A HYBRID METHOD OF RULE-BASED AND STRING MATCHING STEMMER FOR JAVANESE LANGUAGE_

Fatkhul, Amin and Wiwien, Hadikurniawati and Setyawan, Wibisono and Herny, Februariyanti and Jati, Sasongko Wibowo A HYBRID METHOD OF RULE-BASED AND STRING MATCHING STEMMER FOR JAVANESE LANGUAGE_. Journal of Theoretical and Applied Information Technology (JATIT).

PDF Download (2MB)

Abstract

"Language is rich in morphological variations but poor in linguistic computational resources. Ngoko Javanese language is a morphologically rich language that has a different variant form of words. This paper describes an algorithm by which a stem for Ngoko Javanese language. Ngoko Javanese language stemmer is efficiently used in information retrieval. Through this algorithm, we can get a root from its actual word. We use a hybrid rule-based and string matching algorithm. Special rules are created to remove the prefixes and suffixes of the Ngoko Javanese terms. The algorithm has been tested on hundreds of Ngoko Javanese words. Results reveal that the accuracy reaches to about 67%."

Item Type:	Article
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software
Faculty / Institution:	Fakultas Teknologi Informasi > Program Studi Teknik Informatika
Depositing User:	Jati Sasongko Wibowo
Date Deposited:	09 Dec 2022 08:13
Last Modified:	09 Dec 2022 08:13
URI:	https://eprints.unisbank.ac.id/id/eprint/8975

Actions (login required)

View Item