ELECTRICA

DETERMINING EFFECTIVE FEATURES FOR WORD SENSE DISAMBIGUATION IN TURKISH

1.

Istanbul University, Engineering Faculty, Computer Engineering Department 34320, Avcilar, Istanbul, Turkey

ELECTRICA 2005; 5: 1341-1352
Read: 824 Downloads: 487 Published: 28 December 2011

Word sense disambiguation is necessary or at least helpful for many natural language processing applications. This paper deals with the feature selection strategies for word sense disambiguation task in general for all types of words in Turkish language. There are many different features that can contribute to the meaning of a word. These features can vary according to the metaphorical usages, POS of the word, etc. The observations indicated that detecting the critical features can contribute much thanthe contribution of using various current learning methodologies.

 

Files
EISSN 2619-9831