Although many researches on Korean Morphological Analysis and Korean POS (Part-of-Speech) tagging have been performed since 1980s, it is not easy to find a Korean POS tagger for mobile devices. We will introduce a Korean POS tagger that performs well ... Although many researches on Korean Morphological Analysis and Korean POS (Part-of-Speech) tagging have been performed since 1980s, it is not easy to find a Korean POS tagger for mobile devices. We will introduce a Korean POS tagger that performs well on mobile devices such as cellular phones and portable media players. Mobile devices have some critical restrictions that a computing power is low and a working memory capacity is small. Therefore, it may be impossible to perform POS tagging on mobile devices by using traditional methods based on statistical information. Our system performs morphological analysis by using a rule-based method (a modified left-longest-match-preference method), and performs POS tagging by using a statistical method (a Reduced Hidden Markov Model). In the experiments, our system requested smaller memory use and showed faster tagging speed than the system based a traditional Hidden Markov Model. ，韩语论文题目，韩语毕业论文 |
