Troll on the net - testing the vulnerability of different networks structures on misinformation

Author:

Kopczewski Tomasz1,Kocerka Julian2

Affiliation:

1. University of Warsaw

2. Faculty of Economic Sciences, University of Warsaw

Abstract

Abstract Internet trolls are a powerful instrument for spreading misinformation. A troll can be an algorithm that sends a predetermined message to the network and works on the principle: "Repeat a lie often enough, and people will believe it". This study uses a modified DeGroot model of social learning to test the network's vulnerability to messages spread by such trolls. In the study, random graphs generated in a Monte Carlo simulation are infected by a troll placed in a random vertex. The number of interactions needed to fully identify with the troll's message by all network participants will measure the vulnerability of the graphs on misinformation. Multidimensional Scaling and Random Forest machine learning methods applied to data from Monte Carlo simulation will determine which graph and vertex statistics are the best predictors of the network's vulnerability to misinformation. We found that the features of the troll are much stronger predictors than the characteristic of the network.

Publisher

Research Square Platform LLC

Reference30 articles.

1. Authors (2023), online report and codes, https://microeconomics.shinyapps.io/troll/ (accessed 08.06.2023)

2. Brin, S., & Page, L. (1998). The anatomy of a large-scale hypertextual Web search engine. Computer Networks and ISDN Systems, 30(1–7), 107–117. doi:10.1016/s0169-7552(98)00110-x

3. Brown, E (2019), Online fake news is costing us $78 billion globally each year, ZDNet, https://www.zdnet.com/article/online-fake-news-costing-us-78-billion-globally-each-year/ (accessed 08.06.2023)

4. Carroll, J. D., & Arabie, P. (1998). Multidimensional scaling. Measurement, judgment and decision making, 179–250.

5. Reaching a Consensus.;DeGroot MH;Journal of the American Statistical Association,1974

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3