• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.7
• /
• pp.16-19
• /
• 2019

A thinner film has superior electrical properties and a better amorphous structure. Amorphous structures can be effective in improving conductivity through a depletion effect. Research is needed on the Schottky contact, where potential barriers are formed, as a way to identify these characteristics. $SiO_2/SnO_2$ thin films were prepared to examine the amorphous structure and Schottky contact, $SiO_2$ thin films were prepared using Ar = 20 sccm. $SnO_2$ thin films were deposited using mixed gas with a flow rate of argon and oxygen at 20 sccm, and $SnO_2$ thin films were added by magnetron sputtering and treated at $100^{\circ}C$ and $150^{\circ}C$. To identify the conditions under which the amorphous structure was constructed, the XRD patterns were investigated and C-V and I-V measurements were taken to make Al electrodes and perform electrical analysis. The depletion layer was formed by the recombination of electrons and holes through the heat treatment process. $SiO_2/SnO_2$ thin films confirmed that the pores were well formed when heat treated at $100^{\circ}C$ and an electric current was applied over the micro area. An amorphous $SiO_2/SnO_2$ thin film with heat treatment at $100^{\circ}C$ showed no reflection at $33^{\circ}\;2{\theta}$ in the XRD pattern, and a reflection at $44^{\circ}2\;{\theta}$. The macroscopic view (-30 V

• Journal of Korea Water Resources Association
• /
• v.52 no.1
• /
• pp.51-60
• /
• 2019

The major purpose of this study is to construct an in-situ soil moisture verification network employing Frequency Domain Reflectometry (FDR) sensors for Cosmic-ray soil moisture observation system operation as well as long-term field-scale soil moisture monitoring. The test bed of Cosmic-ray and FDR verification network system was established at the Sulma Catchment, in connection with the existing instrumentations for integrated data provision of various hydrologic variables. This test bed includes one Cosmic-ray Neutron Probe (CRNP) and ten FDR stations with four different measurement depths (10 cm, 20 cm, 30 cm, and 40 cm) at each station, and has been operating since July 2018. Furthermore, to assess the reliability of the in-situ verification network, the volumetric water content data measured by FDR sensors were compared to those calculated through the core sampling method. The evaluation results of FDR sensors- measured soil moisture against sampling method during the study period indicated a reasonable agreement, with average values of $bias=-0.03m^3/m^3$ and RMSE $0.03m^3/m^3$, revealing that this FDR network is adequate to provide long-term reliable field-scale soil moisture monitoring at Sulmacheon basin. In addition, soil moisture time series observed at all FDR stations during the study period generally respond well to the rainfall events; and at some locations, the characteristics of rainfall water intercepted by canopy were also identified. The Temporal Stability Analysis (TSA) was performed for all FDR stations located within the CRNP footprint at each measurement depth to determine the representative locations for field-average soil moisture at different soil profiles of the verification network. The TSA results showed that superior performances were obtained at FDR 5 for 10 cm depth, FDR 8 for 20 cm depth, FDR2 for 30 cm depth, and FDR1 for 40 cm depth, respectively; demonstrating that those aforementioned stations can be regarded as temporal stable locations to represent field mean soil moisture measurements at their corresponding measurement depths. Although the limit on study duration has been presented, the analysis results of this study can provide useful knowledge on soil moisture variability and stability at the test bed, as well as supporting the utilization of the Cosmic-ray observation system for long-term field-scale soil moisture monitoring.

• Journal of Intelligence and Information Systems
• /
• v.27 no.3
• /
• pp.175-197
• /
• 2021

Recently, with the development of deep learning technology, research on unstructured data analysis is being actively conducted, and it is showing remarkable results in various fields such as classification, summary, and generation. Among various text analysis fields, text classification is the most widely used technology in academia and industry. Text classification includes binary class classification with one label among two classes, multi-class classification with one label among several classes, and multi-label classification with multiple labels among several classes. In particular, multi-label classification requires a different training method from binary class classification and multi-class classification because of the characteristic of having multiple labels. In addition, since the number of labels to be predicted increases as the number of labels and classes increases, there is a limitation in that performance improvement is difficult due to an increase in prediction difficulty. To overcome these limitations, (i) compressing the initially given high-dimensional label space into a low-dimensional latent label space, (ii) after performing training to predict the compressed label, (iii) restoring the predicted label to the high-dimensional original label space, research on label embedding is being actively conducted. Typical label embedding techniques include Principal Label Space Transformation (PLST), Multi-Label Classification via Boolean Matrix Decomposition (MLC-BMaD), and Bayesian Multi-Label Compressed Sensing (BML-CS). However, since these techniques consider only the linear relationship between labels or compress the labels by random transformation, it is difficult to understand the non-linear relationship between labels, so there is a limitation in that it is not possible to create a latent label space sufficiently containing the information of the original label. Recently, there have been increasing attempts to improve performance by applying deep learning technology to label embedding. Label embedding using an autoencoder, a deep learning model that is effective for data compression and restoration, is representative. However, the traditional autoencoder-based label embedding has a limitation in that a large amount of information loss occurs when compressing a high-dimensional label space having a myriad of classes into a low-dimensional latent label space. This can be found in the gradient loss problem that occurs in the backpropagation process of learning. To solve this problem, skip connection was devised, and by adding the input of the layer to the output to prevent gradient loss during backpropagation, efficient learning is possible even when the layer is deep. Skip connection is mainly used for image feature extraction in convolutional neural networks, but studies using skip connection in autoencoder or label embedding process are still lacking. Therefore, in this study, we propose an autoencoder-based label embedding methodology in which skip connections are added to each of the encoder and decoder to form a low-dimensional latent label space that reflects the information of the high-dimensional label space well. In addition, the proposed methodology was applied to actual paper keywords to derive the high-dimensional keyword label space and the low-dimensional latent label space. Using this, we conducted an experiment to predict the compressed keyword vector existing in the latent label space from the paper abstract and to evaluate the multi-label classification by restoring the predicted keyword vector back to the original label space. As a result, the accuracy, precision, recall, and F1 score used as performance indicators showed far superior performance in multi-label classification based on the proposed methodology compared to traditional multi-label classification methods. This can be seen that the low-dimensional latent label space derived through the proposed methodology well reflected the information of the high-dimensional label space, which ultimately led to the improvement of the performance of the multi-label classification itself. In addition, the utility of the proposed methodology was identified by comparing the performance of the proposed methodology according to the domain characteristics and the number of dimensions of the latent label space.

• Journal of Intelligence and Information Systems
• /
• v.27 no.3
• /
• pp.157-173
• /
• 2021

With the development of information technology, the amount of available information increases daily. However, having access to so much information makes it difficult for users to easily find the information they seek. Users want a visualized system that reduces information retrieval and learning time, saving them from personally reading and judging all available information. As a result, recommendation systems are an increasingly important technologies that are essential to the business. Collaborative filtering is used in various fields with excellent performance because recommendations are made based on similar user interests and preferences. However, limitations do exist. Sparsity occurs when user-item preference information is insufficient, and is the main limitation of collaborative filtering. The evaluation value of the user item matrix may be distorted by the data depending on the popularity of the product, or there may be new users who have not yet evaluated the value. The lack of historical data to identify consumer preferences is referred to as data sparsity, and various methods have been studied to address these problems. However, most attempts to solve the sparsity problem are not optimal because they can only be applied when additional data such as users' personal information, social networks, or characteristics of items are included. Another problem is that real-world score data are mostly biased to high scores, resulting in severe imbalances. One cause of this imbalance distribution is the purchasing bias, in which only users with high product ratings purchase products, so those with low ratings are less likely to purchase products and thus do not leave negative product reviews. Due to these characteristics, unlike most users' actual preferences, reviews by users who purchase products are more likely to be positive. Therefore, the actual rating data is over-learned in many classes with high incidence due to its biased characteristics, distorting the market. Applying collaborative filtering to these imbalanced data leads to poor recommendation performance due to excessive learning of biased classes. Traditional oversampling techniques to address this problem are likely to cause overfitting because they repeat the same data, which acts as noise in learning, reducing recommendation performance. In addition, pre-processing methods for most existing data imbalance problems are designed and used for binary classes. Binary class imbalance techniques are difficult to apply to multi-class problems because they cannot model multi-class problems, such as objects at cross-class boundaries or objects overlapping multiple classes. To solve this problem, research has been conducted to convert and apply multi-class problems to binary class problems. However, simplification of multi-class problems can cause potential classification errors when combined with the results of classifiers learned from other sub-problems, resulting in loss of important information about relationships beyond the selected items. Therefore, it is necessary to develop more effective methods to address multi-class imbalance problems. We propose a collaborative filtering model using CGAN to generate realistic virtual data to populate the empty user-item matrix. Conditional vector y identify distributions for minority classes and generate data reflecting their characteristics. Collaborative filtering then maximizes the performance of the recommendation system via hyperparameter tuning. This process should improve the accuracy of the model by addressing the sparsity problem of collaborative filtering implementations while mitigating data imbalances arising from real data. Our model has superior recommendation performance over existing oversampling techniques and existing real-world data with data sparsity. SMOTE, Borderline SMOTE, SVM-SMOTE, ADASYN, and GAN were used as comparative models and we demonstrate the highest prediction accuracy on the RMSE and MAE evaluation scales. Through this study, oversampling based on deep learning will be able to further refine the performance of recommendation systems using actual data and be used to build business recommendation systems.

• Journal of Bio-Environment Control
• /
• v.28 no.4
• /
• pp.322-327
• /
• 2019

This study was conducted to improve the performance of automatic grafting robot using image recognition technique. The stem diameters of tomatoes and cucumber at the time of grafting were $2.5{\pm}0.3mm$ and $2.2{\pm}0.2mm$ for scions and $3.1{\pm}0.7mm$ and $3.6{\pm}0.3mm$ for rootstocks, respectively. The grafting failure was occurred when the different height between scions and rootstocks were over 4 mm and below 2 mm due to the small contact area of both cutting surface. Therefore, it was found that the height difference at the cutting surface of 3 mm is appropriate. This study also found that grafting failure was occurred when the stem diameters of both scions and rootstocks were thin. Therefore, it was suggested to use at least one stem with thicker than the average stem diameter. Field survey on the cutting angle of stems by hand were ranged from 13 to 55 degree for scions and 15 to 67 degree for rootstocks, respectively, which indicates that this could cause the grafting failure problem. However, the automatic grafting robot developed in this study rotates the seedlings 90 degree and then the stems are cut using a cutting blade. The control part of robot use all images taken from grafting process to determine the distance between a center of both ends of stem and a gripper center and then control the rotation angle of a gripper. Overall, this study found that The performance of automatic grafting robot using image recognition technique was superior with the grafting success rates of cucumber and tomato as $96{\pm}3.2%$ and $95{\pm}4%$, respectively.

• Journal of the Korean Wood Science and Technology
• /
• v.49 no.1
• /
• pp.23-43
• /
• 2021

In this study, we selected two essential oils, Citrus unshiu and Cinnamomum cassia with superior antioxidant effects from the essential oils of 7 wild plants in South Korea and examined their antimicrobial activity against Candida albicans, which causes dermatitis to identify the antimicrobial components in the essential oils. As a result of measuring DPPH (1,1-diphenyl-2-picrylhydrazyl) radical scavenging activity, SC50 value of the Citrus unshiu essential oil was 0.010 mg/mL, while for the Cinnamomum cassia essential oil, SC50 value was 0.09 mg/mL. In addition, when ABTS (2,2'-azinobis (3-ethylbenzothiazoline-6-sulfonic acid) radical scavenging activity was measured, SC50 value of the Citrus unshiu essential oil was 0.09 mg/mL, while for the Cinnamomum cassia essential oil, it was 0.06 mg/mL, exhibiting high antioxidant activity. For the minimum inhibitory concentration (MIC), the essential oil of Cinnamomum cassia was 1.25 mg/mL and that of Citrus unshiu was 5 mg/mL, demonstrating a high antimicrobial activity of the Cinnamomum cassia essential oil. Through the thin layer chromatography (TLC) bioassay, we assessed the antimicrobial activity against C. albicans according to the fraction components of the two essential oils. Also, by using preparative TLC (prep. TLC), we obtained the active fractions, and by performing GC/MS analysis of the components with the same Rf value, we identified the antimicrobial-active components. As a result, the main components having antioxidant and antimicrobial activities were cinnamyl acetate, eucalyptol, linalool, and citral of the Cinnamomum cassia essential oil and linalool from the Citrus unshiu essential oil. Also, based on the analysis of the fractional components that showed antioxidant and antimicrobial activities in both of the two essential oils, it was found that linalool has antioxidant activity, while cinnamyl acetate, eucalyptol, citral, and geranyl acetate have antioxidant and antimicrobial activities.

• Journal of Korea Water Resources Association
• /
• v.53 no.12
• /
• pp.1159-1172
• /
• 2020

This study applied deep convolution neural network based on U-Net and SegNet using long period weather radar data to very short-term rainfall prediction. And the results were compared and evaluated with the translation model. For training and validation of deep neural network, Mt. Gwanak and Mt. Gwangdeoksan radar data were collected from 2010 to 2016 and converted to a gray-scale image file in an HDF5 format with a 1km spatial resolution. The deep neural network model was trained to predict precipitation after 10 minutes by using the four consecutive radar image data, and the recursive method of repeating forecasts was applied to carry out lead time 60 minutes with the pretrained deep neural network model. To evaluate the performance of deep neural network prediction model, 24 rain cases in 2017 were forecast for rainfall up to 60 minutes in advance. As a result of evaluating the predicted performance by calculating the mean absolute error (MAE) and critical success index (CSI) at the threshold of 0.1, 1, and 5 mm/hr, the deep neural network model showed better performance in the case of rainfall threshold of 0.1, 1 mm/hr in terms of MAE, and showed better performance than the translation model for lead time 50 minutes in terms of CSI. In particular, although the deep neural network prediction model performed generally better than the translation model for weak rainfall of 5 mm/hr or less, the deep neural network prediction model had limitations in predicting distinct precipitation characteristics of high intensity as a result of the evaluation of threshold of 5 mm/hr. The longer lead time, the spatial smoothness increase with lead time thereby reducing the accuracy of rainfall prediction The translation model turned out to be superior in predicting the exceedance of higher intensity thresholds (> 5 mm/hr) because it preserves distinct precipitation characteristics, but the rainfall position tends to shift incorrectly. This study are expected to be helpful for the improvement of radar rainfall prediction model using deep neural networks in the future. In addition, the massive weather radar data established in this study will be provided through open repositories for future use in subsequent studies.

• Journal of Korean Society of Forest Science
• /
• v.100 no.4
• /
• pp.609-615
• /
• 2011

Forsythia saxatilis is a Korean endemic plant designated as rare and endangered by the Korea Forest Service (KFS). Growth and physiological characteristics of F. saxatilis were investigated under two different light intensities in order to figure out an appropriate growth environment for conservation and restoration of the species in its natural habitat. Shoot length, leaf size and weight, photosynthetic pigment content and photosynthetic parameters were measured for F. saxatilis grown at two experimental plots under relative light intensities (RLI) of 20% and 60% of the full sun, respectively. Fresh leaf weight of plants grown under high relative light intensities (RLI-60) exceeded that of plants grown at 20% RLI. The ratio of fresh leaf weight to leaf size at RLI-60 was 1.47 times superior comparing to that recorded at RLI-20. The content of photosynthetic pigments such as chlorophyll a, b and carotenoid were higher in plants grown at RLI-60, whereas the ratio of total chlorophyll to carotenoid content was higher in the leaves at RLI-20. Photosynthetic rate, stomatal conductance and transpiration rate at RLI-60 were, respectively, 2.5, 2.65 and 1.79 times higher comparing to those recorded at RLI-20. Water use efficiency, however, was higher at RLI-20. The chlorophyll/nitrogen ratio was 1.83 times higher at RLI-20 than at RLI-60. In contrast, the ratio of net photosynthesis to chlorophyll content at RLI-60 was 2.58 times higher than that of RLI-20. In conclusion, light intensity might be the major factor affecting growth and physiological characteristics of F. saxatilis grown under canopy of tall tree species.

• (The) Research of the performance art and culture
• /
• no.32
• /
• pp.465-488
• /
• 2016

This paper attempts to read in different ways and to interpret newly on Cheoyongrang mhang-hae-sa in "Sam-guk-yu-sa". Ulsan have held Cheoyong Cultural Festival for 47 times according to "Sam-guk-yu-sa". However, there have been a frequent identity crisis about Cheoyong Cultural Festival because of controversial issue about Cheoyong, This paper interpretate Cheoyongrang mhang-hae-sa newly to overcome these crisis, Cheoyong's dancing and retreating was not the resignation and tolerance, but the treat and warning, as the dance of Namsansin god of Posukjeong, Buk-acksin god of Keumkangryung and Jisin god of Dongryejeon was the warning of Silla's ruination. 'The Mhang' of Mhang-he-sa temple should be interpreted not as 'watch' but 'fifteen days'. Mhang-he means the roads buried in darkness and vanished had become a sea. The name of Shin-bhang-sa temple means Gae-un-po province of Ulsan had become 'the newly purified region' because of the inspection of King Heon-ghang. The main keyword of Cheoyongrang mhang-hae-sa is 'Byuk-sa-jin-gyung'. 'Byuk-sa-jin-gyung' means to repel the impious and pray the pleasure. The purpose of the personal Gut and national Gut, Narae, was also 'Byuk-sa-jin-gyung'. The reinvented bridal room with a fresh life was like the world of Byuk-sa-jin-gyung. The dance of God Sa-bhang was, as well the desperate desire to New Silla. Cheoyong was a shaman with a superior authority who set up the power to foresee to the god of smallpox. The image of Cheoyong at is not the resignation and tolerance, but the foresight and authority. Therefore, the slogan of Cheoyong Cultural Festival, the resignation and tolerance, should be reexamined. The new Cheoyong Cultural Festival should adopt the concept of foresight and authority and Byuk-sa-jin-gyung. Cheoyong Cultural Festival, have been held for 49times, often had identity problems. The identity of Cheoyong have been misinterpreted as the resignation and tolerance. The slogan of Cheoyong Cultural Festival should be reexamined. The new Cheoyong Cultural Festival should adopt the concept of foresight and authority and Byuk-sa-jin-gyung.

• The Journal of Korean Philosophical History
• /
• no.32
• /
• pp.309-336
• /
• 2011

There were various linking points that connect Li xue(Neo-Confucianism) to aesthetics in Song Dynasty as following. 1. The traditional moral as "pursuing pleasure of Kong-zi and Yan Hui" 2. Esteem of "life and vitality". Scholars of Li xue in Song regarded the pleasure of acting up to "benevolence" as a beauty, and this benevolence originated in the "heaven and earth; the universe". "Benevolence", that is to say, is name of the nature that continuous reproduction breed in an endless succession by "Yin-Yang the universe", thus the natural "life and vitality" of the "heaven and earth" as the matter of course is the perfect beauty. 3. An idea of "serene contemplation". Originally the "serene contemplation" belongs to discipline of "Li xue", however simultaneously this conception was entirely applicable to aesthetic point of view. 4. Cosmological consciousness. In the same manner, the "pleasure" which is moralistic and moreover aesthetic is indivisible from cosmic contemplation itself. Because of this point, the art and aesthetics of Song Dynasty self-consciously had the cosmological consciousness in its fullness. 5. Respect of beauty of nature. Scholars of "Li xue" considered as : no matter what "Li" or "Qi" that producing all things is "coming of itself", that is by no means artificially operated or prearranged in advance. Such standpoint was applied to creative art and made art of Song Dynasty esteem beauty of nature (coming of itself) exceedingly. 6. Laying stress on "disposition". Scholars of "Li xue" ordinarily valued much of "disposition of a sage", consequently this tendency influenced on aesthetics. "disposition" indicates the whole impression that one who has appearance and the inside(personality, temper, thought, etc.) gives to others. By putting that impression into practice of art and literature, it is to materialize the works of art as a unity of form and subject, also as an expression of human existence that breathed into one's sensibility on the whole. 7. Principles of "completing inquiry", "study the laws of nature by close access" of "Li xue". These principles made art and literature of Song Dynasty take a serious view of "Li" of all over the universe, so made them close investigate things, and after all have achieved very remarkable characteristic in art and literature, especially in paintings of Song Dynasty. Theory of painting in Song Dynasty had occupied considerably high position in Chinese aesthetic history. It was positively superior to former generations no matter what in quantity or in theoretical minuteness and its systematic level. Undoubtedly the Chinese theory of painting had been achieving development time after time since Song Dynasty. However if we could make a comparison it with every single period (ex. Yuan, Ming, and Qing Dynasties), there is no prominent period than Song Dynasty in theory of paintings. Song period had number of essays of Landscape painting.