• Journal of the Korean Society for information Management
• /
• v.27 no.4
• /
• pp.49-69
• /
• 2010

Library classification system is based upon academic disciplines, However, it is difficult to classify for Internet resources due to its lack of up-to-datedness and practicality. Especially, headings of Korean Decimal Classification need to reflect practical aspects and it should be also developed for classification of web based resources. The purposes of this study are to analyze the structures of directory classifications in Internet resources and to suggest additional headings of KDC as a practical library classification as well as a classification system for internet resources. Directory classification systems of Naver, Yahoo, Kyobo Internet book store, Amazon were selected and their food and culture subjects were analyzed for this study. The headings of KDC were compared to them and new possible headings were suggested with reference of NDC and DDC in the area of food and culture. This study provided a way of developing KDC for a classification system for Internet resources as well as library materials.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.12 no.9
• /
• pp.4192-4198
• /
• 2011

Terrestrial LiDAR is a high precision positioning technique to monitor the behavior and change of structures and natural slopes, but it has depended on subjective hand intensive tasks for the classification(surface and vegetation or structure and vegetation) of positioning data. Thus it has a couple of problems including lower reliability of data classification and longer operation hours due to the surface characteristics of various geographical and natural features. In order to solve those problems, the investigator developed a technique of using the NDVI, which is a major index to monitor the changes on the surface(including vegetation), to categorize land covers, combining the results with the terrestrial LiDAR data, and classifying the results according to items. The application results of the developed technique show that the accuracy of convergence was 94% even though there was a problem with partial misclassification of 0.003% along the boundaries between items. The technique took less time for data processing than the old hand intensive task and improved in accuracy, thus increasing its utilization across a range of fields.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.41 no.3
• /
• pp.277-288
• /
• 2021

Building information modeling (BIM) element to industry foundation classes (IFC) entity mappings need to be checked to ensure the semantic integrity of BIM models. Existing studies have demonstrated that machine learning algorithms trained on geometric features are able to classify BIM elements, thereby enabling the checking of these mappings. However, reliance on geometry is limited, especially for elements with similar geometric features. This study investigated the employment of relational data between elements, with the assumption that such additions provide higher classification performance. Neural structured learning, a novel approach for combining structured graph data as features to machine learning input, was used to realize the experiment. Results demonstrated that a significant improvement was attained when trained and tested on eight BIM element types with their relational semantics explicitly represented.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.19 no.2
• /
• pp.389-396
• /
• 2024

Generative AI has recently been utilized across all fields, achieving expert-level advancements in deep data analysis. However, identifying regional names in scientific literature remains a challenge due to insufficient training data and limited AI application. This study developed a standardized dataset for effectively classifying regional names using address data from Korean institution-affiliated authors listed in the Web of Science. It tested and evaluated the applicability of machine learning and deep learning models in real-world problems. The BERT model showed superior performance, with a precision of 98.41%, recall of 98.2%, and F1 score of 98.31% for metropolitan areas, and a precision of 91.79%, recall of 88.32%, and F1 score of 89.54% for city classifications. These findings offer a valuable data foundation for future research on regional R&D status, researcher mobility, collaboration status, and so on.

• Annual Conference of KIPS
• /
• 2011.04a
• /
• pp.1551-1554
• /
• 2011

대용량 데이터 환경에의 적용이 가능한 대용량 학습기반의 자동범주화 기법과 범용적으로 사용할 수 있는 기법은 대량의 정보를 처리해야하는 정보분석 및 정보서비스 환경에 가장 필요한 기술요소라 할 수 있다. 본 논문에서는 대용량의 문서를 단위 컴포넌트로 분할하여 학습하고 이를 동적으로 결합하는 대용량 분류기 생성 기법을 소개하고 자동범주화 성능을 SVM 모델과 비교하여 봄으로써, 본 기술의 활용 가능성을 살펴보도록 한다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.07a
• /
• pp.673-674
• /
• 2020

의식주 중에서 자신을 표현하고 외부와의 교류를 할 수 있는 분야는 패션분야로서 인간 생활과 밀접한 관계를 가지고 있으며 사람들의 개인화된 성향 변화 및 인터넷 환경의 개선으로 트렌드는 빠르게 변화하고 있다. 인공지능 기술의 발전은 단순히 객체의 검출 및 분류에서 벗어나 패션 아이템의 분석 및 세부적인 속성을 분석할 수 있는 수준에 다다랐으며 인공지능 기술을 활용하여 사용자에게 추천할 수 있는 서비스가 출시되고 있다. 패션 트렌드의 빠른 변화 및 인공지능 기술의 발전으로 이를 활용한 플랫폼에 기반을 두어 디자이너에게는 디자인 기술을 향상시킬 수 있으며 사용자에게는 개인화된 제품을 구매할 수 있는 플랫폼 개발이 요구되고 있다. 본 논문에서는 인공지능 기술 기반 패션 분석 기술 개발을 위하여 패션 검출 모듈, 패션 검색 모듈, 패션 검색을 위한 벡터 검색 모듈, 상하의 분리를 위한 세그먼테이션 모듈, 패션 복종 분류 모듈을 개발하여 통합하였으며 패션 검색 정확도는 Top-5 기준 75.28%, 벡터 검색 속도는 벡터당 0.002m sec 이하, 세그먼테이션 추출 정확도 87.6%이상, 패션 검출 결과 IoU 0.5 환경에서 96.2%, 복종분석 90.54%의 성능을 보였다.

• Clean Technology
• /
• v.18 no.3
• /
• pp.307-311
• /
• 2012

Patents is a strong asset. Samsung and Apple's patent lawsuit is a prime example. So many countries reinforce the intellectual property and they lay the emphasis on the patent. Utilizing the patent information efficiently is basic to the patent analysis. Patent information will provide for new science and technology information sources, international code is classified according to the international patent system IPC, being easily accessible. In this paper, analysis of foreign and domestic patents for zeolite technologies analysis using IPC. The current of technology development in such countries as Korea, USA, Japan, China and EU was analyzed by classifying the patents for 1992 through 2011 according to registration country, assignee, calendar year and technology area.

• Management & Information Systems Review
• /
• v.35 no.4
• /
• pp.269-287
• /
• 2016

This paper proposes the newly developed a classification scheme for Management of Technology (MOT) research, by analyzing research papers published in major academic journals of MOT. To refine our proposed classification, we conducted a contents analysis after collecting 696 papers published in 11 major international journals and 2 representative domestic journals including Journal of Korean Technology Innovation Society, from 2006 to 2012. Our findings show that this classification scheme is composed of total 8 methodologies within each research design such as conceptual qualitative, empirical qualitative, conceptual quantitative, and conceptual quantitative researches. In addition, we provide research reviews based on characteristics between 1) domestic and foreign researchers, 2) domestic and international journals, and 3) different major domestic journals as follows. Firstly, the underlying trends of research design and methodologies in MOT research are concentrated in empirical studies regardless of the origin of researchers and journals. However, the proportion of theoretical discussion in the international journals is higher than that of theoretical discussion in the domestic journals. Secondly, there are many more empirical qualitative researches and archival researches written in the Journal of Korean Technology Innovation Society, whereas the proportion of empirical quantitative research and conceptual research published in the Journal of Technology Innovation is higher than that. From the results, this paper will contribute to choosing research designs and methodologies more suitable to research purpose for future MOT researchers and provide the direction for concrete systematic approach on research designs and methodologies.

• Journal of KIISE:Software and Applications
• /
• v.30 no.12
• /
• pp.1124-1134
• /
• 2003

The development of microarray technology has supplied a large volume of data to many fields. In particular, it has been applied to prediction and diagnosis of cancer, so that it expectedly helps us to exactly predict and diagnose cancer. It is essential to efficiently analyze DNA microarray data because the amount of DNA microarray data is usually very large. Since accurate classification of cancer is very important issue for treatment of cancer, it is desirable to make a decision by combining the results of various expert classifiers rather than by depending on the result of only one classifier. Generally combining classifiers gives high performance and high confidence. In spite of many advantages of ensemble classifiers, ensemble with mutually error-correlated classifiers has a limit in the performance. In this paper, we propose the ensemble of neural network classifiers learned from negatively correlated features using three benchmark datasets to precisely classify cancer, and systematically evaluate the performances of the proposed method. Experimental results show that the ensemble classifier with negatively correlated features produces the best recognition rate on the three benchmark datasets.

• Journal of Industrial Convergence
• /
• v.20 no.8
• /
• pp.77-84
• /
• 2022

This study aims to present a method for implementing a decision support system that can be used for selecting emerging technologies by applying a machine learning-based automatic classification technique. To conduct the research, the architecture of the entire system was built and detailed research steps were conducted. First, emerging technology candidate items were selected and trend data was automatically generated using a big data system. After defining the conceptual model and pattern classification structure of technological development, an efficient machine learning method was presented through an automatic classification experiment. Finally, the analysis results of the system were interpreted and methods for utilization were derived. In a DTW-kNN-based classification experiment that combines the Dynamic Time Warping(DTW) method and the k-Nearest Neighbors(kNN) classification model proposed in this study, the identification performance was up to 87.7%, and particularly in the 'eventual' section where the trend highly fluctuates, the maximum performance difference was 39.4% points compared to the Euclidean Distance(ED) algorithm. In addition, through the analysis results presented by the system, it was confirmed that this decision support system can be effectively utilized in the process of automatically classifying and filtering by type with a large amount of trend data.