• 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 D : µ¥ÀÌŸº£À̽º

Á¤º¸°úÇÐȸ ³í¹®Áö D : µ¥ÀÌŸº£À̽º

Current Result Document : 8 / 12 ÀÌÀü°Ç ÀÌÀü°Ç   ´ÙÀ½°Ç ´ÙÀ½°Ç

ÇѱÛÁ¦¸ñ(Korean Title) űװ£ ÀÇ¹Ì ºÐ¼®À» ÀÌ¿ëÇÑ ´ÙÁß ¹®¼­ ¿ä¾à ±â¹ý
¿µ¹®Á¦¸ñ(English Title) Multi-Document Summarization Technique using Semantic Analysis between Tags
ÀúÀÚ(Author) ÇãÁö¿í   ÁÖ¿µµµ   À̵¿È£   Jee-Uk Heu   Young-Do Joo   Dong-Ho Lee  
¿ø¹®¼ö·Ïó(Citation) VOL 39 NO. 01 PP. 0078 ~ 0088 (2012. 02)
Çѱ۳»¿ë
(Korean Abstract)
ÃÖ±Ù ÀÎÅͳÝÀÇ ±Þ¼ÓÇÑ ¹ß´Þ°ú º¸±ÞÀ¸·Î ÀÎÇÏ¿© À¥ »ó¿¡¼­ »ý¼ºµÇ´Â ¹®¼­ÀÇ ¾çÀº ÇÏ·ç°¡ ´Ù¸£°Ô Áõ°¡ÇÏ°í ÀÖ´Ù. ÇÏÁö¸¸ »ç¿ëÀÚ´Â ¸¹Àº ½Ã°£°ú ³ë·ÂÀ» µé¿© ÀÚ½ÅÀÌ ¿øÇÏ´Â Á¤º¸°¡ ´ã±ä ¹®¼­¸¦ ã±â À§ÇÏ¿© °Ë»ö ¿£ÁøÀÇ µµ¿òÀ» ¹Þ´õ¶óµµ °Ë»öµÈ ¹®¼­¸¦ ÀÏÀÏÀÌ °ËÅäÇØ¾ß ÇÑ´Ù´Â ¾î·Á¿òÀÌ Á¸ÀçÇÑ´Ù. »ç¿ëÀÚµéÀÇ ÀÌ·¯ÇÑ ¾î·Á¿òÀ» ÇؼÒÇϱâ À§ÇÏ¿© ¹®¼­ÀÇ ÇÙ½ÉÀ» È¿°úÀûÀ¸·Î ¿ä¾àÇÏ´Â ´ÙÁß ¹®¼­ ¿ä¾à ±â¹ý¿¡ ´ëÇÑ ¿¬±¸°¡ È°¹ßÈ÷ ÁøÇàµÇ¾î¿Ô´Ù. ±×·¯³ª ±âÁ¸ÀÇ ´ÙÁß ¹®¼­ ¿ä¾à ±â¹ýµéÀº ´ëºÎºÐ È®·üÀÌ·Ð ¹× ±â°èÇнÀ¿¡ ±Ù°ÅÇÑ ¹æ¹ýÀ» »ç¿ëÇϱ⠶§¹®¿¡ ÇнÀ ¹× ¿ä¾à °úÁ¤¿¡ ³ôÀº ºñ¿ëÀÌ ¿ä±¸µÇ¸ç »õ·Ó°Ô ÃâÇöÇÏ´Â °íÀ¯¸í»çµé¿¡ ´ëÇÑ ºÐ¼®ÀÌ ¿ëÀÌÇÏÁö ¾Ê´Ù´Â ´ÜÁ¡µéÀ» °¡Áö°í ÀÖ´Ù. º» ³í¹®Àº ÀÌ·¯ÇÑ ´ÜÁ¡µéÀ» ÇØ°áÇϱâ À§ÇÏ¿© °íÀ¯¸í»ç¸¦ Æ÷ÇÔÇÑ ¹®¼­ ³»¿¡ Á¸ÀçÇÏ´Â ´Ü¾îµé¿¡ ´ëÇÑ ÅÂ±× Å¬·¯½ºÅ͸¦ Æø¼Ò³ë¹Ì ½Ã½ºÅÛÀÎ Çø®Ä¿¸¦ ÀÌ¿ëÇÏ¿© ½Ç½Ã°£À¸·Î ȹµæÇÏ¿© ºÐ¼®½Ã°£°ú °è»êºñ¿ëÀ» ÁÙÀÌ°í, ´Ü¾îµé°£ ÀǹÌÀû °ü°è¸¦ ºÐ¼®ÇÏ¿© Áß¿ä ´Ü¾îµéÀ» Ãß·Á³»¸ç À̸¦ ±â¹ÝÀ¸·Î ´Ü¾îÀÇ Áß¿äµµ¿Í´Ù¸¥ ´Ü¾îµé°£ÀÇ ÀǹÌÀûÀÎ °ü°è¸¦ ºÐ¼®ÇÏ¿© Áß¿ä¹®ÀåµéÀ» Ãß·Á³»´Â ´ÙÁß ¹®¼­ ¿ä¾à ±â¹ýÀ» Á¦¾ÈÇÑ´Ù.
¿µ¹®³»¿ë
(English Abstract)
Recently, the amount of documents created on the web has been rapidly increasing day by day, due to the rapid propagation and the development of the internet. In order to find the necessary information, the user has to manually review all of the searched documents in spite of the assistance of search engines, and it requires too much time and effort. To address this problem, various multi-document summarization techniques have been studied to efficiently summarize the core of the original document. However, most of all existing multi-document summarization techniques suffer from a high cost in learning and summarization processes because their methods are mainly based on probability theory and machine learning, and failed to analyze the proper nouns which are emergently upraised. To overcome these drawbacks, we propose a novel multi-document summarization technique that analyze the importance of the key words and the semantic relatedness among them, and then detect the key sentences in the document based on the tag cluster of the each word including proper nouns in the document by exploiting Flickr which is one of the representative folksonomy systems to reduce analysis time and computing cost between each words in the documents.
Å°¿öµå(Keyword) ´ÙÁß ¹®¼­ ¿ä¾à   ÅÂ±× Å¬·¯½ºÅÍ   ÀÇ¹Ì ºÐ¼®   Multi-Document Summarization   Tag Cluster   Semantic Analysis  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå