GDL

Journal from gdlhub / 2017-08-14 14:09:06
Oleh : Younes Jaafar, Driss Namly, Karim Bouzoubaa, Abdellah Yousfi, King Saud University
Dibuat : 2017-04-14, dengan 1 file

Keyword : Arabic stemmingEvaluationBenchmarkEvaluation corpus
Url : http://www.sciencedirect.com/science/article/pii/S1319157816301239
Sumber pengambilan dokumen : web

Many approaches and solutions have been proposed for developing Arabic light stemmers. These stemmers are often used in the context of application-oriented projects, especially when it comes to developing information retrieval (IR) systems. However, Arabic light stemming, as the process of stripping off a set of prefixes and/or suffixes, is a blinded task suffering from problems such as incorrect removal, vocalization ambiguity, single solution, etc. Moreover, each researcher claims that his/her stemmer reached a level of strength and accuracy quite high. However, in most cases, these stemmers are black boxes and it is not possible to access neither their source codes to verify their validity, nor the evaluation corpora that were used to claim such accuracy. Since these stemmers are very important for researchers, their comparison and evaluation is then essential to facilitate the choice of the stemmer to use in a given project. In this paper, we propose a new Arabic stemmer that gives solutions to the above mentioned drawbacks. In addition, we propose an automatic approach for the evaluation and comparison of Arabic stemmers that takes into account metrics related to the accuracy of results as well as the execution time of stemmers.

Beri Komentar ?#(0) | Bookmark

Properti	Nilai Properti
ID Publisher	gdlhub
Organisasi	King Saud University
Nama Kontak	Herti Yani, S.Kom
Alamat	Jln. Jenderal Sudirman
Kota	Jambi
Daerah	Jambi
Negara	Indonesia
Telepon	0741-35095
Fax	0741-35093
E-mail Administrator	elibrarystikom@gmail.com
E-mail CKO	elibrarystikom@gmail.com