Tunnel Geology Prediction Using a Neural Network Based on Instrumented Drilling Test

Round 1

Reviewer 1 Report

Thank you for the opportunity to review. This is one of the best papers that i have been offered for review, it is very clearly presented.

Minor notes:

Line 33 reference Horner or Honer as in references

Line 44 been can be deleted 

Lines 54 and 56 referenced to 9 or 10?

Lines 287-290 should be moved in front of lines284-286?

Line 289 diagonal elements positive or negative gradient on the plot for clarity?

Figure 10 no text on colour graded legend

Figure 11 legend is not discriminated when printed in black and white

Line 446 anomalous format for reference

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

This paper deals with using a neural network to predict geological conditions for tunneling based on instrumented drilling test data. Two neural networks have been developed: basic neural network and deep neural network using a genetic algorithm (GA). Batch normalization and GA optimization approaches are employed in the deep neural network.

The article is interesting, well structured and deserves publication in this journal. In order to improve the quality of the article, the authors should clarify the following:

Why would the proposed much more complex deep neural network be used for geology prediction, when with its application slightly better results are obtained than with the basic neural network? For example: the predictive accuracy of the basic neural network was 90.55%, while that of the deep neural network was 91.68%. The conclusion is that too much of the article was spent on displaying one or another neural network, and that their application success is approximately the same. This is especially important for publication in a journal dealing with applied science.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf