Utilizing Social Media Data for Estimating Transit Performance Metrics

Round 1

Reviewer 1 Report (Previous Reviewer 1)

Comments for author File: Comments.pdf

Need minor English Language corrections.

Author Response

Please see attached document.

Author Response File: Author Response.pdf

Reviewer 2 Report (Previous Reviewer 2)

The article has been improved and with some minor revisions, it is suitable for publication.

- On page 4 when the authors explain sentiment analysis, it is highly suggestible that authors enrich the methodology by drawing a table in which indicators of sentiment analysis are first documented by other similar studies and then finalize them via contextualizing them based on NYC circumstances. 

- Twitter may be changed to X (former Twitter)

- The reference style should be modified according to the sustainability journal style

- 

Author Response

Please see attached document.

Author Response File: Author Response.pdf

Reviewer 3 Report (Previous Reviewer 3)

This manuscript analyzes customers' opinions on the service quality of bus operators by analyzing relevant data on public transportation in Twitter. The methods and conclusions of this study seem to be less innovative.

（1）    According to the author's literature review, many studies have analyzed people's emotional feedback on urban public transportation using Twitter data. What are the characteristics of this study?

（2）    The Twitter and MTA data analyzed in this study is from June 2015 to May 2016. Is it valuable to analyze data from 8 years ago?

（3）    The author extracted high-frequency words from Twitter data to indicate customers' views on public transportation. For example, passengers' experience of public transportation services is mostly negative, and bus users are more picky about waiting times and delays. So what is the value of this analysis? It seems that even without this analysis, public transportation operators should also understand that reducing bus delays is one of the important ways to improve user experience.

Author Response

Please see attached document.

Author Response File: Author Response.pdf

Reviewer 4 Report (Previous Reviewer 4)

This manuscript was previously submitted to the Journal of Sustainability with manuscript number sustainability-2569811. Some previously proposed modifications still need to be done, and corrections are needed. More explanations about the disadvantages and privileges of methodology are essential. The validation of their method and results is not clear. The introduction section could be more organized and more consistent in content. The literature review in the introduction section needs to be more comprehensive, especially in the applications of social networks in industry, economics, and society. It is recommended that the following articles be explained in detail: DOI: 10.1016/j.jad.2020.08.017. DOI: 10.3389/fpubh.2022.875030. DOI: 10.3389/fpubh.2022.831549. DOI: 10.1016/j.jad.2020.08.017. Also, the quality of the figures is low. The texts in the figures are very small and are not readable. Besides, the innovations of the present study should be elaborated and highlighted.

Author Response

Please see attached document.

Author Response File: Author Response.pdf

Reviewer 5 Report (New Reviewer)

I have questions/observations for the authors:

1. the Twitter platform is currently called X. I realize that it was Twitter at the time of data acquisition, but it would be appropriate to clarify this point in the body of the article. In addition, the changes made by Elon Musk, have significantly affected the way X is used, perhaps including issues related to data retrieval and relationships. Every X user has experienced inconveniences related to the new algorithms and how posts are displayed on the feed. It is imperative to clarify this and indicate how the article author's findings relate to the current state - or are they only of archival value?

(2) Was it necessary to filter the downloaded data, in order to avoid noise, and eliminate fake accounts, bots etc., or was it not necessary? How was the authenticity of the accounts surveyed verified? 

3 Why is the data from several years ago? The results of this study may already be outdated and contribute nothing to science. What was the purpose of the study? What was the research question? WHAT did the authors want to achieve through their research? What do they show, what do they prove?

4 How did the authors deal with tweets containing errors and typos? How did they clean up the data in this regard?

5. please make your objectives and hypotheses clear and explicit. if the authors abandoned hypotheses in favour of an exploratory approach, please complete this clearly.

Author Response

Please see attached document.

Author Response File: Author Response.pdf

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

Please find the attached document for comments.

Comments for author File: Comments.pdf

Author Response

The authors thank the reviewers for their thoughtful comments as they are contributing to make the paper stronger. The authors have addressed the comments and followed the suggestions of the reviewers by expanding the introduction, discussions and conclusion, and adding recent literature.

Author Response File: Author Response.pdf

Reviewer 2 Report

Review report for Sustainability

Article code: sustainability-2569811

Summary

The article aims to determine performance metrics customers frequently mentioned when riding the New York Transit system. The authors want to analyze users’ transit experiences by using social media and recognizing trends regarding sentiment with a spatial approach. The results of the work provide valuable implications for transit systems planners and authorities. Therefore, the article is a suitable one for publication consideration. However, there are some problems that should be addressed as follow;

Title

This section suffers from a comprehensive one. For instance, case study has not been mentioned.  

 Introduction

The first sentence ‘It is commonly accepted that…” needs to be backed by some scientific findings particularly recent case studies.

The link between the transport system and urban sustainable development has been missed. Moreover, sustainable urban transportation and the challenge of using transport system by commuters has been overlooked.

This section lacks a compelling argument on the transport system, urban sustainable development, and New York City on the one hand and the importance of investigating users’ perceptions of the functionality of the transport system (problem finding) for urban planning and management guidance on the other hand.  

 Results

The authors refer to a very thing named location in the abstract and methodology sections which is so important in sustainable transport planning as locational analysis could help the related realistic policies to be shaped. However, the paper lacks spatial maps showing the various results on them.

It is highly recommended that authors add a new section named policy implications and suggest some policies for New York transit system authorities regarding the results. Also, giving further research could be useful for the next investigations.  

Author Response

The authors thank the reviewers for their thoughtful comments as they are contributing to make the paper stronger. The authors have addressed the comments and followed the suggestions of the reviewers by expanding the introduction, discussions and conclusion, and adding recent literature.

Author Response File: Author Response.pdf

Reviewer 3 Report

A methodology using text and sentiment analysis of social media data is proposed to determine performance metrics customers frequently mentioned when riding the New York Transit system and how they compare to those reported by the agency.

The following suggestions have been given.

1.What are the innovations of the paper? Please illustrate in abstract and introduction.

2. The second part ‘Literature Review’ should be integrated into the first part ‘Introduction’.

3.What are the advantages in methodology to utilize social media data for estimating transit performance metrics?

4. To clarify the methodology applied in the paper, the author can draw a flow chart in the third part.

5. How long does the entire analysis process of utilizing social media data for estimating transit performance metrics take?

6. The authors mentioned that ‘It is important to understand the different factors and history that distinguish and define transit systems across the world from one another; how these factors relate to the transit rider’s customer experience can influence and support a region-specific study’. Please show the hazards brought by transit systems. The author can refer to the following paper:

Song Y, Chen XJ, Zou BP, Mu JD, Hu RS, Cheng SQ, Zhao SL. Monitoring Study of Long-Term Land Subsidence during Subway Operation in High-Density Urban Areas Based on DInSAR-GPS-GIS Technology and Numerical Simulation. Computer Modeling in Engineering & Sciences, 2023, 134 (2): 1021-1039.

Yu Huang, Yangjuan Bao, Wang Yuhong. Analysis of geoenvironmental hazards in urban underground space development in Shanghai. Natural Hazards, 2015, 75: 2067-2079.

Author Response

The authors thank the reviewers for their thoughtful comments as they are contributing to make the paper stronger. The authors have addressed the comments and followed the suggestions of the reviewers by expanding the introduction, discussions and conclusion, and adding recent literature.

Author Response File: Author Response.pdf

Reviewer 4 Report

In this manuscript, sentiment analysis of New York transit system customers is conducted based on their experience in social media (Twitter data). Also, the effects of key factors such as station cleanliness and train speed on transit performances are evaluated.

1- The main question of this research should be highlighted in the last paragraph of the introduction section.

2- The specific gap in the research subject should be mentioned. What are the innovations of the current study? The authors should discuss what this research adds to the subject field compared with other published reports.

3- What are the privileges and disadvantages of the proposed methodology in sentiment analysis of the collected data from Twitter? The authors are encouraged to add more comments about the characteristics of their methodology.

4- The literature review in the introduction section needs to be more comprehensive, especially in the field of social networks and pandemic effects. It is recommended that the following articles be explained in detail: DOI: 10.1016/j.jad.2020.08.017. DOI: 10.3389/fpubh.2022.875030. DOI: 10.3389/fpubh.2022.831549. DOI: 10.1109/TKDE.2022.3233481. Also, the quality of the figures is low. The texts in the figures are very small and are not readable.

5- It is suggested to reorganize the conclusion section. The main findings should appear point by point next to the paragraph, not inside it.

6- The English written language of the article should be improved.

The English written language of the article should be improved

Author Response

The authors thank the reviewers for their thoughtful comments as they are contributing to make the paper stronger. The authors have addressed the comments and followed the suggestions of the reviewers by expanding the introduction, discussions and conclusion, and adding recent literature.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

No further comments. Accepted in present form.

Accepted in present form.

Reviewer 2 Report

The article has been improved and it now is publishable 

Reviewer 3 Report

1. A methodology using text and sentiment analysis of social media data to extract transit user’s perception about the quality of service was proposed in the paper. But the innovation of the paper is not strong.

2. How to verify the accuracy of the method proposed in the paper?

3.‘Literature Review’ and ‘Introduction’ were divided into two parts, which is generally messy. The performance of the logic is not strong.

4. The authors would like to emphasize that the objective of the study is the proposition of a methodology using text and sentiment analysis of social media data to extract transit user’s perception about the quality of service. So the visualization of the methodology is important. The authors should try to summarize it using a chart flow in the third part.

Minor editing of English language required.

Reviewer 4 Report

Some previously proposed modifications still need to be done, and corrections are needed. The conclusion section needs to be reorganized and summarized. The suggested articles by the reviewer are not focused on the pandemic effects. The assessment methods of the proposed articles are comparable to the considered approach of the present manuscript, and the appraisal and review of the mentioned methods will lead to the improvement of the quality of the present manuscript. Authors are encouraged to consider this subject in their manuscript, especially about the features of their approaches. The colors used in Figures 8 and 9 should be as distinct as possible. More explanations about the disadvantages and privileges of methodology are essential.