2004The KIPS Transactions PartDRequires access

Structure-based Clustering for XML Document Retrieval

Jeong Hee Hwang, Keun Ho Ryu

Open publisher page 0 citations

Abstract

웹에서 효율적인 정보 관리와 데이터 교환을 위해 XML의 중요성이 증가함에 따라 XML의 구조 통합과 구조 검색에 대한 연구가 진행되고 있다. 구조가 정의되어 있는 XML 문서의 구조 검색은 스키마 또는 DTD를 통해 가능하다 그러나 DTD나 스키마가 정의되어 있지 않은 XML 문서에 대한 검색은 기존의 검색 방법을 적용할 수 없다. 그러므로 이 논문에서는 구조 정보가 주어지지 않은 많은 양의 XML 문서를 대상으로 구조를 빠르게 검색하기 위한 기반 연구로써 새로운 클러스터링 기법을 제안한다. 먼저 각 문서로부터 빈발한 구조의 특성을 추출한다. 그리고 추출된 빈발 구조를 문서의 대표 구조로 하여 유사 구조기반의 클러스터링을 수행한다. 이것은 서로 다른 구조의 전체 문서를 대상으로 검색하는 것보다 신속하게 구조 검색을 할 수 있도록 한다. 또한 유사한 구조들로 그룹화되어 있는 클러스터들을 기반으로 XML 문서에 대한 구조 검색을 수행한다. 아울러 구조 검색의 적용 방법을 기술하고, 그에 대한 결과의 예를 보여 제안 기법의 효율성을 증명한다. As the importance or XML is increasing to manage information and exchange data efficiently in the web, there are on going works about structural integration and retrieval. The XML. document with the defined structure can retrieve the structure through the DTD or XML schema, but the existing method can't apply to XML. documents which haven't the structure information. Therefore. in this paper we propose a new clus-tering technique at a basic research which make it possible to retrieve structure fast about the XML documents that haven't the structure information. We first estract the feature of frequent structure from each XML document. And we cluster based on the similar structure by con-sidering the frequent structure as representative structure of the XML document, which makes it possible to retrieve the XML document raster than dealing with the whole documents that have different structure. And also we perform the structure retrieval about XML documents based on the clusters which is the group of similar structure. Moreover, we show efficiency of proposed method to describe how to apply the structure retrieval as well as to display the example of application result.

About this research paper

What this paper is about

웹에서 효율적인 정보 관리와 데이터 교환을 위해 XML의 중요성이 증가함에 따라 XML의 구조 통합과 구조 검색에 대한 연구가 진행되고 있다. 구조가 정의되어 있는 XML 문서의 구조 검색은 스키마 또는 DTD를 통해 가능하다 그러나 DTD나 스키마가 정의되어 있지 않은 XML 문서에 대한 검색은 기존의 검색 방법을 적용할 수 없다. 그러므로 이 논문에서는 구조 정보가 주어지지 않은 많은 양의 XML 문서를 대상으로 구조를 빠르게 검색하기 위한 기반 연구로써 새로운 클러스터링 기법을 제안한다. 먼저 각 문서로부터 빈발한 구조의 특성을 추출한다. 그리고 추출된 빈발 구조를 문서의 대표 구조로 하여 유사 구조기반의 클러스터링을 수행한다. 이것은 서로 다른 구조의 전체 문서를 대상으로 검색하는 것보다 신속하게 구조 검색을 할 수 있도록 한다. 또한 유사한 구조들로 그룹화되어 있는 클러스터들을 기반으로 XML 문서에 대한 구조 검색을 수행한다. 아울러 구조 검색의 적용 방법을 기술하고, 그에 대한 결과의 예를 보여 제안 기법의 효율성을 증명한다. As the importance or XML is increasing to manage information and exchange data efficiently in the web, there are on going works about structural integration and retrieval. The XML. document with the defined structure can retrieve the structure through the DTD or XML schema, but the existing method can't apply to XML. documents which haven't the structure information. Therefore. in this paper we propose a new clus-tering technique at a basic research which make it possible to retrieve structure fast about the XML documents that haven't the structure information. We first estract the feature of frequent structure from each XML document. And we cluster based on the similar structure by con-sidering the frequent structure as representative structure of the XML document, which makes it possible to retrieve the XML document raster than dealing with the whole documents that have different structure. And also we perform the structure retrieval about XML documents based on the clusters which is the group of similar structure. Moreover, we show efficiency of proposed method to describe how to apply the structure retrieval as well as to display the example of application result.

Why it matters

A significance statement is not available in the OpenAlex record.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

웹에서 효율적인 정보 관리와 데이터 교환을 위해 XML의 중요성이 증가함에 따라 XML의 구조 통합과 구조 검색에 대한 연구가 진행되고 있다. 구조가 정의되어 있는 XML 문서의 구조 검색은 스키마 또는 DTD를 통해 가능하다 그러나 DTD나 스키마가 정의되어 있지 않은 XML 문서에 대한 검색은 기존의 검색 방법을 적용할 수 없다. 그러므로 이 논문에서는 구조 정보가 주어지지 않은 많은 양의 XML 문서를 대상으로 구조를 빠르게 검색하기 위한 기반 연구로써 새로운 클러스터링 기법을 제안한다. 먼저 각 문서로부터 빈발한 구조의 특성을 추출한다. 그리고 추출된 빈발 구조를 문서의 대표 구조로 하여 유사 구조기반의 클러스터링을 수행한다. 이것은 서로 다른 구조의 전체 문서를 대상으로 검색하는 것보다 신속하게 구조 검색을 할 수 있도록 한다. 또한 유사한 구조들로 그룹화되어 있는 클러스터들을 기반으로 XML 문서에 대한 구조 검색을 수행한다. 아울러 구조 검색의 적용 방법을 기술하고, 그에 대한 결과의 예를 보여 제안 기법의 효율성을 증명한다. As the importance or XML is increasing to manage information and exchange data efficiently in the web, there are on going works about structural integration and retrieval. The XML. document with the defined structure can retrieve the structure through the DTD or XML schema, but the existing method can't apply to XML. documents which haven't the structure information. Therefore. in this paper we propose a new clus-tering technique at a basic research which make it possible to retrieve structure fast about the XML documents that haven't the structure information. We first estract the feature of frequent structure from each XML document. And we cluster based on the similar structure by con-sidering the frequent structure as representative structure of the XML document, which makes it possible to retrieve the XML document raster than dealing with the whole documents that have different structure. And also we perform the structure retrieval about XML documents based on the clusters which is the group of similar structure. Moreover, we show efficiency of proposed method to describe how to apply the structure retrieval as well as to display the example of application result.

Key concepts: XML validation, Document Structure Description, Efficient XML Interchange, Streaming XML, Computer science, XML Schema Editor, XML Encryption, Information retrieval

Related papers

Back to paper searchBrowse research topicsOriginal source
Structure-based Clustering for XML Document Retrieval — Research Paper | ScholarLens