• 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 : 10 / 11 ÀÌÀü°Ç ÀÌÀü°Ç   ´ÙÀ½°Ç ´ÙÀ½°Ç

ÇѱÛÁ¦¸ñ(Korean Title) »ç¿ëÀÚ ¿µÇâ·Â°ú È°µ¿·Â ºÐ¼®À» Àû¿ëÇÑ Á¤º¸Ãßõ ±â¹ýÀÇ ¿¬±¸
¿µ¹®Á¦¸ñ(English Title) Research on an Information Recommendation Technique based on the Analysis of Influence and Activity of Users
ÀúÀÚ(Author) ¼ÛÁöÇö   ±è°æÁÖ   À̹μö   Jihyun Song   Kyeongjoo Kim   Minsoo Lee  
¿ø¹®¼ö·Ïó(Citation) VOL 33 NO. 01 PP. 0105 ~ 0113 (2017. 04)
Çѱ۳»¿ë
(Korean Abstract)
ÃÖ±Ù ¼Ò¼È ³×Æ®¿öÅ© ÀÌ¿ëÀÚÀÇ ¼ö°¡ ±ÞÁõÇϸ鼭 »ç¿ëÀÚÀÇ ¼±È£µµ¸¦ ºÐ¼®ÇÏ¿© ÀûÀýÇÑ ¾ÆÀÌÅÛÀ» ÃßõÇÏ´Â °³ÀÎÈ­ Ãßõ ¼­ºñ½º°¡ ÇÊ¿äÇÏ´Ù. º» ³í¹®¿¡¼­´Â Ãßõ¿¡ È°¿ëÇÒ ¼ö ÀÖ´Â µÎ °¡ÁöÀÇ ¿ä¼ÒµéÀÎ »ç¿ëÀÚÀÇ ¿µÇâ·Â°ú È°µ¿·Â ¿ä¼Ò¸¦ ºÐ¼®ÇÏ¿© Á¤º¸ Ãßõ ±â¹ýÀ» Á¦¾ÈÇÏ¿´´Ù. »ç¿ëÀÚÀÇ ½Å·Úµµ¸¦ ºÐ¼®ÇÔ¿¡ À־´Â »ç¿ëÀÚÀÇ ¿µÇâ·Â(´Ù¸¥ »ç¿ëÀÚÀÇ ÇàÀ§¸¦ ÃËÁøÇÔ) ¿ä¼Ò¿Í »ç¿ëÀÚÀÇ È°µ¿·Â(°èÁ¤ »ý¼ºÀ¸·ÎºÎÅÍ ÃÖ±Ù±îÁö Á¤º¸»ý¼ºÀÇ ¾ç°ú ÆÐÅÏ) ¿ä¼Ò¸¦ ½ÉÃþÀûÀ¸·Î ºÐ¼®ÇÏ¿© ±× °¡ÁßÄ¡¸¦ °è»êÇÏ¿´´Ù. »ç¿ëÀÚÀÇ ¿µÇâ·ÂÀ» ºÐ¼®ÇÔ¿¡ À־´Â ƯÁ¤ Ä«Å×°í¸®¿¡¼­ »ç¿ëÀÚ°¡ »ý¼ºÇÑ À¯¿ëÇÑ Á¤º¸ÀÇ ¼ö, Àü¹®¼º, ÇØ´ç »ç¿ëÀÚÀÇ Á¤º¸¸¦ ±¸µ¶ÇÏ´Â »ç¿ëÀÚÀÇ ¼ö¿Í ´Ù¸¥ »ç¿ëÀÚµé·ÎºÎÅÍ ¹ÞÀº ±àÁ¤ÀûÀÎ ¸Þ½ÃÁö¸¦ °í·ÁÇÏ¿´´Ù. »ç¿ëÀÚ È°µ¿·ÂÀ» ºÐ¼®ÇÔ¿¡ À־´Â ´Ù¾çÇÑ Æò°¡ºÐÆ÷, °èÁ¤ »ý¼ºÀ¸·ÎºÎÅÍ ÃÖ±Ù±îÁöÀÇ Á¤º¸ »ý¼ºÀÇ ¾ç°ú ÃÖ±Ù È°µ¿À» °í·ÁÇÏ¿´´Ù. Yelpµ¥ÀÌÅ͸¦ ±â¹ÝÀ¸·Î Ãßõ±â¹ýÀ» Àû¿ëÇÏ°í ±¸ÇöÇÏ¿´°í, ÅëÇÕÀûÀ¸·Î ¼º´ÉÆò°¡¸¦ ¼öÇàÇÑ °á°ú ³í¹®¿¡¼­ Á¦¾ÈÇÑ ±â¹ýÀÌ ´ëü·Î Á¤È®ÇÑ ¿¹ÃøÀ» Á¦°øÇÔÀ» Áõ¸íÇÏ¿´´Ù.
¿µ¹®³»¿ë
(English Abstract)
Over the past few years, the number of users for Social Network Service has been exponentially increasing and it is important for the service to analyze applicable data and provide personalized information to users. In this paper, we propose an information recommendation technique based on two specific types of analysis on user behaviors such as the user influence (i.e., affect other person¡¯s behaviour) and the user activity (i.e., amount of information produced and patterns since account creation). The components to measure each of the user influence and user activity are identified. The amount of useful information produced, expertise, number of users subscribing information, and positive messages from other users are considered for analyzing the user influence. The evaluation distribution, amount of information produced since account creation, and recent activities were considered in order to analyze the user activity. The accuracy of the information recommendation is verified using Yelp data and shows significantly promising results.
Å°¿öµå(Keyword) Ãßõ   »ç¿ëÀÚ ¿µÇâ·Â   »ç¿ëÀÚ È°µ¿·Â   Yelp   Social Network Service   Personalized Information   recommendation technique  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå