• 한국생물공학회:학술대회논문집
• /
• 한국생물공학회 2003년도 생물공학의 동향(XIII)
• /
• pp.92-93
• /
• 2003

• 한국산학기술학회논문지
• /
• 제11권2호
• /
• pp.517-524
• /
• 2010

최근의 컴퓨터 기술과 인터넷 기술의 발달로 인해 분석 데이터가 급속도로 증가함에 따라 이들을 다루기 위한 자동분류시스템에 대한 요구가 높다. 문서분류시스템은 감독학습이 필수적이기 때문에 최소한의 전문가의 개입만으로도 높은 정확도가 보장되는 자동화 시스템에 대한 요구가 크다. 반면, 분류할 데이터들은 형식이나 내용상으로 그 복잡도가 높아지고 있어서, 일반적인 분류방법으로는 좋은 분석결과를 얻기 어려운 양상을 보인다. 특히 스팸성 데이터와 같이 어떠한 의도가 반영되어 가공되거나 변형되는 데이터는 분석의 어려움을 가중시킨다. 본 논문에서는 분류알고리즘의 성능향상을 위해 제안한 ETOM과 RPost방법을 구현하였다. 분류의 경계선 상에 있는 스팸문서들에 구현시스템을 적용하여 그 과정을 분석하였다. 실험결과 제안방법에 의한 정확도가 0.795에서 0.93으로 약 16%의 증가하였음을 확인하였다.

• 한국산학기술학회논문지
• /
• 제22권5호
• /
• pp.296-302
• /
• 2021

본 논문에서는 영상정보에 기반한 호흡상태 분류 방법을 제안한다. 호흡신호는 초광대역 레이더 센서를 이용하여 획득하고 호흡신호의 값으로 이루어진 1차원 그래프 대신 그래프의 영상 정보가 담긴 2차원 정보 기반으로 호흡상태를 분류한다. 호흡상태의 분류는 심층신경망 모델을 사용하고, 심층신경망 모델은 호흡신호 그래프가 포함된 2차원 영상의 특징들을 학습하여 영상기반의 호흡상태 분류의 결과를 제공한다. 기존의 레이더 센서 기반 호흡신호의 상태 분류는 1차원 벡터의 구성요소 값 및 그 값들의 변화량을 이용하여 회귀, 심층학습 방법을 적용하였다. 그러나 1차원 그래프 기반의 호흡상태 분류는 다양한 형태의 정상호흡 상태에 대한 분류 성능에서 한계를 보였다. 본 논문에서는 호흡 신호로부터 얻은 그래프의 이미지 자체를 2차원 입력 신호로 사용하여 심층 신경망 모델을 적용하여 분류를 수행하였다. 본 논문에서 제안하는 영상정보 기반의 호흡상태 분류는 기존의 1차원 벡터 기반 호흡상태 분류 대비 호흡상태 분류의 정확도를 약 10% 향상 시켰다. 또한 기존의 두 가지 호흡상태 (정상 및 비정상) 분류에서 확장하여 세 가지 호흡상태 (정상1, 정상2, 비정상) 분류를 수행하였다.

• 한국멀티미디어학회논문지
• /
• 제20권3호
• /
• pp.571-580
• /
• 2017

Recently, there are many communication softwares based on text on various smart devices. Unlike traditional print publishing, mobile publishing and SNS tools tends to utilize more decorative or more emotional fonts so that users can pass some feelings from contents. So font providers have released new fonts which deal with the requirements of the market. Nevertheless being released lots of new fonts, general users have not used them because they searched only by font name or font provider's name. It means that there is no way for users to know and find new things. In this study, we suggest font shape classification rules for font registration system based on font design features. We proved the validity of classification standard study through some experiments with 50 commercial fonts. Also the result of this study was provided for Korea Telecommunication Technology Association and adopted by the Korea industrial standard.

• International Journal of Internet, Broadcasting and Communication
• /
• 제15권3호
• /
• pp.103-117
• /
• 2023

This study focuses on the behavioural lexical classification for extracting animation character actions and the analysis of the character's upper and lower body movements. The behaviour and state of characters in the animation industry are crucial, and digital technology is enhancing the industry's value. However, research on animation motion application technology and behavioural lexical classification is still lacking. Therefore, this study aims to classify the predicates enabling animation motion, differentiate the upper and lower body movements of characters, and apply the behavioural lexicon's motion data. The necessity of this research lies in the potential contributions of advanced character motion technology to various industrial fields, and the use of the behavioural lexicon to elucidate and repurpose character motion. The research method applies a grammatical, behavioural, and semantic predicate classification and behavioural motion analysis based on the character's upper and lower body movements.

• 대한안전경영과학회지
• /
• 제17권3호
• /
• pp.9-14
• /
• 2015

The study is conducted in order to establish validity of industrial safety field's neo-qualification composition by suggesting reengineering procedure of NCS based industrial safety field's qualifying occupational plan. Industrial safety fields are first classified into great classification which are environment energy safety, and then into sub-classification which are industrial safety management, industrial health management, and nondestructive test. NCS based neo-qualification plan is deveopled into progress as following steps; First, planning NCS neo-qualification system. Second, presenting work education combined program certification standard. Third, developing NCS based neo-qualification assessment standard. Fourth, developing work education qualification prototype problem. The process of qualification plan was composed with Worksheet 14 and with 4phase being completing summary of each section, certification and development of assessment standard, and development of prototype problem. As a result, Industrial safety manager was redesigned into Industrial sfety manager and Construction Safety Manager, and Occupational Health Care Manager was redesigned into Occupational Health Care Manager and Industrial Health Manager. This study, in conclusion, suggests development of qualifying occupation that could be applied to the integrated management system and development of future industrial safety field's specialized neo-qualifying occupation.

• 농촌지도와개발
• /
• 제22권3호
• /
• pp.285-292
• /
• 2015

Recently, the demand of rural tourism has been increased to promote farm household income and rural economy. Korean government has supported to promote rural tourism. One of the most difficult tasks in estimating the economic impact of the tourism industry is how the industry should be defined in terms of an economic sector, since tourism is not defined in national Input-Output (I-O) tables or in the Standard Industrial Classification code. Moreover, there is no specified Standard Industrial Classification for rural tourism. The purpose of the study aims to examine specified Standard Industrial Classification of rural tourism using the I-O model analysis to estimate the economic impacts of rural tourism. Results showed that there were two components considered as inputs. One is the inputs that final demand can move to input of rural tourism in I-O tables. The other is one that the final demand was provided by farm household as intermediate inputs.

• Industrial Engineering and Management Systems
• /
• 제14권1호
• /
• pp.58-72
• /
• 2015

A primary concern of national R&D plans is to encourage technological development in private firms and research institutes. For effective R&D planning and program support, it is necessary to assess technological impacts that may exist both directly and indirectly among technology areas within the whole technology system; however, previous studies analyze only direct impacts among technologies, failing to capture both direct and indirect impacts. Therefore, this study proposes an approach based on decision-making trial and evaluation laboratory (DEMATEL) to identifying specific characteristics of technology areas, such as technological impact and degree of cause or effect (DCE). The method employs patent co-classification analysis to construct a technological knowledge flow matrix. Next, to capture both direct and indirect effects among technology areas, it incorporates the modified DEMATEL process into patent analysis. The method helps analysts assess the technological impact and DCE of technology areas, and observe their evolving trajectories over time, thereby identifying relevant technological implications. This study presents a case study using Korean patents registered during 2003-2012. We expect our analysis results to be helpful input for R&D planning, as well as the suggested approach to be incorporated into processes for formulating national R&D plans.

• 대한산업공학회지
• /
• 제43권3호
• /
• pp.192-202
• /
• 2017

This study develops a text spam filtering system for Facebook based on two variable categories: keywords learned from Instagram and meta-information of Facebook posts. Since there is no explicit labels for spam/ham posts, we utilize hash tags in Instagram to train classification models. In addition, the filtering accuracy is enhanced by considering meta-information of Facebook posts. To verify the proposed filtering system, we conduct an empirical experiment based on a total of 1,795,067 and 761,861 Facebook and Instagram documents, respectively. Employing random forest as a base classification algorithm, experimental result shows that the proposed filtering system yield 99% and 98% in terms of filtering accuracy and F1-measure, respectively. We expect that the proposed filtering scheme can be applied other web services suffering from massive spam posts but no explicit spam labels are available.

• Industrial Engineering and Management Systems
• /
• 제16권1호
• /
• pp.64-79
• /
• 2017

Identifying disease genes from human genome is a critical task in biomedical research. Important biological features to distinguish the disease genes from the non-disease genes have been mainly selected based on traditional feature selection approaches. However, the traditional feature selection approaches unnecessarily consider many unimportant biological features. As a result, although some of the existing classification techniques have been applied to disease gene identification, the prediction performance was not satisfactory. A small set of the most important biological features can enhance the accuracy of disease gene identification, as well as provide potentially useful knowledge for biologists or clinicians, who can further investigate the selected biological features as well as the potential disease genes. In this paper, we propose a new stepwise random forests (SRF) approach for biological feature selection and disease gene identification. The SRF approach consists of two stages. In the first stage, only important biological features are iteratively selected in a forward selection manner based on one-dimensional random forest regression, where the updated residual vector is considered as the current response vector. We can then determine a small set of important biological features. In the second stage, random forests classification with regard to the selected biological features is applied to identify disease genes. Our extensive experiments show that the proposed SRF approach outperforms the existing feature selection and classification techniques in terms of biological feature selection and disease gene identification.