Skoči na glavni sadržaj

Izvorni znanstveni članak

https://doi.org/10.2498/cit.1002174

Rule-based Approach for Arabic Root Extraction: New Rules to Directly Extract Roots of Arabic Words

Fatma Abu Hawas ; Department of Computer Science, Yarmouk University, Jordan
Keith E. Emmert ; Tarleton State University, Stephenville, Texas, USA


Puni tekst: engleski PDF 1.013 Kb

str. 57-68

preuzimanja: 1.122

citiraj


Sažetak

Extracting word roots in Arabic language is very problematic due to the specific morphological and structural changes in the language. To address this problem, several techniques have been proposed. This paper continues the problem of identifying and exploiting relationship amongst Arabic letters for Arabic root extraction begun in [1]. Eight different rules that detect the root letters according to other letters in the word have been proposed and tested, four of them benefiting from the idea of morphological substitution (MUTATION). The approach has been evaluated using the Holy Quran words. The evaluation results show a promising root extraction algorithm.

Ključne riječi

rule-based stemmer, word root, suffixes, prefixes, words patterns

Hrčak ID:

123178

URI

https://hrcak.srce.hr/123178

Posjeta: 1.477 *