Estimating the Heating Load of Buildings for Smart City Planning Using a Novel Artificial Intelligence Technique PSO-XGBoost
Round 1
Reviewer 1 Report
The paper deals with new tool able to support smart city planning by using some indices. This topic is interesting enough and method well-developed and discussed. However, in my opinions, several actions should be made in order to help the reader:
1. the general description of buildings is necessary (i.e. how dataset has been defined);
2. despite authors affirm “However, due to the details of SVM, RF, GP, and 94 CART were introduced in many previous kinds of literature [31-35]; so, they were not introduced in 95 the present study.”, in my opinion, some description is required;
3. The previous works should be better discussed;
4. graphs and figures should be commented.
I suggest major revision
Author Response
"Please see the attachment."
Author Response File: 
Author Response.docx
Reviewer 2 Report
The description for PSO is too brief and not self-contained.
The correlation plots have no significant usefulness to the work, you could have plotted the correlation between the inputs and the HL, this might give more insight.
The bench mark methods implementation looks to be biased, since there is no description of how you parameterized the SVM,... and other methods!
The R2 plots do not add values to the results.
You did not show the procedure of how you determined the significance of the inputs.
I was expecting a time series plot.
Generally, the English should be improved.
Author Response
"Please see the attachment."
Author Response File: Author Response.docx
Reviewer 3 Report
1. A time series plot comparing the predicted and the true data is needed.
2 The advantages and the disadvantages of the other methods are not discussed..
3. Future work is not discussed.
4. Τhe structure of the paper must be defined in the introduction section.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
The quality of paper has been significantly improved.
In my opinion, it can be published.
Regards
Author Response
Thank you very much for your recommendation!
Reviewer 2 Report
The amount of your original work in the paper is not commensurate with the work you referenced. The manuscript contains a very extensive description of other works compared to your work amount.
PSO-XGBoost is not a novel technique but you can say a new one.
The literature review section contains a lot of redundant and irrelevant information, what the reader really needs to know is: What are the previous methods, Why do you think they did not work well highlighting the information that you will improve in your work, other wise the reader will get lost, bored and looses focus. Reviewing the relevant work should be in a critical and relevant manner.
The title is not correct: Does the energy efficiency have heating load?
I can not understand the term: HL of Energy Efficiency of Buildings? no need then for the second (of)
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 3
Reviewer 2 Report
_
