Á¤º¸°úÇÐȸ³í¹®Áö (Journal of KIISE)
Current Result Document : 1 / 2
ÇѱÛÁ¦¸ñ(Korean Title) |
½Ã°£Â÷ ÇнÀÀ» ÀÌ¿ëÇÑ ´Ü¾î °¨Á¤ °ª ÃøÁ¤¹ý ¿¬±¸ |
¿µ¹®Á¦¸ñ(English Title) |
Measuring Semantic Orientation of Words using Temporal Difference Learning |
ÀúÀÚ(Author) |
±è¿µ»ï
½ÅÈ¿ÇÊ
Youngsam Kim
Hyopil Shin
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 45 NO. 12 PP. 1287 ~ 1291 (2018. 12) |
Çѱ۳»¿ë (Korean Abstract) |
½Ã°£Â÷(temporal-difference) ÇнÀÀº °ÈÇнÀÀÇ ÇÙ½ÉÀûÀÎ ¾Ë°í¸®ÁòÀ¸·Î ¸¶¸£ÄÚÇÁ üÀÎ ¸ðÇü¿¡¼ »óÅÂÀÇ °¡Ä¡¸¦ ½Ç½Ã°£À¸·Î ÃøÁ¤Çϴµ¥ À¯¿ëÇÑ ¹æ¹ý·ÐÀ» Á¦°øÇÑ´Ù. ÀÌ ¹æ¹ý·Ð¿¡¼ È°¿ëµÇ´Â ¸¶¸£ÄÚÇÁ ¸ðÇüÀº °¨¼â ºñ(discount factor)¸¦ »ç¿ëÇÏ¿© º¸»óÀÌ ÁÖ¾îÁö´Â ½ÃÁ¡°ú °¡±î¿î »óÅÂÀϼö·Ï º¸»ó °ª¿¡ ´ëÇØ ´õ¸¹Àº °¡ÁßÄ¡¸¦ ÁÖ°Ô µÈ´Ù. º» ³í¹®¿¡¼´Â ÅؽºÆ®ÀÇ ¾î¶² ¾îÈÖ°¡ °®´Â °¨Á¤ °ªÀ» ÃøÁ¤Çϴµ¥ ÀÖ¾î ½Ã°£Â÷ ÇнÀÀÌ ±âÁ¸ÀÇ º£ÀÌÁî È®·üÀ» ÀÌ¿ëÇÏ´Â ¹æ¹ýº¸´Ù »ó´ëÀûÀ¸·Î À¯¿ëÇÔÀ» º¸ÀÌ°íÀÚ ÇÑ´Ù. ÀÌ´Â ½Ã°£Â÷ ÇнÀÀÌ º»ÁúÀûÀ¸·Î Á¡ÁõÀû(incremental) ó¸®ÀÌ¸ç °¨¼â ºñ¸¦ ÅëÇØ ºÎ¿©ÇÒ °¨Á¤ °ªÀÇ °¡ÁßÄ¡¸¦ Á¶ÀýÇÒ ¼ö Àֱ⠶§¹®ÀÌ´Ù. º» ³í¹®Àº ¿µÈÆò ÀڷḦ ÀÌ¿ëÇÏ¿© ÀÌ ¹æ¹ýÀÇ È¿°ú¸¦ °£Á¢ÀûÀÎ ¹æ¹ý°ú Á÷Á¢ÀûÀÎ ¹æ¹ý ¸ðµÎ¿¡¼ °ËÁõÇÏ¿´À¸¸ç, ÀÌ ¹æ¹ýÀÌ ´ë¿ë·®ÀÇ ÀÚ·á¿¡ Àû¿ë °¡´ÉÇÔ(scalable)À» º¸À̱â À§ÇØ ºñµ¿±â º´·Äó¸® ¹æ½ÄÀ¸·Îµµ ÀÌ ¹æ¹ýÀÇ È¿°ú°¡ À¯ÁöµÊÀ» °ËÁõÇÏ¿´´Ù.
|
¿µ¹®³»¿ë (English Abstract) |
Temporal-difference(TD) learning is a core algorithm of reinforcement learning, which employs models of Markov process. In the TD methods, rewards are always discounted by a discount factor and states receive these discounted values as their rewards. In this paper, we attempted to estimate a semantic orientation of words in texts using the TD-based methods and examined the effectiveness of the proposed methods by comparing them to existing feature selection methods (indirect approach) and Bayes probabilities (direct approach). The TD-based estimation would be useful for tasks of social opinion mining, since TD learning is inherently an on-line method. In order to show our approach is scalable to huge data, the estimation method is also evaluated using asynchronous parallel processing.
|
Å°¿öµå(Keyword) |
½Ã°£Â÷ ÇнÀ
°ÈÇнÀ
´Ü¾î °¨Á¤ °ª
Á¡ÁõÀû ó¸®
ºñµ¿±â º´·Äó¸®
temporal-difference learning
reinforcement learning
semantic orientation of words
incremental processing
asynchronous parallel processing
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|