KLASTERING DOKUMEN MENGGUNAKAN HIERARCHICAL AGGLOMERATIVE CLUSTERING : Prosiding Seminar Nasional Sistem dan Teknologi Informasi (SNASTI) 2010

Abstract

Document retrieval process stored in document database often produces very large numbers of documents. And many documents are available is not relevant to the desired document. Clustering the documents in database before retrieval is one way to find relevant documents.

This study attempted to document be clustered using Agglomerative Hierarchical Clustering Algorithms. It emphasized clustering to documents written in Indonesian, because today, the needs of users in the homeland of information is increasing. The relationship between documents can be measured by the similarity between the documents (similarity).

This algorithm was tested by using the documents from UII SNATI publications from 2004-2009. The experimental results show that this algorithm can be applied to group documents written in Indonesian. The selection of appropriate keywords will increase the quality of information retrieval to the document. This quality is reflected in the recall rates 0.6 and 0.5 precision.

Disampaikan di Seminar Nasional Sistem dan Teknologi Informasi(SNASTI) , 10 Desember 2010