Predicting Road Crash Severity Using Classifier Models and Crash Hotspots

Round 1

Reviewer 1 Report

Some pictures could be enhanced as they lack sharpness. Besides the period of three years indicate the projec started in 2016. Too bad no further data has been collected.

Author Response

Dear Reviewer 1,

We would like to thank you for the comments on our manuscript titled "Predicting Road Crash Severity using Classifier Models and Crash Hotspots," ID: "applsci-1990566". We have carefully revised the manuscript based on the constructive comments and suggestions you and the reviewers gave. A complete list of the revisions in the revised version of the paper is given in subsequent pages. We would like to thank you for your support in improving both the content and the structure of the article.

Regards,

Md. Kamrul Islam*,  Imran Reza*, Uneb Gazder, Rocksana Akter, Md Arifuzzaman, Muhammad Muhitur Rahman

Author Response File: Author Response.docx

Reviewer 2 Report

   This paper applies machine learning models to analyze the road crash severity. The paper is generally written. However, I have the following comments:

1.       The literature review is not comprehensive. In addition to the common machine learning methods the authors listed, there are also some methods that have been adopted, such as GBDT. Just to name a few,

Economic development, demographic characteristics, road network and traffic accidents in Zhongshan, China: gradient boosting decision tree model. Transportmetrica A: Transport Science, 2020, 16(3), 359-387.

Predicting Bus Passenger Flow and Prioritizing Influential Factors Using Multi-Source Data: Scaled Stacking Gradient Boosting Decision Trees. IEEE Transactions on Intelligent Transportation Systems, 2021, 22(4), 2510-2523.

2.       Given that different methods (RF, XGBoost) can output important features, which one is better?

3.       Fig. 7 and 8, variable->Variable

4.       It is suggested to combined two subfigures of Fig. 5 into one.

Author Response

Dear Reviewer 2,

We would like to thank you for the comments on our manuscript titled "Predicting Road Crash Severity using Classifier Models and Crash Hotspots," ID: "applsci-1990566". We have carefully revised the manuscript based on the constructive comments and suggestions you and the reviewers gave. A complete list of the revisions in the revised version of the paper is given in subsequent pages. We would like to thank you for your support in improving both the content and the structure of the article.

Regards,

Md. Kamrul Islam*,  Imran Reza*, Uneb Gazder, Rocksana Akter, Md Arifuzzaman, Muhammad Muhitur Rahman

Author Response File: Author Response.docx

Reviewer 3 Report

1. It is not fetal, it is Fatal in Figure 3. Similarly, check for English language and spelling corrections.

2. Figure 6, better represent as a table than figure.

3. Other than using classifier model forms, what is the contribution of this research to the literature and what gaps are addressed by this work?

4. Any reasons for selecting such kind of predictive models over others?

5. What are the limitations of the present study?

Author Response

Dear Reviewer 3,

We would like to thank you for the comments on our manuscript titled "Predicting Road Crash Severity using Classifier Models and Crash Hotspots," ID: "applsci-1990566". We have carefully revised the manuscript based on the constructive comments and suggestions you and the reviewers gave. A complete list of the revisions in the revised version of the paper is given in subsequent pages. We would like to thank you for your support in improving both the content and the structure of the article.

Regards,

Md. Kamrul Islam*,  Imran Reza*, Uneb Gazder, Rocksana Akter, Md Arifuzzaman, Muhammad Muhitur Rahman

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

I am happy with the response