A Smart Framework for Managing Natural Disasters Based on the IoT and ML

Round 1

Reviewer 1 Report

Although a lot of research in this direction has already been done. But, here authors have summarized the use of IoT in disaster management by citing the example of flood modelling.  Seeking the relevance of updated summarized literature, I recommend it for the publication with minor suggestion. Authors also need to add some literature on the use of hybrid model, e.g., AI based physical model or integration of two different models in the flood modelling and assessment. 

Author Response

Thanks for reviewing our paper. Thank you for your pertinent comment and support. These were really useful to improve the quality of our paper. Our responses are in line, starting with “Answer:”.
We highlighted the changes in the revised version by yellow color. Then we mentioned to the number page and line of changes beside each comment.

Author Response File: Author Response.pdf

Reviewer 2 Report

A smart framework based on IoT and ML is proposed for managing natural disasters is proposed and the model is tested for real data for the city of Jeddah, Saudi Arabia. The authors should address the following comments in the revision:

In abstract it is mentioned that real data of the city of Jeddah, Saudi Arabia from 2009 to 2014. But in section 4, it is mentioned 2009 to 2013. Mention the appropriate one.

The model is trained with 1827 samples collected for 5 years and the authors claim 99% accuracy. I recommend authors to include dataset from 2014 to 2022. Because Figure 1 is captured on 2022.

At the end of Introduction section, organization of the manuscript is missing.

The expressions (x) and (y) should be presented like equations with variables.

The classification metrics are defined in expressions  (1) to (6), which is unnecessary. These definitions are available in all existing books.

The authors should write a proper justification for ‘how RF is achieving highest performance over other algorithms?’.

The analytical and optimization of parameters in model building is missing.

What about the employment of deep learning for this application?

What is the performance of proposed strategy for other datasets?

Author Response

Thanks for reviewing our paper. Thank you for your pertinent comment and detailed observations. These were really useful to improve the quality of our paper. Our responses are in line, starting with “Answer:”.
We highlighted the changes in the revised version by yellow color. Then we mentioned to the number page and line of changes beside each comment.

Author Response File: Author Response.pdf

Reviewer 3 Report

1.       Please revise abstract by including more details in a brief way.

2.       Figs 2 about general overview seems blur. Have a look at all the figures.

3.       The proposed algorithm is very generic. Please revise by including more details.

4.       Literature review can be improved by including latest studies relevant to AI, ML applications, challenges and bottlenecks such as, Incorporation of Blockchain Technology for Different Smart Grid Applications: Architecture, Prospects, and Challenges and many others in 2022 and 2023.

5.       Focus on results to present in a better way.

6.       Please focus more on the benefits of this study for researchers with future research areas.

 

 

Author Response

Thanks for reviewing our paper. Thank you for your pertinent comment and detailed observations. These were really useful to improve the quality of our paper. Our responses are in line, starting with “Answer:”.
We highlighted the changes in the revised version by yellow color. Then we mentioned to the number page and line of changes beside each comment.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

I have no further comments.