• Proceedings of the KSPS conference
• /
• 1994.02a
• /
• pp.45-53
• /
• 1994

음 변화 규칙을 음성규칙(A)와 음운규칙(B)와 기타(C)로 분류한다면, A.음성규칙의 발생 총계는 B.음운규칙의 발생 총계보다 2배 이상 더 많다. 여기에 가장 발생 빈도가 높은 규칙부터 순서대로 10개만 정리하면 다음과 같다. 순위 규칙부류 명칭 백분율 1 A 음절 말 유성자음의 불파화 31.84, 2 A 유성음화 19.37 3 B 장모음화 10.00, 4 A 설측음화 9.78, 5 B 경음화 6.43, 6 B 음절말 장애음 중화 5.16, 7 A ㅅ 구개음화 3.71, 8 C 음절 조정 규칙 3.10, 9 B 단모음화 1.60, 10 A ㄴ 구개음화 1.42 별도로 진행한 총체적 연구(이상억에 의하면, 음성규칙, 즉 A가 총 6개 있는데 그 중 5규칙이 10위 내에 나타나 상위 쪽에 3번이나 포함되어 있는 점이 주목된다. 특히 규칙 1과 2의 합계는 전체 중 51.2%를 넘는다. 규칙 10위까지의 합계는 전체 중 92.4%를 차지한다. 13.장모음화(10%)는 12.단모음화(1.6%)보다 훨씬 높은 출현빈도를 보인다. 또 각종 구개음화 규칙의 순위는 ㅅ(7위), ㄴ(10위), ㄹ(17위), ㄷ(26위) 구개음화로 분포되어 있음이 밝혀졌다. 한편 삭제규칙(음운탈락 및 단순화) 계열이 14, 15, 19, 21, 25, 28위에 널리 나열되어 있는 데 반해, 삽입규칙은 23위에 ㄴ삽입 하나가 보일 뿐이다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.04b
• /
• pp.136-138
• /
• 2004

XML 이 웹 상에서의 정보 표현, 통합, 교환을 위한 표준이 됨에 따라 다양한 XML 질의 언어들이 제안되었으며, World Wide Web Consortium(W3C)은 XQery를 XML 질의 언어의 표준으로 권고하였다. XQuery는 SQL과 유사하게 중첩 질의를 허용하므로, 중첩된 XQuery 질의를 동일한 의미를 가지면서 보다 효율적으로 실행될 수 있는 질의로 변환하는 정규화 규칙들이 제안되었다. 그러나 제안된 정규화 규칙들은 제한적인 형태의 중첩 질의에만 적용되는 문제점을 가지고 있다 특히, FLWR 표현식의 where 절에 있는 중첩을 처리할 수 없다. 본 논문에서는 SQL 질의의 정규화 규칙들을 확장하여 FLWR 표현식의 모든 절에 나타나는 중첩을 처리할 수 있는 XQuery 질의의 정규화 규칙들을 제안한다 이를 위해 먼저, 상관과 집계의 유무에 따라 XQuery 질의의 중첩 유형을 분류하고, 각 유형 별로 정규화 규칙들을 제안한다 다음으로, 중첩된 XQuery 질의에 정규화 규칙들을 적용하는 세부 알고리즘을 제안한다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.20 no.4
• /
• pp.534-540
• /
• 2010

High dimensionality and highly correlated features are the major characteristics of hyperspectral data. Linear projections such as LDA and its variants have been used in extracting low-dimensional features from high-dimensional spectral data. Regularization of LDA has been introduced to alleviate the overfitting that often occurs in a small-sized training data set and leads to poor generalization performance. Among them, a smoothness regularized LDA seems to be effective in the feature extraction for hyperspectral data due to its capability of utilizing the high correlatedness. This paper studies the performance of the regularized LDA in hyperspectral data classification experimentally with varying conditions of the training data. In addition, a new dual smoothness regularized LDA is proposed and evaluated that makes use of both the spectral-domain and spatial-domain correlations between neighboring pixels.

• Proceedings of the ESK Conference
• /
• 1993.10a
• /
• pp.236-242
• /
• 1993

소비자들이 원하는 제품의 감성을 언어로 표현하여 줄 때, 이를 퍼지추론하여 칼라를 선택해주는 시스템을 개발하였다. 시스템은 감성언어입력, 감성언어퍼지화, 칼라추론, 추론규칙, 출력 등 모두 5개 의 모듈로 구성되어 있다. 시스템은 감성언어를 색상, 채도, 명도로 변환시킨 후 이를 다시 R, G, B 값으로 변환시키게 된다. 이때, 색상, 채도, 명도로의 변환에는 퍼지화규칙이 사용되게 되며, R,G,B 값으로의 변환 에는 칼라추론규칙이 사용되게 된다. 퍼지화규칙을 만들기 위해 S.D.(의미미분)법에 의한 감성언어의 요인 분석을 실시하였으며, 동시에 문헌조사를 통해 얻은 칼라와 관련한 감성정보를 if-then 규칙 형태로 시스템에 구현하였다.

• Journal of the Korean Data and Information Science Society
• /
• v.25 no.6
• /
• pp.1345-1352
• /
• 2014

Data mining is the representative methodology to obtain meaningful information in the era of big data.By Wikipedia, association rule learning is a popular and well researched method for discovering interesting relationship between itemsets in large databases using association thresholds. It is intended to identify strong rules discovered in databases using different interestingness measures. Unlike general association rule, inverse association rule mining finds the rules that a special item does not occur if an item does not occur. If two types of association rule can be simultaneously considered, we can obtain the marketing information for some related products as well as the information of specific product marketing. In this paper, we propose a balanced attributable relative accuracy applicable to these association rule techniques, and then check the three conditions of interestingness measures by Piatetsky-Shapiro (1991). The comparative studies with rule accuracy, relative accuracy, attributable relative accuracy, and balanced attributable relative accuracy are shown by numerical example. The results show that balanced attributable relative accuracy is better than any other accuracy measures.

• The Journal of Korean Association of Computer Education
• /
• v.21 no.1
• /
• pp.51-58
• /
• 2018

Recently artificial neural networks have shown excellent performance in various fields. However, there is a problem that it is difficult for a person to understand what is the knowledge that artificial neural network trained. One of the methods to solve these problems is an algorithm for extracting rules from trained neural network. In this paper, we extracted rules from artificial neural networks using ordered-attribute search(OAS) algorithm, which is one of the methods of extracting rules, and analyzed result to improve extracted rules. As a result, we have found that the distribution of output values of the hidden layer unit affects the accuracy of rules extracted by using OAS algorithm, and it is suggested that efficient rules can be extracted by binarizing hidden layer output values using hidden unit clarification.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2004.10a
• /
• pp.657-660
• /
• 2004

규칙기반의 상품추천시스템은 많은 인터넷 쇼핑몰에서 활용되고 있지만 규칙을 추출할 수 있는 마케팅 전문가 확보와 방대한 양의 고객 데이터 처리의 어려움으로 유용한 규칙을 찾는 것이 매우 어렵다. 본 연구에서는 이러한 규칙기반 상품추천시스템의 단점을 보완할 수 있는 방법으로 전역 최적화 기법의 하나인 유전자 알고리즘을 활용하여 고객정보를 토대로 추천 규칙을 도출할 수 있는 방안을 제시한다. 또한 본 연구에서 제안한 유전자 알고리즘에 기반한 추천 규칙들이 장착된 웹 기반의 개인화된 상품추천시스템의 프로토타입을 개발하고 이에 대한 실제 사용자들의 이용 만족도를 확인함으로써 본 연구에서 제안한 방법론의 유용성을 확인하고자 한다.

• Journal of the Korea Society of Computer and Information
• /
• v.16 no.4
• /
• pp.127-135
• /
• 2011

Efficient ordering of decision variables is one of the methods that find solutions quickly in the depth first search using backtracking. At this time, development of variables ordering algorithms considering dynamic and static properties of the problems is very important. However, to exploit optimal variable ordering algorithms appropriate to the problems. In this paper, we propose a problem classifying rule which provides problem type based on variables' properties, and use this rule to predict optimal type of variable ordering algorithms. We choose frequency allocation problem as a DS-type whose decision variables have dynamic and static properties, and estimate optimal variable ordering algorithm. We also show the usefulness of problem classifying rule by applying base station problem as a special case whose problem type is not generated from the presented rule.

• Journal of KIISE:Software and Applications
• /
• v.31 no.3
• /
• pp.369-378
• /
• 2004

In partially free word order languages like Korean and Japanese, the rule-based method is effective for text chunking, and shows the performance as high as machine learning methods even with a few rules due to the well-developed overt Postpositions and endings. However, it has no ability to handle the exceptions of the rules. Exception handling is an important work in natural language processing, and the exceptions can be efficiently processed in memory-based teaming. In this paper, we propose a hybrid of rule-based method and memory-based learning for Korean text chunking. The proposed method is primarily based on the rules, and then the chunks estimated by the rules are verified by memory-based classifier. An evaluation of the proposed method on Korean STEP 2000 corpus yields the improvement in F-score over the rules or various machine teaming methods alone. The final F-score is 94.19, while those of the rules and SVMs, the best machine learning method for this task, are just 91.87 and 92.54 respectively.

• Journal of KIISE:Computing Practices and Letters
• /
• v.10 no.5
• /
• pp.419-433
• /
• 2004

As XML becomes a standard for data representation, integration, and exchange on the Web, several XML query languages have been proposed. World Wide Web Consortium(W3C) has proposed XQuery as a standard for the XML query language. Like SQL, XQuery allows nested queries. Thus, normalization rules have been proposed to transform nested XQuery queries to semantically equivalent ones that could be executed more efficiently. However, previous normalization rules are applicable only to restricted forms of nested XQuery queries. Specifically, they can not handle FLWR expressions having nested expressions in the where clause. In this paper, we propose normalization rules for XQuery queries by extending those for SQL queries. Our proposed rules can handle FLWR expressions haying nested expressions in every clause. The major contributions of this paper are as follows. First, we classily nesting types of XQuery queries according to the existence of correlation and aggregation. We then propose normalization rules for each nesting type. Second, we propose detailed algorithms that apply the normalization rules to nested XQuery queries.