Monolingual and Cross-language Information Retrieval Approaches for Malay and English Language Document

This thesis concerns a Malay-English monolingual and cross-language information retrieval system. It presents a pioneer work in the aspects that are important for the development of Malay-English information retrieval system. An improved Malay stemming algorithm has been developed to stem the var...

Full description

Saved in:

Bibliographic Details
Main Author:	Abdullah, Muhamad Taufik
Format:	Thesis
Language:	English
Published:	2006
Subjects:	Bilingualism - Malay Bilingualism - English
Online Access:	http://psasir.upm.edu.my/id/eprint/5869/1/FSKTM_2006_1%20IR.pdf
Tags:	Add Tag No Tags, Be the first to tag this record!

id	my-upm-ir.5869
record_format	uketd_dc
spelling	my-upm-ir.58692022-01-13T02:54:28Z Monolingual and Cross-language Information Retrieval Approaches for Malay and English Language Document 2006-02 Abdullah, Muhamad Taufik This thesis concerns a Malay-English monolingual and cross-language information retrieval system. It presents a pioneer work in the aspects that are important for the development of Malay-English information retrieval system. An improved Malay stemming algorithm has been developed to stem the various word forms into their common root for the purpose of indexing and retrieving of Malay documents. The new stemming approaches have been introduced for Malay language, namely Rules-Frequency-Order (RFO), Minimum-Rules-Frequency-Order (MRFO), Rules- Frequency-Application-Order (RFAO), and Rules-Application-Frequency-Order (RAFO). The performance of the new Malay stemming algorithm and approaches are tested using the first two chapters of the Malay translation of the Quranic documents. The results show that the new stemming algorithm and approaches are superior to the previous stemming algorithm and approach. The retrieval effectiveness of the stemming algorithm and approaches are then tested on the actual Quranic collection using vector space model and latent semantic indexing. The results show that there is an improvement in performance from non-stemmed Malay to stemmed Malay, and also from previous stemming algorithm to the new stemming algorithm. Since the employment of the new stemming algorithm and approaches achieved good performance results in Malay monolingual information retrieval, a Malay-English cross-language information retrieval experiment has been performed. The results again show that there is an improvement in performance from non-stemmed Malay to stemmed Malay, and from previous stemming algorithm to the new stemming algorithm. In addition, the results reveal that the new stemming in Malay has performed better than the English stemming in retrieving relevant document. The results can be a reference to forthcoming similar experiments and research for cross language testing of documents retrieval. Bilingualism - Malay Bilingualism - English 2006-02 Thesis http://psasir.upm.edu.my/id/eprint/5869/ http://psasir.upm.edu.my/id/eprint/5869/1/FSKTM_2006_1%20IR.pdf text en public doctoral Universiti Putra Malaysia Bilingualism - Malay Bilingualism - English Computer Science and Information Technology Ahmad, Fatimah
institution	Universiti Putra Malaysia
collection	PSAS Institutional Repository
language	English
advisor	Ahmad, Fatimah
topic	Bilingualism - Malay Bilingualism - English
spellingShingle	Bilingualism - Malay Bilingualism - English Abdullah, Muhamad Taufik Monolingual and Cross-language Information Retrieval Approaches for Malay and English Language Document
description	This thesis concerns a Malay-English monolingual and cross-language information retrieval system. It presents a pioneer work in the aspects that are important for the development of Malay-English information retrieval system. An improved Malay stemming algorithm has been developed to stem the various word forms into their common root for the purpose of indexing and retrieving of Malay documents. The new stemming approaches have been introduced for Malay language, namely Rules-Frequency-Order (RFO), Minimum-Rules-Frequency-Order (MRFO), Rules- Frequency-Application-Order (RFAO), and Rules-Application-Frequency-Order (RAFO). The performance of the new Malay stemming algorithm and approaches are tested using the first two chapters of the Malay translation of the Quranic documents. The results show that the new stemming algorithm and approaches are superior to the previous stemming algorithm and approach. The retrieval effectiveness of the stemming algorithm and approaches are then tested on the actual Quranic collection using vector space model and latent semantic indexing. The results show that there is an improvement in performance from non-stemmed Malay to stemmed Malay, and also from previous stemming algorithm to the new stemming algorithm. Since the employment of the new stemming algorithm and approaches achieved good performance results in Malay monolingual information retrieval, a Malay-English cross-language information retrieval experiment has been performed. The results again show that there is an improvement in performance from non-stemmed Malay to stemmed Malay, and from previous stemming algorithm to the new stemming algorithm. In addition, the results reveal that the new stemming in Malay has performed better than the English stemming in retrieving relevant document. The results can be a reference to forthcoming similar experiments and research for cross language testing of documents retrieval.
format	Thesis
qualification_level	Doctorate
author	Abdullah, Muhamad Taufik
author_facet	Abdullah, Muhamad Taufik
author_sort	Abdullah, Muhamad Taufik
title	Monolingual and Cross-language Information Retrieval Approaches for Malay and English Language Document
title_short	Monolingual and Cross-language Information Retrieval Approaches for Malay and English Language Document
title_full	Monolingual and Cross-language Information Retrieval Approaches for Malay and English Language Document
title_fullStr	Monolingual and Cross-language Information Retrieval Approaches for Malay and English Language Document
title_full_unstemmed	Monolingual and Cross-language Information Retrieval Approaches for Malay and English Language Document
title_sort	monolingual and cross-language information retrieval approaches for malay and english language document
granting_institution	Universiti Putra Malaysia
granting_department	Computer Science and Information Technology
publishDate	2006
url	http://psasir.upm.edu.my/id/eprint/5869/1/FSKTM_2006_1%20IR.pdf
_version_	1747810498637725696

Monolingual and Cross-language Information Retrieval Approaches for Malay and English Language Document

Similar Items