Decision List Algorithm for Word Sense Disambiguation for TELUGU Natural Language Processing
Durga Prasad Palanati, Ramakrishna Kolikipogu · 2013
Word Sense Disambiguation plays a key role in Information Retrieval Applications. Telugu language contains more ambiguous words when compared to other European languages. Word complexity is high for Telugu and Information Retrieval in a Telugu language is a challenging task. This paper focus on assigning a particular sense for a word based on the context of use. Word cooccurance probability over the corpus is measured with the help of Decision list algorithm and assigned the sense. This algorithm is proven to be accurate for English text corpus about 90%. We adopted the same decision list algorithm to disambiguate sense of Telugu words with unsupervised learning process. The performance of Telugu Information Retrieval is measured with and without WSD. The results are more accurate with WSD when compared to word sense ambiguity.