• 한국전자거래학회지
• /
• 제4권1호
• /
• pp.21-36
• /
• 1999

Using user monitoring techniques on web, marketing decision makers in cyber shopping malls can gather customer behavior data as well as sales transaction data and customer profiles. In this paper, we present a marketing rule extraction technique for customer behavior analysis in cyber shopping malls, The technique is an application of market basket analysis which is a representative data mining technique for extracting association rules. The market basket analysis technique is applied on a customer behavior log table, which provide association rules about web pages in a cyber shopping mall. The extracted association rules can be used for mall layout design, product packaging, web page link design, and product recommendation. A prototype cyber shopping mall with customer monitoring features and a customer behavior analysis algorithm is implemented using Java Web Server, Servlet, JDBC(Java Database Connectivity), and relational database on windows NT.

• 한국중재학회지:중재연구
• /
• 제20권2호
• /
• pp.67-90
• /
• 2010

Electronic discovery refers to the discovery of electronically stored information. The differences between producing paper documents and electronic information can be categorized into seven groups: massive volume, persistence, dynamic and changeable contents, metadata, environment-dependence, dispersion and searchability. Since these differences make the discovery more expensive and less expeditious, it is necessary to limit the scope of discovery. Accordingly, a number of arbitration institutions have already introduced rules, guidelines or protocols on electronic discovery. ICDR guidelines take a minimal approach and address only the proper form of electronic document. CIArb Protocol is intended to act as a checklist for discovery of electronic data. CPR Protocol offers four modes of discovery of electronic documents ranging from minimal to extensive among which the parties may choose the way of electronic discovery. IBA Rules on Evidence and ICC Rules are silent on the issue of electronic discovery, however, working parties of the ICC are considering updates to the rules to deal with electronic discovery. It is disputed whether rules, guidelines or protocols on electronic discovery is necessary or appropriate. Although some have suggested that existing rules can make adequate provision for electronic discovery, it is more desirable to prepare new rules, guidelines or protocols to make arbitrators and counsels be familiar with electronic discovery process, to provide an adequate standard for electronic discovery and to limit the time and cost of electronic discovery. Such rules on electronic discovery should include provisions regarding the form of electronic document production, conference between parties regarding electronic discovery, keyword search, bearing the expenses to reduce disputes over electronic discovery.

• 한국산업보건학회지
• /
• 제16권1호
• /
• pp.11-16
• /
• 2006

There are many counting rules for analyzing man-made mineral fibers. The representatives are the NIOSH Method 7400 A and B counting rules. The two rules have different rules of length-to-width ratio(aspect ratio) and diameter. The A rule counts only fibers $>5{\mu}m$ in length, and only fibers with aspect ratio >3:1. The B rule counts only ends of fibers $>5{\mu}m$ in length and $<3{\mu}m$ in diameter, and only fibers with aspect ratio ${\geq}5:1$. The A counting rule had been used before the B counting rule was introduced. The purpose of this study is to compare the A and B counting rules for airborne fibers from various man-made mineral fibers(glass wool fibers, rock wool fibers, refractory ceramic fibers, and continuous filament glass fibers) industries. There were significantly differences between the paired counts of A and B rules in all types of fibers(p<0.05). A rule counts/B rule counts(A/B ratios) were 1.52 for glass fibers, 1.53 for rock wool fibers, 1.19 for RCF, and 1.82 for continuous filament glass fibers. The counting results by A and B counting rules were highly correlated in glass wool fibers, rock wool fibers and refractory ceramic fibers(RCF) samples (r=0.96 for all types of fibers) except continuous filament glass fibers(r=0.82). Regression equations to correct for the differences between counting rules were presented in this paper.

• 한국중재학회지:중재연구
• /
• 제16권1호
• /
• pp.121-151
• /
• 2006

As globalisation extends its effect and particularly following China's accession to the World Trade Organization(WTO) in 2001, ever greater numbers of international transactions will feature a Chinese party. China has certainly made efforts in recent years to rectify law problem. While conducting business in China, foreign companies occasionally find themselves embroiled in disputes with Chinese individuals and companies. As foreign businesses invest in the extraordinary market opportunities in China, international arbitration has also become the preferred method for handling disputes with Chinese partners or with other foreign corporation over operations in China. The new Arbitration Rules of the International Economic and Trade Arbitration Commission(CIETAC) came into force on 1 May 2005. The new rules represent a major overhaul of CIETAC arbitration procedures and are sure to enhance CIETAC's position as a leading player in the resolution of China-foreign business disputes. The changes are significant for all companies doing business in China. So, this article investigated some amendments on the basis of 2000 Rules.

• Korean Journal of Hydrosciences
• /
• 제4권
• /
• pp.65-79
• /
• 1993

Selection of the best operating rule among a set of alternatives for a multipurpose reservoir system operation requires to evaluate many minor criteria in addition to the major objectives assessed to the system. These problems are sufficiently complex and difficult that they are beyond heuristic decision rules and experiences in case several noncommensurable multiple criteria are included in the evaluation. With the assistance of multicriterion decision analysis techniques, it is possible to select the best one among various alternatives by systematically comparing and ranking the alternatives with respect to the criteria of choice. Evaluation criteria for multipurpose reservoir system operating rules were identified and defined, and the multicriterion decision analysis techniques were applied to evaluate the four existing operating rules of the Chungju multipurpose project according to the identified nine multiple criteria. The application results show that the methodology is very efficient to select the best operation alternative among a finite number of operating rules with many evaluation criteria for a large-scale reservoir system operation.

• 한국중재학회지:중재연구
• /
• 제8권1호
• /
• pp.353-375
• /
• 1998

This study is focused on the rules for resolving documentary credits disputes. First, International Chamber of Commerce published Documentary Credit Dispute Expertise Rules on October 1, 1997. The DOCDEX Rules are the International Chamber of Commerce(ICC) response to a clear call from the international banking community for a rapid, cost effective, expert-based dispute resolution mechanism for documentary credit practice, including bank-to-bank reimbursement issues. Next, The International Center for Letter of Credit Arbitration was established in September 1996. The Center was founded as a result of an initiative from within the letter of credit community and has been co-sponsored by the United States Council on International Banking(USCIB) and the Institute of International Banking Law and Practice Inc. In September, ICLOCA adopted its "Rules of Arbitration for Letter of Credit Disputes." Therefore, parties to letter of credit disputes should choose a appropriate dispute resolution mechanism under the circumstances in the future.

• 지능정보연구
• /
• 제19권2호
• /
• pp.39-54
• /
• 2013

전자상거래의 폭발적 증가는 소비자에게 더 유리한 많은 구매 선택의 기회를 제공한다. 이러한 상황에서 자신의 구매의사결정에 대한 확신이 부족한 소비자들은 의사결정 절차를 간소화하고 효과적인 의사결정을 위해 추천을 받아들인다. 온라인 상점의 상품추천시스템은 일대일 마케팅의 대표적 실현수단으로써의 가치를 인정받고 있다. 그러나 사용자의 기호를 제대로 반영하지 못하는 추천시스템은 사용자의 실망과 시간낭비를 발생시킨다. 본 연구에서는 정확한 사용자의 기호 반영을 통한 추천기법의 정교화를 위해 데이터마이닝과 다중모형조합기법을 이용한 상품추천시스템 모형을 제안하고자 한다. 본 연구에서 제안하는 모형은 크게 두 개의 단계로 이루어져 있으며, 첫 번째 단계에서는 상품군 별 우량고객 선정 규칙을 도출하기 위해서 로지스틱 회귀분석 모형, 의사결정나무 모형, 인공신경망 모형을 구축한 후 다중모형조합기법인 Bagging과 Bumping의 개념을 이용하여 세 가지 모형의 결과를 조합한다. 두 번째 단계에서는 상품군 별 연관관계에 관한 규칙을 추출하기 위하여 장바구니분석을 활용한다. 상기의 두 단계를 통하여 상품군 별로 구매가능성이 높은 우량고객을 선정하여 그 고객에게 관심을 가질만한 같은 상품군 또는 다른 상품군 내의 다른 상품을 추천하게 된다. 제안하는 상품추천시스템은 실제 운영 중인 온라인 상점인 'I아트샵'의 데이터를 이용하여 프로토타입을 구축하였고 실제 소비자에 대한 적용가능성을 확인하였다. 제안하는 모형의 유용성을 검증하기 위하여 제안 상품추천시스템의 추천과 임의 추천을 통한 추천의 결과를 사용자에게 제시하고 제안된 추천에 대한 만족도를 조사한 후 대응표본 T검정을 수행하였으며, 그 결과 사용자의 만족도를 유의하게 향상시키는 것으로 나타났다.

• 대한안전경영과학회:학술대회논문집
• /
• 대한안전경영과학회 2000년도 추계학술발표논문집
• /
• pp.299-306
• /
• 2000

General definition of data mining is the knowledge discovery or is to extract hidden necessary information from large databases. Its technique can be applied into decision making, prediction, and information analysis through analyzing of relationship and pattern among data. One of the most important work is to find association rules in data mining. The objective of this paper is to find customer's trend using association rule from analysis of database and the result can be used as fundamental data for CRM(Customer Relationship Management). This paper uses Apriori algorithm and FoodMart data in order to find association rules.

• 지능정보연구
• /
• 제9권1호
• /
• pp.195-209
• /
• 2003

데이터마이닝의 목적은 알려지지 않은 유용한 정보를 얻어 경영에 활용하려는 것이다. 데이터마이닝의 한 기법인 연관규칙도 마찬가지의 목적을 가지고 있다. 하지만 연관규칙 기법으로 생성된 결과는 모두 같기 때문에 타겟 마케팅에 적용 시 그 마케팅 전략은 같을 수밖에 없다. 본 논문에서는 연관규칙을 이용하여 타겟 마케팅에 적용 시에 데이터마이닝의 목적에 부합하는 새로운 규칙을 제안한다. 새로운 규칙이란 연관규칙과 같이 고객이 구매한 항목에 대해 관심을 가져 구매 항목간의 규칙을 생성하는 것뿐만 아니라 구매하지 않은 항목에 대해서도 관심을 가짐으로써 구매하지 않은 항목간의 규칙을 생성하여 타겟 마케팅에 필요한 정보를 생성하는 것을 말한다. 이러한 정보를 이용하면 타겟 항목을 바로 마케팅 하는 전략뿐만 아니라 타겟 항목을 판매하기 위하여 다른 항목을 마케팅 하는 전략도 가능하게 된다. 그 이유는 새로운 규칙 에 의해 생성된 다른 항목의 마케팅이 타겟 항목의 판매를 촉진시키기 때문이다. 본 논문에서는 타겟 마케팅에 적용하기 위한 새로운 규칙과 전략을 설명하고 예제를 통하여 실제 데이터베이스에서 기존 전략 외에 새로운 전략을 생성시키는 과정을 설명한다.

• 한국시뮬레이션학회논문지
• /
• 제19권4호
• /
• pp.59-66
• /
• 2010

이 논문에서는 온톨로지로 표현한 트랜잭션으로부터 연관 규칙을 생성하고 이를 기반으로 추론을 수행하여 개인화 의류 추천을 제공하는 시스템을 제안한다. Onto-Apriori 알고리즘을 이용한 연관 규칙 생성은 유행에 따른 구매성향 변동을 능동적으로 분석할 수 있다. 생성된 규칙은 온톨로지에 메타 노드로 표현하고 이를 기반으로 추론함으로써 사용자의 질의에 맞는 추천 항목을 찾아낼 수 있다. 시스템을 평가하기 위하여 추론 소요시간과 추천 정확도 2가지 요소를 기준으로 시뮬레이션을 수행하여 유효성을 증명하였다.