A Schelling Extended Model in Networks—Characterization of Ghettos in Washington D.C.
Round 1
Reviewer 1 Report
The authors used GIS information to define the network to run an extended Schelling model on it, to obtain the location of ghettos in Washington. To determine which parts of the city are segregated, machine learning is used and the results are compared.
The paper should be extended with some additional definitions, examples, and descriptions, such as the definition of ghetto, explanation of the segregation and description of Schelling model. The purpose of the network model should be in the introduction or methods section.
The explanation about happiness and satisfaction of agents are somewhat confusing. If understood correct, the people are happier if they are satisfied with the housing prices based on their incomes, which are affordable to them, except from those in the ghettos that are not satisfied but they cannot afford better housing and relocate to another place.
It can be assumed that a high correlation between income and race exists, highlighting the importance of the financial gap. The paper concluded that ghettos are mostly occupied by the black people which implies racial issues. Are the black people only of African origin, or includes other ethnic groups that are not of European origin, such as people from Latin America, Asia, etc. The paper also concludes that white people are economically favoured. It would be interesting to explore reasons behind this. Is the segregation the cause or effect of this economic aspect?
If ML model is considered correct in detecting ghettos, why is it necessary to develop network model which is less accurate? Is there a comparison with the official classification about ghettos in Washington, if such classification exists?
The authors emphasised the importance of the financial gap, which controls the affordable zones where segregated people can live and suggest that a decline in this gap could favour the relocation of people, diminishing the typical isolation effect of ghettos. This can be further researched.
There are some spelling errors such as finacial, highliting or disminishing. Thorough spellcheck is advised.
Author Response
Dear Reviewer 1,
Thank yor for your comments and suggestions.
Please see the PDF attachment, where they have been adressed under the "Reviewer 1".
Best regards,
Diego Ortega.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Paper is a brief presentation of a classification of administrative units in a binary ghetto/non ghetto using a neighborhood graph on which agents change according to a evolutionary process involving tolerance and other controls reflecting on acceptance / change.
While the methodology has its merits, there is no literature review section. This is indispensable to get a clear view on the state of the art and where this research adds some value to the identification of ghettos.
The role of GIS is exacerbated, as GIS is only used to extract the neighborhood graph upon which the calculations with the model are performed. There is no spatial analysis whatsoever, justifying the application of GIS.
Moreover, as the results are highly dependent on the minimum spatial units, one of the topics that should be addressed is the sensitivity of the method to the modifiable areal unit problem (MAUP), as it introduces a statistical bias that can significantly reflect on the results of the classification even in opposite interpretations.
Whitout a strong literature review, and without the implications of the modifiable areal unit problem incorporated or discussed, the manuscript seems an exploratory or tentative step that lacks scientific soundness and completeness.
Author Response
Dear Reviewer 2,
Thank yor for your comments and suggestions.
Please see the PDF attachment, where they have been adressed under the "Reviewer 2".
Best regards,
Diego Ortega.
Author Response File: Author Response.pdf
Reviewer 3 Report
This article entitled " A Schelling extended model in Networks. Characterization of Ghettos in Washington D.C." describes very interesting research, and the authors have done a lot of meaningful work. Although the topic is very interesting, some of the method descriptions need to be improved and clarified.
Comments for author File: 
Comments.docx
Author Response
Dear Reviewer 3,
Thank yor for your comments and suggestions.
Please see the PDF attachment, where they have been adressed under the "Reviewer 3" title.
Best regards,
Diego Ortega.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
The authors have provided explanations and clarifications from the previous round of review, although changes could have been highlighted in order to be more easily noticed. The impression is still that the methods, findings, related work and discussion are written briefly and may be presented more thoroughly. It would be good to emphasise the contribution and distinction of this method with previously developed methods in the related work section.
Author Response
Dear Reviewer 1,
Thank yor for your comments and suggestions.
Please see the PDF attachment, where they have been answered under the "Reviewer 1" heading.
Best regards,
Diego Ortega.
Author Response File: Author Response.pdf
Reviewer 2 Report
The revised manuscript corrected the major deficiencies detected in the first version. There is now a simple (while short) literature review that provides some contextualization. These literature review should be expanded, in particular Section 2.3 (related works), so the readers can understand the extent to which the methodology applied here adds value to the topic. A short description of each of the application of Schelling models in these cited works would provide such explanation. The last sentence in section 2.3 is not clear.
Please correct the city name in the caption of Figure 3.
Author Response
Dear Reviewer 2,
Thank yor for your comments and suggestions.
Please see the PDF attachment, where they have been answered under the "Reviewer 2" heading.
Best regards,
Diego Ortega.
Author Response File: Author Response.pdf
Reviewer 3 Report
This is a very meaningful study, but there are still some major deficiencies in the author's writing and experiment, and the revised version has not been fundamentally changed. Therefore, I suggest the author to completely revise and resubmit.
Author Response
Dear Reviewer 3,
Thank yor for your comments and suggestions.
Please see the PDF attachment, where they have been answered under the "Reviewer 3" heading.
Best regards,
Diego Ortega.
Author Response File: Author Response.pdf
