Identification of singular samples in near infrared spectrum correction set by using Monte Carlo cross validation
- 10.2991/icmmita-16.2016.58How to use a DOI?
- Near infrared spectroscopy; Monte Carlo cross validation; Partial least square; singular sample
Identification of singular samples is the basis of the robustness of the calibration model for near infrared spectra. By using the Monte Carlo method of cross validation (MCCV) ,this experiment identifies the singular samples in the calibration set of starch samples. By using the method of partial least squares (PLS) ,modeling of starch samples coming from the before and after eliminating singular samples of calibration set , which is validated by test set, finally ,comparing the stability and prediction accuracy of the model before and after eliminating abnormal samples by using root mean square error of cross validation (RMSECV), correlation coefficient (R) and root mean square error of prediction (RMSEP) as evaluation index. The results show that the near infrared singular sample recognition algorithm of MCCV can effectively eliminate the singular sample,which can make the correction model have lower RMSECV, RMSEP and higher R, and significantly improve the stability and prediction ability of the model.
- © 2017, the Authors. Published by Atlantis Press.
- Open Access
- This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).
Cite this article
TY - CONF AU - Aixiao Zou AU - Yaxin Qu PY - 2017/01 DA - 2017/01 TI - Identification of singular samples in near infrared spectrum correction set by using Monte Carlo cross validation BT - Proceedings of the 2016 4th International Conference on Machinery, Materials and Information Technology Applications PB - Atlantis Press SP - 328 EP - 333 SN - 2352-538X UR - https://doi.org/10.2991/icmmita-16.2016.58 DO - 10.2991/icmmita-16.2016.58 ID - Zou2017/01 ER -