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

»çÀÌÆ®¸Ê

Loading..

Please wait....

±¹³» ÇÐȸÁö

Ȩ Ȩ > ¿¬±¸¹®Çå > ±¹³» ÇÐȸÁö > µ¥ÀÌÅͺ£À̽º ¿¬±¸È¸Áö(SIGDB)

µ¥ÀÌÅͺ£À̽º ¿¬±¸È¸Áö(SIGDB)

Current Result Document : 11 / 11

ÇѱÛÁ¦¸ñ(Korean Title) À§Å°Çǵð¾ÆÀÇ ¹é¸µÅ© Á¤º¸¸¦ ÀÌ¿ëÇÑ °³³ä °èÃþ Æ®¸®ÀÇ ÀÚµ¿ ±¸Ãà ±â¹ý
¿µ¹®Á¦¸ñ(English Title) Automated Development of Concept Hierarchy Tree using Backlink Information of Wikipedia
ÀúÀÚ(Author) ÀÌ°¡Èñ   ±èÇÑÁØ   Ga-Hee Lee   Han-Joon Kim  
¿ø¹®¼ö·Ïó(Citation) VOL 31 NO. 01 PP. 0040 ~ 0049 (2015. 04)
Çѱ۳»¿ë
(Korean Abstract)
º» ³í¹®Àº À§Å°Çǵð¾Æ µ¥ÀÌÅÍÀÇ ¹é¸µÅ© Á¤º¸¸¦ È°¿ëÇÏ¿© °³³äµé °£ÀÇ °èÃþ ±¸Á¶¸¦ ÀÚµ¿ ±¸¼ºÇÏ´Â »õ·Î¿î ±â¹ýÀ» Á¦¾ÈÇÑ´Ù. °³³ä °èÃþ ±¸Á¶´Â Á¤º¸°Ë»ö ¹× ÅؽºÆ®¸¶ÀÌ´× ±â¹ýÀÇ °³¼±À» À§ÇØ Áß¿äÇÑ ÀڷᱸÁ¶·Î »ç¿ëµÈ´Ù. ½ÇÁ¦·Î ´ëÇ¥ÀûÀÎ °³³ä °èÃþ ±¸Á¶·Î Open Directory Project(ODP)°¡ Àִµ¥ ÀÌ´Â ÃÖÀûÀÇ °Ë»ö °á°ú¸¦ ¾ò±â À§ÇÑ °¡Àå Æ÷°ýÀûÀÎ À¥ µð·ºÅ丮·Î Æò°¡ ¹Þ°í ÀÖ´Ù. ÇÏÁö¸¸ ODPÀÇ ±¸ÃàÀ» Æ÷ÇÔÇÏ¿© ±âÁ¸ °³³ä °èÃþ ±¸Á¶ÀÇ ±¸Ãà ±â¹ýµéÀº »ç¶÷ÀÇ ¼öÀÛ¾÷¿¡ ÀÇÁ¸Çϰųª ÅؽºÆ® ¹®¼­¿¡ ÃâÇöÇÏ´Â ÁÖ¿ä ´Ü¾î¸¦ °³³äÀ¸·Î ½ÂÈ­ÇÏ¿© ÀÌµé °£ÀÇ °³³äÀû À§»ó°ü°è¸¦ µµÃâÇÑ´Ù. º» ³í¹®Àº °íÇ°ÁúÀÇ °³³äÆ®¸®¸¦ ±¸¼ºÇϱâ À§ÇØ À§Å°Çǵð¾Æ ÆäÀÌÁö ÁýÇÕÀ» °³³ä °ø°£À¸·Î °£ÁÖÇÏ´Â Àü·«À» ÃëÇÑ´Ù. ±âº» ¾ÆÀ̵ð¾î´Â º¸Æí¼ºÀ» °¡Áö´Â À§Å°ÆäÀÌÁö¸¦ ¼±º°ÇÏ¿© ÀÌµé °£ÀÇ ¹é¸µÅ© °ü°è¸¦ È®·üÀûÀ¸·Î Çؼ®ÇÏ¿© °³³äÀû Æ÷ÇÔ °ü°è¸¦ »êÃâÇÏ´Â °ÍÀÌ´Ù. Á¦¾È ±â¹ý¿¡ ÀÇÇØ »ý¼ºµÈ °³³äÆ®¸®´Â ¼öÀÛ¾÷À¸·Î ±¸ÃàµÈ ODPÀÇ °èÃþ °ü°è¸¦ ÀÚµ¿À¸·Î µµÃâÇÒ ¼ö ÀÖÀ» »Ó¸¸ ¾Æ´Ï¶ó ÇöÀç ½ÃÁ¡ÀÇ °³³äµé°£ÀÇ °ü°è±îÁö µµÃâÇÏ´Â ¸Å¿ì ½Ç¿ëÀû ±â¹ýÀ̶ó ÇÒ ¼ö ÀÖ´Ù.
¿µ¹®³»¿ë
(English Abstract)
In this paper, we propose a novel method of automatically building hierarchical concept tree using Wikipedia backlink information. The concept tree is used as an important data structure to improve information retrieval and text mining techniques. In fact, a typical hierarchy of concepts, called Open Directory Project (ODP) has been recognized as the most comprehensive web directory used to obtain the best search results. However, conventional methods of building concept trees such as ODP are dependent upon manual job, or regard each of words occurring in textual documents as a concept and then derive conceptual hierarchical relationships among concepts. Our strategy is that a set of particular Wikipedia pages are considered to be a concept space in order to build up high quality concept trees. The key idea behind the proposed method is that after selecting only Wikipedia pages (shortly, wikipages) with general senses, the conceptual subsumption relationships among the wikipages are statistically calculated by using their backlink information. The concept tree produced by the proposed method contains not only hierarchical relationships of ODP that is manually built up, but also newly defined hierarchical relationships that reflects the current trends.
Å°¿öµå(Keyword) ÅؽºÆ® ¸¶ÀÌ´×   Á¤º¸°Ë»ö   °³³ä Æ®¸®   À§Å°Çǵð¾Æ   Text mining   Information retrieval   Concept tree   Wikipedia  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå