Predicting the Choice of Online or Offline Shopping Trips Using a Deep Neural Network Model and Time Series Data: A Case Study of Tehran, Iran
Round 1
Reviewer 1 Report
1. The introduction section is somewhat redundant, and the innovative points are not clearly highlighted
2. When reviewing literature, it is not recommended to separate each literature into a paragraph. This disrupts the connection between literature, so the author needs to rewrite it
3. The author needs to supplement the reliability and validity tests of the questionnaire
4. Some derivation formulas for the network used by the author need to be listed
5. The use of indicators such as MSE and RMSE requires literature as a basis, and this literature is recommended https://doi.org/10.3390/systems11080392
6. The author uses more traditional comparative models and needs to supplement some deep network models, such as LSTM, for verification to be more convincing
7. The article lacks a discussion section, which needs to indicate theoretical and methodological contributions
Author Response
"Please see the attachment."
Author Response File: Author Response.pdf
Reviewer 2 Report
The authors proposed a CNN to predict whether people shop online or offline.
1. You need to use one-hot encoding for categorical variables in most machine learning algorithms – most certainly in CNN. Otherwise, the CNN will think that these are continuous variables – which completely falsifies the learning of the CNN. Employment status, for example, is most certainly not a continuous variable from 1 to 7.
2. If the CNN is supposed to predict “yes” or “no” as to whether the person shops online or offline, how can you calculate a mean squared error (MSE)? MSE is only used for regression problems (i.e. calculating the difference between the predicted values and the actual values).
3. Please report the f1 score, precision and recall of your model as otherwise, it is impossible to judge the prediction accuracy of your models.
4. Why did you choose a CNN and not any other NN? I am asking because CNNs are the first choice for image classification, which is not what you are doing.
5. In Figure 1, you have an arrow from the test data to data validation. I hope that there is no 'data spilling' from the test data to the validation data occurring. Please clarify this figure.
6. Please attach an English translation of the survey you used.
7. How is low, medium, and high delivery time defined? How can this question be answered by people who shop in person?
There are various grammatical mistakes throughout the paper.
Author Response
"Please see the attachment."
Author Response File: Author Response.pdf
Reviewer 3 Report
The topic is interesting and the paper is developed adequately. I think it could be improved by making the following changes:
Presentation and Style
The manuscript is formatted according to APA edition. However, several issues need to be corrected:
1. Lines 90 to 95, any reason for the small font? If not, provide the proper font size and format for the consistency.
2. Table 1 should have the proper size of the table name and content inside of the table. Any specific reason for a bigger font inside of the table than the table name? Use APA guidelines when formatting the tables
Originality (Relevance and Contribution)
This study investigates the factors influencing the choice of online and offline shopping trips and their impacts on urban transportation, environment, and economy in Tehran, Iran. As my biggest concern about this manuscript is all about the outdated data you collected and pieces of evidence you provided (e.g., Table 1 using the data from 2020, and the whole data set from 2021) I am unable to assess the technical aspect of the manuscript for the most recent academic research quality. Any specific reason for this outdated data? Please provide it for the academic readers.
Rigor and Depth of Research
As important information, the connection between the introduction and literature review is missing (e.g., all the issues you raised in the introduction section without reference or previous studies, you provided the statistics or number to support your idea, e.g., online and offline shopping behavior studies regarding shopping trips in the literature review section without connecting to the introduction section) I am unable to assess the aspect of the manuscript.
The topic is interesting and the paper is developed adequately. I think it could be improved by making the following changes:
Presentation and Style
The manuscript is formatted according to APA edition. However, several issues need to be corrected:
1. Lines 90 to 95, any reason for the small font? If not, provide the proper font size and format for the consistency.
2. Table 1 should have the proper size of the table name and content inside of the table. Any specific reason for a bigger font inside of the table than the table name? Use APA guidelines when formatting the tables
Originality (Relevance and Contribution)
This study investigates the factors influencing the choice of online and offline shopping trips and their impacts on urban transportation, environment, and economy in Tehran, Iran. As my biggest concern about this manuscript is all about the outdated data you collected and pieces of evidence you provided (e.g., Table 1 using the data from 2020, and the whole data set from 2021) I am unable to assess the technical aspect of the manuscript for the most recent academic research quality. Any specific reason for this outdated data? Please provide it for the academic readers.
Rigor and Depth of Research
As important information, the connection between the introduction and literature review is missing (e.g., all the issues you raised in the introduction section without reference or previous studies, you provided the statistics or number to support your idea, e.g., online and offline shopping behavior studies regarding shopping trips in the literature review section without connecting to the introduction section) I am unable to assess the aspect of the manuscript.
Author Response
"Please see the attachment."
Author Response File: Author Response.pdf
Reviewer 4 Report
This study looks at the elements that influence online and offline buying decisions, as well as their effects on Tehran's public transportation, environment, and economy. In late 2021, a survey of 1,000 e-commerce customers in certain Tehran districts was done to obtain data on successful orders in both settings. A deep neural network model predicted shopping trip types using characteristics such as age, gender, automobile ownership, delivery cost, and product price. The deep neural network outperformed MLP, decision trees, and KNN with an accuracy of 95.63%. Delivery cost, delivery time, and product pricing were all important factors in trip selection. This study provides insights for transportation planners, e-commerce managers, and politicians to create solutions for lowering transportation costs, emissions, and urban congestion while increasing customer happiness and promoting sustainable development.
Paper Comments:
The following statements are some comments about the paper:
1. The introduction section can be furnished with some new papers like:
About other deeplearning application : https://doi.org/10.3390/futuretransp3010012 and About deep learning structure and advantages : A. J. . Moshayedi, A. S. . Roy, A. Kolahdooz, and Y. . Shuxin, “Deep Learning Application Pros And Cons Over Algorithm”, EAI Endorsed Trans AI Robotics, vol. 1, p. e7, Feb. 2022.
2. Paper writing method and Quality
· The motivation of the paper should be improved. And please write your research contribution with number order
· Please check the whole manuscript for types and grammar errors. Language of the paper should be improved.
· Some minor grammatical mistakes are there, read carefully and correct them.
· In the conclusion part please write the exact improvement number by using your proposed method.
3. Figures:
· Redraw the figure 1 (Flowchart), make it look better, the text inside should be more visible and bigger.
· Make it colorfull to have a better look.
4. Tables:
· Explain in details the table 4.
5. Equations:
· No equations or algorithms used in the paper.
6. References:
· You could use some more references in the introduction part.
7. Authors are requested to make typesetting strictly according to the paper template on the conference website.
This study has merit for publication. However, I would recommend a major revision to improve the quality of the manuscript.
This study looks at the elements that influence online and offline buying decisions, as well as their effects on Tehran's public transportation, environment, and economy. In late 2021, a survey of 1,000 e-commerce customers in certain Tehran districts was done to obtain data on successful orders in both settings. A deep neural network model predicted shopping trip types using characteristics such as age, gender, automobile ownership, delivery cost, and product price. The deep neural network outperformed MLP, decision trees, and KNN with an accuracy of 95.63%. Delivery cost, delivery time, and product pricing were all important factors in trip selection. This study provides insights for transportation planners, e-commerce managers, and politicians to create solutions for lowering transportation costs, emissions, and urban congestion while increasing customer happiness and promoting sustainable development.
Paper Comments:
The following statements are some comments about the paper:
1. The introduction section can be furnished with some new papers like:
About other deeplearning application : https://doi.org/10.3390/futuretransp3010012 and About deep learning structure and advantages : A. J. . Moshayedi, A. S. . Roy, A. Kolahdooz, and Y. . Shuxin, “Deep Learning Application Pros And Cons Over Algorithm”, EAI Endorsed Trans AI Robotics, vol. 1, p. e7, Feb. 2022.
2. Paper writing method and Quality
· The motivation of the paper should be improved. And please write your research contribution with number order
· Please check the whole manuscript for types and grammar errors. Language of the paper should be improved.
· Some minor grammatical mistakes are there, read carefully and correct them.
· In the conclusion part please write the exact improvement number by using your proposed method.
3. Figures:
· Redraw the figure 1 (Flowchart), make it look better, the text inside should be more visible and bigger.
· Make it colorfull to have a better look.
4. Tables:
· Explain in details the table 4.
5. Equations:
· No equations or algorithms used in the paper.
6. References:
· You could use some more references in the introduction part.
7. Authors are requested to make typesetting strictly according to the paper template on the conference website.
This study has merit for publication. However, I would recommend a major revision to improve the quality of the manuscript.
Author Response
"Please see the attachment."
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Overall, the author has revised all the content and is worthy of recognition. But seeing a paragraph in the reply was a bit confusing
The author mentioned“Therefore, I have deleted MSE and RMSE and added precision, recall, and f1 score instead. I have also revised the corresponding text in new version of the manuscript lines 271-279 to explain my choice of indicators”
Although the new indicators added by the author are correct, I am confused as to why the author deleted the original indicators (MSE and RMSE). Is the experimental result not good?
Author Response
"Please see the attachment."
Author Response File: Author Response.pdf
Reviewer 2 Report
The authors added the recomeded metrics to asses the model and corrected model.
Author Response
"Please see the attachment."
Author Response File: Author Response.pdf
Reviewer 3 Report
I have confirmed the significant improvement of the manuscript, however, some re-writing issues need to be solved
1. Table 2, keep the consistency of writing - Evaluation…, Assessing…, Analysis…, you may have one format of noun for each sentence
2. Table 2, keep the consistency of writing - The ability to…, They only investigated…, you may have one type for each part
Author Response
"Please see the attachment."
Author Response File: Author Response.pdf
Reviewer 4 Report
please put the paper objective before paper structure
please put the paper objective before paper structure, the authors are respond to all my questions
Author Response
"Please see the attachment."
Author Response File: Author Response.pdf