Research in the domain of fake newsspreader detection has not been explored much from a network analysisperspective . In this paper, we propose a graph neural network based approach to identify nodes that are likely to become spreaders of false information . Using topology and interaction based trust properties of nodes in real-world Twitternetworks, we are able to predict false information spreaders with an accuracy of over 90% . Using the community health assessment model and interpersonal trust we propose aninductive representation learning framework to predict nodes ofdensely-connected community structures that are most likely to spread fakenews .
Author(s) : Bhavtosh Rath, Aadesh Salecha, Jaideep SrivastavaLinks : PDF - Abstract
Code :
https://github.com/afansi/multiscalegnn
Keywords : spreaders - nodes - propose - information - community -
- Bioinformatics Specialization on Coursera
- Machine Learning Nanodegree
- Practical Deep Learning for Coders – Part 1
- Introduction to Python
- Neural Networks from Scratch with Python by Sentdex
- Machine Learning A-Z
- How to Win a Data Science Competition on Coursera
- Hands-On Machine Learning with Scikit-Learn and TensorFlow