• Proceedings of the Korea Database Society Conference
• /
• 2000.11a
• /
• pp.101-108
• /
• 2000

기업에서 보유한 전문 지식 정보가 급속도로 증가함에 따라 대량의 문서에 저장된 지식 정보를 효과적으로 탐색하여 기업 경영에 활용하기 위한 지식경영시스템 도입이 확산되고 있다. 이러한 지식경영시스템에서 핵심적인 구성 요소는 전문 분야의 지식 정보를 체계적으로 분류하고 효율적으로 검색하기 위한 지식 탐사 기법이다. 본 논문에서는 데이타마이닝 기법을 이용하여 문서를 자동적으로 분류하기 위한 새로운 모델을 제안하였다. 연관 규칙 탐사 알고리즘을 이용하여 학습용 문서 집합으로부터 세부 분야를 대표하는 색인어 집합을 구성하였다. 세부 분야별 색인어 집합에 대하여 전체 문서에 대한 비중에 따라 가중치 배열을 구성하여 문서를 자동으로 분류하기 위한 기준으로 삼았다. 임의의 문서를 자동적으로 분류하는 실험을 통하여 제안된 방법의 효율성을 검정하였다.

• Annual Conference of KIPS
• /
• 2001.04a
• /
• pp.43-46
• /
• 2001

본 논문에서는 데이타마이닝 기법을 이용하여 지능형 전문 검색엔진을 설계하고 사용자 인터페이스를 구현하였다. 먼저, 컴퓨터 분야의 전문 용어에 대하여 연관 규칙 탐사 알고리즘을 이용하여 의미적으로 연관된 용어들끼리 클러스터로 구성하였다. 전문 용어별로 구성된 클러스터는 본 논문에서 제안한 지식베이스 테이블에 저장하여 의미적으로 연관된 용어를 포함하는 웹 문서를 검색하는 과정에서 이용하였다. 검색과정에서는 사용자가 제시한 키워드와 관련된 전문 용어들간의 연관정도를 가중치로 부여하여 연관 정도가 높은 웹 문서순으로 출력하였다. 제안된 방법을 통하여 사용자가 제시한 키워드와 의미적으로 연관된 웹 문서를 효과적으로 검색할 수 있었다.

• Journal of the Korean Society for information Management
• /
• v.17 no.1
• /
• pp.129-148
• /
• 2000

Experiments were performed on three subsets of a Korean test collection in order to determine whether 2-Poisson model's Z value is a good measure for selecting subject words from a document to be indexed. It was found that subject word selection based on the Z value was effective for only one subset with short texts, i.e., the Science and Technology subset. Correlation analyses between 2-Poisson model's Z and TF.IDF weight for the three subsets showed that the correlation was relatively high for two test subsets with short texts, i.e., the Science and Technology subset and the Newspaper subset.

• Journal of the Korean Society for Library and Information Science
• /
• v.32 no.4
• /
• pp.69-86
• /
• 1998

The purpose of velop the indexing algorithm of secondary legal information by the study of characteristics of legal information, to compare the indexing system using controlled vocabulary to the indexing system using natural language in the secondary legal information full-text databases, and to prove propriety and superiority of the indexing system using controlled vocabulary. The results are as follows; 1)The indexing system using controlled vocabulary in the secondary legal information full-text databases has more effectiveness than the indexing system using natural language, in the recall rate, the precision rate, the distribution of propriety, and the faculty of searching for the unique proper-records which the indexing system using natural language fans to find 2)The indexing system which adds more words to the controlled vocabulary in the secondary legal information full-text databases does not better effectiveness in the retail rate, the precision rate, comparing to the indexing system using controlled vocabulary. 3)The indexing system using word-added controlled vocabulary with an extra weight in the secondary legal information full-text databases does not better effectiveness in the recall rate, the precision rate, comparing to the indexing system using word-added controlled vocabulary without an extra weight. This study indicates that it is necessary to have characteristic information the information experts recognize - that is to say, experimental and inherent knowledge only human being can have built-in into the system rather than to approach the information system by the linguistic, statistic or structuralistic way, and it can be more essential and intelligent information system.

• Journal of KIISE:Software and Applications
• /
• v.33 no.7
• /
• pp.636-645
• /
• 2006

A tenn with more domain specific information has higher level of term specificity. We propose new specificity calculation methods of terms based on information theoretic measures using compositional and contextual information. Specificity of terms is a kind of necessary conditions in tenn hierarchy construction task. The methods use based on compositional and contextual information of terms. The compositional information includes frequency, $tf{\cdot}idf$, bigram and internal structure of the terms. The contextual information of a tenn includes the probabilistic distribution of modifiers of terms. The proposed methods can be applied to other domains without extra procedures. Experiments showed very promising result with the precision of 82.0% when applied to the terms in MeSH thesaurus.

• The Journal of the Korea Contents Association
• /
• v.15 no.10
• /
• pp.387-399
• /
• 2015

Recently collaboration network is generated to find out experts in their field as potential collaborators in health care sector. In this paper, the co-author network of a National Cancer Center researcher was generated for identifying each researcher's role and collaborative research pattern. The co-author network of 2,437 authors was extracted from 1,194 SCI(E) publications from 2000 to 2010 and author's role was analyzed by author's centrality value. Centrality reflecting only the number of papers and centrality weighted by the paper number, impact factor, and authorship contribution was evaluated. On the comparison with simple degree centrality value and the weighted degree centrality, difference of value was statistically significant(t=11.66, p=0.00). Co-author network considering various variables of the paper provides more objective figure of researcher's role. This suggests that co-author network could be more effective in identifying potential collaborators.

• Journal of the korean academy of Pediatric Dentistry
• /
• v.34 no.1
• /
• pp.43-52
• /
• 2007

The supply and demand planning the pediatric dentists is earnest, because of the start of the dental specialist system on 2008 and aging society with low fertility. Therefore in order to develop the model, that is adequate to estimate demand for the pediatric dentists, a studies on the supply and demand planing of other health manpower were reviewed. The obtained results were as follows : 1. The health demand method was appropriate for demand estimation of the pediatric dentists. 2. There was independent variables needed for demand estimation model: prevalence, utilization rate, referral rate, fertility rate, productivity, annual working days, and so on. 3. Since statistical data for application of these variables was insufficient as result of searching, questionnaire researching and discussion of specialist may be necessary. 4. Each independent variables should be inducted into an equation by using a adequate regression model and then estimated.

• The Journal of Sustainable Design and Educational Environment Research
• /
• v.17 no.3
• /
• pp.9-16
• /
• 2018

This research is intended to propose future research directions by identifying differences between Korea and China's public facilities at the time of introduction and presenting improvement measures by comparing the criteria for green building certification. The study focuses on the comparison of Korea's G-SEED 2016 and China's ESGB 2014. For data related to green building certification in Korea, refer to the Construction Technology Research Institute Green Building Certification Criteria 2016 v1.2 Guide for New Housing. A study on the Green Building Certification System in China referred to the Green Building Assessment Standards. Comparisons were made between G-SEED 2016 general building certification review criteria and ESGB 2014 public facility certification criteria, and certification methods, essential items and specialties for each area.

• Annual Conference on Human and Language Technology
• /
• 1994.11a
• /
• pp.374-377
• /
• 1994

본 논문에서는 한글 문서 검색 시스템에서 자연어 질의어로 검색할경우, 질의어를 주제어와 참조어로 나누어 재구성하여 검색하는 방법을 제시하였다. 먼저 주제어로 전문검색을 하여 후보 카드들을 추출한 후 비주제어로 다시 본문 탐색을 하여 추출된 카드의 가중치를 재조정함으로써 카드추출의 정확성을 높였다. 이 논문에 제시된 방법의 실험은 한국전자통신연구소 언어정보연구실에서 개발한 멀티미디어 전자 백과 사전의 자연어 검색모듈에서 행하여 졌다. 이 방법으로 별다른 검색속도의 저하나, 저장공간의 추가가 없이 기존의 검색 방법에서보다 약 58%정도의 검색의 정확성이 올라갔다. 본 논문에서 제시한 검색의 방법은 여러가지 응용의 자연어 인터페이스에서 데이타를 검색하는 정보검색의 분야에 적용되어 정확성을 높일 수 있을 것이다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.16 no.1
• /
• pp.183-190
• /
• 2012

This study constructs an expert search system which has a mutual cooperation function based on thesis and author profile. The proposed methodology is as follows. First, we propose weighting method which can search a keyword and the most relevant keyword. Second, we propose a method which can search the experts efficiently with this weighting method. On the preferential basis, keywords and author profiles are extracted from the papers, and experts can be searched through this method. This system will be available to many fields of social network. However, this information is distributed to many systems. We propose a method using multi-ontology to integrate distributed data. The multi-ontology is composed of meta ontology, instance ontology, location ontology and association ontology. The association ontology is constructed through analysis of keyword association dynamically. An expert network is constructed using this multi-ontology, and this expert network can search expert through association trace of keyword. The expert network can check the detail area of expertise through the research list which is provided by the system.