Source positioning method based on maximum likelihood under independent cascade model

Abstract

One or more embodiments of the invention provide a source positioning method based on maximum likelihood under an independent cascade model. The method comprises the steps: carrying out diffusion of a seed node set on an IC model, forming an infection network, calculating a propagation probability through employing an independent path, calculating a likelihood function with a minimum influence range, and selecting a seed set through employing a greedy strategy, and finally outputting an infection source. Compared with the traditional source positioning method which can only solve the problem of single influence source positioning, the method provided by the invention has the advantage that work of solving the problem of complex multi-source positioning by using the simple IC model is less. According to the method, under the condition that only probability factors are considered, the single-source and multi-source source positioning problems can be solved at the same time, and a corresponding basis is provided for the research that the likelihood probability is applied to the multi-source problem in the future. According to the technology, the efficiency of identifying the source nodes which influence propagation in the social network can be improved, and the application range and the practicability of the technology in the field of source positioning problems are widened.

Description

technical field [0001] The invention belongs to a method for source location by using independent cascading models and likelihood function calculations in complex networks, and particularly relates to a method for source location based on independent path technology and a maximum likelihood estimation algorithm based on influence coverage method. Background technique [0002] The continuous development of Internet technology is turning the world into a global village, and people's social networks continue to expand. These provide faster and wider channels for the dissemination of information, but it also provides convenience for the dissemination of some bad information, such as some rumors. , viruses, etc. are widely disseminated through these mobile clients. Some unreasonable people are easily influenced by these rumors and easily believe these rumors. negatively affect social stability. Therefore, how to locate the communication source accurately and quickly is an impor...

AI Technical Summary

Problems solved by technology

[0004] Most of the existing source location algorithms deal with single-source problems, while multi-source problems are relatively complex, so there are still relatively few methods for dealing with multi-source problems, and most of the current source location research methods are based on SI, SIR models, etc. The infectious disease model relies on the time factor as a consideration, and there is less research on the probability model of IC

Insufficient efficiency in identifying source nodes of influence propagation in social networks

Images