
<?xml version="1.0" encoding="UTF-8"?>
<record>
  <title>Morphological Analyzer for Kannada Inflectional Words Using Hybrid Approach</title>
  <journal>International Journal of Computational Linguistics Research </journal>
  <author>Prathibha, R J, Padma, M C</author>
  <volume>7</volume>
  <issue>4</issue>
  <year>2016</year>
  <doi></doi>
  <url>http://www.dline.info/jcl/fulltext/v7n4/jclv7n4_3.pdf</url>
  <abstract>Morphological analyzer is a tool which extracts morphological features and grammatical categories 
associated with the given inflectional word. Kannada is an inflectional and morphologically rich 
language. This paper presents the design of morphological analyzer for Kannada inflectional words 
using affix-stripping, rule-based, paradigm-based and questionnaire-based approaches. The proposed 
model comprises of text normalizer, morphological stemmer, morphological analyzer and a module for 
semi-automatic insertion of lexical details of unanalyzed words into Kannada monolingual lexicon. 
The performance of proposed model is tested on 7 different data sets that are manually created by 
collecting Kannada text from government official circular documents Kannada newspapers, legal documents 
and news from All India Radio. The experimental result shows that the accuracy of proposed model on these 
7 different data sets is around 90%.
</abstract>
</record>
