• Proceedings of the Korea Information Processing Society Conference
• /
• 2003.11c
• /
• pp.1617-1620
• /
• 2003

본 논문은 사용자 질의가 가지는 특정 클래스로부터 개념적으로 서로 연관있는 컴포넌트를 검색하기 위하여 퍼지 시소러스를 통한 질의 평가 방법을 이용하였다. 시소러스에 의한 사용자 질의 확장과정은 용어 불일치 문제를 해결함으로써 검색에 대한 일정한 정확도를 보장하면서 재현율을 향상시킬 수 있게 한다. 필의 확장과정의 효율성을 평가하기 위하여 시뮬레이션을 통한 최적의 검색 효율을 나타내는 임계치를 설정하고 재현율과 정확도를 비교하였다.

• Management & Information Systems Review
• /
• v.35 no.3
• /
• pp.173-193
• /
• 2016

The aim of this paper is to extend the TOPSIS(Technique for Order Preference by Similarity to Ideal Solution) to the fuzzy environment for solving the disaster recovery priority decision problem in credit bureau business information system. In this paper, the rating of each information systems and the weight of each criterion are described by linguistic terms which can be expressed in trapezoidal fuzzy numbers. Then, a vertex method is proposed to calculate the distance between two trapezoidal fuzzy numbers. According to the concept of the TOPSIS, a closeness coefficient is defined to determine the ranking order of all information systems. The combination between the fuzzy set and TOPSIS brings several benefits when compared with other approaches, such that the fuzzy TOPSIS require few fuzzy judgements to parameterization, which contributes to the agility of the decision process, it does not limit the number of alternatives simultaneously evaluated, and it does not cause the ranking reversal problem when a new alternative is included in the evaluation process. This paper is demonstrated with a real case study of a credit rating agency involving 9 evaluation criteria and 9 credit bureau business information systems assessed by 6 evaluators, and provide the systematic disaster recovery framework for BCP(Business Continuity Planning) to practitioner. Finally, this paper show that the procedure of the proposed fuzzy TOPSIS method is well suited as a decision-making tool for the disaster recovery priority decision problem in credit bureau business information system.

• Journal of Digital Convergence
• /
• v.14 no.9
• /
• pp.229-239
• /
• 2016

The aim of this paper is to extend the TOPSIS(Technique for Order Performance by Similarity to Ideal Solution) to the fuzzy environment for solving the new professor selection problem in a university. In order to achieve the goal, the rating of each candidate and the weight of each criterion are described by linguistic terms which can be expressed in trapezoidal fuzzy numbers. In this paper, a vertex method is proposed to calculate the distance between two trapezoidal fuzzy numbers. According to the concept of the TOPSIS, a closeness coefficient is defined to determine the ranking order of all candidates. This research derived; 1) 4 evaluation criteria(research results, education and research competency, personality, major suitability) for new professor selection, 2) the 5 step procedure of the proposed fuzzy TOPSIS method for the group decision, 3) priorities of 4 candidates in the new professor selection case. The results of this paper will be useful to practical expert who is interested in analyzing fuzzy data and its multi-criteria decision-making tool for personal selection problem in personal management. Finally, the theoretical and practical implications of the findings were discussed and the directions for future research were suggested.

• Journal of Intelligence and Information Systems
• /
• v.10 no.3
• /
• pp.39-57
• /
• 2004

Many dairy producers periodically receive information about their bulk tank milk with reference to bulk tank somatic cell counts, standard plate counts, and preliminary incubation counts. This information, when collected over a period of time, in combination with bulk tank mastitis culture reports can become a significant knowledge base. Several guidelines have been proposed to interpret farm bulk tank milk bacterial counts. However many of the suggested interpretive criteria lack validation, and provide little insight to the interrelationship between different groups of bacteria found in bulk tank milk. Also the linguistic terms describing bulk tank milk quality or herd management status are rather vague or fuzzy such as excellent, good or unsatisfactory. The objective of this paper was to develop a set of fuzzy descriptors to evaluate bulk tank milk quality and herd's milking practice based on bulk tank milk microbiology test results. Thus, fuzzy logic based reasoning methodologies were developed based on fuzzy inference engine. Input parameters were bulk tank somatic cell counts, standard plate counts, preliminary incubation counts, laboratory pasteurization counts, non agalactiae-Streptococci and Streptococci like organisms, and Staphylococcus aureus. Based on the input data, bulk tank milk quality was classified as excellent, good, milk cooling problem, cleaning problem, environmental mastitis, or mixed with mastitis and cleaning problems. The results from fuzzy reasoning would provide a reference regarding a good management practice for milk producers, dairy health consultants, and veterinarians.

• Journal of Korea Society of Industrial Information Systems
• /
• v.10 no.1
• /
• pp.28-37
• /
• 2005

In our previous works, we suggested a method that extracts representative keywords from a few positive documents and assigns weights to them. To show the usefulness of the method, in this paper, we evaluate the performance of a famous classification algorithm called GIS(Generalized Instance Set) when it is combined with our method. In GIS algorithm, generalized instances are built from learning documents by a generalization function and then the K-NN algorithm is applied to them. Here, our method is used as a generalization function. For comparative works, Rocchio and Widrow-Hoff algorithms are also used as a generalization function. Experimental results show that our method is better than the others for the case that only positive documents are considered, but not when negative documents are considered together.

• Journal of the Korea Computer Industry Society
• /
• v.4 no.10
• /
• pp.587-598
• /
• 2003

This paper construct two-level word similarity relation matrices about title and to scientific treatise. As guide keyword similarity relation matrices which is constructed to co-occurrence frequency base same time keeps recall rater by query expansion by tolerance relation, it is index structure to improve the precision rate by two-level contents base retrieval. Therefore, draw area knowledge through subject analysis and reasoned user's information request and area knowledge to fuzzy logic base. This research is research to improve vocabulary mismatch problem and information expression having essentially on query.

• Journal of the Korean Society for information Management
• /
• v.18 no.2
• /
• pp.7-30
• /
• 2001

The characteristics of the short document representatives and short queries of OPAC systems need the different ranking algorithms from IR systems. This study tested and analyzed the effectiveness of four sorting schemes and four ranking algorithms and the six effectiveness measurements for the ranked Boolean OPAC systems. The sorting by publication year was better but without significant difference. The cover density ranking was significantly better than the frequency-based ranking of the Fuzzy or DNF models. The simple effectiveness measurement based on the average rank of relevant documents retrieved was as good as the others and better than the precision P.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2004.05a
• /
• pp.385-388
• /
• 2004

인공지능분야에서의 온톨로지란 기본적인 개념의 규정과 개념들 사이의 관계를 표현한 용어들의 분류(classification)를 의미한다. 온톨로지를 만들기 위해서는 많은 온톨로지 관련 언어가 있다. 그 중 최근의 연구 방향은 DAML+OIL과 OWL로 작성된 온톨로지를 이용한 추론, 인텔리전트 서비스 분야이다. 본 논문에서는 웹 상의 미술 작품 온톨로지에 대해 기존의 키워드 매칭 검색 대신에 추론엔진을 이용한 시맨틱 기반의 확장된 검색 방법을 소개한다. 향후 연구는 퍼지 개념을 도입하여 기존의 결과 보다 정확한 검색 결과를 얻기 위한 연구를 할 것이다.

• Journal of the Korea Society of Computer and Information
• /
• v.16 no.3
• /
• pp.17-25
• /
• 2011

Green IT, composed term with Green and Information Technology(IT), use IT for energy savings and carbon emission reductions. Green IT went beyond the scope of greening IT, and recently it's concept is expanded as far as counterplan of climate change including greening other industries by IT. 85% of total greenhouse gas emissions from the energy sector and 20% of them comes from transport parts, so it is time to research IT for automotive industry. In this paper, we take up the knowledge based fuzzy logic to provide life cycle analysis associated with greenhouse gas emissions for industry produced warranty claims frequently such as automobile industry. We propose a analysis method of warranty claims using expert knowledge about the warranty in car exhaust systems related to greenhouse gas emissions, past test results of malfunction, analysis of past field data, and warranty data. Furthermore, we propose life knowledge-based GWS (Greenery Warranty System). We demonstrate the applicability of IT in eco-friendly automotive industry by implementing knowledge-based fuzzy logic and applying.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.2
• /
• pp.142-149
• /
• 2004

It is important to retrieve exact information coinciding with user's need from lots of Usenet news and filter desired information quickly. Differently from email system, we must previously register our interesting news group if we want to get the news information. However, it is not easy for a novice to decide which news group is relevant to his or her interests. In this work, we present a service classifying user preferred news groups among various news groups by the use of Kohonen network. We first extract candidate terms from example documents and then choose a number of representative keywords to be used in Kohonen network from them through fuzzy inference. From the observation of training patterns, we could find the sparsity problem that lots of keywords in training patterns are empty. Thus, a new method to train neural network through reduction of unnecessary dimensions by the statistical coefficient of determination is proposed in this paper. Experimental results show that the proposed method is superior to the method using every dimension in terms of cluster overlap defined by using within cluster distance and between cluster distance.