Abstract
In the field of chemometrics, an important issue in multivariate calibration is model updating. Model updating is the adaption process in which a model obtained for a given set of samples and measurement conditions (primary) is updated to predict the analyte in new samples and measurement conditions (secondary). The calibration method partial least squares is applied with two new updating approaches. In one approach, only one updated model is obtained to predict the analyte amount in both primary and secondary conditions. The other approach forms two updated models in which one model is used to predict in primary conditions and second model based on the first model is used to predict in secondary conditions. Both approaches are evaluated with near-infrared spectral datasets. Datasets include spectra of soil, corn, olive oil adulterated with sunflower and pharmaceutical tablets. Fusion process and single merits are used to select models. Model selection methods are evaluated based on prediction errors using selected models.
Regularization Adaption Processes for Multivariate Calibration Maintenance
In the field of chemometrics, an important issue in multivariate calibration is model updating. Model updating is the adaption process in which a model obtained for a given set of samples and measurement conditions (primary) is updated to predict the analyte in new samples and measurement conditions (secondary). The calibration method partial least squares is applied with two new updating approaches. In one approach, only one updated model is obtained to predict the analyte amount in both primary and secondary conditions. The other approach forms two updated models in which one model is used to predict in primary conditions and second model based on the first model is used to predict in secondary conditions. Both approaches are evaluated with near-infrared spectral datasets. Datasets include spectra of soil, corn, olive oil adulterated with sunflower and pharmaceutical tablets. Fusion process and single merits are used to select models. Model selection methods are evaluated based on prediction errors using selected models.
Comments
Poster #W53