Learning Ideological Embeddings from Information Cascades

Abstract

Modeling information cascades in a social network through the lenses of the ideological leaning of its users can help understanding phenomena such as misinformation propagation and confirmation bias, and devising techniques for mitigating their toxic effects.