³í¹®¸í |
°¡¿ì½Ã¾È ÇÁ·Î¼¼½º ¸ðµ¨¿¡ ´ëÇÑ µ¥ÀÌÅÍ ÇÊÅ͸µ ±â¹ý Àû¿ë / Data Filtering for a Gaussian Process Model / Ãß°è-05. °ÇÃàȯ°æ ¹× ¼³ºñ |
¼ö·Ï»çÇ× |
´ëÇѰÇÃàÇÐȸ Çмú¹ßÇ¥´ëȸ ³í¹®Áý, v.34 n.2 (2014-10) |
ÆäÀÌÁö |
½ÃÀÛÆäÀÌÁö(385) ÃÑÆäÀÌÁö(2) |
ÁÖÁ¦¾î |
°¡¿ì½Ã¾È ÇÁ·Î¼¼½º ; µ¥ÀÌÅÍ ±â¹Ý ¸ðµ¨ ; µ¥ÀÌÅÍ ÇÊÅ͸µ ; ·£´ý »ùÇà ÄÁ¼¾¼½º ; Gaussian Process ; Data-driven Model ; Data Filtering ; RANSAC |
¿ä¾à2 |
This study reports a data filtering method for a GP(Gaussian Process) model for an existing building. The GP model is explained in another paper presented by the authors. This paper deals with how to deal with outliers usually appearing in a training data set for the GP model. In this study, RANSAC (RANdom SAmple Consensus) is selected for detecting the outliers from the training data. The RANSAC method can be beneficially applied to improve the accuracy of the GP model. |