A HYBRID STATISTICAL APPROACH TO STEMMING IN TURKISH: AN AGGLUTINATIVE LANGUAGE

Authors : Tarık KIŞLA, Bahar KARAOĞLAN

Pages : 401-412

Doi:10.18038/btda.31812

View : 13 | Download : 9

Publication Date : 2016-07-14

Article Type : Research Paper

Abstract :Finding Stem is a complicated and important issue for agglutinative languages like Turkish where theoretically infinite number of surface forms can be obtained from a single lexeme. Both analytical and statistical approaches have been tried for stemming Turkish words. Two main problems apparent with these approaches are the involvement of a dictionary which enforces the assumption of closed vocabulary and the disambiguation of the actual stem among the numerous candidates. Here, we present a method that exploits the simple fact that nouns and verbs have different suffix patterns. Statistical methods are used for stripping off the suffixes. Based on the suffix pattern PoS is determined which then enables the decision for the stem boundary. Thus, the major contribution of the study is the avoiding the disambiguation problem and not using a regular dictionary for stemming. The performance rate for proposed method on golden standard PoS tagged Turkish corpus is 93.83%.
Keywords : Stemming, Natural Language Processing, Turkish, Agglutinative Language