• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2003년도 컴퓨터소사이어티 추계학술대회논문집
• /
• pp.255-258
• /
• 2003

This paper defines Normalized Term Frequency Weighting method for automatic text categorization by using Box-Cox, and then it applies automatic text categorization. Box-Cox transformation is statistical transformation method which makes normalized data. This paper applies that and suggests new term frequency weighting method. Because Normalized Term Frequency is different from every term compared by existing term frequency weighting method, it is general method more than fixed weighting method such as log or root. Normalized term frequency weighting method's reasonability has been proved though experiments, used 8000 newspapers divided in 4 groups, which resulted high categorization correctness in all cases.

• 한국정보처리학회논문지
• /
• 제7권4호
• /
• pp.1150-1159
• /
• 2000

This paper presents an automatic text categorization model that improves the accuracy by combining statistical and knowledge-based categorization methods. In our model we apply knowledge-based method first, and then apply statistical method on the text which are not categorized by knowledge-based method. By using this combined method, we can improve the accuracy of categorization while categorize all the texts without failure. For statistical categorization, the vector model with Inverted Category Frequency (ICF) weighting is used. For knowledge-based categorization, Phrasal Patterns and Keyword Sets are introduced to represent sentence patterns, and then pattern matching is performed. Experimental results on new articles show that the accuracy of categorization can be improved by combining the tow different categorization methods.

• 한국언어정보학회지:언어와정보
• /
• 제13권1호
• /
• pp.1-20
• /
• 2009

Summarization is a complex cognitive task and its simulation is very difficult for machines. This paper presents an automatic summarization strategy that is based on a discourse categorization of the textual information. This categorization is carried out by the automatic identification of discourse markers in texts. We defend here the use of discourse methods in automatic summarization. Two evaluations of the summarization strategy are presented. The summaries produced by our strategy are evaluated with summaries produced by humans and other applications. These two evaluations display well the capacity of our application, based on EXCOM, to produce summaries comparable to the summaries of other applications.

• 한국도서관정보학회지
• /
• 제32권3호
• /
• pp.307-330
• /
• 2001

본 연구에서는 kNN분류기를 이용한 범주화 방법에 대한 성능 실험을 하였다. kNN분류기와 같은 대부분의 예제기반 자동 분류기법은 학습문서집단의 자질을 축소하게 되는데 자질을 몇 퍼센트 축소함으로써 높은 성능을 얻을 수 있는지를 알아보고자 하였다. 또한, kNN분류기는 학습문서집단에서 검증문서와 가장 유사한 k개의 학습문서를 찾아야 하는데, 이때 가장 적합한 k값은 얼마인지를 실험을 통하여 검증하여 보고자 하였다.

• 정보관리연구
• /
• 제35권4호
• /
• pp.81-105
• /
• 2004

본 연구에서는 웹기반 질의문을 자동 범주화하는 방안에 대해 조사하였다. 질의문 범주화에 대한 실험은 SVM-light를 사용하여 범주자질로써 유사어 사전을 부여하기 전과 후를 비교하였다. 유사어는 학습을 통해 수작업으로 대상문서에서 713개를 추출하였다. 전체적으로 유사어 부여전과 부여후의 결과는 6개 범주에서 정도율은 -0.01%로 거의 변화가 없었으며 재현율은 8.53%가 향상되었다. F1-Measure 값도 4.58%가 향상되었다. 특히 범주내 정도율과 재현율의 표준편차가 18.39%나 개선되어 적정한 검색효율을 확보할 수 있었다.

• 한국정보처리학회논문지
• /
• 제7권9호
• /
• pp.2867-2875
• /
• 2000

본 논문은 인터넷 문서 자동 분류 시스템의 구현에 대하여 논한다. 문서 자동분류 알고리즘을 설정하고, 역전파 학습 모델을 이용하여 문서의 범주화를 수행하는 시스템을 구축한다. 문서학습을 위해서 범주별 인터넷 문서들을 수집하고 수집한 문서에 대하여 카이제곱($\chi^2$)검정을 수행함으로써 범주화 자질을 추출한다. 이 범주화 자질을 바탕으로 하여 학습 및 분류 벡터 집합을 생성한다. 실험 결과의 평가로부터 본 논문에서 구현한 시스템이 유사도 계산을 이용한 문서의 분류 시스템보다 성능이 향상된 것을 알 수 있었다.

• 정보관리연구
• /
• 제33권2호
• /
• pp.19-32
• /
• 2002

최근 디지털 도서관이 등장하고 인터넷이 폭 넓게 보급되어 온라인 상에서 얻을 수 있는 텍스트 정보의 양이 급증함에 따라 효율적인 정보 관리 및 검색이 요구되고 있다. 자동 문서 범주화란 문서의 내용에 기반하여 미리 정의되어 있는 범주에 문서를 자동으로 할당하는 작업으로써 효율적인 정보 관리 및 검색을 가능하게 하는 동시에 방대한 양의 수작업을 감소시키는데 그 목적이 있다. 문서 분류를 위해서는 문서들을 가장 잘 표현할 수 있는 자질들을 정하고, 이러한 자질들을 통해 분류할 문서를 색인 과정을 통해 표현한다. 또한, 문서 분류기를 통해 문서를 목적에 맞게 분류한다. 본 논문에서는 자동 문서 범주화를 수행하기 위한 각 단계를 소개하고 각 수행 단계에서 사용되는 여러 가지 기법들을 소개하고자 한다.

• Journal of Information Processing Systems
• /
• 제14권2호
• /
• pp.418-434
• /
• 2018

With the increasing use of the Internet and electronic documents, automatic text categorization becomes imperative. Several machine learning algorithms have been proposed for text categorization. The k-nearest neighbor algorithm (kNN) is known to be one of the best state of the art classifiers when used for text categorization. However, kNN suffers from limitations such as high computation when classifying new instances. Instance selection techniques have emerged as highly competitive methods to improve kNN through data reduction. However previous works have evaluated those approaches only on structured datasets. In addition, their performance has not been examined over the text categorization domain where the dimensionality and size of the dataset is very high. Motivated by these observations, this paper investigates and analyzes the impact of instance selection on kNN-based text categorization in terms of various aspects such as classification accuracy, classification efficiency, and data reduction.

• Computers and Concrete
• /
• 제9권5호
• /
• pp.375-387
• /
• 2012

The objective of this investigation was to develop rules for automatic categorization of concrete quality using selected artificial intelligence methods based on machine learning. The range of tested materials included concrete containing a new waste material - solid residue from coal combustion in fluidized bed boilers (CFBC fly ash) used as additive. The rapid chloride permeability test - Nordtest Method BUILD 492 method was used for determining chloride ions penetration in concrete. Performed experimental tests on obtained chloride migration provided data for learning and testing of rules discovered by machine learning techniques. It has been found that machine learning is a tool which can be applied to determine concrete durability. The rules generated by computer programs AQ21 and WEKA using J48 algorithm provided means for adequate categorization of plain concrete and concrete modified with CFBC fly ash as materials of good and acceptable resistance to chloride penetration.

• The Journal of the Acoustical Society of Korea
• /
• 제15권1E호
• /
• pp.29-33
• /
• 1996

A categorization for cluster is necessary when an unsupervised classfication is used for remote sensing image classification. It is desirable that this method is performed automatically, because manual categorization is a highly time consuming process. In this paper, several automatic determination methods were proposed and evaluated. They are four methods. a) maximum number method : which assigns the tharget cluster to the category which occupies the largest area of that cluster b) maximum percentage method : which assigns the target cluster to the category which shows the maximum percentage within the category in that cluster. c) minmun distance method : which assigns the target cluster to the category having minmum distance with that cluster d) element ratio matching method : which assigns local regions to the category having the most similar element ratio of that region From the results of the experiments, it was certified that the result of minimum distance method was almost the same as the result made by a human operator.