Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

 

Specific comments:

The manuscript was comprehensive in both structure and content. However, I have one minor suggestion to revise and improve the paper quality: 

Title:

1)    Does it better no to use abbreviations in title?

Abstract and keywords:

2)    What is the meaning of "sustainable commercial street development"? Does it have a scientific definition? This phrase is used multiple times throughout the manuscript and needs to be clearly defined.

3)    Don’t use keywords which repeated in title and abstract.

Introduction:

4)    The introduction section is comprehensive. If possible, please clarify the study's objective and confirm that the variables used for predicting commercial street locations are backed by the literature review in this section, specifically regarding the influencing factors utilized for prediction.

Materials and Methods:

5)    If you are referring to streets, it would be more appropriate to highlight them as line features on the maps. However, you have used point locations. If your analysis is based on line features, why did you convert them to points? Alternatively, if you are analyzing point locations, why did you refer to them as commercial streets instead of commercial places? For example, check Fig.1.

6)    You applied Ripley’s K function to assess the clustering pattern of point features. Please provide a brief description of the method and its purpose in your study in one sentence. See, line 128.

2.3. Research Methods:

7)    Some researchers use appropriate methods but only present the formulas without explaining how they were applied in their studies. If you believe these methods are widely recognized, there's no need to introduce them in detail. However, it is crucial for readers to understand how the methods were applied, step by step, in a summarized form. Please include this explanation in the section. For example, if you have any code, consider attaching it to your study.

 

Results:

8)    3.1. Model Design and Evaluation/ Evaluation of Prediction Results: I believe there are numerous machine learning algorithms that can be used to assess prediction/model accuracy. It is crucial to apply the appropriate methods or statistics to validate the accuracy of the model's prediction results.

 

Discussion:

9)    What other machine learning algorithms can be applied in future studies using Points of Interest (POI) for predictions? Please suggest some options.

Conclusion:

10) Please summarize this section and revise it to align with your main findings.

   Thank you.

 

 

 

 

Author Response

Article Revision Instructions

 

Title of manuscript: Prediction of Commercial Street Location Based on POI Big Data and Machine Learning

Journal Name: IJGI

Manuscript ID: ijgi-3209508

 

 

1. Response to Reviewer 1 Comments

1.1. Summary

Thank you for your valuable comments during the review of our manuscript. Your thorough review and constructive feedback have been essential for us to improve this work. We greatly appreciate your identification of key issues and have made every effort to address and enhance these aspects. Your professional insights not only help elevate the quality of this paper but also provide guidance for our future research directions. Please be assured that we will carefully address each of your recommendations point by point in revision.

1.2. Point-by-point response to Comments and Suggestions for Authors

Comments 1:

Does it better no to use abbreviations in title?

Response 1:

We sincerely appreciate your astute observation regarding the use of abbreviations in the title. Upon careful consideration of your suggestion, we have revised the original title to include the full phrase "Point of Interest (POI)" instead of the abbreviation "POI." This modification enhances clarity and ensures that readers unfamiliar with the acronym can immediately grasp the subject matter. We believe this change aligns with best practices in academic writing, promoting accessibility and precision in scientific communication.

Comments 2:

What is the meaning of "sustainable commercial street development"? Does it have a scientific definition? This phrase is used multiple times throughout the manuscript and needs to be clearly defined.

Response 2:

We are grateful for your insightful comment regarding the need for a clear definition of "sustainable commercial street development." In response to your valuable feedback, we have added a comprehensive description of this concept in the "Introduction" section. The revised text now states:

"In this study, 'sustainable commercial street development' refers to an efficient and holistic strategy for commercial street growth, emphasizing scientific and rational site selection and layout to maximize market demand and consumer experience. This approach focuses on optimizing resource utilization and selecting optimal locations through rigorous data analysis, ultimately achieving sustainable growth of commercial activities and promoting regional economic prosperity [3,4].

This concept encompasses several key elements:

• Economic viability: Ensuring long-term profitability and resilience of businesses.
• Environmental responsibility: Minimizing ecological footprint and promoting green practices.
• Social inclusivity: Creating spaces that cater to diverse community needs and foster social interaction.
• Urban integration: Seamlessly blending commercial development with existing urban fabric and infrastructure.
• Adaptability: Designing flexible spaces that can evolve with changing market trends and consumer preferences.

By incorporating these principles, sustainable commercial street development aims to create vibrant, enduring commercial spaces that contribute positively to urban ecosystems while meeting the dynamic needs of businesses and consumers alike."

Comments 3:

Don’t use keywords which repeated in title and abstract.

Response 3:

We sincerely appreciate your astute observation regarding the selection of keywords. Your suggestion has led us to critically reassess our keyword choices to ensure they provide maximum value to potential readers and accurately represent the core aspects of our research. In light of your feedback, we have implemented the following changes:

Removed: "commercial street location; points of interest" as these terms are indeed prominently featured in the title and abstract.

Added: "urban planning; site selection; commercial street prediction"

These new keywords have been carefully selected to:

1) Broaden the scope of our study's discoverability within the field of urban studies.

2) Highlight the methodological approach (site selection and prediction) used in our research.

3) Emphasize the interdisciplinary nature of our work, bridging urban planning with data-driven decision-making.

We believe these revised keywords offer a more comprehensive and accurate representation of our study, enhancing its visibility and relevance to researchers across various related disciplines. Thank you for prompting this valuable improvement to our manuscript.

Comments 4:

The introduction section is comprehensive. If possible, please clarify the study's objective and confirm that the variables used for predicting commercial street locations are backed by the literature review in this section, specifically regarding the influencing factors utilized for prediction.

Response 4:

We have strengthened the description of the research objectives in "1. Introduction," stating: “With the acceleration of urbanization and the increasing diversification of consumer demand, commercial streets, as important carriers of urban ecological economic activities, face increasingly complex and critical site selection decisions. Rational site selection can not only enhance the competitiveness of commercial facilities but also optimize urban resource allocation and promote regional economic development. However, traditional site selection methods mainly rely on empirical judgment and qualitative analysis, making it difficult to comprehensively and accurately reflect the complex relationship between commercial facilities and their surrounding environments.” We have also supplemented the basis for the influencing factors used in the prediction, stating: “Among the studies targeting site selection, Feilong Hao et al. analyze POIs in Changchun City to reveal the polycentric and stratified trends of retail shop location patterns and their interactions with factors such as consumer behavior, retail formats, and governmental planning, which provide a basis for urban planning and the allocation of commercial facilities [17]. Colaco Rui et al. propose a new cluster analysis and a small number of variables based on a new commercial classification system, thus strengthening the link between commercial classification and location modeling [18]. Colaco Rui et al. also used cellular automata (CA), incorporating centrality indicators in spatial syntax as drivers, to simulate commercial land use changes in order to better understand future land use dynamics [19].”

Comments 5:

If you are referring to streets, it would be more appropriate to highlight them as line features on the maps. However, you have used point locations. If your analysis is based on line features, why did you convert them to points? Alternatively, if you are analyzing point locations, why did you refer to them as commercial streets instead of commercial places? For example, check Fig.1.

Response 5:

We sincerely appreciate your astute observation regarding the representation of commercial streets in our study. Your comment has prompted us to clarify our methodological approach and rationale.

In "2.1. Research Area," we have expanded our explanation for using point data:

"The decision to represent commercial streets as point locations rather than line features was made after careful consideration of both the available data and the nature of commercial activity in urban environments. Due to the lack of a clear definition of the actual boundaries of commercial streets, which often blend seamlessly into surrounding areas, we opted to use representative points to signify the location of each commercial street.

These representative points are determined as the center or key nodes of the commercial street, based on a comprehensive review and evaluation of multiple data sources: Government data on zoning and commercial districts; Merchant registration information; Street view data providing visual confirmation of commercial activity; User-uploaded data from location-based services.

This point-based approach allows us to:

1) Capture the essence of commercial activity concentration without arbitrary delineation of street boundaries

2) Facilitate more precise spatial analysis, particularly when applying clustering algorithms

3) Accommodate the varying scales and layouts of commercial streets, from small clusters to extensive linear developments

While we acknowledge that this method may not capture the full linear extent of some commercial streets, we believe it provides a more flexible and analytically robust representation of commercial activity for the purposes of our study. This approach aligns with similar methodologies in urban studies literature, such as the work of Li et al. (2020) on POI-based urban function recognition. We have ensured that our terminology throughout the paper consistently refers to 'commercial street locations' or 'commercial activity centers' to accurately reflect this point-based representation."

Comments 6:

You applied Ripley’s K function to assess the clustering pattern of point features. Please provide a brief description of the method and its purpose in your study in one sentence. See, line 128.

Response 6:

In "2.1. Research Area," we explain the use of Ripley's K function, stating: “Using Ripley's K function for analysis, we observed that the observed K values consistently exceed the expected K values and fall above the confidence interval. This indicates a significant spatial clustering phenomenon in the distribution of commercial street locations in Foshan, particularly in the core commercial areas of the central and eastern parts of the city, as shown in Figure 2. These regions exhibit dense commercial activities, with commercial street locations highly concentrated, forming a clear commercial core. Conversely, the commercial streets in the city's edge areas show relative sparsity, where commercial development is limited and still in the developmental phase. The spatial distribution characteristics of commercial streets in Foshan provide important insights for future commercial development and urban planning, as identifying clustering phenomena can help optimize resource allocation and enhance the effectiveness of commercial street site selection.”

Comments 7:

Some researchers use appropriate methods but only present the formulas without explaining how they were applied in their studies. If you believe these methods are widely recognized, there's no need to introduce them in detail. However, it is crucial for readers to understand how the methods were applied, step by step, in a summarized form. Please include this explanation in the section. For example, if you have any code, consider attaching it to your study.

Response 7:

Thank you for your suggestion. We have added a step-by-step description of the research methods in "2.3. Research Methods," stating: “This study utilizes POI big data to first categorize the data, extract features, and balance samples, constructing training and testing sets. Next, the ID3 algorithm is employed to generate a decision tree model, using information gain or entropy to split nodes, and the model is visualized. Based on this model, we predict 2,157 potential commercial street grids in Foshan. Finally, by combining logical chains and factor analysis, we propose targeted policy recommendations to optimize the layout and site selection strategies for commercial streets.”

Comments 8:

In general, I could say that the methodology is well developed and described. Nevertheless, results should be verified comparing them with literature and other kind of research approaches.

Response 8:

Thank you for your suggestions. We have addressed this issue in "3.1. Model Design and Evaluation" and "5. Conclusion." The specific descriptions are as follows: “The model's performance was evaluated by comparing the true values and predicted values on the test set, achieving an accuracy of 83%. This high accuracy indicates the credibility and reliability of the prediction results. Only 2022 data has been used to validate the model accuracy, and due to the long period of time required for the construction and development of the commercial street, we will validate the forecasting model again in another 5 or 10 years to optimize the forecasting model.” “Compared with traditional methods, this study not only provides more refined and objective site selection recommendations, but also provides strong data support for urban planning and business decision-making.”

Comments 9:

What other machine learning algorithms can be applied in future studies using Points of Interest (POI) for predictions? Please suggest some options.

Response 9:

In "4. Discussion," we have added suggestions for future optimization research, stating: “This study exclusively employs decision tree algorithms and does not incorporate other machine learning methods such as random forests, support vector machines, gradient boosting decision trees, neural networks, or spatial econometric models for further optimization of the site selection prediction model in Foshan. The primary reason is that the development cycle of commercial streets is relatively long, and changes in short-term predictions may not be significant. Additionally, since the main objective of this research is to explore the preliminary application of big data and machine learning techniques in commercial street site selection, using a simple and interpretable decision tree model facilitates a more intuitive understanding of the model's impact on site selection factors. In the future, we plan to incorporate various machine learning models after five or ten years, combining long-term observational data to improve prediction accuracy and fully consider dynamic factors in both time and space, aiming to provide more comprehensive decision support for the planning of commercial streets in Foshan.”

Comments 10:

Please summarize this section and revise it to align with your main findings.

Response 10:

Based on your suggestion, we have made supplementary adjustments to "5. Conclusion," stating: “The research results indicate that commercial street locations in Foshan exhibit significant spatial clustering in the core commercial areas of the central and eastern parts of the city, while they are relatively sparse in the edge areas, where commercial development is limited.”

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The thesis is based on POI big data and machine learning methods to predict the optimal location of potential commercial street development in Foshan City, and analyzes the impact of different types of POI data on the location of sustainable commercial street. Overall, the thesis has a certain amount of workload, strong innovation, but there are still some problems with the research content, the specific comments are as follows:

 

1.       the research significance is not clear enough. From the authors' research, the research method, the data is more innovative. However, the study area selected Foshan City, and did not further explain the special significance and representativeness of the study of commercial street site selection. The author's research wants to solve what problems in the study area is not expressed clearly. The significance of the study is not clear.

 

2.       The literature review needs to be further organized around the research topic. The authors mentioned a lot of existing research on commercial streets in the literature review, and the research vein ranges from the center ground theory to the empirical research using big data since the 21st century to the quantitative research using POI and machine learning. However, the authors' literature review is insufficient on the site selection of commercial streets, and the literature cited in the literature review section only ranges from [7]-[23], which is too few in number. The author's review in the literature review is insufficient and is only a list of literature.

 

3.       Limitations of data sources. The POI data used in the study is only a single year 2022, which does not take into account the effect of temporal changes. And the data is not enough multi-source, which can be combined with more data sources, such as social media data, traffic flow data, etc., to improve the accuracy and comprehensiveness of the prediction.

 

4.       The depth and presentation of the research content need to be further improved. In the specific analysis of the research results, the authors' analysis is shallow and not in place, for example, in part 3.2, the authors' results show that medical care is the main factor affecting the location of commercial streets, which is a relatively innovative conclusion, but the authors did not analyze the reasons in detail.

 

5.       The value of the paper needs to be further explored. The authors in the discussion of influencing factors put forward the findings of influencing factors and the conclusions reached in previous studies, the innovation of the authors' research in addition to the research methodology, but also to draw any conclusions with research value, please add the description. In addition, what is the difference between the research on the location of commercial streets in Foshan City and other provinces and cities, and how can the conclusions of the research be generalized.

 

6.       Other possible problems. The thesis cartography is not standardized enough. It is suggested to supplement the picture frame, it is not necessary to keep 9 decimals in the legend, and some pictures are not clear enough. There is a latitude and longitude network does not need a compass, the scale choose integer. It is suggested to supplement the related literature on commercial street siting and influencing factors.

 

To summarize, there are some problems with the thesis, so it is recommended to revise it and review it again.

 

 

Author Response

Article Revision Instructions

 

Title of manuscript: Prediction of Commercial Street Location Based on POI Big Data and Machine Learning

Journal Name: IJGI

Manuscript ID: ijgi-3209508

 

2. Response to Reviewer 2 Comments

2.1. Summary

Thank you for your valuable comments during the review of our manuscript. Your thorough review and constructive feedback have been essential for us to improve this work. We greatly appreciate your identification of key issues and have made every effort to address and enhance these aspects. After analyzing your comments, we have made improvements to the article in the following sections.

2.2. Point-by-point response to Comments and Suggestions for Authors

Comments 1:

The research significance is not clear enough. From the authors' research, the research method, the data is more innovative. However, the study area selected Foshan City, and did not further explain the special significance and representativeness of the study of commercial street site selection. The author's research wants to solve what problems in the study area is not expressed clearly. The significance of the study is not clear.

Response 1：

We sincerely appreciate your insightful feedback regarding the clarity of our research significance and the justification for our study area selection. In response to your valuable comments, we have substantially expanded and refined the explanation of our research significance in the "1. Introduction" section:

"The rapid pace of urbanization and the increasingly diverse consumer demands have rendered the location decisions for commercial streets—vital carriers of urban economic activities—increasingly complex and critical. Rational site selection not only enhances the competitiveness of commercial facilities but also optimizes urban resource allocation and promotes regional economic development. However, traditional site selection methods, predominantly relying on empirical judgment and qualitative analysis, often fall short in comprehensively and accurately reflecting the intricate relationship between commercial facilities and their surrounding environment.

This study aims to address several key challenges in the field of urban commercial development:

• The need for data-driven, quantitative approaches to commercial street site selection that can capture the complexity of modern urban environments.
• The integration of big data and machine learning techniques to enhance the accuracy and objectivity of location decisions.
• The development of a replicable methodology that can be adapted to various urban contexts, contributing to the broader field of urban planning and commercial geography.

Our research seeks to bridge the gap between traditional qualitative site selection methods and the data-rich realities of contemporary urban landscapes, offering a novel approach that combines POI data analysis with machine learning algorithms."

Regarding the selection of Foshan City as our study area, we have elaborated in "2.1. Research Area" as follows:

"Foshan City was chosen as the focus of this study for several compelling reasons:

• Economic Significance: Foshan boasts a robust economic foundation, ranking among the top cities in China for GDP per capita. This economic vitality provides a dynamic environment for commercial street development, offering rich data for analysis.
• Urban Innovation: In recent years, Foshan has been at the forefront of urban planning and commercial development innovation. This progressive approach provides an excellent policy support and market environment for the evolution of commercial streets, making it an ideal testbed for our predictive model.
• Representativeness: As a rapidly developing second-tier city in China, Foshan represents a common urban typology in the country. Insights gained from this study can potentially be generalized to similar urban contexts, both within China and internationally.
• Data Availability: Foshan's advanced digital infrastructure and open data policies provide access to rich, high-quality POI datasets, crucial for the successful implementation of our methodology.
• Unique Urban Structure: The city's polycentric urban structure, with a mix of traditional and modern commercial areas, offers an opportunity to study diverse patterns of commercial street development within a single urban entity.

By focusing on Foshan, we aim to develop and test our predictive model in a context that combines economic dynamism, urban innovation, and representativeness, potentially yielding insights applicable to a wide range of urban environments undergoing rapid commercial development."

Comments 2:

The literature review needs to be further organized around the research topic. The authors mentioned a lot of existing research on commercial streets in the literature review, and the research vein ranges from the center ground theory to the empirical research using big data since the 21st century to the quantitative research using POI and machine learning. However, the authors' literature review is insufficient on the site selection of commercial streets, and the literature cited in the literature review section only ranges from [7]-[23], which is too few in number. The author's review in the literature review is insufficient and is only a list of literature.

Response 2：

Thank you very much for your suggestions. We have supplemented the literature on commercial street site selection in "1. Introduction" and provided a review of the current research status. The specific descriptions are as follows: “Scholars around the world have long been studying shopping streets, constantly exploring their multidimensional characteristics such as spatial layout, consumer behavior, urban vitality, and store-pedestrian interaction.” “Among the studies targeting site selection, Feilong Hao et al. analyze POIs in Changchun City to reveal the polycentric and stratified trends of retail shop location patterns and their interactions with factors such as consumer behavior, retail formats, and governmental planning, which provide a basis for urban planning and the allocation of commercial facilities [15]. Colaco Rui et al. propose a new cluster analysis and a small number of variables based on a new commercial classification system, thus strengthening the link between commercial classification and location modeling [16]. Colaco Rui et al. also used cellular automata (CA), incorporating centrality indicators in spatial syntax as drivers, to simulate commercial land use changes in order to better understand future land use dynamics [17].” “Although existing studies have achieved rich results in the spatial layout, commercial structure, vitality shaping, and spatial distribution of commercial streets, there are still some limitations in site selection prediction. Traditional commercial street siting research mainly relies on qualitative analysis, lacking data support and quantitative methods, which makes it difficult to adapt to the increasingly complex urban planning and commercial development needs. Most of the existing siting studies focus on analyzing the current spatial distribution of commercial street sites, with less attention to predictive simulation.”

Comments 3:

Limitations of data sources. The POI data used in the study is only a single year 2022, which does not take into account the effect of temporal changes. And the data is not enough multi-source, which can be combined with more data sources, such as social media data, traffic flow data, etc., to improve the accuracy and comprehensiveness of the prediction.

Response 3：

Thank you very much for your suggestions. We have addressed these points in "4. Discussion." The specific description is as follows: “To address these limitations, future research should optimize the proposed methodology by incorporating more data sources, such as socio-economic indicators, transportation networks, urban form, and other multi-source big data, in order to gain a more comprehensive understanding of the factors affecting the development of sustainable shopping streets. In addition, the robustness and applicability of the proposed methodology can be further enhanced by expanding the study area, comparing different cities, and analyzing the temporal changes of POI data for a more detailed, accurate, and generalizable study.”

Comments 4:

The depth and presentation of the research content need to be further improved. In the specific analysis of the research results, the authors' analysis is shallow and not in place, for example, in part 3.2, the authors' results show that medical care is the main factor affecting the location of commercial streets, which is a relatively innovative conclusion, but the authors did not analyze the reasons in detail.

Response 4：

Thank you for your suggestions. We have made the corresponding adjustments in the "3.2. Decision Rule Analysis" section. The specific description is as follows: “Medical care is often an important part of a city's functionality and can reflect the level of infrastructure and services in the area. Areas with good healthcare services tend to attract more residents and tourists, thus promoting commercial activities and the formation of commercial streets. However, the detailed mechanism of its role in influencing the location of commercial streets needs to be further studied.”

Comments 5:

The value of the paper needs to be further explored. The authors in the discussion of influencing factors put forward the findings of influencing factors and the conclusions reached in previous studies, the innovation of the authors' research in addition to the research methodology, but also to draw any conclusions with research value, please add the description. In addition, what is the difference between the research on the location of commercial streets in Foshan City and other provinces and cities, and how can the conclusions of the research be generalized.

Response 5：

We have addressed this issue in the "4. Discussion" section. The specific description is as follows: “Despite the significant contribution of this study, its limitations and the need for further research must be recognized. The distribution of commercial streets in Foshan City is mainly characterized by polycentricity, with traditional and modern commercial streets co-existing. The scope of the study is limited to Foshan City, which may affect the generalizability of the conclusions. Future research should explore the transferability of the proposed methodology to other urban contexts and conduct comparative analyses between different cities to identify the deeper reasons for the differences and similarities in urban findings.”

In the "5. Conclusion" section, we have supplemented the research value of this study. The specific description is as follows: “Compared with traditional methods, this study not only provides more refined and objective site selection recommendations, but also provides strong data support for urban planning and business decision-making.”

Comments 6:

Other possible problems. The thesis cartography is not standardized enough. It is suggested to supplement the picture frame, it is not necessary to keep 9 decimals in the legend, and some pictures are not clear enough. There is a latitude and longitude network does not need a compass, the scale choose integer. It is suggested to supplement the related literature on commercial street siting and influencing factors.

Response 6：

Thank you for your valuable suggestions. We have adjusted the figure legends to retain three decimal places, increased the image resolution to 330 DPI, removed unnecessary compass elements, and adjusted the scale. Additionally, we have supplemented the relevant literature on commercial street site selection and influencing factors, including:

• Hao, F.; Yang, Y.; Wang, S. Patterns of Location and Other Determinants of Retail Stores in Urban Commercial Districts in Changchun, China. Complexity 2021, 2021, e8873374, doi:10.1155/2021/8873374.
• Colaco, R.; de Abreu e Silva, J. Commercial Classification and Location Modelling: Integrating Different Perspectives on Commercial Location and Structure. Land 2021, 10, 567, doi:10.3390/land10060567.
• Colaco, R.; Silva, J. de A. e Commercial Land Use Change and Growth Processes - An Assessment of Retail Location in Lisbon, Portugal, 1995-2020. Urban Manag. 2024, 13, 157–170, doi:10.1016/j.jum.2023.11.005.

 

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

See in the attached file.

In general I could say that the methodology is well developed and described. Nevertheless, results should be verified comparing them with literature and other kind of research approaches. 

Comments for author File: Comments.pdf

Author Response

Article Revision Instructions

 

Title of manuscript: Prediction of Commercial Street Location Based on POI Big Data and Machine Learning

Journal Name: IJGI

Manuscript ID: ijgi-3209508

 

3. Response to Reviewer 3 Comments

3.1. Summary

Thank you for your detailed review and valuable comments on our paper. Your suggestions have played a significant role in deepening our research ideas and enhancing the quality of the manuscript. We have made corresponding adjustments and additions based on your feedback, and we hope to further meet your expectations.

Comments 1

I believe there are numerous machine learning algorithms that can be used to assess prediction accuracy. It is crucial to apply the appropriate methods or statistics to validate the accuracy of the model's prediction results.

Response 1：

Thank you for your comments. We have made adjustments in "3.1. Model Design and Thank you for your comments. We have made adjustments in "3.1. Model Design and Evaluation." The specific description is as follows: “The model's performance was evaluated by comparing the true values and predicted values on the test set, achieving an accuracy of 83%. This high accuracy indicates the credibility and reliability of the prediction results. Only 2022 data has been used to validate the model accuracy, and due to the long period of time required for the construction and development of the commercial street, we will validate the forecasting model again in another 5 or 10 years to optimize the forecasting model.”

Conclusion

Finally, we sincerely thank you and your team for the support and guidance provided throughout the review process. We are grateful to the three reviewers; your valuable feedback and suggestions are crucial for enhancing the quality of our research and academic standards. We truly appreciate you taking the time from your busy schedules to carefully review our manuscript and provide helpful insights on the issues within it. We also want to thank the editor for their hard work and all the staff at the journal for their support and assistance!

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

I don't think the author did a good job of modifying the comments I mentioned in the last round

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors

The first review wasn't take into account. Please, see the comments in the attached  file

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.docx