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

»çÀÌÆ®¸Ê

Loading..

Please wait....

Çмú´ëȸ ÇÁ·Î½Ãµù

Ȩ Ȩ > ¿¬±¸¹®Çå > Çмú´ëȸ ÇÁ·Î½Ãµù > Çѱ¹Á¤º¸Åë½ÅÇÐȸ Çмú´ëȸ > 2019³â Ãß°èÇмú´ëȸ

2019³â Ãß°èÇмú´ëȸ

Current Result Document : 4 / 29 ÀÌÀü°Ç ÀÌÀü°Ç   ´ÙÀ½°Ç ´ÙÀ½°Ç

ÇѱÛÁ¦¸ñ(Korean Title) ÄÁÅÙÃ÷ Âü½Å¼º ºÐ¼®¿¡ ±â¹ÝÇÑ °³ÀÎÈ­ °¡Ä¡Á¤º¸ÀÇ Ãßõ ±â¹ý
¿µ¹®Á¦¸ñ(English Title) Recommendation Technique of Personalized Qualitative Information Based on Content Novelty Analysis
ÀúÀÚ(Author) ±è¸íÈÆ   Myeong-hun Kim  
¿ø¹®¼ö·Ïó(Citation) VOL 23 NO. 02 PP. 0269 ~ 0272 (2019. 10)
Çѱ۳»¿ë
(Korean Abstract)
Ãßõ ±â¹ýÀº °³ÀÎÀÇ °ü½É»ç¿Í »óȲÀÎÁö ±â¹ÝÀÇ °³ÀÎÈ­µÈ ¾ÆÀÌÅÛÀ» Á¦°øÇÔÀ¸·Î½á ¾ÆÀÌÅÛÀÇ ¼Òºñ°úÁ¤¿¡¼­ ¹ß»ýÇÏ´Â ºÎÇϸ¦ ÁÙ¿©ÁÖ°í Á¤º¸ ¼ÒºñÀÇ È¿À²¼ºÀ» Áõ´ë½ÃÅ°´Âµ¥ Áß¿äÇÑ ¿ªÇÒÀ» ÇÑ´Ù. ÀϹÝÀûÀ¸·Î °³ÀÎÈ­µÈ Á¤º¸´Â °³ÀÎÀÇ ´©ÀûµÈ °ü½É»ç¿Í À¯»çµµ°¡ ³ôÀº °æÇâÀ» °¡Áø´Ù°í °¡Á¤ÇÏ´Â ÀüÅëÀûÀÎ Content-Based(CB)±â¹ýµé°ú´Â ´Þ¸®, º» ¿¬±¸¿¡¼­´Â °³ÀÎÀÌ »õ·Î¿î ÄÁÅÙÃ÷¸¦ °¡Áø Âü½Å¼ºÀÌ ³ôÀº Á¤º¸¿¡ ´õ ³ôÀº ¹ÝÀÀ°ú Á¤º¸ È®»ê Á¤µµ¸¦ º¸ÀÓÀ» Áõ¸íÇÏ°í ÀÌ »ç½ÇÀ» ¹ÙÅÁÀ¸·Î ÄÁÅÙÃ÷ Âü½Å¼® ºÐ¼®¿¡ ±â¹ÝÇÑ °³ÀÎÈ­µÈ °¡Ä¡Á¤º¸ÀÇ Ãßõ ±â¹ýÀ» Á¦½ÃÇÑ´Ù. Á¤º¸ Àͼ÷µµ(Familiarity)¿Í Á¤º¸¿¡ ´ëÇÑ ¹ÝÀÀµµÀÇ °ü°è¸¦ Social Network¿¡¼­ ½ÇÁ¦ Á¤º¸ È帧À» °üÂûÇÏ¿© º» °¡Ä¡Á¤º¸ Ãßõ ±â¹ýÀÌ ÀüÅëÀûÀÎ CB ±â¹ÝÀÇ ±â¹ýµéº¸´Ù ´õ °íµµÈ­µÈ °³ÀÎÈ­ °¡Ä¡Á¤º¸¸¦ Á¦°øÇÒ ¼ö ÀÖÀ½À» Áõ¸íÇÏ´Â °ÍÀÌ º» ¿¬±¸ÀÇ ¸ñÀûÀÌ´Ù.
¿µ¹®³»¿ë
(English Abstract)
Recommendation technique plays a significant role in providing personalized information to users, with enhanced satisfaction and reduced overload for consuming content. In this paper, we prove that people in online social network tends to have interest in information which has novel contents and spread it more actively while traditional recommendation technique based on content-based algorithms insist personal interest in the present has close relationship(or high similarity) with people¡¯s past interest. Also, we prove that low familiarity (same as high novelty in this research) of content means high degree of diffusing information by observing information flow in huge online social network such as Facebook. The goal of this research is to provide more personalized information(qualitative information) with Content-Novelty analysis than traditional algorithms based on CB method, and define personalized information is not the high similarity but high novelty of content.
Å°¿öµå(Keyword) Recommendation Technique   Qualitative Information   Content Novelty   Content Similarity   Personalized Information  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå