• Journal of Digital Convergence
• /
• v.19 no.9
• /
• pp.181-188
• /
• 2021

Research and importance of legal services applied with AI so that it can be easily understood and predictable in difficult legal fields is increasing. In this study, based on the decision of the Tax Tribunal in the field of tax law, a model was built through self-learning through information collection and data processing, and the prediction results were answered to the user's query and the accuracy was verified. The proposed model collects information on tax decisions and extracts useful data through web crawling, and generates word vectors by applying Word2Vec's Fast Text algorithm to the optimized output through NLP. 11,103 cases of information were collected and classified from 2017 to 2019, and verified with 70% accuracy. It can be useful in various legal systems and prior research to be more efficient application.

• Journal of the Korea Convergence Society
• /
• v.12 no.10
• /
• pp.365-372
• /
• 2021

The purpose of this study was to examine the statements characteristics of the pressure ulcer nursing record by natural langage processing and assess the prediction accuracy for each pressure ulcer stage. Nursing records related to pressure ulcer were analyzed using descriptive statistics, and word cloud generators (http://wordcloud.kr) were used to examine the characteristics of words in the pressure ulcer prevention nursing records. The accuracy ratio for the pressure ulcer stage was calculated using deep learning. As a result of the study, the second stage and the deep tissue injury suspected were 23.1% and 23.0%, respectively, and the most frequent key words were erythema, blisters, bark, area, and size. The stages with high prediction accuracy were in the order of stage 0, deep tissue injury suspected, and stage 2. These results suggest that it can be developed as a clinical decision support system available to practice for nurses at the pressure ulcer prevention care.

• Journal of Platform Technology
• /
• v.9 no.2
• /
• pp.46-54
• /
• 2021

Recently, interest in the safety of Self-driving has been increasing. Self-driving have been studied and developed by many universities, research centers, car companies, and companies of other industries around the world since the middle 1980s. In this study, we propose the automatic extraction method of the threatening obstacle on the Road for the Self-driving. A threatening obstacle is defined in this study as a comparatively large object at center of the image. First of all, an input image and its decreased resolution images are segmented. Segmented areas are classified as the outer or the inner area. The outer area is adjacent to boundaries of the image and the other is not. Each area is merged with its neighbors when adjacent areas are included by a same area in the decreased resolution image. The Obstacle area and Non Obstacle area are selected from the inner area and outer area respectively. Obstacle areas are the representative areas for the obstacle and are selected by using the information about the area size and location. The Obstacle area and Non Obstacle area consist of the threatening obstacle on the road. Through experiments, we expect that the proposed method will be able to reduce accidents and casualties in Self-driving.

• Knowledge Management Research
• /
• v.22 no.3
• /
• pp.91-106
• /
• 2021

With the recent rapid development of deep learning technology, the demand for analyzing huge text documents in the national R&D field from various perspectives is rapidly increasing. In particular, interest in the application of a BERT(Bidirectional Encoder Representations from Transformers) language model that has pre-trained a large corpus is growing. However, the terminology used frequently in highly specialized fields such as national R&D are often not sufficiently learned in basic BERT. This is pointed out as a limitation of understanding documents in specialized fields through BERT. Therefore, this study proposes a method to build an R&D KoBERT language model that transfers national R&D field knowledge to basic BERT using further pre-training. In addition, in order to evaluate the performance of the proposed model, we performed classification analysis on about 116,000 R&D reports in the health care and information and communication fields. Experimental results showed that our proposed model showed higher performance in terms of accuracy compared to the pure KoBERT model.

• Journal of the Korean BIBLIA Society for library and Information Science
• /
• v.32 no.3
• /
• pp.265-284
• /
• 2021

This study is to present improvement plans by examining how Korean literature house websites provide information services. Seventy-nine Korean literature houses out of eighty-eight members of the Korean Literature House Association were studied, except nine that did not construct websites. Three core elements, including website style, literary works information and writer information, together with thirteen sub-elements, were derived from precedent studies. As a result, it was found that 90% of the literature houses were operating websites, but the classification criteria for the literary works and cataloguing rules were not unified, and literature information was not provided sufficiently. Thus, this study suggested improvement plans such as support to build a website, developing cataloging guidelines for literature houses, providing more full-text literature and providing information about literary works and writer.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.44 no.3
• /
• pp.1-9
• /
• 2021

This research examines deep learning based image recognition models for beef sirloin classification. The sirloin of beef can be classified as the upper sirloin, the lower sirloin, and the ribeye, whereas during the distribution process they are often simply unified into the sirloin region. In this work, for detailed classification of beef sirloin regions we develop a model that can learn image information in a reasonable computation time using the MobileNet algorithm. In addition, to increase the accuracy of the model we introduce data augmentation methods as well, which amplifies the image data collected during the distribution process. This data augmentation enables to consider a larger size of training data set by which the accuracy of the model can be significantly improved. The data generated during the data proliferation process was tested using the MobileNet algorithm, where the test data set was obtained from the distribution processes in the real-world practice. Through the computational experiences we confirm that the accuracy of the suggested model is up to 83%. We expect that the classification model of this study can contribute to providing a more accurate and detailed information exchange between suppliers and consumers during the distribution process of beef sirloin.

• The Journal of Korean Institute of Next Generation Computing
• /
• v.15 no.4
• /
• pp.7-17
• /
• 2019

Machine learning is one of the most accurate techniques to predict and analyze various phenomena. K-means clustering is a kind of machine learning technique that classifies given data into clusters of similar data. Because it is desirable to perform an analysis based on a lot of data for better performance, K-means clustering can be performed in a model with a server that calculates the centroids of the clusters, and a number of clients that provide data to server. However, this model has the problem that if the clients' data are associated with private information, the server can infringe clients' privacy. In this paper, to solve this problem in a model with a number of clients, we propose a privacy-preserving K-means clustering method that can perform machine learning, concealing private information using homomorphic encryption.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.11
• /
• pp.139-147
• /
• 2021

BIBFRAME, which is attracting attention as an alternative to the machine-readable catalog format (MARC) in the library world, presents a new bibliographic data model in the open web environment while maintaining compatibility with existing data. To convert KORMARC(Korean data model of MARC) records into BIBFRAME, we extract 25 key fields by analyzing the latest 5,000 bibliographic data from Nowon-gu Library in Seoul. The extracted core fields are classified into three types according to the compatibility of MARC 21, and define conversion rules for each type. In addition, implement an open source-based converter to perform KORMARC to BIBFRAME conversion. As a basic study on KORMARC to BIBFRAME conversion, this study is meaningful in that it analyzes the latest KORMARC information actually used, defines conversion rules, and attempts BIBFRAME conversion.

• The Korean Journal of Archival Studies
• /
• no.4
• /
• pp.131-161
• /
• 2001

This paper is to examine the bibliographical description of the revolutionary historical archives and the standardization of the archives management in China. The standardization in the field of records and archives management was not initiated on its own way but as a part of national - leveled standardization. As a first step National Committee on Technical Standardization of Literature was established, followed by the establishing of Committee on Technical Standardization of Micro - filming and Committee of Technical Standardization of Paper Form. The standardization of the records and archives management was carried out in the context of functions of these three committees. In 1983 the standardization in the sphere of records and archives management speeded up, when the National Archives Administration formed small organizations which led the standardization work all over the country. A committee of standardization originated from small organizations and it brought a great progress of the standardization. If some opinions on standardization were submitted from records offices or related offices, they were examined by the committee of standardization. The opinions that were submitted by the committee of standardization were examined by the National Archives Administration which proclaimed it officially. The Chinese government commenced to establish the bibliographical data centre for historical archives and materials on the basis of this process of standardization. In the case of the revolutionary historical archives the description was made on the level of sources(provenance), which was sent to the bibliographical data centre for historical archives and materials. The Chinese government set digitalizing as a goal in records and archives management in the middle of 1990's. It was regulated that the description of records item that should be transferred to the center must be digitalized. However, the description of the file level was not made separately being reflected in the process of description of item level. (The second part of the paper will be released in the next volume).

• Journal of the Korea Institute of Construction Safety
• /
• v.2 no.1
• /
• pp.9-15
• /
• 2019

Despite the efforts for enhancing the safety record, construction industry has been suffered from higher fatalities than other industries. The poor record of safety in construction industry means that there is a clear need for an effective countermeasure. As mentioned in previous studies, it is important to identify the type of activities or risks that are likely to cause accidents and to develop appropriate safety measures. Considering the large number of accident cases on the temporary installations including work platforms and work passages, the temporary installations should be managed first. To support it, this study aims to analyze falling accident cases on construction working platforms and passages which can lead to develop proper safety measures. Through the analysis of 1663 accident cases in the perspective of cost, progress, activity, and type of workers, this study identifies how the recent accidents occur and what is the cause of the accident occurrence. The identified causes of accident occurrence will help us to improve current construction safety.