Home > Explore Content > All Issues > Article View
Automatic Construction of Korean Unknown Word Dictionary using Occurrence Frequency in Web Documents
Journal of The Korea Society of Computer and Information
Abbr :
JKSCI
2008, 13(3), pp.27-34
Publisher : The Korean Society Of Computer And Information
Research Area : Computer Science
2008, 13(3), pp.27-34
Publisher : The Korean Society Of Computer And Information
Research Area : Computer Science
-
In this paper, we propose a method of automatically constructing a dictionary by extracting unknown words from given eojeols in order to improve the performance of a Korean morphological analyzer. The proposed method is composed of a dictionary construction phase based on full text analysis and a dictionary construction phase based on web document frequency. The first phase recognizes unknown words from strings repeatedly occurred in a given full text while the second phase recognizes unknown words based on frequency of retrieving each string, once occurred in the text, from web documents. Experimental results show that the proposed method improves 32.39% recall by utilizing web document frequency compared with a previous method.
184 Viewed
Print this page
Download PDF