Á¤º¸°úÇÐȸ³í¹®Áö (Journal of KIISE)
Current Result Document : 215 / 216
ÇѱÛÁ¦¸ñ(Korean Title) |
Çѱ¹¾î ÇüÅÂ¼Ò ºÐ¼® ¹× Ç°»ç űëÀ» À§ÇÑ µö ·¯´× ±â¹Ý 2´Ü°è ÆÄÀÌÇÁ¶óÀÎ ¸ðµ¨ |
¿µ¹®Á¦¸ñ(English Title) |
A Deep Learning-based Two-Steps Pipeline Model for Korean Morphological Analysis and Part-of-Speech Tagging |
ÀúÀÚ(Author) |
À±ÁØ¿µ
ÀÌÀ缺
Jun Young Youn
Jae Sung Lee
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 48 NO. 04 PP. 0444 ~ 0452 (2021. 04) |
Çѱ۳»¿ë (Korean Abstract) |
Àΰø½Å°æ¸ÁÀ» È°¿ëÇÑ ÃÖ±ÙÀÇ Çѱ¹¾î ÇüÅÂ¼Ò ºÐ¼® ¹× ÅÂ±ë ¿¬±¸´Â ÁַΠǥÃþÇü¿¡ ´ëÇØ ÇüÅÂ¼Ò ºÐ¸®¿Í Ç°»ç űëÀ» ¸ÕÀúÇÏ°í, ¿øÇü º¹¿ø »çÀüÀ» ÀÌ¿ëÇÏ¿© ÈÄ󸮷ΠÇüÅÂ¼Ò ¿øÇüÀ» º¹¿øÇØ¿Ô´Ù. º» ¿¬±¸¿¡¼´Â ÇüÅÂ¼Ò ºÐ¼® ¹× Ç°»ç űëÀ» µÎ ´Ü°è·Î ³ª´©¾î, sequence-to-sequence¸¦ ÀÌ¿ëÇÏ¿© ÇüÅÂ¼Ò ¿øÇüÀ» ¸ÕÀú º¹¿øÇÏ°í, ÃÖ±Ù ÀÚ¿¬¾îó¸®ÀÇ ´Ù¾çÇÑ ºÐ¾ß¿¡¼ ¿ì¼öÇÑ ¼º´ÉÀ» º¸ÀÌ´Â BERT¸¦ ÀÌ¿ëÇÏ¿© ÇüÅÂ¼Ò ºÐ¸® ¹× Ç°»ç űëÀ» ÇÏ¿´´Ù. µÎ ´Ü°è¸¦ ÆÄÀÌÇÁ¶óÀÎÀ¸·Î Àû¿ëÇÑ °á°ú, º°µµÀÇ ±ÔÄ¢À̳ª º¹ÇÕ ÅÂ±× Ã³¸® µîÀÌ ÇÊ¿äÇÑ ÇüÅÂ¼Ò ¿øÇü º¹¿ø »çÀüÀ» »ç¿ëÇÏÁö ¾Ê°íµµ ¿ì¼öÇÑ ÇüÅÂ¼Ò ºÐ¼® ¹× ÅÂ±ë °á°ú¸¦ º¸¿´´Ù.
|
¿µ¹®³»¿ë (English Abstract) |
Recent studies on Korean morphological analysis using artificial neural networks have usually performed morpheme segmentation and part-of-speech tagging as the first step with the restoration of the original form of morphemes by using a dictionary as the postprocessing step. In this study, we have divided the morphological analysis into two steps: the original form of a morpheme is restored first by using the sequence-to-sequence model, and then morpheme segmentation and part-of-speech tagging are performed by using BERT. Pipelining these two steps showed comparable performance to other approaches, even without using a morpheme restoring dictionary that requires rules or compound tag processing.
|
Å°¿öµå(Keyword) |
ÇüÅÂ¼Ò ºÐ¼®
ÇüÅÂ¼Ò Ç°»ç űë
ÆÄÀÌÇÁ¶óÀÎ
morphological analysis
part-of-speech tagging
sequence-to-sequence
BERT
pipeline
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|