<?xml version="1.0" standalone="yes"?>
<Paper uid="C00-1042">
  <Title>Statistical Morphological Disambiguation for Agglutinative Languages</Title>
  <Section position="2" start_page="0" end_page="0" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> Re(:ent advances in (:onltmter har(lware and availal)ility of very large corpora have made (;t1(`- al)plication of s(;atistical techniques to natural language processing a t)asible and a very at)pealing resem'ch area. Many useflll results have 1)cell obtained by applyilig these techniques to English (and similar languages) in parsing, word sense dismnbiguation, part-of speech (POS) tagging, speech recognition, et;c. However, languages like Turkish, Czech, Hungarian and Finnish, displ W a substantially different behavior than English. Unlike English, these languages have agglutinative or inflective morphology and relatively free constituent order. Such languages have. received little previous attention in statistical processing.</Paragraph>
    <Paragraph position="1"> In this lmper, we t)resent our work on modeling Turkish using statistical methods, and present resuits on morphological disainbiguation. The methods developed here are certainly al)plicable to other agglutinative languages, especially those involving productive derivational phenomena. The Iml)er is organized as follows: After a brief overview of related previous work, we smnma.rize relevant aspects of Turkish and present details of various statistical models for nlorlfliological disanfl/iguation for Turkish. We then present results and analyses fronl our experiments.</Paragraph>
  </Section>
class="xml-element"></Paper>