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

»çÀÌÆ®¸Ê

Loading..

Please wait....

±¹³» ³í¹®Áö

Ȩ Ȩ > ¿¬±¸¹®Çå > ±¹³» ³í¹®Áö > Çѱ¹Á¤º¸Ã³¸®ÇÐȸ ³í¹®Áö > Á¤º¸Ã³¸®ÇÐȸ ³í¹®Áö ¼ÒÇÁÆ®¿þ¾î ¹× µ¥ÀÌÅÍ °øÇÐ

Á¤º¸Ã³¸®ÇÐȸ ³í¹®Áö ¼ÒÇÁÆ®¿þ¾î ¹× µ¥ÀÌÅÍ °øÇÐ

Current Result Document : 4 / 4

ÇѱÛÁ¦¸ñ(Korean Title) Çâ»óµÈ ±³Â÷ ¹öÀü °áÇÔ ¿¹ÃøÀ» À§ÇÑ ÀÌÁö¾È ÃÖÀûÈ­ ÇÁ·¹ÀÓ¿öÅ©
¿µ¹®Á¦¸ñ(English Title) Bayesian Optimization Framework for mproved Cross-Version Defect Prediction
ÀúÀÚ(Author) ÃÖÁ¤È¯   ·ù´ö»ê   Jeongwhan Choi   ksan Ryu  
¿ø¹®¼ö·Ïó(Citation) VOL 10 NO. 09 PP. 0339 ~ 0348 (2021. 09)
Çѱ۳»¿ë
(Korean Abstract)
ÃÖ±Ù ¼ÒÇÁÆ®¿þ¾î °áÇÔ ¿¹Ãø ¿¬±¸´Â ±³Â÷ ÇÁ·ÎÁ§Æ® °£ÀÇ °áÇÔ ¿¹Ãø»Ó¸¸ ¾Æ´Ï¶ó ±³Â÷ ¹öÀü ÇÁ·ÎÁ§Æ® °£ÀÇ °áÇÔ ¿¹Ãø ¶ÇÇÑ ÀÌ·ç¾îÁö°í ÀÖ´Ù. Á¾·¡ÀÇ ±³Â÷ ¹öÀü °áÇÔ ¿¹Ãø ¿¬±¸µéÀº WP(Within-Project)·Î °¡Á¤ÇÑ´Ù. ÇÏÁö¸¸, CV(Cross-Version) ȯ°æ¿¡¼­´Â ÇÁ·ÎÁ§Æ® ¹öÀü °£ÀÇ ºÐÆ÷ Â÷ÀÌÀÇ Á߿伺À» °í·ÁÇÑ ¿¬±¸µéÀÌ ¾ø´Ù. º» ¿¬±¸¿¡¼­´Â ´Ù¸¥ ¹öÀü °£ÀÇ ºÐÆ÷ Â÷À̱îÁö °í·ÁÇÏ´Â ÀÚµ¿È­µÈ º£ÀÌÁö¾È ÃÖÀûÈ­ ÇÁ·¹ÀÓ¿öÅ©¸¦ Á¦¾ÈÇÑ´Ù. À̸¦ ÅëÇØ ºÐÆ÷ Â÷ÀÌ¿¡ µû¶ó ÀüÀÌ ÇнÀ(Transfer Learning) ¼öÇà ¿©ºÎ¸¦ ÀÚµ¿À¸·Î ¼±ÅÃÇÏ¿© ÁØ´Ù. ÇØ´ç ÇÁ·¹ÀÓ¿öÅ©´Â ¹öÀü °£ÀÇ ºÐÆ÷ Â÷ÀÌ, ÀüÀÌ ÇнÀ°ú ºÐ·ù±â(Classifier)ÀÇ ÇÏÀÌÆÛÆĶó¹ÌÅ͸¦ ÃÖÀûÈ­ÇÏ´Â ±â¹ýÀÌ´Ù. ½ÇÇèÀ» ÅëÇØ ÀüÀÌ ÇнÀ ¼öÇà ¿©ºÎ¸¦ ºÐÆ÷Â÷ ±âÁØÀ¸·Î ÀÚµ¿À¸·Î ¼±ÅÃÇÏ´Â ¹æ¹ýÀÌ È¿°úÀûÀ̶ó´Â °ÍÀ» ¾Ë ¼ö ÀÖ´Ù. ±×¸®°í ÃÖÀûÈ­¸¦ ÀÌ¿ëÇÏ´Â °ÍÀÌ ¼º´É Çâ»ó¿¡ È¿°ú°¡ ÀÖÀ¸¸ç ÀÌ·¯ÇÑ °á°ú ¼ÒÇÁÆ®¿þ¾î ÀνºÆå¼Ç ³ë·ÂÀ» °¨¼ÒÇÒ ¼ö ÀÖ´Ù´Â °ÍÀ» È®ÀÎÇÒ ¼ö ÀÖ´Ù. À̸¦ ÅëÇØ ±³Â÷ ¹öÀü ÇÁ·ÎÁ§Æ® ȯ°æ¿¡¼­ ½Å±Ô ¹öÀü ÇÁ·ÎÁ§Æ®¿¡ ´ëÇÏ¿© È¿°úÀûÀÎ Ç°Áú º¸Áõ È°µ¿ ¼öÇàÀ» Áö¿øÇÒ °ÍÀ¸·Î ±â´ëµÈ´Ù.
¿µ¹®³»¿ë
(English Abstract)
In recent software defect prediction research, defect prediction between cross projects and cross-version projects are actively studied. ross-version defect prediction studies assume WP(Within-Project) so far. However, in the CV(Cross-Version) environment, the previous ork does not consider the distribution difference between project versions is important. In this study, we propose an automated Bayesian ptimization framework that considers distribution differences between different versions. Through this, it automatically selects whether o perform transfer learning according to the difference in distribution. This framework is a technique that optimizes the distribution ifference between versions, transfer learning, and hyper-parameters of the classifier. We confirmed that the method of automatically electing whether to perform transfer learning based on the distribution difference is effective through experiments. Moreover, we can ee that using our optimization framework is effective in improving performance and, as a result, can reduce software inspection effort. his is expected to support practical quality assurance activities for new version projects in a cross-version project environment.
Å°¿öµå(Keyword) Software Defect Prediction   Bayesian Optimization   Transfer Learning   Cross-Version Defect Prediction  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå