• 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 : 2 / 2

ÇѱÛÁ¦¸ñ(Korean Title) »ïºÐ ±×·¡ÇÁ ±â¹Ý ¿Ü½Ä Ãßõ ±â¹ý
¿µ¹®Á¦¸ñ(English Title) A Tripartite Graph based Restaurant Recommendation Technique
ÀúÀÚ(Author) ÀÓǪ¸§   ±èÇÑÁØ   Pu-reum Lim   Han-joon Kim  
¿ø¹®¼ö·Ïó(Citation) VOL 34 NO. 01 PP. 0069 ~ 0079 (2018. 04)
Çѱ۳»¿ë
(Korean Abstract)
Á¤º¸È­ ½Ã´ë¿¡ ¼Ò¼È µ¥ÀÌÅ͸¦ È°¿ëÇÑ Ãßõ ¾Ë°í¸®ÁòÀº ±× Çʿ伺ÀÌ ¸Å¿ì Áõ°¡ÇÏ¿´´Ù. ÇÏÁö¸¸ ¼Ò¼È µ¥ÀÌÅ͸¦ È°¿ëÇÑ ±âÁ¸ÀÇ Ãßõ ¾Ë°í¸®ÁòÀº ¸¸Á·ÇÒ¸¸ÇÑ ¼º´ÉÀ» º¸ÀÌÁö ¸øÇÑ´Ù. ÀÌ´Â ¼Ò¼È µ¥ÀÌÅÍ°¡ Àΰ£ÀÇ ÀÚ¿¬¾î ÅؽºÆ® µ¥ÀÌÅÍ À̹ǷΠÇÔÃàÀû Àǹ̳ª ¹è°æÁö½Ä µîÀÌ µ¥ÀÌÅÍ ¼Ó¿¡ ¼û°ÜÁ® Àֱ⠶§¹®ÀÌ´Ù. µû¶ó¼­ ¼Ò¼È µ¥ÀÌÅ͸¦ È°¿ëÇÑ Ãßõ ¾Ë°í¸®ÁòÀÌ ¸¸Á·ÇÒ¸¸ÇÑ Ãßõ ¼º´ÉÀ» ³»±â À§Çؼ­´Â µ¥ÀÌÅÍ ¼Ó¿¡ ¼û°ÜÁø Á¤º¸¸¦ ÃßÃâÇÏ¿© Ãßõ¿¡ ¹Ý¿µÇؾ߸¸ ÇÑ´Ù. À̸¦ ÇØ°áÇϱâ À§ÇØ º» ³í¹®¿¡¼­´Â »ïºÐ ±×·¡ÇÁ(Tripartite Graph)¸¦ ÀÌ¿ëÇÑ »õ·Î¿î Ãßõ ¾Ë°í¸®ÁòÀ» Á¦¾ÈÇÑ´Ù. º» ³í¹®ÀÇ Ãßõ ¾Ë°í¸®ÁòÀº »ïºÐ ±×·¡ÇÁÀÇ ±¸Á¶Àû Ư¼ºÀ» È°¿ëÇÑ ¿¬»ê °úÁ¤À» ÅëÇØ µ¥ÀÌÅÍ ¼Ó¿¡ ½ÉÃþÀûÀ¸·Î ¼û°ÜÁ® ÀÖ´Â Á¤º¸¸¦ Ãßõ¿¡ Àû¿ëÇÒ ¼ö ÀÖ´Ù. ¶ÇÇÑ Á¦¾È ±â¹ýÀº »ó´ëÀûÀ¸·Î ÀÛÀº Á¤º¸¿¡¼­µµ ±âÁ¸ÀÇ Ãßõ ¾Ë°í¸®Áòº¸´Ù À¯¿ëÇÑ Á¤º¸¸¦ ´õ ¸¹ÀÌ ÃßÃâÇÒ ¼ö ÀÖ´Ù.
¿µ¹®³»¿ë
(English Abstract)
In the modern information age, the need for recommendation algorithms using social data has significantly increased. However, conventional recommendation algorithms using social data have not achieved reasonable performance. This is because social data is natural language data, and thus background knowledge and implicit words are hidden in the data. Therefore, to develop recommendation systems using social data, hidden information in the social data should be reflected in the recommendation algorithm. To this end, we propose a novel recommendation algorithm based upon the tripartite graph. The proposed algorithm utilizes the structural characteristics of the tripartite graph, and thus it can extract hidden information in the social data. As a result, the proposed algorithm can extract more useful information than other conventional recommendation algorithms even though the amount of given data is relatively small.
Å°¿öµå(Keyword) ÅؽºÆ® ¸¶ÀÌ´×   Ãßõ ¾Ë°í¸®Áò   ÀÚ·á ±¸Á¶   µ¥ÀÌÅͺ£À̽º   text mining   recommendation algorithm   data structure   database  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå