Mining Top K Spread Sources for a Specific Topic and a Given Node.

Area of Science:

• Social Network Analysis
• Information Diffusion
• Computational Social Science

Background:

• Users in social networks are influenced by groups of nodes, not just individuals.
• Identifying the source of topic influence is a challenging NP-hard problem.

Purpose of the Study:

• To propose an effective framework for identifying influential nodes in topic propagation.
• To develop an efficient algorithm for probability inference in social network influence.

Main Methods:

• Representing topic propagation using Bayesian networks.
• Constructing a propagation model based on a variant of the voter model.
• Developing the Probability Transition Matrix (PTM) algorithm for probability inference.

Main Results:

• The PTM algorithm demonstrates a complexity of O(θ(3)log2θ), where θ is the number of nodes.
• Extensive experiments on real datasets validate the effectiveness and efficiency of the PTM algorithm.
• The proposed framework successfully addresses the NP-hard problem of identifying influential nodes.

Conclusions:

• The PTM algorithm provides an effective and efficient solution for identifying influential nodes in social networks.
• The framework offers a robust method for understanding and modeling topic spread.
• This research contributes to the field of social network analysis and information diffusion studies.