期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2017
卷号：95
期号：15
页码：3560
出版社：Journal of Theoretical and Applied
摘要：Stemming reduces numerous variant forms of a word to its base, stem or root form which is essential for different language processing applications including Urdu IR. Urdu is a resource poor and morphologically rich language. Multilingual Urdu vocabulary is very challenging to process due to its complex morphology. Research of Urdu stemming has an age of a decade. However, there has not been any work reported on dictionary based Urdu stemming. The present work introduces a dictionary based Urdu stemmer with improved performance as compared to the existing Urdu stemmers. The significance of the study is the identification of dictionary-based approach for Urdu stemming as the most promising approach, especially with dictionary update feature. Testing shows 94.85% overall accuracy on test data and results can be further improved by cleaning test data and dictionary updates.
关键词：Dictionary based stemming; dictionary updates; infixes; Fused classification
