• Smart Media Journal
• /
• v.8 no.2
• /
• pp.58-65
• /
• 2019

Automatic text summarization is a process of shortening a text document by either extraction or abstraction. The abstraction approach inspired by deep learning methods scaling to a large amount of document is applied in recent work. Abstractive text summarization involves utilizing pre-generated word embedding information. Low-frequent but salient words such as terminologies are seldom included to dictionaries, that are so called, out-of-vocabulary(OOV) problems. OOV deteriorates the performance of Encoder-Decoder model in neural network. In order to address OOV words in abstractive text summarization, we propose a copy mechanism to facilitate copying new words in the target document and generating summary sentences. Different from the previous studies, the proposed approach combines accurate pointing information and selective copy mechanism based on bidirectional RNN and bidirectional LSTM. In addition, neural network gate model to estimate the generation probability and the loss function to optimize the entire abstraction model has been applied. The dataset has been constructed from the collection of abstractions and titles of journal articles. Experimental results demonstrate that both ROUGE-1 (based on word recall) and ROUGE-L (employed longest common subsequence) of the proposed Encoding-Decoding model have been improved to 47.01 and 29.55, respectively.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.36 no.6
• /
• pp.589-599
• /
• 2018

The land cover map has been produced by using satellite and aerial images. However, these two images have the limitations in spatial resolution, and it is difficult to acquire images of a area at desired time because of the influence of clouds. In addition, it is costly and time-consuming that mapping land cover map of a small area used by satellite and aerial images. This study used multispectral camera-based drone to acquire multi-temporal images for orthoimages generation. The efficiency of produced land cover map was evaluated using time series analysis. The results indicated that the proposed method can generated RGB orthoimage and multispectral orthoimage with RMSE (Root Mean Square Error) of ${\pm}10mm$, ${\pm}11mm$, ${\pm}26mm$ and ${\pm}28mm$, ${\pm}27mm$, ${\pm}47mm$ on X, Y, H respectively. The accuracy of the pixel-based and object-based land cover map was analyzed and the results showed that the accuracy and Kappa coefficient of object-based classification were higher than that of pixel-based classification, which were 93.75%, 92.42% on July, 92.50%, 91.20% on October, 92.92%, 91.77% on February, respectively. Moreover, the proposed method can accurately capture the quantitative area change of the object. In summary, the suggest study demonstrated the possibility and efficiency of using multispectral camera-based drone in production of land cover map.

• Journal of Physiology & Pathology in Korean Medicine
• /
• v.34 no.6
• /
• pp.326-333
• /
• 2020

Nitrate-nitrite-nitric oxide (NO) pathway is a major alternative source of NO and is essential for NO - dependent physiological functions in body. Food supplements having nitrate/nitrite can improve metabolic syndromes including hypertension through antioxidant activity or vasodilation. The purpose of this study was to observe the effects of fermented garlic (F. garlic) having high concentration of NO2- on changes in blood flow and nitric oxide synthesis in the cerebral cortex of rodents. The generation of nitric oxide detected by a chemi-luminescence detector was higher in F. Garlic compared with NaNO2 solution under artificial gastric juice with pH 2.0. Ether F. garlic or NaNO2 diluted with artificial cerebrospinal fluid was directly applied into around the needle probe of laser Doppler flow meter that was located on epidural surface of the cortex. Direct application of F. garlic resulted in increase of cerebral blood flow detected by a laser Doppler flow meter with a dose-dependent manner. Compared with NaNO2 solution, F. garlic produced changes in cerebral blood flow at lower concentration of NO2-. Pretreatment of methylene blue, a guanylyl cyclase inhibitor prevented upregulation of cerebral blood flow by the treatment of F. garlic. In addition, the application of F. garlic with 250, 500ppm of NO2- caused significantly the production of NO in the cortical tissue but NaNO2 solution with 500ppm of NO2- did not. In summary, these results suggested that F. garlic with high content of NO2- induce increase in cerebral blood flow through nitric oxide-dependent signal pathway.

• Journal of the Korean Society for Library and Information Science
• /
• v.56 no.3
• /
• pp.241-264
• /
• 2022

Bibliographic metadata can help researchers effectively utilize essential publications that they need and grasp academic trends of their own fields. With the manual creation of the metadata costly and time-consuming. it is nontrivial to effectively automatize the metadata construction using rule-based methods due to the immoderate variety of the article forms and styles according to publishers and academic societies. Therefore, this study proposes a two-step extraction process based on rules and deep neural networks for generating bibliographic metadata of scientific articlles to overcome the difficulties above. The extraction target areas in articles were identified by using a deep neural network-based model, and then the details in the areas were analyzed and sub-divided into relevant metadata elements. IThe proposed model also includes a model for generating reference summary information, which is able to separate the end of the text and the starting point of a reference, and to extract individual references by essential rule set, and to identify all the bibliographic items in each reference by a deep neural network. In addition, in order to confirm the possibility of a model that generates the bibliographic information of academic papers without pre- and post-processing, we conducted an in-depth comparative experiment with various settings and configurations. As a result of the experiment, the method proposed in this paper showed higher performance.

• Journal of Life Science
• /
• v.32 no.5
• /
• pp.391-410
• /
• 2022

Human microbiota is a community of microorganisms, including bacteria, fungi, and viruses, that inhabit various locations of the body, such as the gut, oral, and skin. Along with the development of metabolomic analysis and next-generation sequencing techniques for 16S ribosomal RNA, it has become possible to analyze the population for subtypes of microbiota, and with these techniques, it has been demonstrated that bacterial microbiota are involved in the metabolic and immunological processes of the hosts. While specific bacteria of microbiota, called commensal bacteria, positively affect hosts by producing essential nutrients and protecting hosts against other pathogenic microorganisms, dysbiosis, an abnormal microbiota composition, disrupts homeostasis and thereby has a detrimental effect on the development and progression of various types of diseases. Recently, several studies have reported that oral and gut bacteria of microbiota are involved in the carcinogenesis of gastrointestinal tumors and the therapeutic effects of anticancer therapy, such as radiation, chemotherapy, targeted therapy, and immunotherapy. Studying the complex relationships (bacterial microbiota-cancer-immunity) and microbiota-related carcinogenic mechanisms can provide important clues for understanding cancer and developing new cancer treatments. This review provides a summary of current studies focused on how bacterial microbiota affect gastrointestinal cancer and anticancer therapy and discusses compelling possibilities for using microbiota as a combinatorial therapy to improve the therapeutic effects of existing anticancer treatments.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.3
• /
• pp.731-736
• /
• 2023

Recently, the medical field has been applying mandatory Electronic Medical Records (EMRs) and Electronic Health Records (EHRs) systems that computerize and manage medical records, and distributing them throughout the entire medical industry to utilize patients' past medical records for additional medical procedures. However, the conversations between medical professionals and patients that occur during general medical consultations and counseling sessions are not separately recorded or stored, so additional important patient information cannot be efficiently utilized. Therefore, we propose an electronic medical record system that uses speech recognition and natural language processing deep learning to store conversations between medical professionals and patients in text form, automatically extracts and summarizes important medical consultation information, and generates electronic medical records. The system acquires text information through the recognition process of medical professionals and patients' medical consultation content. The acquired text is then divided into multiple sentences, and the importance of multiple keywords included in the generated sentences is calculated. Based on the calculated importance, the system ranks multiple sentences and summarizes them to create the final electronic medical record data. The proposed system's performance is verified to be excellent through quantitative analysis.

• Journal of Fashion Business
• /
• v.1 no.1
• /
• pp.71-83
• /
• 1997

The purpose of this study is to analyze the present fashion phenomenon by considering the types of street fashion, the center of avant-grade modern fashion, that shows the origination background and special feature concretely. The times was defined through the 1980's from World War II that street style originated, so the range of study was the 1990's when the street style was influenced by that of the past and was revived. The ways of study were to analyze the records of ideology, art and music connected with the street style from World War II to the present when it has risen. The summary of result is as follows. (1) In the 1940's, Zooties was the jet of desire suppressed by African-Americans that couldn't receive favors socially and economically and Hipsters pursued reformative bebop that made up of soft Jazz. In the 1950's, Modernists were running after Cool Jazz to the minimum. In the 1970's, Funk appered in the sexual desire and erotic strength, and was surfaced from Negro Getto. In the 1980's B-boys & Flygirls showed the street style by the scribble art of slum in the New York. As mentioned above, In the 1990's, Acid Jazz influenced by the Jazz of Negro has been the fashion added to the tradition of musical form that come from eclecticism of Jazz tended Neo-Jazz. (2) In the 1940's, Western style dreamed the country life because of rapid urbanization. In the 1950's, Beat obtained the feeling of liberation from the dissolute life and activity. In the 1960's, Psychedelics showed the freedom affected by the Pop-art and Op-art, and Hippies pursued the true individuality as 'love & peace' life style and the return to nature. In the 1990's, Grunge look influenced by the above has been fashion that shows the practical use of second-hand clothing or patchwork contrary to elitism. (3) In the 1940's, Caribbean style appered in the typical textile color with the center of West Indies. In the 1960's, Rude boys showed the magnificence and difficulty of Jamaica, and Rastafarians had a tendency to come back to the ancient civilization of America. In the 1970's, Two-tone was the simple clothing for harmonizing among human races. In the 1990's, Jamaica look influenced by the above has been the Lege fashion introduced to a high fashion, appearing in the special bright color, applique, unique hair style, and so on. (4) In the 1950's, Sufers pursued natural rhythm, getting out of everything. In the 1970' s, Skaters enjoyed the speed on the paved road. In the 1980's, Casuals emphasized the spirit of cooperation of young-things. In the 1990's, Casual look Influenced by the above has been the fashion that forms the activity, function and strong spirit of cooperation by pursuing comfortable life and sports in the tension of life and variety of modern society. (5) It was hard for Bikers to adjust themselves in society after the war. In the 1950's, Coffee bar cowboys were the reckless running boys in the leather jacket. In the 1960's, Rockers created the group originality as disobedient outsiders and Greasers imitated Rolling Stones. In the 1980's, Punks resisted the viewpoint of the old generation in offensive fashion. In the 1990's, Cyberpunk influenced by the above has pursued the classless structure, electronic music and metallic clothing that forebodes gloomily as the computer generation of ultra-modern science times. Accordingly, in understanding a complex modern fashion phenomenon, it was analyzed that the street styles of the past, from World War II to the 1980's, were reflected in that of the 1990's dividing into the five types in a word, namely Acid Jazz, Grunge look, Jamaica look, Casual look and Cyberpunk.

• Journal of Venture Innovation
• /
• v.1 no.2
• /
• pp.13-34
• /
• 2018

In this case, research has been conducted to verify that the purpose of establishing a reputation for establishing a reputation for identifying brand equity is research and investigation in the context of establishing a reputation for establishing a reputation, and conducting research on CSR activities that are essential to the essential activities of companies. The survey conducted a survey on behalf of the National Center for Adult Women and Girls and Girls ' Generation, using the final round of Section 305 to develop a series of findings, including the analysis of the feasibility analysis, feasibility analysis, reliability, correlation analysis, and factors analysis. A summary of the effects of research on corporate reputation on corporate reputation according to corporate CSR activities is as follows. 1st. Brand equity will have a positive effect on the Company's reputation. The hypothesis had a significant impact on brand recognition, brand loyalty, and perceived quality, but did not have significant impact on the brand image. 2st. Brand equity will have a positive effect on ethical responsibility. The hypothesis had a significant impact on brand recognition, brand loyalty, and perceived quality, but did not have significant impact on the brand image. 3st, the brand equity will have a positive effect on the benefit of the benevolent. The hypothesis had a significant impact on brand recognition, brand loyalty, and perceived quality, but did not have significant impact on the brand image. 4st, The theory that the influence of positive(+) will affect the company's reputation has a significant impact on both ethical and philanthropic factors. 5st. The ethical responsibility was found to have no effect on the usefulness of the brand between brand assets and corporate reputation. 6st. The philanthropic responsibility was partly attributable to the fact that there was a substitution between brand equity and corporate reputation. In sum, the company needs to prioritize its brand recognition before establishing its reputation, and the reason why it should be implemented is that the other elements of the brand equity should be evaluated with the presence of other elements of the brand equity, thereby ensuring continued compliance with continuous CSR activities. As a result, consumers expect to see the performance-based role of the company as a strategic and long-term perspective, as consumers want to see the CSR activity and the corporate reputation in a long-term manner, as opposed to the financial responsibility and legal responsibilities of the past, as opposed to the past.

• Journal of Life Science
• /
• v.27 no.5
• /
• pp.509-516
• /
• 2017

Perilla frutescens (L.) Britton var. sprouts (PFS) is a plant of the labiatae family. The purpose of this work was to assess the preventive effects of PFS ethanolic extracts (PFSEs) on cytokine-induced ${\beta}$-cell damage. Cytokines, which are released by the infiltration of inflammatory cells around the pancreatic islets, are involved in the pathogenesis of type 1 diabetes mellitus. The combination of interleukin-$1{\beta}$ (IL-1), interferon-${\gamma}$ (IFN-${\gamma}$), and tumor necrosis factor-${\alpha}$ (TNF-${\alpha}$) induced formation of reactive oxygen species (ROS). Accumulation of intracellular ROS led to ${\beta}$-cell dysfunction and apoptosis. PFSEs possess antioxidant activity and thus lead to downregulation of ROS generation. Cytokines decrease cell viability, stimulate the expression of inducible nitric oxide synthase (iNOS) and cyclooxygenase-2 (COX-2), and induce the production of nitric oxide (NO). PFSEs prevented cytokine-induced cell viability in a dose-dependent manner. Incubation with PFSE resulted in significant reduction in cytokine-induced NO production that correlated with reduced levels of the iNOS and COX-2 protein expression. Furthermore, PFSE significantly decreased the activation of nuclear factor ${\kappa}B$ (NF-${\kappa}B$) by inhibition of $I{\kappa}B{\alpha}$ phosphorylation in RINm5F cells. In summary, our results suggest that the protective effects of PFSE might serve to counteract cytokine-induced ${\beta}$-cell destruction. Findings indicate that consumption of Perilla frutescens (L.) Britton var. sprouts alleviates hyperglycemia-mediated oxidative stress and pro-inflammatory cytokine-induced ${\beta}$-cell damage and thus has beneficial anti-diabetic effects.

• 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.