http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
A Kernel Approach to Discriminant Analysis for Binary Classification
신양규,Shin, Yang-Kyu The Korean Data and Information Science Society 2001 한국데이터정보과학회지 Vol.12 No.2
We investigate a kernel approach to discriminant analysis for binary classification as a machine learning point of view. Our view of the kernel approach follows support vector method which is one of the most promising techniques in the area of machine learning. As usual discriminant analysis, the kernel method can discriminate an object most likely belongs to. Moreover, it has some advantage over discriminant analysis such as data compression and computing time.
신양규,Shin, Yang-Kyu 한국데이터정보과학회 1995 한국데이터정보과학회지 Vol.6 No.2
We propose a framework for representing and processing uncertain knowledge on the basis of constraint satisfaction. A system of equations and/or inequalities can be considered as a set of constraints that should be solved, and each constraint in the set is transformed into a corresponding logical formula which can be solved through a constraint solving program. Most of rule-based systems, for instance, use a simple probabilistic theory in order to maintain uncertain knowledge, therefore uncertain knowledge can be represented and processed in the constraint satisfaction program quite efficiently.
Bootstrap Confidence Cones for Spherical Data
신양규,Shin, Yang-Kyu Korean Data and Information Science Society 1992 한국데이터정보과학회지 Vol.3 No.1
The set of eigenvectors of the second moment matrix and the mean vector are the measures of orientation for a distribution supported on the unit sphere. Bootstrap confidence cone for the eigenvector is constructed and the consistency of this method is discussed. The performance of our bootstrap cone for the eigenvector is compared with that of the asymptotic confidence cones for two measures under the parametric assumptions for the underlying distributions and that of the bootstrap cone for the mean vector by Monte Carlo simulation.
신양규,Shin, Yang-Kyu 한국데이터정보과학회 1997 한국데이터정보과학회지 Vol.8 No.2
In oriental medicine, it is required that correct medical knowledge should be maintained for medical expert system which analyzes and diagnoses patients symptoms. Typical medical expert system has a knowledge base as its core, and the knowledge base contains a domain specific knowledge about patients records. However, oriental medicine diagnostic knowledge is formed mostly as qualitative data, knowledge could be ambiguous and uncertain. In this paper, we looked at quantification methods and propose a method for quantifying the oriental medicine diagnostic knowledge, which is improving the knowledge base of an oriental medicine expert system.
한의 진단 모델의 추론 과정에서 발생하는 불확실한 진단 지식의 처리
신양규,Shin, Yang-Kyu 한국데이터정보과학회 1997 한국데이터정보과학회지 Vol.8 No.1
The inference process for medical expert system is mostly formed by diagnostic knowledge on the if-then rule base. Oriental medicine diagnostic knowledge, however, may involve uncertain knowledge caused by ambiguous concept. In this paper, we analyze an oriental medicine diagnostic process by a rule-based inference system, and propose a method for representing and processing uncertain oriental medicine diagnostic knowledge using CLP( R ) which is a kind of constraint satisfaction program.
Bootstrap Confldence Cones for Spherical Data
신양규(Yang Kyu Shin) 한국데이터정보과학회 1992 한국데이터정보과학회지 Vol.3 No.1
The set of eigenvectors of the second moment matrix and the mean vector are the measures of orientation for a distribution supported on the unit sphere. Bootstrap confidence cone for the eigenvector is constructed and the consistency of this method is discussed. The performance of our bootstrap cone for the eigenvector is compared with that of the asymptotic confidence cones for two measures under the parametric assumptions for the underlying distributions and that of the bootstrap cone for the mean vector by Monte Carlo simulation.
김진상,신양규,Kim, Jin-Sang,Shin, Yang-Kyu 한국데이터정보과학회 1999 한국데이터정보과학회지 Vol.10 No.2
본 논문에서는 의료용 전문가 시스템에 사용 가능한 의료지식을 수리논리적으로 표현하고 이에 대한 연역 및 진단추론을 행하는 방법을 제시하였다. 문제해결을 위해서는 연역추론을 행하며 원인의 규명을 위해서는 진단추론을 행하지만 일차논리 언어로 표현된 의료지식에서 두 종류의 추론을 병행할 수 있다. 그리고 의료지식에 자주 발생하는 시간에 따라 가변적인 결과의 추론방법도 함께 고찰하였다. We investigate a logical approach to represent medical knowledge, reason deductively and diagnostically. It is suggested that medical knowledge-bases can be formulated as a set of sentences stated in classical logic where each sentence reflects a doctor's knowledge about the human anatomy or his/her view of patient's symptoms. It is also suggested that a form of temporal reasoning can be captured within the same framework because each sentence can have a different truth value based on time. We apply our logical framework to formalize diagnostic reasoning, where the primary cause of illness is chosen among the set of minimal causation on the basis of abductive hypotheses. Most of our examples are given in the context of medical expert systems.
김진상,신양규,Kim, Jin-Sang,Shin, Yang-Kyu 한국데이터정보과학회 2001 한국데이터정보과학회지 Vol.12 No.1
본 논문은 논리문장으로 표현된 지식을 처리하는 정리증명 과정에서 증명이 완료되기 전에 잠정적 결론을 유도하는 확률추론 기법을 제시한다. 정리증명 과정 중에 베이지안 해석을 이용하여 지식을 갱신하는 방법을 제시하고, 의사결정 방법을 사용하여 시간에 민감한 사안에 대해 신속하게 대처할 것인지 아니면 고의로 미룰 것인지를 결정하는 방법을 밝힌다. We present a probabilistic reasoning method for inferring knowledge about mathematical truth before an automated theorem prover completes a proof. We use a Bayesian analysis to update beleif in truth, given theorem-proving progress, and show how decision-theoretic methods can be used to determine the value of continuing to deliberate versus taking immediate action in time-critical situations.
김진상,신양규,Kim, Jin-Sang,Shin, Yang-Kyu 한국데이터정보과학회 2000 한국데이터정보과학회지 Vol.11 No.1
As the number of online documents increases enormously with the expansion of information technology, the importance of automatic document classification is greatly enlarged. In this paper, an automatic document classification method is investigated and applied to UseNet 20 newsgroup articles to test its efficacy. The classification system uses Naive Bayes classification algorithm and the experimental result shows that a randomly selected newsgroup arcicle can be classified into its own category over 77% accuracy. 정보통신기술의 비약적인 발전은 온라인으로 생성되는 전자문서의 양을 폭발적으로 증가시키고 있다. 따라서 수동으로 문서를 분류하던 종래의 방법 대신 문서의 자동분유 기술 개발이 특별히 요구되고 있다. 본 논문에서는 베이지안 학습 기법을 이용하여 문서를 자동으로 분류하는 방법을 연구하고, 20개의 유즈넷 뉴스그룹 문서들을 분류하도록 시험하였다. 사용한 알고리즘은 Naive Bayes Classifier이며, 구현한 시스템을 이용해 유즈넷 문서를 대상으로 자동분류를 실험한 결과 분류의 정확률이 약 77%로 나타났다.