• 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 : 8 / 11 ÀÌÀü°Ç ÀÌÀü°Ç   ´ÙÀ½°Ç ´ÙÀ½°Ç

ÇѱÛÁ¦¸ñ(Korean Title) ³×Æ®¿öÅ© µ¥ÀÌÅÍ ºÐ¼®À» À§ÇÑ ¸µÅ© ±×·¡ÇÁ ±â¹Ý Áßø Ä¿¹Â´ÏƼ Ž»ö ¹æ¾È
¿µ¹®Á¦¸ñ(English Title) Overlapping Community Detection Based on Link-Graphs for Network Analysis
ÀúÀÚ(Author) È«Áö¿ø   ÀÌÀ¯Áø   ±è»ó¿í   Jiwon Hong   Yujin Lee   Sang-Wook Kim  
¿ø¹®¼ö·Ïó(Citation) VOL 34 NO. 02 PP. 0089 ~ 0098 (2018. 08)
Çѱ۳»¿ë
(Korean Abstract)
Çö½Ç ¼¼°èÀÇ Çö»ó Áß ´Ù¼ö´Â ³×Æ®¿öÅ© µ¥ÀÌÅÍ·Î ¸ðµ¨¸µµÉ ¼ö ÀÖ´Ù. ƯÈ÷, ÃÖ±Ù¿¡´Â ¼Ò¼È ³×Æ®¿öÅ© ¼­ºñ½º(SNS)ÀÇ ¹ß´Þ·Î ´Ù¾çÇÑ ÇüÅÂÀÇ ¼Ò¼È ³×Æ®¿öÅ©°¡ »õ·ÎÀÌ µîÀåÇÏ°í ÀÖ´Ù. ÀÌ·¯ÇÑ ³×Æ®¿öÅ© µ¥ÀÌÅÍ¿¡¼­ ¼­·Î ±íÀº °ü°è¸¦ °®´Â Ä¿¹Â´ÏƼ(community)µéÀ» ã¾Æ³»´Â Ä¿¹Â´ÏƼ Ž»ö ±â¼úÀº Áß¿äÇÑ ºÐ¼® µµ±¸ÀÇ Çϳª·Î »ç¿ëµÈ´Ù. Ä¿¹Â´ÏƼ´Â ³ëµå °£ÀÇ °ü°èÀÇ ÁýÇÕÀ¸·Î Á¤ÀÇµÉ ¼ö ÀÖ´Ù. º» ³í¹®¿¡¼­´Â ³ëµå °£ÀÇ °ü°è¸¦ ³ªÅ¸³»´Â ¿§Áö°¡ ´Ù¼öÀÇ Ä¿¹Â´ÏƼ¿¡ ¼ÓÇÒ ¼ö ÀÖ´Ù´Â Á¡¿¡ Âø¾ÈÇÏ¿© ¸µÅ© ±×·¡ÇÁ(link-graph)¸¦ ¹ÙÅÁÀ¸·Î seed È®Àå ±â¹ýÀ» ÀÌ¿ëÇÏ¿© ¿§Áö Áß½ÉÀûÀÌ¸ç °¢ ¿§Áö°¡ ´Ù¼öÀÇ Ä¿¹Â´ÏƼ¿¡ ¼ÓÇÏ´Â °ÍÀ» Çã¿ëÇÏ´Â Áßø Ä¿¹Â´ÏƼ Ž»ö ¹æ¾ÈÀ» Á¦¾ÈÇÑ´Ù. À̸¦ ¹ÙÅÁÀ¸·Î ±âÁ¸ Áßø Ä¿¹Â´ÏƼ Ž»ö ¹æ¾ÈÀÌ Ä¿¹Â´ÏƼ °£ÀÇ ¾èÀº Áßø¸¸À» Çã¿ëÇÏ´Â ¹®Á¦¸¦ ÇØ°áÇÏ°í ´õ Á¤È®ÇÑ Áßø Ä¿¹Â´ÏƼ Ž»öÀ» ´Þ¼ºÇÒ ¼ö ÀÖ´Ù. ¶ÇÇÑ, ½ÇÇèÀ» ÅëÇØ Á¦¾ÈÇÏ´Â ¹æ¹ýÀÌ Áßø Ä¿¹Â´ÏƼ Ž»ö¿¡ È¿°úÀûÀÓÀ» Áõ¸íÇÏ¿´´Ù.
¿µ¹®³»¿ë
(English Abstract)
A number of phenomena in the real world can be modeled as network data. In particular, various types of social networks are emerging recently thanks to the rapid growth of social network services (SNS). Community detection is one of the most important analytic tools, which finds community structures that are densely connected set of nodes. A community can be defined as a set of relationships between nodes. In this paper, we focus on the fact that an edge that represents relationship between nodes can be part of multiple communities. Using link-graph, we adopt seed expansion method to achieve edge-centric overlapping community detection that allows an edge to be in multiple communities. With our proposed method, we can remedy the problem that some of existing overlapping community detection method only allows swallow overlapping and thus achieve more accurate overlapping community detection. We prove that the proposed method is efficient for searching overlapping communities from real world networks via a set of experiments.
Å°¿öµå(Keyword) ¼Ò¼È ³×Æ®¿öÅ© ºÐ¼®   Ä¿¹Â´ÏƼ Ž»ö   Áßø Ä¿¹Â´ÏƼ Ž»ö   Social network analysis   community detection   overlapping community detection  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå