[Response] Thanks for your comments. We have added more
details as suggested. Each recorded spectrum was obtained by averaging
16 scans at a resolution of 4 cm-1. Each sample was
measured three times. The average of the three spectra obtained from
each same sample was used in subsequent analysis. The spectra file of
samples was pretreated through normalization to reduce measurement
error. Each spectrum was aligned to the baseline using instrumentation
software. Line139-143.
The
external validation was not properly conducted. The author used a
subset of the data that was collected in the same experiment and
labelled it as test set, which is not correct. ANN is notorious for
overfitting and the only way to prove that the model is working is by
conducting proper external validation. To do so, a new batch of oil
and potatoes should be subjected to the same experimental procedure
and conditions as was previously done to generate the modeling data
and then predict the response of interest from the new set of spectra
using the developed model.
[Response] Thanks for your
comments, we have added
the external validation
in the Manuscript as suggested. Line 256-267
5. The author compared the ANN model to PLS one; however, he/she did not
mention any details related to the development of
the PLS model. Was it also
developed using only three FTIR peaks? Or were the full spectra of the
samples used to develop the model?