Road Intersection Extraction Based on Low-Frequency Vehicle Trajectory Data
Round 1
Reviewer 1 Report
This paper studied road intersection extraction based on low-frequency vehicle trajectory data, it designed filtering rules to fit vehicle turning points, then used the Clustering algorithm using the local Direction Centrality algorithm CDC） for vehicle turning points to obtain intersections, alculated intersection center locations based on the density-based spatial clustering of applications with noise (DBSCAN) algorithm to obtain urban road intersections.some experimental results are provided, it appears that the overall writing and organization is good, some basic comparison results are also provided, however the only problem is the innovation of the paper, DBSCAN is a traditional and classic method for vehicle trajectory data extraction , this method was not originally proposed by the author, In fact, a large number of improved DBSCAN methods such as adaptive DBSCAN methods have been developed in vehicle trajectory data extraction, so I suggest that the author carefully consider the original innovation of the paper;
Some grammar there needs to be corrected
Author Response
Dear Reviewer：
My changes to your comments specifically please see the attachment.
Sincerely,
Jiusheng Du
Author Response File: 
Author Response.pdf
Reviewer 2 Report
This manuscript addresses the efficient extraction of GNSS-related information. The authors aim to accurately extract urban road network intersections and hub locations from low-frequency GNSS track data. They use clustering for this.
I went through the manuscript carefully. At the detailed level, the following notes are my suggestions:
1) Although the ABSTRACT structure is good, I suggest that the values of the numerical improvements be written in the final sentences. Also, the philosophy of using the proposed method should be explained.
2) In my opinion, the INTRODUCTION section needs to be revised. In this section there should be three points: 1) motivation, 2) a summary of the challenges of previous studies, and 3) contribution. Also, the research contributions should be mentioned in a bullet-form at the end of the INTRODUCTION.
3) It is not clear to me which formulas were invented by the authors themselves and which ones are derived from other references. I found evidence that some formulas are derived from other references and there are similarities.
4) Regarding edge/fog computing, which is the subject of this article, important references have been published recently. I recommend the authors consider them to make the article richer. Some examples are:
https://dl.acm.org/doi/abs/10.1145/3603703
https://www.tandfonline.com/doi/abs/10.1080/13658816.2018.1510124?journalCode=tgis20
5) It is necessary that the analyses related to the final results are justified and based on the relationships and formulas of the previous sections of the article. Currently, the analyses are very superficial.
6) It is better to analyze the time complexity of the proposed method in the worst case.
7) There are still some grammatical errors in the manuscript. Authors should use software such as Grammarly for proof-checking.
8) The tense of the verbs in the CONCLUSION section must be past tense. In this section, the most important numerical improvements of the proposed method should be mentioned and marginal explanations should be avoided. Also, the suggestions mentioned for further research should be presented in a new paragraph.
There are still some grammatical errors in the manuscript. Authors should use software such as Grammarly for proof-checking.
Author Response
Dear Reviewer：
My changes to your comments specifically please see the attachment.
Sincerely,
Jiusheng Du
Author Response File: Author Response.pdf
Reviewer 3 Report
The paper proposed a method, which can perform turn detection and identify road intersections based on low-frequency, low-accuracy GNSS trajectory data. The previously work is well presented. The model used is presented clearly and the graphic representations help the reader. However, the paper only refers to previous researches carried out only by researchers from one area of the world. To increase its quality, I recommend the authors to connect the current research with the authors' achievements from other parts of the world.
Author Response
Dear Reviewer：
My changes to your comments specifically please see the attachment.
Sincerely,
Jiusheng Du
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
I havenot futher comments.
I havenot futher comments.
Reviewer 2 Report
The authors satisfied all my comments.
