sunlight reaches it, photosynthesis does not take place within it, and it contains no organic matter. In addition, its temperature is maintained at a stable low level throughout the year, so it does not get mixed with the sea water on the surface. It contains a large amount of nutritious salts, whose cleanness is maintained. It is a marine resource that has matured for a long period of time. Research into deep sea water, which started in the 1970s, has been made around the whole world, including the USA and Japan. In Korea, research has been active in this area since 2000. As there has been a good amount of research into industrial applications for deep sea water, since 1993, patents for the relevant technologies have been applied. This paper intends to provide a resource to researchers of deep sea water, by summarizing of all domestic deep sea water-related patents applied with Korean Intellectual Property Office from 1993 to 2008. This research was conducted using a computer and KIPRIS Database owned by the Korea Institute of Patent Information. 'Deep sea water' was used as the search keyword. A total of 222 Korean patents relating to deep sea water have been registered on the basis of IPC. Of these, 126 patents relate to the manufacturing and the treatment of foods, foodstuffs, or non-alcoholic beverages(A23L), while 50 patents relate to the production for medical, dental, or cosmetic purposes(A61K). 38 patents relate to water purification, treatment of wastewater, sewage and sludge (C02F), while 8 patents relate to fishery and farming(A01K). In summary, it was found that studies for the practical use of deep sea water have been conducted in relation to the manufacturing and the treatment of foods, foodstuffs, beverages, and cosmetics.

• Journal of Korean Medicine for Obesity Research
• /
• v.8 no.2
• /
• pp.25-35
• /
• 2008

Objectives This study was performed to find out the characters about distribution of abdominal fat(especially superficial and deep subcutaneous fat) in obese premenopausal Korean women. Methods 39 obese premenopausal women were recruited in 2008. Anthropometry and body impedance analysis have been done and abdominal fat distribution had been assessed by computed tomography scan at the level of L4-5. Blood test and questionnaires about depression, eating attitude and physical activity were underwent. Result Abdominal total fat area, subcutaneous fat area including superficial and deep were significantly correlated with anthropometry and BIA result while visceral fat was correlated only with age and waist circumference. In blood profile, only visceral fat area was correlated with HDL cholesterol and triglyceride. And there were no correlation among questionnaires and abdominal fat. There were no difference between superficial and deep subcutaneous fat. Conclusion Abdominal subcutaneous fat including superficial and deep did not have any correlation with heart risk factor. superficial and deep subcutaneous fat had no differences with each other and they did not show any correlation with visceral fat in obese perimenopausal Korean women.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2019.11a
• /
• pp.180-181
• /
• 2019

This paper proposes a rate control algorithm for intra coding frame in HEVC encoder using a deep learning approach. The proposed algorithm is designed for CTU level bit allocation in intra frame by considering visual features spatially and temporally. Our features are generated using visual geometry group (VGG-16) with deep convolutional layers, then it is used for bit allocation per each CTU within an intra frame. According to our experiments, the proposed algorithm can achieve -2.04% Luma component BD-rate gain with minimal bit accuracy loss against the HM-16.20 rate control model.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2004.10a
• /
• pp.1013-1016
• /
• 2004

High level radioactive wastes, such as spent fuels generated from nuclear power plant, will be disposed in a deep geological repository. To maintain the integrity of the disposal canister and to carry out the process effectively, the emplacement process for the canister system in borehole of disposal tunnel should be well defined. In this study, the concept of the disposal canister emplacement process for deep geological disposal was established. To do this, the spent fuel arisings and disposal rate were reviewed. Also, not only design requirements, such canister and disposal depth but also preliminary repository layout concept were reviewed. Based on the requirements and the other bases, the canister emplacement process in the borehole of the disposal tunnel was established. The established concept of the disposal canister emplacement process will be improved continuously with the future studies. And this concept can be effectively used in implementing the reference repository system of our own case.