• 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¸°úÇÐȸ ³í¹®Áö > Á¤º¸°úÇÐȸ ³í¹®Áö C : ÄÄÇ»ÆÃÀÇ ½ÇÁ¦

Á¤º¸°úÇÐȸ ³í¹®Áö C : ÄÄÇ»ÆÃÀÇ ½ÇÁ¦

Current Result Document : 5 / 6 ÀÌÀü°Ç ÀÌÀü°Ç   ´ÙÀ½°Ç ´ÙÀ½°Ç

ÇѱÛÁ¦¸ñ(Korean Title) »çȸ°ü°è¸Á ÀÚµ¿ ±¸ÃàÀ» À§ÇÑ »çȸ °ü°è ÀνºÅϽº ÃßÃâ
¿µ¹®Á¦¸ñ(English Title) Extraction of Instances with Social Relations for Automatic Construction of a Social Network
ÀúÀÚ(Author) ÃָͽĠ  ±èÇмö   Maengsik Choi   Harksoo Kim  
¿ø¹®¼ö·Ïó(Citation) VOL 17 NO. 10 PP. 0548 ~ 0552 (2011. 10)
Çѱ۳»¿ë
(Korean Abstract)
º» ³í¹®¿¡¼­´Â ´ë¿ë·®ÀÇ ¸»¹¶Ä¡·ÎºÎÅÍ »çȸ¸ÁÀ» ÀÚµ¿ ±¸ÃàÇϱâ À§ÇÑ ½Ã½ºÅÛÀ» Á¦¾ÈÇÑ´Ù. Á¦¾È ½Ã½ºÅÛÀº ÀÔ·Â ¹®ÀåÀÇ µÎ ÀÎ¸í »çÀÌÀÇ »çȸ °ü°è Æ÷ÇÔ ¿©ºÎ¸¦ °áÁ¤ÇÏ´Â ºÎºÐ°ú »çȸ °ü°è¸¦ Ç¥ÇöÇÏ´Â ¾îÀýÀ» ÃßÃâÇÏ´Â µÎ ºÎºÐÀ¸·Î ±¸¼ºµÈ´Ù. È¿°úÀûÀ¸·Î ½Ã½ºÅÛÀ» ±¸ÇöÇϱâ À§ÇØ SVMÀ» ÀÌ¿ëÇÏ¿© ÀÇÁ¸ Æ®¶óÀ̱׷¥ Ä¿³ÎÀ̶ó´Â »õ·Î¿î Æ®¸® Ä¿³ÎÀ» ¸¸µé¾ú´Ù. ½Å¹®±â»ç¸¦ ÀÌ¿ëÇÑ ½ÇÇè¿¡¼­, Á¦¾ÈÇÑ ½Ã½ºÅÛÀÌ Àß ¾Ë·ÁÁ® ÀÕ´Â ÀÇÁ¸ Ä¿³ÎÀ» ±â¹ÝÀ¸·Î ÇÏ´Â ÀÌÀü ½Ã½ºÅÛº¸´Ù ƯÈ÷ ÀçÇöÀ²¿¡¼­ ´õ ÁÁÀº ¼º´ÉÀ» º¸¿´´Ù. ½ÇÇè °á°ú¿¡ ÀÇÇØ Á¦¾ÈÇÑ ½Ã½ºÅÛÀÌ »çȸ¸Á ÀÚµ¿ ±¸ÃàÀ» À§ÇÑ µµ±¸·Î½á È°¿ëµÉ ¼ö ÀÖÀ½À» ¾Ë ¼ö ÀÖ¾ú´Ù.
¿µ¹®³»¿ë
(English Abstract)
We propose a social relation extraction system for automatically constructing a social network from a large amount of text. The proposed system consists of two subsystems: One is a subsystem to determine whether an input sentence includes a social relation between two people, and the other is a subsystem to determine which eojeol (a spacing unit in Korean) describes a social relation. We design new tree kernels called dependency trigram kernels for effectively implementing these subsystems by using support vector machines (SVMs). In the experiments with news articles, the proposed system showed better performances especially in recall rates than the previous system based on the well-known dependency kernel. On the basis of experimental results, we suggest that the proposed system can be a useful tool for automatically constructing social networks from text.
Å°¿öµå(Keyword) ÀÇÁ¸ Æ®¶óÀ̱׷¥ Ä¿³Î   »çȸ °ü°è ÃßÃâ   »çȸ¸Á ±¸Ãà   dependency trigram kernel   social relation extraction   social network construction  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå