Mutual Influence of Users Credibility and News Spreading in Online Social Networks

Round 1

Reviewer 1 Report

In this paper, authors model user credibility in Online Social Networks. The main drawback of the paper is the general scenario in which is collocated. Indeed, I don't think that this method could be applied in Social networks, probably only on some of them. So, which ones? Furthermore, authors published other papers of the same topic, which are more or less the same approach presented in this paper. It is not clear what is the novelty of this paper comparing with the others of the authors on user credibility. Authors should explain that this paper is an extension of others and which novelty is presented here. Finally, results are based on a simulation, which can not be accepted for a publication on a journal considering also other papers on this topic. Authors should propose at least an evaluation with a real dataset, and a comparison with other approaches.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

Dear authors,

I want to congratulate you for the effort put in this work.

The current research entitled “Future Internet - Mutual influence of users credibility and news spreading in Online Social Networks” is interesting, the topic being actual, trendy, and challenging. This study investigates how trustworthiness among online social network users influences news spreading, but also how this spreading may affect the users’ credibility. The authors present a set of existing algorithms/models – like Tidal Trust, SUNNY, RN-Trust, etc – known in the literature and use a custom Susceptible-Infected-Recovered model for simulation.

Please delay the results exposure from the introduction section.

The algorithm’s mechanism can be depicted in a more readable way with a flow chart or diagram. This recommendation is not mandatory. Moreover, it is not clear to me where the data come from. In the introduction and other sections, you mention something about simulations since Data Availability Statement/Acknowledgments mentions the thanks for a dataset. Do you use a preexistent dataset or you generate it? Please be more clear about the experiments you have performed.

I suggest including some limitations of the study.

Use consistently neutral way when addressing with “he /she”: “when he post news”. Post or posts?

The caption for Figure 2 can be shorten. Use a brief description for a) and b) then you may detail in one or a few sentences.

The paragraphs “OSNs also were used to detect information…”, “In this paper, we evaluate how trustworthiness…”, “During the last two decades, various…”, “Note that results for case A and C …” are too short.

The references are relatively new. With one exception, all are from 2002-2021 since most of them fall in the second half of this period.

 

I wish you good luck in publishing your paper.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

The aim of this paper is represented by the evaluation of how the trustworthiness among users, interacting on Online Social Networks (OSNs), can influence the news spreading process.

The authors propose a model, by extending their previous works adding an extra dimension of user interactions, based on credibility, as a weighted directed layer (CN layer), to an OSN layer, to study how the CN layer can influence the news spreading and vice versa.

The paper investigates a fascinating problem in nowadays network science, however, there are minor issues that need to be addressed. Although the present work tackles a really interesting topic and the results appear promising, I recommend this paper for publication but not in this form.

I suggest the authors address the following comments:

#Concern #1

The abstract section presents a general background of the issue and it fails in explaining the main target of the paper. The proposal is almost absent, it lacks in fact of some key aspects of the model as the spreading of news and the introduction of mathematical tools and representation as a multiplex network. Furthermore, the target is reduced to the last sentence and a synthesis of the foremost findings is lacking.

The abstract needs to be rewritten.

#Concern #2

The authors decide to include five keywords. Four of them are not specifically cited neither in the abstract nor in the introduction. The keywords represent the tools used in the paper to develop the proposed model for the spreading of the information (real or fake) on the multiplex network. However, these tools in the rest of the paper are either absent or rarely discussed. 

I suggest the authors include more information and references about the theoretical and analytical tools.

#Concern #3

In the Introduction section, several pivotal points of this work are not clearly expressed. For example, the authors first introduce the gate-watching phenomenon of the OSN that takes into consideration the credibility of the content, and then they move towards credibility based on interactions among nodes. The introduction section lacks key aspects about the mathematical tools used. Furthermore, at the end of this section, it is included a brief discussion of findings that is not a common thing to be highlighted in the introduction.

The introduction needs to be improved.   

 

#Concern #4

In the Related Works section, the authors underlined that several models combine elements from game theory and social networks. This sentence lacks references.

In this section the authors discuss mainly the trust models and algorithms in the literature,  then they explore the epidemic topic linked to the spreading of information in different community structures. They referred to very old references, and they totally overlook the spreading in the multiplex network, which is an approach used in the paper, after all. Furthermore, most of the model is referred to the spreading of fake news in the multiplex and the paper lacks very important references in recent literature about misinformation spreading as about collective attention and spreading in the multiplex. 

As an example, the reference [42] that is a paper published in 2007 can be replaced with:

• He, Xingsheng, and Yu-Ru Lin. "Measuring and monitoring collective attention during shocking events." EPJ Data Science 6.1 (2017): 30.

 

I encourage the authors to review some papers before improving their work.

 

• Bródka, Piotr, Katarzyna Musial, and Jaroslaw Jankowski. "Interacting spreading processes in  multilayer networks: a systematic review." IEEE Access 8 (2020): 10316-10341.
• Cinelli, Matteo, et al. "The covid-19 social media infodemic." Scientific Reports 10.1 (2020): 1-10.
• Yu, Xiuli, et al. "Information Spreading on Two-Layered Multiplex Networks With Limited Contact." IEEE Access 8 (2020): 104316-104325.
• Battiston, Federico, Vincenzo Nicosia, and Vito Latora. "The new challenges of multiplex networks: Measures and models." The European Physical Journal Special Topics 226.3 (2017): 401-416.

 

The related works need to be improved, in particular, I suggest introducing aspects on spreading in multiplex, adding more recent scientific literature.

 

#Concern5 #

There are 50 references, and the majority of them are very old.

The [12] is not coherent with the sentence where is cited in the paper.

 

#Concern 6#

Specify and cite the software package used to generate the figures.

 

#Concern 7#

Written English should be improved in my opinion. Please proofread it again

 

#Concern 8#

In the conclusion and future work section, I suggest the authors extend the final discussion about the findings, checking also some mistakes related to numbers of nodes (5000?) and the number of news (500?) simulated. There is an incongruity with the numbers indicated in the Simulations and Results section.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Unfortunately, all the suggestions were not included in the paper. For this reason, I don't think that the paper could be accepted for a publication.

Author Response

No comment required

Round 3

Reviewer 1 Report

No further comments.

Author Response

--