Pengindeksan Ontologi Dokumen Bahasa Indonesia Menggunakan Latent Semantic Analysis

Abstract

Ontology is a collection of concepts that can be applied to the information retrieval system that is able to represent knowledge of a document in a specific domain, so that ontology is considered to improve the performance of information retrieval systems. Therefore the construction of ontology concept must be able to solve semantic problem in corpus. This research proposes ontology indexing of document Bahasa Indonesia using latent semantic analysis. Latent semantic analysis (LSA) is a theory and method which can be considered to solve the problem of semantic words in the corpus by extracting and representing contextual usage meaning of the word by using statistical method. In this research, ontology which has been constructed is applied to the information retrieval systems and evaluated using 20 pairs of queries. As the result of this experiment a recall value of 96.5% and average precision value of 83.4% were obtained.