A Comparative Analysis of Stemming Strategies in Arabic Broken Plural Classification Using Logistic Regression.
Tobruk University Journal of Engineering Sciences · 30.12.2025 · 6 (2) · 166-176
DOI: 10.64516/kmkhyp28 ↗Kurz gefasst
Fachartikel in der Zeitschrift Tobruk University Journal of Engineering Sciences, online erschienen am 30.12.2025. Verfasst von Yousuf Maneetah. Die Arbeit nennt 11 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
The paper compares the efficacy of various machine learning models at identifying broken plurals in Arabic text. We compare Logistic Regression with and without stemming strategies, utilizing Tashaphyne and ISRI stemmers. The evaluation is based on parameters such as accuracy, precision, recall, and F1-score. Our data show that Logistic Regression without stemming performed the best, implying that for this particular dataset, the raw word characteristics were adequate for the machine to develop good categorization patterns. It indicates that stemming may have contributed noise or eliminated critical information, reducing the algorithms' capacity to learn accurate classifications. A confusion matrix was used to evaluate the accuracy of Logistic Regression. The results validated the superior performance of the Logistic Regression classifier in Arabic broken plural identification tasks.
Zitieren
APA
Maneetah, Y. (2025). A Comparative Analysis of Stemming Strategies in Arabic Broken Plural Classification Using Logistic Regression. Tobruk University Journal of Engineering Sciences, 6(2), 166-176. https://doi.org/10.64516/kmkhyp28
Vancouver
Maneetah Y. A Comparative Analysis of Stemming Strategies in Arabic Broken Plural Classification Using Logistic Regression. Tobruk University Journal of Engineering Sciences. 2025;6(2):166-176. doi:10.64516/kmkhyp28
BibTeX
@article{Maneetah2025,
author = {Maneetah, Yousuf},
title = {A Comparative Analysis of Stemming Strategies in Arabic Broken Plural Classification Using Logistic Regression},
journal = {Tobruk University Journal of Engineering Sciences},
year = {2025},
volume = {6},
number = {2},
pages = {166--176},
doi = {10.64516/kmkhyp28}
}
EUVIMED