Travel-To-School Mode Choice Modelling Employing Artificial Intelligence Techniques: A Comparative Study
Round 1
Reviewer 1 Report
I reccomend to insert in the end of introduction the structure of the paper and to improve the conclusion specifying better theoretical and pratical implications of the paper.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 2 Report
This study compares three different machine learning tools (MLT) to predict the mode choice behavior of the school-goers. The research topic is quite interesting and methodology, results, and discussions of this study are relevant. However, few major issues should be improved before publication.
First of all, it is the weakest point of this study that the theoretical contribution is relatively weak. Only applying existing representative ML tools to mode selection behavior and comparing performance between them is not enough for academic papers. It is necessary to present at least the ensemble method of these or to derive a way for its performance improvement.
Second, there is a lack of background on the study of the mode choice behavior of the school-goers. In order to understand the research background and to understand the need for research, it is necessary to separate the chapter explaining the research background and country-specific information for travel mode selection. Additional explanations of why this study is important should also be added.
Finally, there is a lack of information on the variables used in the model, which is the core of the research, and basic statistics. Specific information such as the type of variables used, descriptive statistics, and adjusted parameters should be presented to increase the reliability of the research results.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
Authors well modified the manuscript according to the suggestions.