Á¤º¸°úÇÐȸ³í¹®Áö (Journal of KIISE)
Current Result Document :
ÇѱÛÁ¦¸ñ(Korean Title) |
±àÁ¤ µ¥ÀÌÅÍ ºÐÆ÷¸¦ ¹Ý¿µÇÑ ´ÙÁß ÀνºÅϽº ÁöÁö º¤ÅÍ ±â°è ÇнÀ |
¿µ¹®Á¦¸ñ(English Title) |
Learning Multiple Instance Support Vector Machine through Positive Data Distribution |
ÀúÀÚ(Author) |
ȲÁß¿ø
¹Ú¼º¹è
ÀÌ»óÁ¶
Joong-Won Hwang
Seong-Bae Park
Sang-Jo Lee
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 42 NO. 02 PP. 0227 ~ 0234 (2015. 02) |
Çѱ۳»¿ë (Korean Abstract) |
º» ³í¹®¿¡¼´Â µ¥ÀÌÅÍ ºÐÆ÷¸¦ °í·ÁÇÑ ´ÙÁß ÀνºÅϽº ÁöÁö º¤ÅÍ ±â°è ÇнÀ ¾Ë°í¸®ÁòÀ» Á¦¾ÈÇÑ´Ù. ±âÁ¸ÀÇ ¹æ¹ýÀº ±àÁ¤ °¡¹æ ¾È¿¡¼ ¡°°¡Àå ±àÁ¤¡±ÀÎ ÀνºÅϽº¸¸ °í·ÁÇÏ¿© ¸¶ÁøÀ» ã´Â´Ù. ÀϹÝÀûÀ¸·Î ´ÙÁß ÀνºÅϽº·Î Ç¥ÇöµÈ µ¥ÀÌÅÍ¿¡¼, ±àÁ¤ °¡¹æ¿¡ Æ÷ÇÔµÈ ÀνºÅϽºµé Áß ½ÇÁ¦·Î ±àÁ¤À» ³ªÅ¸³»´Â ÀνºÅϽºµéÀº ÀÚÁú °ø°£ »ó¿¡¼ ¼·Î À¯»çÇÑ °÷¿¡ À§Ä¡ÇØ ÀÖ´Ù. Á¦¾ÈÇÑ ¹æ¹ýÀº ±âÁ¸ÀÇ ´ÙÁß ÀνºÅϽº ÁöÁö º¤ÅÍ ±â°è ÇнÀ ¾Ë°í¸®Áò Áß¿¡¼ ±àÁ¤ ÀνºÅϽºµéÀÇ ±³Â÷Á¡À» ã¾Æ ÀÌ ±³Â÷Á¡°ú °Å¸®¸¦ °è»êÇÏ¿© ¡°°¡Àå ±àÁ¤¡±ÀÎ ÀνºÅϽº¸¦ ¼±ÅÃÇÑ´Ù. ±àÁ¤ ÀνºÅϽºµéÀÇ ±³Â÷Á¡ÀÎ Çǹþ Æ÷ÀÎÆ®¸¦ ±¸ÇÏ´Â ¹æ½ÄÀº µÎ °¡ÁöÀÌ´Ù. ¸ÕÀú, ÇнÀ°úÁ¤ Áß ÃßÁ¤µÈ ±àÁ¤ ÀνºÅϽºµéÀÇ Áß½ÉÁ¡À» »ç¿ëÇÏ´Â ¹æ¹ý°ú ÇнÀ ½ÃÀÛ ½Ã¿¡ °¡Àå ±àÁ¤ÀÏ °ÍÀ¸·Î ¿¹»óµÇ´Â ±àÁ¤ ÀνºÅϽºµéÀÇ Áß½ÉÁ¡À» ã´Â ¹æ¹ýÀ¸·Î ³ª´¶´Ù. ÃÑ 12°³ÀÇ º¥Ä¡¸¶Å© ´ÙÁß ÀνºÅϽº µ¥ÀÌÅÍ ¼ÂÀ» ÅëÇØ Á¦¾ÈÇÑ ¹æ¹ýÀÌ ±âÁ¸ÀÇ ÇнÀ ¾Ë°í¸®Áò¿¡ ºñÇØ ´õ ÁÁÀº ¼º´ÉÀ» º¸ÀÓÀ» º¸ÀδÙ.
|
¿µ¹®³»¿ë (English Abstract) |
This paper proposes a modified MI-SVM algorithm by considering data distribution. The previous MI-SVM algorithm seeks the margin by considering the ¡°most positive¡± instance in a positive bag. Positive instances included in positive bags are located in a similar area in a feature space. In order to reflect this characteristic of positive instances, the proposed method selects the ¡°most positive¡± instance by calculating the distance between each instance in the bag and a pivot point that is the intersection point of all positive instances. This paper suggests two ways to select the ¡°most positive¡± pivot point in the training data. First, the algorithm seeks the ¡°most positive¡± pivot point along the current predicted parameter, and then selects the nearest instance in the bag as a representative from the pivot point. Second, the algorithm finds the ¡°most positive¡± pivot point by using a Diverse Density framework. Our experiments on 12 benchmark multi-instance data sets show that the proposed method results in higher performance than the previous MI-SVM algorithm.
|
Å°¿öµå(Keyword) |
´ÙÁß ÀνºÅϽº ÇнÀ
´ÙÁß ÀνºÅϽº ÁöÁö º¤ÅÍ ±â°è
ÁöÁö º¤ÅÍ ±â°è
µ¥ÀÌÅÍ ºÐÆ÷
multi-instance learning
MI-SVM
support vector machine
data distribution
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|