Factor Identification and Prediction for Teen Driver Crash Severity Using Machine Learning: A Case Study
Round 1
Reviewer 1 Report
This study identifies factors associated with crashes involving teenagers and uses two approaches: explanatory data analysis and machine learning procedure. It seems to me to be interested but the authors modify several issues following below.
why did you use an EDA? Was is used for identifying factors? Four machine learning techniques were used for both identifying factors and predicting crash injury severity. Please be clarify on this. In the literature review, the authors mentioned that after a careful investigation of the literatures, it was found ... in west Texas has not been performed yet. How about other sites? The manuscript under review in this journal stands for academic aspects not for spatial coverage. Be specific for the results of the four machine learning algorithms on both identifying factors and predictions: how many factors were different among the algorithms? Did you use demographic, trends data in the machine learning procedure? In specific, recommend the results from your analysis can be applied and discussed in the detailed classified locations in west Texas.Author Response
Thanks for the comments and please see the uploaded PDF file for our point-by-point response.
Author Response File: Author Response.pdf
Reviewer 2 Report
The paper illustrates a statistical analysis of the data extracted from the TxDOT database, carried out to identify the risk factors related to accidents involving teens in Texas.
The bibliography should be integrated taking into consideration the recent article on Sensors: Micucci et al., relationship between Turning Signal Detection and Motorcycle Driver's Characteristics.
The main causes identified (unsafe speed, failed to control speed, failed to yield right of way, driver inattention, fatigue or Asleep, and Animal on road) are already known and concern the entire population. It is unclear whether there is a difference between adults and teenagers. Furthermore, it appears that the causes do not figure driving under the influence of alcohol and / or drugs.
A number of aspects related to crash severity prediction are identified:
road class, speed limit, first harmful event, number of lanes, traffic control type, right shoulder type, person age, left shoulder use, construction zone, light condition, roadway type, school zone, weather condition, person restraint used, and manner of collision. However, there is no quantitative correlation.
In lines 134-137 a personal form is used (we...) while in the remaining text an impersonal form is used. It would be better if the whole article was in an impersonal form.
Overall, I agree with the publication, provided that the aspects indicated above are reviewed.
Author Response
Thanks for the comments and please see the uploaded PDF file for our point-by-point response.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
The authors modified this manuscript with my comments. Clearly defined and modified. 1) Please discuss which factors identified from the EDA and how to this. 2) why did you select the machine learning algorithms to conduct this study? 3) I am still not sure what's the contribution for this study. Please discuss more in the literature review.
Author Response
Dear Reviewer,
thanks for your further comments and we believe these constructive comments will continuously improve our paper quality. Please find our detailed responses point-by-point in the uploaded response letter and the revised manuscript.
Best,
Dayong Wu (Corresponding Author)
Assistant Research Scientist
Research & Implementation Division
TEXAS A&M TRANSPORTATION INSTITUTE
The Texas A&M University System
Author Response File: Author Response.pdf
Reviewer 2 Report
I have reviewed the new version of the manuscript. The authors accepted the reports and improved the content. In my opinion, the article can be published, except for some minor adjustments. In fact, I would like to point out that the numbering of the bibliography must be sequential (ie line 78, after [13] it is necessary [14], no [34]...).
Author Response
Dear Reviewer:
Manuscript ID: applsci-702879
Title:“Factor Identification and Prediction for Teen Driver Crash Severity Using Machine Learning: A Case Study on West Texas Rural Areas”
Thanks for pointing out this error . We totally agree and have already updated the numbering sequence of the bibliography by EndNote as to make the right number sequence automatically.
We wish to express our very deep appreciation, and the appreciation of all of us, to your great efforts and approval of our manuscript. In the future, we will work harder and more seriously.
Yours sincerely,
Authors