I have the data of total dissolved soilds of apple as references (y-variable).
I also have near-infared spectra data as predictors (x-variables).
I have the StatSoft Statistica software for the analysis.
you had better used matalb software. and you can use apple wet chemistry data as reference Y and spectral data as x and than first pre-processed the data and check non-linearity of the data because ANN is non-linear logarithm and build model. by higher Rc, Rp and you can predicted it
Different related software's can be used for building ANN predictive model using NIR spectra. My suggestion are Unscrambler and IBM modeler. Before ANN modeling, PCA should be done to reduce large spectra variables to PC1, PC2 .... . after calibrating ANN, you have predicted values by calibrated model and you have also reference values for each sample. Now use RPD index to realizing the goodness of your calibrated model.
Firstly, you should chose one of tools, such as some machine learning modules in matlab or python. Then divide all the data into at least three parts for calibration, validation and test. And then build the ANN with different structures, transfer functions and so on. By many repeated comparison between the evaludation indexes during the periods of calibration and validation respectively, a best ANN model should be confirmed, which could be reliable for final purpose.
- Badges
- Science topic
- Similar topics
- Biological Science
- Zoology
- Invertebrate Zoology
- Arthropods
- Insects