Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This study presents a comprehensive exploration of the interplay between Transit-Oriented Development (TOD) economic performance and social equity in RSAs, employing advanced methodologies like eXtreme Gradient Boosting (XGBoost) and SHapley Additive exPlanations (SHAP) to decipher the complex relationships between TOD characteristics and social equity outcomes. It's an interesting topic. The article is of good quality and well structured. My comments are as follows.

(1)     The introduction and Literature review sections are not thick enough. A more comprehensive literature review and an in-depth discussion and comparison with previous studies should be conducted. I suggest author(s) to read and summarize more latest academic research regarding the theme of this paper，and update and supplement the literature review part as needed.

(2)     A short paragraph should be supplemented at the end of the article to discuss the Limitations and future research directions.

(3)     Using cities of different sizes for empirical analysis may bring different results. It is suggested that the author(s) expand the sample for empirical analysis to improve the robustness of the study.

(4)     Although this paper discusses the reasons for cross section analysis, a longitudinal study is still necessary.

Comments on the Quality of English Language

Moderate editing of English language required

Author Response

Response to the Comments of Reviewer 1

Dear editor and Reviewer 1:

Thank you very much for your constructive comments and suggestions about our manuscript, which have helped us to improve the quality of the manuscript. We have carefully considered the valuable comments and suggestions you provided and made corresponding revisions to the manuscript based on them. The modified content is presented in RED font in the document so that you can quickly identify the changed parts. Meanwhile, we have highlighted in YELLOW in the comments to clearly indicate the correspondence between each modification and your review comments. The details of all comments are listed below point by point. If you have any questions, please don’t hesitate to contact us.

We appreciate your earnest work and hope that the correction will be approval.

Yours sincerely,

Authors

Detailed answers to Reviewer 1

Comment 1: The introduction and Literature review sections are not thick enough. A more comprehensive literature review and an in-depth discussion and comparison with previous studies should be conducted. I suggest author(s) to read and summarize more latest academic research regarding the theme of this paper，and update and supplement the literature review part as needed?

Reply: We greatly value your insightful feedback and suggestions concerning the Introduction and Literature Review sections of our manuscript. We concur that expanding the literature review and providing in-depth comparisons with prior studies are crucial for deepening the relevance and scholarly rigor of our work. In response to your suggestions, we have meticulously revised the manuscript's introduction section as follows:

• Background Explanation: We have elaborated further on the context and objectives of TOD , emphasizing the potential conflicts between economic benefits and social equity. This enhanced contextual analysis is designed to provide readers with a more comprehensive understanding of the study's significance.
• Problem Statement and Research Motivation: We have articulated the issue of uneven distribution of economic opportunities and equity in TOD implementation. By citing pertinent literature, we aim to bolster the persuasiveness and academic validity of our research motivation.
• Literature Review and Identification of Research Gaps: We have identified and discussed the gaps in existing research more explicitly and detailed how this study addresses these gaps. This refined literature review aims to clarify the gaps, thereby underscoring the novelty and significance of our research.
• Introduction to Research Methods and Models: We offer a thorough explanation of the research methodologies and models employed, including the NPE model, XGBoost, and the SHAP framework. Our goal is to enhance the transparency and scientific robustness of our study, thereby improving its credibility and reproducibility.
• Contributions and Limitations of the Study: We have explicitly outlined the contributions and limitations of this study, particularly regarding its role in addressing existing research gaps. We intend for this comprehensive discussion to provide readers with a clearer perspective on the expected outcomes and potential limitations of the study.

We hope these revisions satisfactorily address your concerns and contribute to the enhanced clarity and depth of the manuscript. For an in-depth discussion in Section 1 of the introduction, please refer to note 01 for Reviewer 1 on lines 43-83.

We have implemented significant enhancements in the literature review section, focusing on both structural and content-based improvements:

• Structure and Organization: We have refined the structure to improve the clarity and logical flow of the literature review. It now starts with a comprehensive overview of TOD’s economic performance, progresses to a detailed analysis of its multidimensional impacts, and concludes with a critique of existing research alongside suggestions for future studies.
• Depth and Detail: Our revised analysis delves deeper into the economic aspects of TOD, including real estate appreciation, job creation, and business activities. We also explore the dual impacts on social equity, emphasizing both the positive outcomes and potential adverse effects such as gentrification and the displacement of low-income residents.
• Innovations and Contributions: We introduce a significant enhancement to the existing NP model by incorporating employment data, housing cost indicators, and socio-demographic factors. This innovative approach allows for a more holistic assessment of both the economic performance and equity implications of the TOD model, offering fresh insights and methodologies to the field.
• Critical Analysis: We have expanded our critical review of the literature, identifying research biases, model limitations, and data constraints. Additionally, we critically examine the social challenges that may arise from the TOD model. This approach not only underscores the need for a more balanced and comprehensive TOD policy but also contributes to a deeper understanding of its broader implications.

These amendments aim to address your insightful feedback comprehensively, enhancing the scholarly depth and relevance of our manuscript. For an in-depth discussion in Section 2 of the literature review, please refer to note 01 for Reviewer 1 on lines 123-261.

Comment 2: A short paragraph should be supplemented at the end of the article to discuss the Limitations and future research directions.

Reply: Thank you for your valuable feedback. In response, we have integrated a comprehensive discussion on research limitations and directions for future research at the conclusion of our article. This addition aims to provide readers with a clearer vision of how future inquiries could build upon our findings and further the field. Specific enhancements include:

• Sample Diversity and Size Expansion: We acknowledge the need to broaden our sample size and diversify the range of cities included in our study. By extending our analysis to encompass a variety of urban environments, from small towns to megacities, we aim to enhance the external validity of our findings and ensure they are representative of different urban dynamics.
• Longitudinal Studies: We emphasize the critical importance of conducting longitudinal studies to trace the temporal effects of TOD on social equity and address causal relationships. This approach is essential for understanding the long-term impacts of TOD initiatives.
• Methodological Enhancements: We propose the adoption of advanced methodologies, including agent-based modeling and geospatial analysis, to deepen our understanding of TOD impacts. Additionally, we aim to expand our conceptual framework to incorporate factors such as health, environmental quality, and travel behavior. We also plan to leverage technological advancements in big data, machine learning, and remote sensing to enhance the precision of our TOD impact assessments. Furthermore, our research emphasizes the critical need to investigate socio-economic stratification within TOD areas and the mechanisms by which TOD affects social mobility. This examination is essential for evaluating the inclusiveness of TOD initiatives and their potential to promote equitable urban development. In future studies, we also intend to explore the effects of walk lengths on residents' travel choices and the accessibility of TOD areas. This will involve analyzing how different walking distances influence resident travel patterns, residential preferences, and economic activities within TOD zones, providing insights into how these factors vary with walk distance.

We trust these modifications will provide a robust foundation for understanding and extending the research in this field. Please refer to note 02 For the Reviewer 1（lines 867-907）.

Comment 3: Using cities of different sizes for empirical analysis may bring different results. It is suggested that the author(s) expand the sample for empirical analysis to improve the robustness of the study.

Reply: We greatly appreciate your insightful suggestions and concur with your observation regarding the potential variability in empirical results across cities of different sizes. In response to your recommendation, we have included a detailed discussion on the limitations of our current study and proposed directions for future research in the concluding sections of our manuscript.

We acknowledge that the validity and generalizability of our findings are constrained by the limited sample size and the uniformity of city selection involved in our study. Recognizing these limitations, we intend to broaden the scope of our city sample in future empirical analyses. This expansion will encompass cities varying not only in size but also in characteristics such as economic development levels, population density, geographic location, and cultural contexts. Such diversification is aimed at enhancing the external validity of our research and ensuring that our findings are applicable across a more varied spectrum of urban environments.

By addressing these aspects, we hope to provide a more robust foundation for understanding the dynamics of urban development and the impact of related policies. Please refer to note 03 For the Reviewer 1（lines 869-877）.

Comment 4: Although this paper discusses the reasons for cross section analysis, a longitudinal study is still necessary.

Reply: We are grateful for your insightful comments highlighting the need for longitudinal studies despite our current reliance on cross-sectional analysis. We fully acknowledge this perspective and have accordingly discussed the limitations of our study and outlined future research directions in the concluding sections of our manuscript.While our research provided insights into the relationship between TOD and social equity via cross-sectional analysis, we recognize that this approach does not adequately capture the long-term dynamics of urban development. To address this gap, we propose that future studies employ longitudinal methodologies to monitor changes in property values, rental markets, and demographic shifts over extended periods. This approach will be invaluable in providing empirical evidence of the long-term effects of TOD on urban social equity. Such data are crucial for aiding policymakers and urban planners in crafting sustainable TOD strategies that preemptively address issues like gentrification and displacement.

By transitioning to longitudinal research, we aim to cultivate a more profound understanding of the enduring impacts of TOD policies and planning. This shift will significantly enhance our capacity to support the formulation of holistic and effective urban development strategies.We appreciate your constructive feedback, which has been instrumental in refining our research approach and deepening our analytical framework.

 Please refer to note 04 for Reviewer 1（lines 879-888）.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The work has more merits from which the following can be mentioned as important:

- A wide database is used for analyses performed,

- The authors have solid skills in statistical analysis.

However, a few aspects remain questionable, as follows:

-It is not specified with the necessary clarity, the way of anonymizing the data obtained from that mobile operator,

-Walking to a railway station is a feature of travel habits and people's cultural characteristics (It is hard to believe that instead of walking for 20 minutes, some people do not use other alternatives). From here appears the question related to the generality of the analysis or in other words: how general is the analysis? Can this analysis method be transferred elsewhere, outside the Chinese space? What are the results if it is considered another walking length to the railway station?

Author Response

Response to the Comments of Reviewer 2

Dear editor and Reviewer 2:

Thank you very much for your constructive comments and suggestions about our manuscript, which have helped us to improve the quality of the manuscript. We have carefully considered the valuable comments and suggestions you provided and made corresponding revisions to the manuscript based on them. The modified content is presented in RED font in the document so that you can quickly identify the changed parts. Meanwhile, we have highlighted in BLUE in the comments to clearly indicate the correspondence between each modification and your review comments. The details of all comments are listed below point by point. If you have any questions, please don’t hesitate to contact us.

We appreciate your earnest work and hope that the correction will be approval.

Yours sincerely,

Authors

Detailed answers to Reviewer 2

Comment 1: The work has more merits from which the following can be mentioned as important:

- A wide database is used for analyses performed,

- The authors have solid skills in statistical analysis..

Reply: Thank you for your constructive feedback. We dedicated significant effort to constructing and utilizing an extensive database for our analysis. We believe that this approach not only added depth and breadth to the study but also enhanced the reliability and generalizability of our findings. We ensured the accuracy and representativeness of our data through rigorous screening and validation methods. Our team possesses a robust skill base in statistical analysis. In this study, we employed advanced statistical methods to ensure the accuracy and validity of our results. We conducted a thorough exploratory analysis of the data and applied appropriate statistical tests to confirm the statistical significance of our findings. We are committed to maintaining the highest standards of quality and academic rigor in our research. We will continue our efforts to ensure that our work meets these standards..

Comment 2: However, a few aspects remain questionable, as follows: It is not specified with the necessary clarity, the way of anonymizing the data obtained from that mobile operator.

Reply: Thank you for your insightful comments concerning our paper's data anonymization practices. We understand the importance of maintaining rigorous ethical standards and ensuring data privacy. We are happy to provide a more detailed explanation of the anonymization process employed in our study:

• Initial Anonymization Processing: The mobile operators conducted an initial anonymization of the data before transferring it to us. They removed all personally identifiable information (PII), including phone numbers, device IDs, and other identifiable details. This step ensured that the data received was preliminarily anonymized.
• Generation of Random Identifiers: To replace the original identifiers, we employed the cryptographic hash function SHA-256 to generate random identifiers. These identifiers are dissociated from any PII and are not susceptible to reverse engineering, thereby safeguarding participant anonymity.
• Consistency Maintenance: Throughout the data processing and analysis phases, we ensured that each user was consistently associated with the same random identifier to maintain data integrity.
• Data Storage and Processing: All data was stored and processed using these unique, non-identifiable random identifiers. This measure ensured that participants' identities remained anonymous throughout the analytical process. Our analysis utilized only these anonymized datasets, and all results were presented in an aggregated form to prevent any possibility of individual participant identification.

We appreciate the opportunity to clarify these procedures and will revise our manuscript accordingly to include a comprehensive description of these anonymization methods. We believe that enhancing the transparency of our data handling practices will address your concerns and strengthen the credibility of our study.

Please refer to note 01 For Reviewer 2 (lines 381-398) .

Comment 3: Walking to a railway station is a feature of travel habits and people's cultural characteristics (It is hard to believe that instead of walking for 20 minutes, some people do not use other alternatives). From here appears the question related to the generality of the analysis or in other words: how general is the analysis? Can this analysis method be transferred elsewhere, outside the Chinese space? What are the results if it is considered another walking length to the railway station?

Reply: Thank you for your insightful comments and questions. We value your engagement with our research and appreciate your constructive feedback. Here are our detailed responses to your points:

• Generalizability of the Analysis: Our analysis, based on data from Dalian, China, utilizes a robust design approach that is adaptable to various contexts. The core principles and analytical framework, including the extraction and processing of mobile location data, spatial aggregation, and anonymization techniques, are designed to be universally applicable. By implementing these methods in diverse geographic and cultural settings, researchers can obtain comparable insights, thereby enhancing the generalizability of our findings. The decision to use a 20-minute walking radius as the spatial scope for station area studies is grounded in extensive literature review and widely accepted norms regarding the maximum walking distance, ensuring a comprehensive coverage of TOD economic performance and demographic characteristics. Recognizing the variability in individual walking habits and preferences across different regions, our analytical framework is inherently flexible. It allows for adjustments to spatial units and walking distances based on local data and cultural traits, ensuring its effectiveness in diverse settings outside China.
• Transferability of Analytic Methods: While our methodology is technically transferable, adaptations may be required to align with local cultural, economic, and social contexts. For instance, adjustments to spatial units and walking distances can tailor the framework to specific regional conditions. Additionally, we have employed Mapbox to derive walking isochrones, with a focus on temporal rather than spatial distances to increase the flexibility and applicability of our approach.
• Considering Different Walk Lengths: The choice of walk length can significantly influence the outcomes of our study. Shorter walk lengths might encourage greater pedestrian traffic and commercial activity, whereas longer distances could lead to the separation of residential and commercial zones, impacting the area's economic vitality. Although variations in walking time could alter our results, we can refine our research hypotheses and validate our findings by adjusting study parameters and collecting new data. Therefore, in the concluding paragraphs, we make a recommendation for a more in-depth study to explore the specific impacts of varying walking times on TOD areas.

In conclusion, although our study was initially focused on Dalian, the underlying methodology is versatile and can be customized to account for local travel habits and cultural nuances. Acknowledging the importance of testing various parameters, such as walk length, to affirm the robustness and generalizability of our results, we aim to incorporate these aspects into future research.

Please refer to note 02 for Reviewer 2 (lines 278-291 and lines 903-907).

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The paper investigates the relation between social equity and node, place and economic characteristics of rail transit station areas. The literature review is well documented and the authors reveals their contribution in the field. The methodology and data are extensively presented. The results are thoroughly interpreted and suport the discussions and conclusions.

Still, there are some issues the authors should clarify:

1. The RSA are characterised by three dimension: node, place and economic variables. But the XGBoost is used to analyse only the non-linear relation between the economic indicators and social equity. Why the relation between node and place variables is not analysed ? (line 88 and paragraph 3.5.1).

2. Line 173: The citation Zhou et al. [30] is incorrect. Reference [30] has other authors.

3. Figure 2: The legend contains a rectangle "City center", but the map does not contains it.

4. Figure 2 contains the name of the districts of the study area. I recommend to label the districts in figure 4 and 5 to ease the reading of the results. It is very difficult to identify the spatial location of RSA without the district delineation.

5. Table 1: The title refers to economic performance variables solely, but it also includes node and place variables. A change of the table's title has to be done.

6. Table 1: There are physical variables presented, but the unit measures are not included (for distance, time, prices)

7. Figure 11: The top representation from the left column and the third representation from the right column refers to the same interaction pair: weekend residential population area vs. facilitation of bus transfer, but changing the axes. Still the graphical representation is different and not rotated. Please recheck the pairs interactions.

8. The amount of data collection is huge. As a reviewer from European Union, where data collection and personal data manipulation is subject of regulation for individual protection, and in accordance to the journal general policy, a clear statement concerning personal data collecting process, the agreement from individuals and the data protection protocol has to be included. In EU, every research that uses individuals' data is subject of approval. This observation do not minimise the authors work, but is a protection for different cultural and political perspectives that could arise from the worldwide readers.

Author Response

Response to the Comments of Reviewer 3

Dear editor and Reviewer 3:

Thank you very much for your constructive comments and suggestions about our manuscript, which have helped us to improve the quality of the manuscript. We have carefully considered the valuable comments and suggestions you provided and made corresponding revisions to the manuscript based on them. The modified content is presented in RED font in the document so that you can quickly identify the changed parts. Meanwhile, we have highlighted in GREEN in the comments to clearly indicate the correspondence between each modification and your review comments. The details of all comments are listed below point by point. If you have any questions, please don’t hesitate to contact us.

We appreciate your earnest work and hope that the correction will be approval.

Yours sincerely,

Authors

Detailed answers to Reviewer 3

Comment 1: The RSA are characterised by three dimension: node, place and economic variables. But the XGBoost is used to analyse only the non-linear relation between the economic indicators and social equity. Why the relation between node and place variables is not analysed ? (line 88 and paragraph 3.5.1).

Reply: We would like to express our gratitude for your valuable comments. In your review, you pointed out that RSA models typically encompass three dimensions: node, place, and economic variables. However, you noted that the XGBoost model employed in our study appears to analyze only the nonlinear relationship between economic indicators and social equity, potentially overlooking the relationships involving node and place variables

We would like to clarify that the reference to "economic performance" in line 88 and paragraph 3.5.1 does not pertain to the economic dimensions of the RSA model. Instead, it specifically refers to the set of indicators utilized to assess economic performance within the context of our study. Our study indeed employs the XGBoost model to analyze the nonlinear relationship between economic performance indicators and social equity, encompassing the complex interactions among node, location, and economic variables as they relate to social equity.

We acknowledge the necessity of being more explicit in the manuscript regarding how the XGBoost model integrates these three dimensions to predict social equity. To address this, we will include a detailed explanation in the revised manuscript, elucidating how the XGBoost model incorporates node, location, and economic variables to analyze their combined impact on social equity.

Please refer to note 01 For Reviewer 3 (lines 416-419).

Comment 2: Line 173: The citation Zhou et al. [30] is incorrect. Reference [30] has other authors.

Reply: Thank you for your careful review of the references section of our manuscript. We acknowledge the citation error in Line 173 regarding reference [30] and sincerely apologize for any confusion it may have caused. The original sentence incorrectly used "et al." to indicate a single author's literature. We will correct this to ensure the accuracy of the citation:

• Original sentence (error): "This endeavor builds upon the research conducted by Chen et al. [29], Su et al. [17], and Zhou et al. [30], among others."
• Corrected sentence: "This endeavor builds upon the research conducted by Chen et al. [29], Su et al. [17], and Zhang [30], among others."

We recognize that accurate literature citation is essential for academic integrity and research transparency. We will make this correction immediately in the revised manuscript and ensure that all other citations are also scrutinized to avoid similar errors.

Please refer to note 02 For Reviewer 3 (line 246).

Comment 3: Figure 2: The legend contains a rectangle "City center", but the map does not contains it.

Reply: Thank you for your observation regarding the representation of the city center in our figure. We acknowledge that the initial depiction using a rectangle in the legend was unclear and potentially misleading. To provide clarity, we have revised the figure by removing the rectangle representation from the legend. Instead, the city center is now clearly indicated by an arrow at its actual location, as shown in Figure 2. This adjustment more accurately reflects the city center's irregular shape and enhances the figure's representational fidelity.

Please refer to note 03 For Reviewer 3 (line 276).

Comment 4: Figure 2 contains the name of the districts of the study area. I recommend to label the districts in figure 4 and 5 to ease the reading of the results. It is very difficult to identify the spatial location of RSA without the district delineation.

Reply: Thank you for your observation regarding the labeling of the administrative districts in central Dalian. We have labeled the four administrative central districts—Zhongshan District, Xigang District, Shahekou District, and Ganjingzi District—in Figures 4 and 5. This labeling aims to enhance the clarity and comprehensibility of our figures, providing a more accurate geographical context for our analysis.

Please refer to note 04 For Reviewer 3(line 512 and line 540).

Comment 5: Table 1: The title refers to economic performance variables solely, but it also includes node and place variables. A change of the table's title has to be done.

Reply: Thank you for your feedback. The economic performance variables here do not refer to the economic dimension of the three Node-Place-Economic dimensions in the station areas. Specifically, this study measures the economic performance characteristics of the station area through the three node, place, and economic dimensions. To avoid any misunderstandings, we have modified the table headings as follows:

• Original table title: Table 1 Descriptive characteristics of economic performance variables
• Revised title: Table 1 Descriptive characterization of three-dimensional Node-Place-Economic variables in TOD economic performance

Please refer to note 05 For Reviewer 3(lines 381-382).

Comment 6: Table 1: There are physical variables presented, but the unit measures are not included (for distance, time, prices).

Reply: Thank you for your observation regarding the absence of unit measures in Table 1. We have taken note of your feedback and have now incorporated the specific units for each relevant indicator within the 'Indicator' column. This update ensures that the table is more informative and the data is presented with the necessary precision for a comprehensive understanding.

Please refer to note 06 For Reviewer 3 (lines 382-383).

Comment 7: Figure 11: The top representation from the left column and the third representation from the right column refers to the same interaction pair: weekend residential population area vs. facilitation of bus transfer, but changing the axes. Still the graphical representation is different and not rotated. Please recheck the pairs interactions.

Reply: Thank you for your observations regarding the graphical representations in Figure 11. To address your concerns, we provide a detailed analysis of the two figures referenced, which showcase the interaction effects between "weekend residential population density" and "facilitation of bus transfers" on the Shapley interaction values.

Figure 1 (the third from the right column):

• Horizontal Axis: Facilitation of bus transfers
• Vertical Axis: Shapley interaction value
• Color Variable: Weekend residential population density
• Explanation: This graph illustrates how the Shapley interaction values change at different levels of bus transfer facilitation, with color coding to indicate the impact of weekend residential population density. As depicted, higher densities (shown in red) tend to cluster at lower and medium facilitation levels. At high facilitation levels (greater than 50), the interaction values are more variable and associated with lower densities (shown in blue).

Figure 2 (the top figure from the left column):

• Horizontal Axis: Weekend residential population density
• Vertical Axis: Shapley interaction value
• Color Variable: Facilitation of bus transfers
• Explanation: This graph shows the variation in Shapley interaction values across different levels of residential population density, with color coding to reflect the impact of bus transfer facilitation. It reveals that areas with higher facilitation (shown in red) are concentrated around moderate residential densities (approximately 0.25 to 0.35), where interaction values are relatively high. Lower facilitation levels (shown in blue) are more prevalent in areas with both low and high residential densities.

Combined Analysis:

Both figures illustrate a trend where Shapley interaction values increase with both bus transfer facilitation and residential population density, albeit with some fluctuations. The interaction between bus transfer facilitation and residential population density is complex. Higher interaction values are observed at moderate levels of both variables, indicating a significant impact on social equity under these conditions.

Reasons for Differences in Results:

• Different Horizontal Variables: Figure 1 focuses on the impact of changes in bus transfer facilitation on the Shapley values, whereas Figure 2 focuses on changes in residential population density.
• Different Color Coding: Figure 1 uses residential population density for color coding, allowing observation of how bus transfer facilitation impacts Shapley values across different residential densities within the same graph. Conversely, Figure 2 uses bus transfer facilitation for color coding, allowing observation of how residential density impacts Shapley values across different levels of facilitation.

These figures present the interaction effects from different perspectives, where Figure 1 emphasizes the influence of bus transfer facilitation under varying residential densities, and Figure 2 highlights how residential population density influences the outcomes under different levels of facilitation. This approach aims to provide a comprehensive understanding of the complex relationships between these variables and their impact on social equity. Therefore, despite involving the same variables, Figures 1 and 2 produce distinct results and are not simply rotations of the axes.

Comment 8: The amount of data collection is huge. As a reviewer from European Union, where data collection and personal data manipulation is subject of regulation for individual protection, and in accordance to the journal general policy, a clear statement concerning personal data collecting process, the agreement from individuals and the data protection protocol has to be included. In EU, every research that uses individuals' data is subject of approval. This observation do not minimise the authors work, but is a protection for different cultural and political perspectives that could arise from the worldwide readers.

Reply: Thank you for your insightful comments regarding the data collection and personal data manipulation practices associated with our research. We understand and respect the European Union's stringent data protection regulations and the importance of ensuring the ethical handling of personal data globally.

In this study, the data utilized were provided by a mobile network operator and were pre-anonymized before they were transmitted to us. The initial anonymization process conducted by the mobile operator involved removing all personally identifiable information (PII), including phone numbers, device IDs, and any other data that could potentially identify individual users. As a result, the dataset received and used in our analysis did not contain any PII, ensuring that individual privacy was maintained.

Furthermore, our additional steps of anonymization, including the use of the SHA-256 cryptographic hash function to generate non-reversible random identifiers, reinforce the protection of individual identities. These measures comply with the general principles of ethical research practices, the anonymization protocols employed ensure that the data cannot be traced back to any individual, thereby aligning with international standards for data privacy and security.

We have included a comprehensive Data Description section (Section 3.4) in the revised manuscript. This section elaborates on the details of the data processing procedures to ensure full transparency and to address concerns regarding data privacy and protection. It specifically highlights the rigorous anonymization protocols employed, which adhere to international data protection standards, thereby safeguarding individual privacy.

This addition is particularly aimed at reassuring all readers, especially those from regions with stringent data protection regulations like the European Union, about the ethical considerations integral to our research. We are thankful for this opportunity to enhance our manuscript with such critical information, ensuring it meets the highest standards of ethical and academic rigor.

Please refer to note 07 For Reviewer 3 (lines 389-395).

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

The authors answered all my remarks and took proper measures to improve/correct the manuscript.