2010³â Ãá°èÇмú´ëȸ
Current Result Document : 1 / 1
ÇѱÛÁ¦¸ñ(Korean Title) |
ÁöÁöº¤Åͱâ°è¿Í Ä«ÀÌÁ¦°ö Åë°è·®À» ÀÌ¿ëÇÑ ½ºÆÔ ºí·Î±×(Splog) ÆǺ° ½Ã½ºÅÛ |
¿µ¹®Á¦¸ñ(English Title) |
A Splog Detection System Using Support Vector Machines and ¥ö2 Statistics |
ÀúÀÚ(Author) |
À̼º¿í
Songwook Lee
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 14 NO. 01 PP. 0905 ~ 0908 (2010. 05) |
Çѱ۳»¿ë (Korean Abstract) |
º» ¿¬±¸ÀÇ ¸ñÀûÀº À¥ ȯ°æ¿¡¼ ½ºÆÔ ºí·Î±×(Splog)¸¦ ÀÚµ¿À¸·Î ÆǺ°ÇÏ´Â ½Ã½ºÅÛÀ» °³¹ßÇÏ´Â °ÍÀÌ´Ù. ¸ÕÀú ºí·Î±×ÀÇ HTMLÀ» Á¦°ÅÇÑ ÈÄ Ç°»ç¸¦ ºÎÂøÇÏ¿´´Ù. ¾îÈÖ/Ç°»ç ½ÖÀ» ÀÚÁú·Î »ç¿ëÇÏ¿´À¸¸ç Ä«ÀÌÁ¦°ö Åë°è·®À» ÀÌ¿ëÇÏ¿© À¯¿ëÇÑ ÀÚÁúÀ» ¼±ÅÃÇÏ¿´´Ù. ¼±ÅÃµÈ ÀÚÁúÀÇ °¡ÁßÄ¡¸¦ º¤ÅͷΠǥÇöÇÑ ÈÄ, ÁöÁöº¤Åͱâ°è(Support Vector Machines)¸¦ ÇнÀÇÏ¿© ÀÚµ¿À¸·Î ½ºÆÔ ºí·Î±×¸¦ ÆǺ°ÇÏ´Â ½Ã½ºÅÛÀ» Á¦¾ÈÇÏ¿´À¸¸ç, SPLOG µ¥ÀÌÅÍ ÁýÇÕÀ¸·Î ½ÇÇèÇÑ °á°ú F1ôµµ·Î 90.5%ÀÇ Á¤È®·üÀ» ¾ò¾ú´Ù.
|
¿µ¹®³»¿ë (English Abstract) |
Our purpose is to develope the system which detects splogs automatically among blogs on Web environment. After removing HTML of blogs, they are tagged by part of speech(POS) tagger. Words and their POS tags information is used as a feature type. Among features, we select useful features with ¥ö2 statistics and train the SVM with the selected features. Our system acquired 90.5% of F1 measure with SPLOG data set.
|
Å°¿öµå(Keyword) |
spam blog detection
splog
support vector machine
chi square statistics
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|