• 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 :

ÇѱÛÁ¦¸ñ(Korean Title) ´ë±Ô¸ð ¹®¼­ÁýÇÕ¿¡¼­ Å°¿öµå ÃßÃâ ±â¹ý
¿µ¹®Á¦¸ñ(English Title) Extracting Keywords from large numbers of Documents
ÀúÀÚ(Author) Putu Y. Kusmawan   ±ÇÁØÈ£   Joonho Kwon  
¿ø¹®¼ö·Ïó(Citation) VOL 30 NO. 02 PP. 0003 ~ 0014 (2014. 08)
Çѱ۳»¿ë
(Korean Abstract)
ÀÎÅͳÝÀ» ÅëÇÑ È°µ¿ÀÌ Áõ°¡ÇÔ¿¡ µû¶ó, µðÁöÅÐ ÅؽºÆ® ¹®¼­ÀÇ »ç¿ëÀÌ ÀϹÝÈ­µÇ°í ÀÖ´Ù. µû¶ó¼­ ÀÎÅͳÝÀ» ÅëÇؼ­ ´Ù¾çÇÑ Çü½ÄÀÇ ¾öû³­ ¼öÀÇ ¹®¼­µéÀ» ÀÌ¹Ì ¼Õ½±°Ô Á¢ÇÒ ¼ö ÀÖ´Ù. ÀÌ·± »óȲ¿¡¼­ ÁÖ¿äÇÑ µµÀü ¹®Á¦´Â Àº ¹®¼­µéÀ» ºü¸£°Ô ºÐ¼®ÇÏ°í, ¹®¼­µéÀ» ´õ ÁÁÀº ÇüÅ·ΠǥÇöÀ» ÇÏ´Â ½Ã½ºÅÛÀ» °³¹ßÇÏ´Â °ÍÀÌ´Ù. ÀÌ ³í¹®Àº ¸¹Àº ¼öÀÇ ¹®¼­µé·ÎºÎÅÍ ´Ü¾îµé °£ÀÇ ¿¬°ü °ü°è ºÐ¼®À» ÅëÇØ Å°¿öµå¸¦ ÃßÃâÇÏ´Â ½Ã½ºÅÛÀ» Á¦½ÃÇÑ´Ù. ´Ü¾îµé °£ÀÇ ¿¬°ü °Ë»ç´Â ÃâÇö ºóµµ, ÀÎÁ¢ ´Ü¾î °£ÀÇ ¿¬°ü °ü°è, ÀÌÇà ¿¬°ü °ü°è µîÀ» °í·ÁÇÏ¿© °è»êÇÑ´Ù. »ç¿ëÀÚÀÇ ¹®¼­¿¡ ´ëÇÑ ÀÌÇظ¦ µ½±â À§ÇÏ¿©, ±×·¡ÇÁ ±¸Á¶¸¦ »ç¿ëÇÏ¿© ÃßÃâÇÑ Å°¿öµå°£ÀÇ °ü°è¸¦ Ç¥ÇöÇÑ´Ù. ¶ÇÇÑ Å« »çÀÌÁîÀÇ ¹®¼­µéÀ» ó¸®ÇÒ ¼ö ÀÖµµ·Ï ¸Ê/¸®µà½º ÇÁ·¹ÀÓ¿öÅ©¿¡ ±â¹ÝÇÑ ¾Ë°í¸®ÁòÀ» Á¦½ÃÇÑ´Ù. ½ÇÇè °á°úµéÀº Á¦½ÃÇÑ ±â¹ýÀÌ ¸¹Àº ¼öÀÇ ¹®¼­·ÎºÎÅÍ È¿À²ÀûÀ¸·Î Å°¿öµå¸¦ ÃßÃâÇÒ ¼ö ÀÖÀ½À» º¸¿´´Ù.
¿µ¹®³»¿ë
(English Abstract)
As the growing of Internet activity, the usage of digital text document is becoming popular. An enormous number of documents in various forms are now freely available over the Internet. The main challenge of this situation is to make a system which can be used to quickly analyse a large set of documents and express them in a better form of representation. In this paper, we propose a new technique to extract keywords from large set of documents by the word correlation analysis. This analysis method considers frequencies of words, correlation between adjacent words, transitive correlation between words in a document. Then, we represent them as a graph for providing a better visualization for the document. Moreover, we also make our technique scalable by implementing it using Map/Reduce algorithms. Experimental results shows that our method can effectively extract kewords from large sets of documents.
Å°¿öµå(Keyword) ´Ü¾î ¿¬°ü °Ë»ç   Å°¿öµå ÃßÃâ   ¸Ê/¸®µà½º   ±×·¡ÇÁ ±¸Á¶   Word Correlation Analysis   Keyword Extraction   Map/Reduce   Graph structure  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå