Á¤º¸°úÇÐȸ ³í¹®Áö B : ¼ÒÇÁÆ®¿þ¾î ¹× ÀÀ¿ë
ÇѱÛÁ¦¸ñ(Korean Title) |
Çϸð´Ï Ž»ö ¾Ë°í¸®ÁòÀ» ÀÌ¿ëÇÑ ¼ÒÇÁÆ®¿þ¾î ½Å·Ú¼º ¼ºÀå ¸ðµ¨ÀÇ ¸Å°³º¯¼ö ÃßÁ¤ ±â¹ý |
¿µ¹®Á¦¸ñ(English Title) |
A Parameter Estimation Method of Software Reliability Growth Model using Harmony Search Algorithm |
ÀúÀÚ(Author) |
±èÅÂÇö
¹éÁ¾¹®
Taehyoun Kim
Jongmoon Baik
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 41 NO. 03 PP. 0183 ~ 0190 (2014. 03) |
Çѱ۳»¿ë (Korean Abstract) |
¼ÒÇÁÆ®¿þ¾î ½Å·Ú¼º ¼ºÀå ¸ðµ¨Àº Å×½ºÆ® ´Ü°è¿¡¼ ¹ß°ßµÇ´Â °áÇÔ Çö»óÀ» ±â¹ÝÀ¸·Î ¼ÒÇÁÆ®¿þ¾î ½Å·Ú¼ºÀ» ÃßÁ¤ÇÏ´Â ´ëÇ¥ÀûÀÎ ¸ðµ¨ Áß ÇϳªÀÌ´Ù. ¼ÒÇÁÆ®¿þ¾î ½Å·Ú¼º ¼ºÀå ¸ðµ¨Àº °¢ ¸ðµ¨ÀÇ Æ¯¼ºÀ» ³ªÅ¸³»´Â ¸Å°³º¯¼öµéÀ» Æ÷ÇÔÇÏ°í ÀÖÀ¸¸ç ÀÌ¿¡ ´ëÇÑ ÃßÁ¤Àº ¸ðµ¨ÀÇ Á¤È®¼ºÀ» °áÁ¤Áþ´Â Áß¿äÇÑ ¿ä¼ÒÀÌ´Ù. ÀϹÝÀûÀ¸·Î ÃÖ´ë ¿ìµµ ÃßÁ¤ ȤÀº ÃÖ¼Ò Á¦°ö ÃßÁ¤°ú °°Àº ¼öÄ¡Àû ±â¹ýµéÀÌ ¼ÒÇÁÆ®¿þ¾î ½Å·Ú¼º ¼ºÀå ¸ðµ¨ÀÇ ¸Å°³º¯¼ö¸¦ ÃßÁ¤ÇÏ´Â µ¥ ÀÌ¿ëµÇ°í ÀÖÁö¸¸, ÀÌ·¯ÇÑ ±â¹ýµéÀº ¼ÒÇÁÆ®¿þ¾î ½Å·Ú¼º ¼ºÀå °î¼±°ú °°Àº ºñ¼±Çü ÇÔ¼öÀÇ ¸Å°³º¯¼ö ÃßÁ¤¿¡´Â ÀûÇÕÇÏÁö ¾Ê´Ù´Â ¹®Á¦Á¡ÀÌ Á¸ÀçÇÑ´Ù. µû¶ó¼ º» ³í¹®¿¡¼´Â ±âÁ¸ÀÇ ¼öÄ¡Àû ±â¹ýµéÀÇ ¹®Á¦¸¦ ÇØ°áÇϱâ À§ÇØ Çϸð´Ï Ž»ö ¾Ë°í¸®ÁòÀ» ÀÌ¿ëÇÑ ¼ÒÇÁÆ®¿þ¾î ½Å·Ú¼º ¼ºÀå ¸ðµ¨ÀÇ ¸Å°³º¯¼ö ÃßÁ¤ ±â¹ýÀ» Á¦¾ÈÇϵµ·Ï ÇÑ´Ù. ¶ÇÇÑ, ÇØ´ç ±â¹ýÀÇ Á¤È®¼º°ú ¾ÈÁ¤¼ºÀ» Æò°¡Çϱâ À§ÇØ 8°³ÀÇ µ¥ÀÌÅÍ¿¡ ´ëÇØ ±âÁ¸ÀÇ ¼öÄ¡Àû ±â¹ý ¹× À¯ÀüÀÚ ¾Ë°í¸®ÁòÀ» ÀÌ¿ëÇÑ ¸Å°³º¯¼ö ÃßÁ¤ ±â¹ý°úÀÇ ºñ±³¸¦ ¼öÇàÇÏ¿´´Ù.
|
¿µ¹®³»¿ë (English Abstract) |
Software Reliability Growth Model (SRGM) is one of the famous models to estimate software reliability based on a failure detection phenomenon during a testing phase. SRGM has some parameters which represent characteristics of each model and the estimation of these parameters is an important factor determining the accuracy of SRGM. In general, numerical methods, such as Maximum Likelihood Estimation (MLE) or Least Square Estimation (LSE), are used to estimate the parameters of SRGM. However, such methods are not suitable for SRGM of which the cumulative density function is a non-linear function. In this paper, we propose a parameter estimation method of SRGM using Harmony Search algorithm (HS). We conducted experiments on eight datasets to compare the accuracy and stability of the proposed method with the existing numerical method and Genetic Algorithm (GA).
|
Å°¿öµå(Keyword) |
¼ÒÇÁÆ®¿þ¾î ½Å·Ú¼º
¼ÒÇÁÆ®¿þ¾î ½Å·Ú¼º ¼ºÀå ¸ðµ¨
¸Å°³º¯¼ö ÃßÁ¤
Çϸð´Ï Ž»ö ¾Ë°í¸®Áò
software reliability
software reliability growth model
parameter estimation
harmony search algorithm
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|