• 한국지능정보시스템학회:학술대회논문집
• /
• 한국지능정보시스템학회 2001년도 The Pacific Aisan Confrence On Intelligent Systems 2001
• /
• pp.336-342
• /
• 2001

One of the most important problems on rule induction methods is that they cannot extract rules, which plausibly represent experts decision processes. On one hand, rule induction methods induce probabilistic rules, the description length of which is too short, compared with the experts rules. On the other hand, construction of Bayesian networks generates too lengthy rules. In this paper, the characteristics of experts rules are closely examined and a new approach to extract plausible rules is introduced, which consists of the following three procedures. First, the characterization of decision attributes (given classes) is extracted from databases and the classes are classified into several groups with respect to the characterization. Then, two kinds of sub-rules, characterization rules for each group and discrimination rules for each class in the group are induced. Finally, those two parts are integrated into one rule for each decision attribute. The proposed method was evaluated on a medical database, the experimental results of which show that induced rules correctly represent experts decision processes.

• Communications for Statistical Applications and Methods
• /
• 제13권3호
• /
• pp.701-718
• /
• 2006

In cancer microarray experiments, the experimenter or patient which is nested in each experimenter often shows quite heterogeneous error variability, which should be estimated for identifying a source of variation. Our study describes a Bayesian method which utilizes clinical information for identifying a set of DE genes for the class of subtypes as well as assesses and examines the experimenter effect and patient effect which is nested in each experimenter as a source of variation. We propose a Bayesian multilevel mixed effect model based on analysis of covariance (ANACOVA). The Bayesian multilevel mixed effect model is a combination of the multilevel mixed effect model and the Bayesian hierarchical model, which provides a flexible way of defining a suitable correlation structure among genes.

• Asia pacific journal of information systems
• /
• 제6권1호
• /
• pp.223-240
• /
• 1996

The paper proposes an approach, called OPNets, for modeling and validating manufacturing systems that is based on object-oriented high-level Petri nets. In OPNets, modeling components of Petri net are constructed into hierarchical objects that communicate with each other by passing messages. To enhance the reusability and maintainability, OPNets organizes a system into hierarchical objects that inherit attributes and behavioral properties from the object of super class and object-interaction relations are separated from the internal structure of object. The modeling scheme of OPNets tries to resolve the complexity problems of Petri net. To illustrate the modeling schemes of OPNets, a storage/retrieval example has been proposed.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2002년도 합동 추계학술대회 논문집 정보 및 제어부문
• /
• pp.521-524
• /
• 2002

This paper deals with the rule generation from data for control system and data mining using rough set. If the cores and reducts are searched for without consideration of the frequency of data belonging to the same equivalent class, the unnecessary attributes may not be discarded, and the resultant rules don't represent well the characteristics of the data. To improve this, we handle the inconsistent data with a probability measure defined by support, As a result the effect of uncertainty in knowledge reduction can be reduced to some extent. Also we construct the rule base in a hierarchical structure by applying core as the classification criteria at each level. If more than one core exist, the coverage degree is used to select an appropriate one among then to increase the classification rate. The proposed method gives more proper and effective rule base in compatibility and size. For some data mining example the simulations are performed to show the effectiveness of the proposed method.

• 한국대기환경학회지
• /
• 제9권3호
• /
• pp.207-215
• /
• 1993

In order to statistically predict $O_3$ levels in Seoul, the study used the TMS (telemeted air monitoring system) data from the Department of Environment, which have monitored at 20 sites in 1989 and 1990. Each data in each site was characterized by 6 major criteria pollutants ($SO_2, TSP, CO, NO_2, THC, and O_3$) and 2 meteorological parameters, such as wind speed and wind direction. To select proper variables and to determine each pollutant's behavior, univariate statistical analyses were extensively studied in the beginning, and then various applied statistical techniques like cluster analysis, regression analysis, and expert system have been intensively examined. For the initial study of high level $O_3$ prediction, the raw data set in each site was separated into 2 group based on 60 ppb $O_3$ level. A hierarchical cluster analysis was applied to classify the group based on 60 ppb $O_3$ into small calsses. Each class in each site has its own pattern. Next, multiple regression for each class was repeatedly applied to determine an $O_3$ prediction submodel and to determine outliers in each class based on a certain level of standardized redisual. Thus, a prediction submodel for each homogeneous class could be obtained. The study was extended to model $O_3$ prediction for both on-time basis and 1-hr after basis. Finally, an expect system was used to build a unified classification rule based on examples of the homogenous classes for all of sites. Thus, a concept of high level $O_3$ prediction model was developed for one of $O_3$ alert systems.

• 한국생활과학회지
• /
• 제13권6호
• /
• pp.833-845
• /
• 2004

This study examines different individual and environmental factors that affect children's self-control. For an analysis, locus of control, perceived competence, and achievement motivation were all included in individual variables. For family variables, mothers' parenting and patents' marriage conflict were examined. For classroom psycho-social environment, teacher support, peer relationship, class involvement, and teachers' supervision were used. The sample consisted of 548 fifth and sixth grade children. Statistics and methods used for the data analysis were Cronbach's alpha, frequency, percentage, Pearson's correlation, and Hierarchical Regression. Several major results were found from the analysis: First, locus of control, perceived competence, and achievement motivation had a positive correlation with children's self-control. Second, mothers' affective parenting had a positive correlation with children's self-control. However, mothers' controlling parenting and parents' marriage conflict had a negative correlation with it. Third, teacher support, peer relationship, and class involvement had a positive correlation with children's self-control. In addition, teacher supervision had a positive correlation with girls' self-control. Fourth, class involvement, locus of control, and academic competence were important variables predicting boys' self-control. On the other hand, Class involvement, achievement motivation, academic competence, teacher's supervision, and mothers' controlling parenting were important variables predicting girl's self-control.

• 지능정보연구
• /
• 제17권1호
• /
• pp.111-125
• /
• 2011

클러스터링(Clustering)은 유사한 문서나 데이터를 묶어 군집화해주는 프로세스이다. 클러스터링은 문서들을 대표하는 개념별로 그룹화함으로써 사용자가 자신이 원하는 주제의 문서를 찾기 위해 모든 문서를 검사할 필요가 없도록 도와준다. 이를 위해 유사한 문서를 찾아 그룹화하고, 이 그룹의 대표되는 개념을 도출하여 표현해주는 기법이 요구된다. 이 상황에서 문제점으로 대두되는 것이 복합 개념(Complex Concept)의 탐지이다. 복합 개념은 서로 다른 개념의 여러 클러스터에 속하는 중복 개념이다. 기존의 클러스터링 방법으로는 문서를 클러스터링할 때 동일한 레벨에 있는 서로 다른 개념의 클러스터에 속하는 중복된 복합 개념의 클러스터를 찾아서 표현할 수가 없었고, 또한 복합 개념과 각 단순 개념(Simple Concept) 사이의 의미적 계층 관계를 제대로 검증하기가 어려웠다. 본 논문에서는 기존 클러스터링 방법의 문제점을 해결하여 복합 개념을 쉽게 찾아 표현하는 방법을 제안한다. 기존의 계층적 클러스터링 알고리즘을 변형하여 동일 레벨에서 중복을 허용하는 계층적 클러스터링(Hierarchical Overlapping Clustering, HOC) 알고리즘을 개발하였다. HOC 알고리즘은 문서를 클러스터링하여 그 결과를 트리가 아닌 개념 중복이 가능한 Lattice 계층 구조로 표현함으로써 이를 통해 여러 개념이 중복된 복합 개념을 탐지할 수 있었다. HOC 알고리즘을 이용해 생성된 각 클러스터의 개념이 제대로 된 의미적인 계층 관계로 표현되었는지는 특징 선택(Feature Selection) 방법을 적용하여 검증하였다.

• 한국대기환경학회지
• /
• 제22권6호
• /
• pp.808-820
• /
• 2006

In regard to indoor air quality patterns, the government introduced various polices that were about managing and monitoring quality of indoor air as a major assignment, and also executed 'Indoor Air Quality Management Act' which was presented in the May, 2004. However, among the multi-usage facilities controlled by the Act, the school was not included yet. This study goal was to investigate PM 10 pollution patterns of the high school classrooms using a pattern recognition method based on cluster analysis and disjoint principal component analysis, and further to survey levels of inorganic elements in May, June, and September, 2004. A hierarchical clustering method was examined to obtain possible objects in pseudo homogeneous sample classes by transformation raw data and by applying various distance. Following the analysis, the disjoint principal component analysis was used to define homogeneous sample class after deleting outliers. Then three homogeneous Patterns were obtained as follows: the first class had been separated and objects in the class were considered to be sampled under semi-open condition. This class had high concentration of Ca, Fe, Mg, K, Al, and Na which are related with a soil and a chalk compounds. The second class was obtained in which objects were sampled while working air-conditioners and was identified low concentration of PM 10 and elements. Objects in the last class were assigned during rainy day. A chalk, soil element and various types of anthropogenic sources including combustions and industrial influenced the third class. This methodology was thought to be helpful enough to classify indoor air quality patterns and indoor environmental categories when controlling an indoor air quality.

• 대한원격탐사학회지
• /
• 제27권4호
• /
• pp.435-444
• /
• 2011

고해상도 위성영상의 다중분광자료만을 이용하여 토지 피복도를 제작할 경우, 낮은 분광해상도와 단일 토지 피복 내에 존재하는 불균질성으로 인해 분류 결과의 정확도가 저하되는 문제가 발생한다. 특히 식생 클래스의 경우 단일 토지 피복임에도 불구하고 절감 특성에 따라 해당 영역 안에 산림, 초지, 농업지역 등이 함께 분류되는 문제가 두드러진다. 본 연구에서는 이러한 문제를 개선하기 위해 광학 영상 기반의 사전분류를 수행한 후 식생으로 분류된 영역에 대해 고해상도 위성영상의 다중분광정보와 SAR 영상 산란 정보를 통합하고 식생을 세분류하였다. 사전 분류와 식생분류는 최대우도 감독분류를 통해 수행되었으며 식생 세분류 결과와 사전 분류결과 중 비식생 클래스의 융합을 통해 계층적 분류 방법을 제안하였다. 제안 기법은 SAR 영상이나 GLCM 질감 정보를 영상 전체에 걸쳐 단순 통합한 분류결과뿐만 아니라 GLCM 질감 정보를 식생 지역에 적용한 계층적 분류결과에 비해 높은 정확도를 보였으며 특히 식생과 비식생의 분류 정확도가 모두 높게 나타났다.

• 한국컴퓨터정보학회논문지
• /
• 제11권4호
• /
• pp.243-250
• /
• 2006

순환관계란 동일 클래스 내의 엔티티 간의 관계를 말하는데, 이중 N:M 순환관계는 자재소요명세서 구조를 기술하는데 사용할 수 있다. 자재소요명세서란 제조분야에서 자주 쓰이는 것으로 계층형의 특수 데이터 구조로 되어있다. 비즈니스 차원은 거의 대부분 계층구조로 되어있다. 본 연구에는 자재소요명세서 유형의 계층차원을 다음과 같이 설계하고 있다. 먼저 일반적인 N:M 순환관계에서와 마찬가지로 교차 테이블을 만든 다음 이를 OLAP 모델에서의 차원으로 변환시킨다. 즉 교차 테이블의 첫 번째 컬럼은 이 차원의 가장 낮은 수준으로, 두 번째 컬럼은 이 차원의 유일한 상위 수준으로 변환시키는 것이다. 이렇게 설계한 차원을 사용한 다차원 사례 정보시스템도 아울러 구축하고 있다.