• Àüü
  • ÀüÀÚ/Àü±â
  • Åë½Å
  • ÄÄÇ»ÅÍ
´Ý±â

»çÀÌÆ®¸Ê

Loading..

Please wait....

±¹³» ³í¹®Áö

Ȩ Ȩ > ¿¬±¸¹®Çå > ±¹³» ³í¹®Áö > Çѱ¹Á¤º¸Ã³¸®ÇÐȸ ³í¹®Áö > Á¤º¸Ã³¸®ÇÐȸ ³í¹®Áö ¼ÒÇÁÆ®¿þ¾î ¹× µ¥ÀÌÅÍ °øÇÐ

Á¤º¸Ã³¸®ÇÐȸ ³í¹®Áö ¼ÒÇÁÆ®¿þ¾î ¹× µ¥ÀÌÅÍ °øÇÐ

Current Result Document : 2 / 4

ÇѱÛÁ¦¸ñ(Korean Title) Á¢»ç ±¸Á¶ ºÐ¼®°ú ±â°è ÇнÀ¿¡ ±â¹ÝÇÑ Çѱ¹¾î ÀÇ¹Ì ¿ª °áÁ¤
¿µ¹®Á¦¸ñ(English Title) Korean Semantic Role Labeling Based on Suffix Structure Analysis and Machine Learning
ÀúÀÚ(Author) ¼®¹Ì¶õ   ±èÀ¯¼·   Miran Seok   Yu-Seop Kim  
¿ø¹®¼ö·Ïó(Citation) VOL 05 NO. 11 PP. 0555 ~ 0562 (2016. 11)
Çѱ۳»¿ë
(Korean Abstract)
ÀÇ¹Ì ¿ª °áÁ¤Àº ÇÑ ¹®Àå¿¡¼­ ¼ú¾î¿Í ±×°ÍÀÇ ³íÇ×°£ÀÇ ÀÇ¹Ì °ü°è¸¦ °áÁ¤ÇØÁÖ´Â °ÍÀ» ¸»ÇÑ´Ù. ÇÑÆí Çѱ¹¾î ÀÇ¹Ì ¿ª °áÁ¤Àº ¿µ¾î¿Í´Â ´Ù¸¥ Çѱ¹¾î °íÀ¯ÀÇ Æ¯ÀÌÇÑ ¾ð¾î ±¸Á¶ ¶§¹®¿¡ ¸¹Àº ¾î·Á¿òÀ» °¡Áö°í Àִµ¥, ÀÌ·¯ÇÑ ¾î·Á¿ò ¶§¹®¿¡ Áö±Ý±îÁö Á¦¾ÈµÈ ´Ù¾çÇÑ ¹æ¹ýµéÀ» °ð¹Ù·Î Àû¿ëÇϱ⿡ ¾î·Á¿òÀÌ ÀÖ¾ú´Ù. ´Ù½Ã ¸»ÇÏÀÚ¸é, Áö±Ý±îÁö Á¦¾ÈµÈ ¹æ¹ýµéÀº ¿µ¾î³ª Áß±¹¾î¿¡ Àû¿ëÇßÀ» ¶§¿¡ ºñÇؼ­ Çѱ¹¾î¿¡ Àû¿ëÇÏ¸é ³·Àº ¼º´ÉÀ» º¸¿©ÁÖ¾ú´ø °ÍÀÌ´Ù. ÀÌ·¯ÇÑ ¾î·Á¿òÀ» ÇØ°áÇϱâ À§ÇÏ¿© º» ¿¬±¸¿¡¼­´Â Á¶»ç³ª ¾î¹Ì¿Í °°Àº Á¢»ç±¸Á¶¸¦ ºÐ¼®ÇÏ´Â °Í¿¡ ÃÊÁ¡À» ¸ÂÃß¾ú´Ù. Çѱ¹¾î´Â ÀϺ»¾î¿Í °°Àº ±³Âø¾îÀÇ ÇϳªÀε¥, ÀÌµé ±³Âø¾î¿¡¼­´Â ¸Å¿ì Àß Á¤¸®µÇ¾î ÀÖ´Â Á¢»ç±¸Á¶°¡ ¾îÈÖ¿¡ ¹Ý¿µµÇ¾î ÀÖ´Ù. ±³Âø¾î´Â ¹Ù·Î À̵é Àß Á¤ÀÇµÈ Á¢»ç ±¸Á¶ ¶§¹®¿¡ ¸Å¿ì ÀÚÀ¯·Î¿î ¾î¼øÀÌ °¡´ÉÇÏ´Ù. ¶ÇÇÑ º» ¿¬±¸¿¡¼­´Â ´ÜÀÏ ÇüżҷΠÀÌ·ç¾îÁø ³íÇ×Àº ±âÃÊ Åë°è·®À» ±âÁØÀ¸·Î ÀÇ¹Ì ¿ª °áÁ¤À» ÇÏ¿´´Ù. ¶ÇÇÑ ÁöÁö º¤ÅÍ ±â°è(Support Vector Machine: SVM)°ú Á¶°ÇºÎ ¹«ÀÛÀ§Àå(Conditional Random Fields: CRFs)¿Í °¯Àº ±â°è ÇнÀ ¾Ë°í¸®ÁòÀ» »ç¿ëÇÏ¿© ¾Õ¿¡¼­ °áÁ¤µÇÁö ¸øÇÑ ³íÇ×µéÀÇ ÀÇ¹Ì ¿ªÀ» °áÁ¤ÇÏ¿´´Ù. º» ³í¹®¿¡¼­ Á¦½ÃµÈ ¹æ¹ýÀº ±â°è ÇнÀ Á¢±Ù ¹æ½ÄÀÌ Ã³¸®ÇØ¾ß ÇÏ´Â ³íÇ×ÀÇ ¹üÀ§¸¦ ÁÙ¿©ÁÖ´Â ¿ªÇÒÀ» Çϴµ¥, ÀÌ´Â ±â°è ÇнÀ Á¢±ÙÀº »ó´ëÀûÀ¸·Î ºÒÈ®½ÇÇÏ°í ºÎÁ¤È®ÇÑ ÀÇ¹Ì ¿ª °áÁ¤À» Çϱ⠶§¹®ÀÌ´Ù. ½ÇÇè¿¡¼­´Â º» ¿¬±¸´Â 15,224 ³íÇ×À» »ç¿ëÇÏ¿´´Âµ¥, ¾à 83.24%ÀÇ f1 Á¡¼ö¸¦ ¾òÀ» ¼ö ÀÖ¾ú´Âµ¥, ÀÌ´Â Çѱ¹¾î ÀÇ¹Ì ¿ª °áÁ¤ ¿¬±¸¿¡ À־ ÇØ¿Ü¿¡¼­ ¹ßÇ¥µÈ ¿¬±¸ Áß °¡Àå ³ôÀº ¼º´ÉÀ¸·Î ¾Ë·ÁÁø °Í¿¡ ºñÇØ ¾à 4.85%ÀÇ Çâ»óÀ» º¸¿©ÁØ °ÍÀÌ´Ù.
¿µ¹®³»¿ë
(English Abstract)
Semantic Role Labeling (SRL) is to determine the semantic relation of a predicate and its argu-ments in a sentence. But Korean semantic role labeling has faced on difficulty due to its different language structure compared to English, which makes it very hard to use appropriate approaches developed so far. That means that methods proposed so far could not show a satisfied perfor-mance, compared to English and Chinese. To complement these problems, we focus on suffix information analysis, such as josa (case suffix) and eomi (verbal ending) analysis. Korean lan-guage is one of the agglutinative languages, such as Japanese, which have well defined suffix structure in their words. The agglutinative languages could have free word order due to its de-veloped suffix structure. Also arguments with a single morpheme are then labeled with statistics. In addition, machine learning algorithms such as Support Vector Machine (SVM) and Condi-tional Random Fields (CRF) are used to model SRL problem on arguments that are not labeled at the suffix analysis phase. The proposed method is intended to reduce the range of argument instances to which machine learning approaches should be applied, resulting in uncertain and inaccurate role labeling. In experiments, we use 15,224 arguments and we are able to obtain approximately 83.24% f1-score, increased about 4.85% points compared to the state-of-the-art Korean SRL research.
Å°¿öµå(Keyword) ÀÇ¹Ì ¿ª °áÁ¤   Á¢»ç ±¸Á¶ ºÐ¼®   Á¶»ç   ¾î¹Ì   ±â°è ÇнÀ   ÁöÁö º¤ÅÍ ±â°è   Á¶°ÇºÎ ¹«ÀÛÀ§Àå   Semantic Role Labeling   Suffix Structure Analysis   Josa   Eomi   Machine Learning   Support Vector Machine   Conditional Random Fields  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå