Identify Optimal Pedestrian Flow Forecasting Methods in Great Britain Retail Areas: A Comparative Study of Time Series Forecasting on a Footfall Dataset
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThis study is based on the evaluation of various optimal forecasting methods for predicting pedestrian footfall in various retail areas in Great Britain.
Very good research work and interesting methodology, however, it is recommended that the following observations be addressed before publication:
1. In section 2 of Materials and Methods there are incorrect citations containing a ? sign.
2. In section 2 in Materials and Methods, in the pre-processing stage, it is suggested to justify the non-use of noise filters such as the Kalman filter to deal with white noise.
3. In section 2 under Materials and Methods within the pre-processing stage it is suggested to indicate the type of interpolation performed based on time, distance, speed or other method and justify its selection.
4. In section 2 in Materials and Methods it is suggested to expand the section on Spatial Distribution Analysis by describing the procedure for integrating spatial features into the obtained forecasts.
5. In section 4 under Discussion, it is suggested to compare with other work of similar characteristics.
6. In section 4 under Discussion, it is suggested to indicate in the limitations section how the use of an interpolation mechanism helped you in your work.
Comments on the Quality of English Language
minor corrections are necessary
Author Response
Please see the attachment
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsThis paper identified optimal pedestrian flow forecasting methods in Great Britain retail areas by using footfall dataset. The paper has clear logic and reasonable structure, which has certain theoretical significance and practical value. However, the article still needs to solve the following problems.
1. It is necessary to introduce the dataset in the abstract.
2. The Literature review is not sufficient. Please add some new literature, such as:
[1] Luo W, Wang Y, Jiao P, et al. Improvement Strategy at Pedestrian Bottleneck in Subway Stations. Discrete Dynamics in Nature & Society, 2022.
[2] Luo W, Jiao P, Wang Y. Pedestrian Arching Mechanism at Bottleneck in Subway Transit Hub. Information (Switzerland), 2021, 12(4):164.
3. How does the authors ensure the accuracy of Wi-Fi Sensors data?
4. How are the training and testing data divided? What are the respective proportions of the total sample size?
5. In Table 1, what methods do Navie, MA, and HW represent respectively? When abbreviations first appear, their full names should be indicated.
Comments on the Quality of English LanguageThe English could be improved to more clearly express the research.
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsArticle ID : ijgi-3306921
Article Name : Identify optimal pedestrian flow forecasting methods in Great Britain retail areas: A comparative study of time series forecasting on footfall dataset
In this paper, They examined the best pedestrian footfall forecasting method for varied retail districts in Great Britain. Six representative time series forecasting models showed that the LSTM model outperformed traditional approaches in most areas. Other algorithms predicted pedestrian numbers better at certain areas with specific spatial attributes.
These are the following points need to be address.
1) In Section 1, authors should add contributions of the paper in subsection under introduction section and clearly mention the contribution of the present research paper.
2)The literature review fails. Please offer one table that summarizes current advances on the topic.
2) Table 1 is not properly derived. Pl derive it properly with from 2018 to till date in ascending order with another 2-3 additional attributes.
3) Redraw the figure 1 which is not derived properly with all the components with detail specification.
4)Pl describe the dataset properly with in one subsection (preferable is in section 3)
5) Subsection Machine Learning approaches is not defined properly. Pl define it with proper references.
6) Pl derive the eq 1 to 6 properly with its parameters
7) Pl derive the figure 8 , 9, 10 , 11 and 12 properly with their subfigures.
8) in the present research, Pl mention with suitable table with comparison of LSTM model with other existing model.
9) Rewrite the conclusion section with proper derived work with future recommendation.
Author Response
Please see the attachment
Author Response File: Author Response.pdf
