Next Article in Journal
Spatial Distribution, Migration, and Ecological Risk of Cd in Sediments and Soils Surrounding Sulfide Mines—A Case Study of the Dabaoshan Mine of Guangdong, China
Previous Article in Journal
An Improved Model for Water Quality Management Accounting for the Spatiotemporal Benthic Flux Rate
 
 
Article
Peer-Review Record

Ant Colony Based Artificial Neural Network for Predicting Spatial and Temporal Variation in Groundwater Quality

Water 2023, 15(12), 2222; https://doi.org/10.3390/w15122222
by Ravinder Bhavya, Kaveri Sivaraj and Lakshmanan Elango *
Reviewer 1: Anonymous
Reviewer 2: Anonymous
Water 2023, 15(12), 2222; https://doi.org/10.3390/w15122222
Submission received: 3 May 2023 / Revised: 3 June 2023 / Accepted: 5 June 2023 / Published: 13 June 2023
(This article belongs to the Section Water Quality and Contamination)

Round 1

Reviewer 1 Report

1.Line -6: “Data-driven models based on artificial intelligence are efficiently used to solve complex  problems”. This sentence can be moved to line -10, after: manual labor.

2. Line: 15,what is frequency of data, is it monthly or twice a year ??

Line: 37, “ANNs has also proved its applicability in handling problems in agriculture, medical science, education, finance, cyber security, and 38 trading commodity”. Provide reference for each sector where ML is applied.

There some type errors for example at “.” In line :161   after network

Line: 192-193, not clear- it can be rewritten

Line: 194, what is interval of periodicity

Line: 199, in general ion balance error  < 5% accepts, what is range in your case, otherwise provide reference for 10%.

 

In the methods section, author may provide the details about where these ANN were deployed??

Line:337, “there is not connection between any well”-  rewrite this sentence.

There are some typo errors, authors can do the proof reading carefully. 

Author Response

Please see attachment

Author Response File: Author Response.docx

Reviewer 2 Report

Comments to the authors:

In this study, the authors applied an ant colony-based artificial neural network to predict spatial and temporal variation in multiple groundwater quality parameters. Overall, the research was methodological sound with promising results. Here are some comments to be addressed:

(i) Abstract: Highlight the research gap and why/how important machine learning prediction used for groundwater quality and its management

(ii) Introduction: Huge modifications are necessary for the introduction section. The research gaps and significance of the study are not shown. Moreover, the authors should provide more precise research objectives in this study.

(iii) Connection between machine learning prediction and groundwater prediction is slightly lacking. Please rewrite this section. Some relevant studies should be incorporated into the manuscript.

(a) doi:10.1007/s10661-020-08543-4.

(b) doi: 10.1007/s10661-021-09202-y

(iv) Submitted paper's formatting required revision.

(v) Coefficient of Determination should be 2.5.1 Coefficient of Determination (R2). Same to RMSE, NSE, MAE

(vi) Why these performance criteria were used?

(vii) Move Figure 3 to the methodology section. It is not part of the result

(viii) What is the meaning of asterisks in Table 1

(ix) Provide an explanation for formulae of variables in the performance criteria

(x) Table 2: Add a column of the unit instead of putting the unit side by side

(xi) what are the input parameter and output parameter? Quite unclear.

(xii) Discussion is lacking in this study. Did it improve the prediction accuracy? Also, what are the proposed mitigations? 

(xiii) Conclusion section seems to be a repetition of the results section. Huge modifications are required. Please provide insights into this study and what can be further done in the future.

(ixx) Implications for future research may also be included in the conclusion at the end.

Moderate English editing is required

Author Response

Please see attachment

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

The authors have substantially addressed comments that I raised during the first review. Overall, it reads better, however still minor comments to be addressed:

1.English Editing: Minor English revision is required.
2.Introduction section reads better: Include the recent literature to enrich your introduction section:

(i)  Applications of Deep Learning in Water Quality Management: A State-of-the-Art Review

(ii) Water quality index modeling using random forest and improved SMO algorithm for support vector machine in Saf-Saf river basin

3. Provide supporting information for further discussing this statement: In general, the models performed well when the standard deviation of the data was comparatively high. That is, the MLPNN-ACO learnt the data better when the standard deviation is higher. For such datasets, the model performance was considerably higher as indicated by the performance metrics. In the case of water quality parameters which did not vary much with respect to time such as pH, F- and U2+, the learning rate was poor. 

Goob job!

 

Minor English revision is required.
Do not use "Such" for the starting of a sentence in academic writing.

 

Author Response

Please see attachment

Author Response File: Author Response.docx

Back to TopTop