An Agent-Based Model of Task-Allocation and Resource-Sharing for Social Internet of Things
Round 1
Reviewer 1 Report
I-General comments
-It would be better if you add a use case scenario about the usefulness of the proposed model in the context of SIoT.
-Under "3.2 design concept", the subsubsection 3.2.x paragraphs should be extended (you can not have title with only two-three lines), otherwise you can change it to a bulleted list.
-Figure 3, figure 4 and figure 5 should be changed, the three graphs are tightly overlapped which make it difficult to observe their values, change the graph drawing to clearer drawing.
II-related works
-add the following works to the related work section:
Siot: Giving a social structure to the internet of things, Atzori et al, IEEE communications letters. (related to SIoT)
A Social Relationships Based Service Recommendation System For SIoT Devices, Khelloufi et al,IEEE Internet of Things Journal. (related to social relationships between devices)
STLF: Spatial-temporal-logical knowledge representation and object mapping framework, Dhelim et al, 10.1109/SMC.2016.7844459(related to social object mobility)
III-Language and grammar
-define the full name only the first time you when you give its abbreviation, after that use the abbreviation instead.
Ex. in Line-18 "Social internet of things (SioT)", in Line-38 should be 'SIoT' not 'Social IoT"
the same note goes for all other abbreviations.
Ex. IoT in Line-107
- in Line-43 "a framework with above-mentioned characteristics", what characteristics are you referring to,
i guess you wanted to say "the bellow-mentioned"
- remove the bold font in the word 'mobility" in Line-278, the same goes for the sentences in Line-354 and Line-380
- add a space before the reference 51 in Line-297VI-References
-. the name of the journal is missing in reference 8.
-. there is no need to write the whole title in capital letters for reference 19
-remove ref 15 and 16 from the reference list, and add them a footnotes in the page you cited these two refs.
Author Response
Dear Reviewer,
Thank you for all these valuable comments. Indeed, we appraise your insight and knowledge. We have addressed all the issues that were mentioned. In the following, we provide a point-to-point rebuttal of the comments.
-It would be better if you add a use case scenario about the usefulness of the proposed model in the context of SIoT.
We have added a paragraph in the Introduction where we have contextualized SIoT with the challenges of fog computing and how the challenges faced in fog computing can have answers if objects of SIoT are socialized. Please see paragraph from line 66 to 77 in the new draft. We have also updated section 2.4 (SIoT Applications) to emphasize the social objects and how these can be useful in the future. Please see lines 126-141.
-Under "3.2 design concept", the subsubsection 3.2.x paragraphs should be extended (you can not have title with only two-three lines), otherwise you can change it to a bulleted list.
We have reworked whole section 3 and merged smaller subsections into main sections.
-Figure 3, figure 4 and figure 5 should be changed, the three graphs are tightly overlapped which make it difficult to observe their values, change the graph drawing to clearer drawing.
All Figures (graphs) are drawn again and they are in good shape now.
-add the following works to the related work section:
Siot: Giving a social structure to the internet of things, Atzori et al, IEEE communications letters. (related to SIoT)
A Social Relationships Based Service Recommendation System For SIoT Devices, Khelloufi et al,IEEE Internet of Things Journal. (related to social relationships between devices)
STLF: Spatial-temporal-logical knowledge representation and object mapping framework, Dhelim et al, 10.1109/SMC.2016.7844459(related to social object mobility)
All missing related work is added now. See lines 147-150, lines 98-100, and lines 187-188.
-define the full name only the first time you when you give its abbreviation, after that use the abbreviation instead.
Ex. in Line-18 "Social internet of things (SioT)", in Line-38 should be 'SIoT' not 'Social IoT"
the same note goes for all other abbreviations.
Ex. IoT in Line-107
- in Line-43 "a framework with above-mentioned characteristics", what characteristics are you referring to,
i guess you wanted to say "the bellow-mentioned"
- remove the bold font in the word 'mobility" in Line-278, the same goes for the sentences in Line-354 and Line-380
- add a space before the reference 51 in Line-297VI-References
-. the name of the journal is missing in reference 8.
-. there is no need to write the whole title in capital letters for reference 19
-remove ref 15 and 16 from the reference list, and add them a footnotes in the page you cited these two refs.
All the above modifications are done. We appreciate this level of keenness to make our contribution error-free.
Reviewer 2 Report
The overall presentation of the paper is good but required some revision. Please see my comments below:
- The title does not reflect the content of the paper because you proposed and evaluate a model which is not present in the title.
- The research motivation is not very clear in the introduction, please try to refine it.
- If you can add the main contributions of the paper in the introduction that will be great.
- Some subsections are too short especially in section 3, a careful revision is recommended.
- In all graphs, I can see time and service units that are not defined clearly in the paper.
- Many people claim that Fog is more suitable for IoT applications, I would suggest adding a section to discuss if it is possible. You can see the following papers to know more about Fog:
https://doi.org/10.3390/bdcc2020010
https://doi.org/10.1109/ACCESS.2018.2866491
- It will be good if you can add some future research direction in the conclusion section.
Author Response
Dear Reviewer,
Thank you for all these valuable comments. Indeed, we appraise your insight and knowledge. We have addressed all the issues that were mentioned. In the following, we provide a point-to-point rebuttal of the comments.
The title does not reflect the content of the paper because you proposed and evaluate a model which is not present in the title.
The title has been changed. Thanks for this very valid comment and we appreciate it.
The research motivation is not very clear in the introduction, please try to refine it.
Section 2 changed quite substantially to address this concern. Please see the highlighted text in the revised version of the paper.
If you can add the main contributions of the paper in the introduction that will be great.
We have also made changes in the Introduction. Now lines 55-66 describe the main contribution of the paper.
Some subsections are too short especially in section 3, a careful revision is recommended.
We have reworked whole section 3 and merged smaller subsections into main sections.
In all graphs, I can see time and service units that are not defined clearly in the paper.
The definitions are more clearly given in section 3.2: Observations.
Many people claim that Fog is more suitable for IoT applications, I would suggest adding a section to discuss if it is possible. You can see the following papers to know more about Fog:
https://doi.org/10.3390/bdcc2020010
https://doi.org/10.1109/ACCESS.2018.2866491
This turned out to be a very learned and good suggestion for this paper. We have added a paragraph in the Introduction where we have contextualized SIoT with the challenges of fog computing and how the challenges faced in fog computing can have answers if objects of SIoT are socialized. Please see paragraph from line 66 to 77 in the new draft. We were really not aware of this potential dimension of SIoT. We are very thankful for this direction.
It will be good if you can add some future research direction in the conclusion section.
The above also helped us frame our future work (see the last paragraph). In fact, the above papers open up new dimensions in which this work can be extended. It also provided much-needed technological relevance.
This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.
Round 1
Reviewer 1 Report
The proposed model seams be sound, but it can be used in every type of context where agents must cooperate. In particular, authors should justify what of new their model offer and why it is suitable for IoT applications. Related work should be extended taking into account, for example, of the works of Fortino and Russo. Authors should compare their results with the ones of related work. Conclusions should highlight the main contributions of the paper, in particular, for the IoT research and applications.
Author Response
Thanks for your valuable comments.
In particular, authors should justify what of new their model offer and why it is suitable for IoT applications. Related work should be extended taking into account, for example, of the works of Fortino and Russo.
Section 2 has been updated to address both points.
Conclusions should highlight the main contributions of the paper, in particular, for the IoT research and applications.
The conclusion has been updated to address this point.
Reviewer 2 Report
The paper though tackling interesting topic didn't give the perspective in which this idea would be implemented. The specification of the objects and Social objects of the model is not clear. More important main achievement of the paper is also not clear
Results were presented in small compact figures which made it difficult to understand
Captions for figures were very long this should be in moved to the discussion, and leave a short line as a caption.
Discussion of the Results was poorly presented and didn’t highlight clearly what the results mean for different scenarios, most parameter were lightly taken in the discussion.
References format are no consistent, ad following different referencing schemes, some of them are incomplete e.g. reference 6, 8,12, 15, 38 etc.
Author Response
Thanks for your valuable comments.
The paper though tackling interesting topic didn't give the perspective in which this idea would be implemented.
Section 2 has been updated to address this point.
The specification of the objects and Social objects of the model is not clear.
We have put our previous paper as a reference. A more detailed description of the specification of social objects is given there. To avoid redundancy, we opted not to include them in this paper.
More important main achievement of the paper is also not clear
Section 2 has been updated to address this point.
Results were presented in small compact figures which made it difficult to understand
We have increased the size of all the figures.
Captions for figures were very long this should be in moved to the discussion, and leave a short line as a caption.
Done as suggested.
Discussion of the Results was poorly presented and didn’t highlight clearly what the results mean for different scenarios, most parameter were lightly taken in the discussion.
The results section has been changed with important points highlighted now.
References format are no consistent, ad following different referencing schemes, some of them are incomplete e.g. reference 6, 8,12, 15, 38 etc.
Done as suggested.
Reviewer 3 Report
This paper focus the impact of profile-based mobility and is based on work published earlier. The research is still on-going, and the data presented, although of interest, is based on simulation using NetLogo, and as the title properly refers, the experiments are meant to be towards and agent-based model of social IoT.
Introduction, background and motivation (sections 1 and 2) are well written. Subsection 2.5 outlines the proposed model and points previous works, depicting what is the novel content on this paper.
Section 3 describes the model. Figures 1 and 2 were previously used in reference 9, and as such, it should be indicated that reference. This aspect is important because the figures presented are exactly the same shown in the previous publication.
Section 4 presents the novel simulation and its results. I suggest that figure 3 could be presented in a more perceptive way: perhaps a table showing numbers or graphics is another format? This, because the explanation in the caption, which is OK, is not, at least to me, of easy comprehension. And figures 4, 5, and 6, what is the purpose? At some point the reader might get lost in the way. And this is the core of the research, and as such, should be emphasised the most as possible. The discussion subsection 4.3 is OK, but given the graphs presented earlier, it was not completely clear to me, how the results prove the conclusions. And Section 5 would benefit of a slight less synthetic explanation, and going further, for instance, explaining how this results could be used in the future agent-based model of social IoT that appears to be under development.
So, although the paper is fairly well written and organised, would largely benefit of some improvements in sections 4 and 5.
Author Response
Section 3 describes the model. Figures 1 and 2 were previously used in reference 9, and as such, it should be indicated that reference. This aspect is important because the figures presented are exactly the same shown in the previous publication.
Done as suggested.
Section 4 presents the novel simulation and its results. I suggest that figure 3 could be presented in a more perceptive way: perhaps a table showing numbers or graphics is another format? This, because the explanation in the caption, which is OK, is not, at least to me, of easy comprehension.
Done as suggested.
And figures 4, 5, and 6, what is the purpose? At some point the reader might get lost in the way. And this is the core of the research, and as such, should be emphasised the most as possible. The discussion subsection 4.3 is OK, but given the graphs presented earlier, it was not completely clear to me, how the results prove the conclusions. And Section 5 would benefit of a slight less synthetic explanation, and going further, for instance, explaining how this results could be used in the future agent-based model of social IoT that appears to be under development.
So, although the paper is fairly well written and organised, would largely benefit of some improvements in sections 4 and 5.
Section 4 and 5 are changed substantially to address the above points.
Round 2
Reviewer 1 Report
The proposed model and the simulation results are reasonable. However, the focus of the paper is not IoT objects. In fact, the paper copes with a more general problem, i.e., where agents may represent IoT objects, people and other kinds of entity and several researchers propose solutions for such a general problem. Authors should highlight the aspects that make the proposed model and simulation specific for IoT objects.
Author Response
Dear Reviewer,
Thanks for these valuable comments. As a response to the recommendation that "Authors should highlight the aspects that make the proposed model and simulation specific for IoT objects", we have updated paragraph 5 in the Introduction to have a more specific linkage between paragraphs 3 and 4 and paragraph 5. After describing what kind of social capabilities are required by social objects (of SIoT) and corresponding computing paradigms, we have rewritten paragraph 5 as follows:
"The above-mentioned characteristics provide the foundation for our quest to examine the impact of societal features of IoT. Therefore, the framework presented in this paper is of combinatorial nature, combining societal aspects and features of distributed computing paradigm. That is, ..."
I hope this makes Introduction more synchronous while answering the comments given.
Reviewer 2 Report
The results are poorly presented, what is the unit of time in the results figures? why 4-6 digit numbers needed to represent the time, why these are not evenly spaced intervals? what is the unit of the service axis?
Conclusion should mean conclusion not a summary, and its poorly written.
The references are not following a particular standard, these are thrown randomly in the end f the paper, with many still incomplete
Author Response
Dear Reviewer,
First, we thank you for these value able comments. We have done all the suggested changes.
The results are poorly presented, what is the unit of time in the results figures?
The unit of time is number of iterations. It has now been added to all figures
why 4-6 digit numbers needed to represent the time, why these are not evenly spaced intervals?
These numbers represent the iterations and we have these many iterations. As the simulation data is huge, it is not possible to display all values on x-axis, so selected data has been shown on x-axis label. However, We double checked the spaces of x-axis labels and found that they are equally spaced.
what is the unit of the service axis?
It has been explained in the text now. To address the concern of another’s reviewer, we avoided putting long and detailed captions for the y-axis.
Conclusion should mean conclusion not a summary, and its poorly written.
The conclusion is written to address the recommendation.
The references are not following a particular standard, these are thrown randomly in the end f the paper, with many still incomplete
We double checked the references. They are not randomly placed. Rather they are placed in the order of citation in the text which is the MDPI citation style. We also checked that references are complete and nothing is missing.
Round 3
Reviewer 1 Report
I asked to highlight the aspects that make the proposed model and simulation specific for IoT objects, but the revision did not satisfy the request.
