[Website Utama ITS]           
  » Login
Username :
Password :
  » Dosen Per Fakultas
» FMIPA
» FTI
» FTSP
» FTK
» FTIf
» Pasca Sarjana
» PENS
» PPNS
  » Publikasi Ilmiah Per Fakultas
» FMIPA
» FTI
» FTSP
» FTK
» FTIf
» Pasca Sarjana
» PENS
» PPNS
  » Pencarian
  
  
 
 
  Modification of Khoja Stemmer for Searching Arabic Text  
Diupload oleh : Diana Purwitasari, S.Kom., M.Sc.
Pengarang : Purwitasari, D., Najibullah, A., Arifin, A.Z.
Tahun : 2012
Dipublikasikan di : The First Indonesian - Japanese Conference on Knowledge Creation and Intelligent Computing (KCIC)2012, ITS 13-14 Maret 2012
Jenis Jurnal : International Conference
Eksternal Link : http://
Bidang Penelitian : Information Retrieval
Abstrak : N-gram model could be used to provide searchers with morphological variants of search terms for improving searching performance. Implementation of automated affix removal to simplify morphological variants into its stem is known as stemming algorithms. This paper discussed about n-gram model of size two called digram which is used in searching process with Arabic text. Since stemmer usage could enhance searching performance, we have observed performance of Khoja Stemmer. As a result, in this paper we also introduced addition of morphological rules correspond to terms which are not well recognized by Khoja stemmer. Our experiments showed that additional rules for Khoja Stemmer gave better reduction rate of indexed terms. We also showed that Khoja Stemmer and our additional rules as stemming algorithm in preprocessing can really enhance performance of Arabic text searching with digram model for similarity calculation.
File : [ PDF ]
Copyright © 2010 - Institut Teknologi Sepuluh Nopember        Desain dan Perawatan: Tim Webmaster ITS (webadmin[at]its.ac.id)