Open AccessOrigin Estimation Model and Algorithm of Rumor Source in Social Networks
Author(s) -
Keyi Pan
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1881/2/022079
Subject(s) - rumor , computer science , monte carlo method , function (biology) , convergence (economics) , algorithm , estimation , tracking (education) , data mining , mathematical optimization , theoretical computer science , mathematics , statistics , engineering , psychology , pedagogy , public relations , systems engineering , evolutionary biology , political science , economics , biology , economic growth
In the network model, few put effort to study the methodology to find the source of diffusion while many topics on information diffusion are discussed in depth. To find the source of rumor is a different story from other problem like how to estimate the effect of rumor. The rumor source tracking is a substantially important problem since it is closely connected to the web security and social steadiness during convulsion. When the existing theorem based on diffusion problem cannot do much help on tracking rumor source, some novel methodologies using mathematical modeling and monte-carlo simulation can tackle the problem systematically. In this paper, we first introduce a target variable, a error function, to evaluate the estimation by comparing the difference of of infected spots which affected by the rumor. Then we proved the complexity of source backtracking problem. We imported the Apporxmation Algorithm to optimize the monte-carlo trials for the convergence of minimized error function. For our experiments on email records of European research institutions, we implement our error function checking criteria to verify the effectiveness and suitability of our framework, as well as the evaluation of some special situation and advice for further improvement.