XML retrieval untuk dokumen Bahasa Indonesia

Abstract

XML (eXtensible Markup Language) retrieval is the content based retrieval of documents structured with XML and aims to implement focused retrieval strategies aiming at returning document components, which is XML elements instead of whole documents in response to a user query. Query languages for XML retrieval can be classified into content only (CO) and content and structure (CAS) query languages. Content only queries usually used for information retrieval where user does not know structure of a document but can get a specific answer from the query. Whereas, for content and structure queries used for XML retrieval systems which aims to answering user query on retrieve specific part of a document from content and structure XML documents. In this research queries are use content and structure (CAS) character with XML query additional fragments tagging of documents structured with XML. The purpose of this reasearch is to implement XML retrieval for document in Bahasa Indonesia using XML query additional fragments tagging with SIMNOMERGE similarity. The testing used 2 documents i.e. agricultural document and medicinal plants document. For the testing agricultural documents used 30 queries and 1000 documents. Whereas for medicinal plants documents used 13 queries and 93 documents. The testing result showed that the perfomance of information search engine is better when we use the medicinal plants documents than agricultural document. However, the performance of search engine with XML query additional fragment tagging with content and disease in madicinal plants document gave good result with around 0.8589 average precision and for XML query additional fragment tagging with text in agricultural document gave result with around 0.6156 average precision