EURO 2024 Copenhagen
Abstract Submission

EURO-Online login

3067. Evidential reasoning-based inference of early-stage information propagation source in large-scale social networks

Invited abstract in session TB-6: Advancements of OR-Analytics in Statistics, Machine Learning and Data Science 13, stream Advancements of OR-analytics in statistics, machine learning and data science.

Tuesday, 10:30-12:00
Room: 1013 (building: 202)

Authors (first author is the speaker)

1. Yu-wang Chen
Alliance Manchester Business School, The University of Manchester
2. Tao Wen
Alliance Manchester Business School, The University of Manchester

Abstract

The emergence of social media has transcended the constraints of time and space in communication, facilitating instant interaction among individuals. However, the influx of massive information has led to the widespread dissemination of fake news and deep fakes across online social networks. Identifying the source of fake news promptly and accurately has become a focal point of research. To mitigate the escalation of such situations at an early stage, only information from a limited number of observers can be utilized. Hence, this work first optimizes the deployment of observers in large-scale social networks, achieving maximized coverage and sufficient information collection. Moreover, each observer can provide only limited information, including the time of receiving information and their respective location. In this work, multi-source spatiotemporal data from pairs and a series of observers are integrated by the evidential reasoning algorithm, with weight coefficients determined by the structural characteristics of observers. Through simulation experiments conducted on real-world networks and information propagation paths extracted from Twitter, the proposed method accurately traces the propagation source, demonstrated by superior accuracy and error distance compared to typical algorithms. Such advancements can be further applied in other contexts, such as the management and intervention of infectious diseases within epidemic networks.

Keywords

Status: accepted

Back to the list of papers