Sensor Fusion for Occupancy Estimation: A Study Using Multiple Lecture Rooms in a Complex Building

Round 1

Reviewer 1 Report

Interesting work. I only miss information on ventilation in the tested facilities. Using it or not using it may alter test results and give wrong results.

There is no information about the ventilation used in the tested rooms !!

This aspect should be noted in this study. It can only be theoretically.

Author Response

We thank you very much for your constructive criticism.

Please see the attachment for our responses.

Author Response File: Author Response.pdf

Reviewer 2 Report

1. Results: Recommend to be Major revisions   

This paper uses various Machine Learning methods to explore the combination of multiple sensors for quality improvement. As known that a reliable occupancy estimation can help in many different cases and applications. Particularly for the containment of the SARS-CoV-2 virus, room occupancy is a major factor. The estimation can benefit visitor management systems in real time, but also predictive for room reservation strategy. Using different terminal and non-terminal sensors in different premises of variable size, the goal of this paper is to estimate room occupancy. In the process, the proposed models are trained with different combinations of rooms in training and testing to examine distinctions in the infrastructure of the considered building. The results indicate, that the estimation benefits with the combination of different sensors. Additionally, it is found that a model should be trained with data from every room in a building and cannot be transferred to other rooms.

This paper is with minor merits for Machine Learning and Knowledge Extraction, i.e., lacking of strong theoretical supports to clearly demonstrate the very findings to reveal its valuable contributions. It requires some major revisions.

Firstly, for Section 1, authors should provide the comments of the cited papers after introducing each relevant work. What readers require is, by convinced literature review, to understand the clear thinking/consideration why the proposed approach can reach more convinced results. This is the very contribution from authors. In addition, authors also should provide more sufficient critical literature review to indicate the drawbacks of existed approaches, then, well define the main stream of research direction, how did those previous studies perform? Employ which methodologies? Which problem still requires to be solved? Why is the proposed approach suitable to be used to solve the critical problem? We need more convinced literature reviews to indicate clearly the state-of-the-art development.

For Section 2, authors should also introduce their proposed research framework more effective, i.e., some essential brief explanation vis-à-vis the text with a total research flowchart or framework diagram for each proposed algorithm to indicate how these employed models are working to receive the experimental results. It is difficult to understand how the proposed approaches are working.

For Section 3, authors should use more alternative models as the benchmarking models, authors should also conduct some statistical test to ensure the superiority of the proposed approach, i.e., how could authors ensure that their results are superior to others? Meanwhile, authors also have to provide some insight discussion of the results. Authors can refer the following recommended paper,

A hybrid approach for forecasting ship motion using CNN-GRU-AM and GCWOA. Applied Soft Computing, 2022, 114, 108084.

Author Response

We thank you very much for your constructive criticism.

Please see the attachment for our responses.

Author Response File: Author Response.pdf

Reviewer 3 Report

Interesting and relevant paper. Please consider the following remarks:

* Your research questions are simple yes/no questions. Please provide quantifiable research questions.

* Equations (1), (2), and (3): whilst you describe all variables, it is unclear to read how you derived the equations and what you are trying to calculate / measure. E.g., for (1): how do measure the distance and distance from which point? What are E(t) and P(t) supposed to be?

* Chapter 2.4: you simply list a plethora of machine learning algorithms. Which of them did you use and for which goals / applications? How did the algorithms perform? Also, Linear Regression is traditionally classified as a machine learning method.

* Figure 3: what doe BLE, H, T, .... etc. mean?

All in all, I think the article is relevant for the journal. However, more care should be applied to highlighting the comparison of approaches (state-of-the-art, compared algorithms) and why your approach worked in particular.

Author Response

We thank you very much for your constructive criticism.

Please see the attachment for our responses.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Authors have completely addressed all my concerns.

Author Response

Dear Reviewer,

we thank you very much for your feedback!

Reviewer 3 Report

The suggestions of the previous review round have been met.

After some spelling and grammar check as well as provision of a camera-ready version, the article could be a relevant addition of the journal.

Author Response

Dear Reviewer,

we thank you very much for your feedback.

We made some little changes during an English grammar check and uploaded our final version.