DyDiff-VAE: A Dynamic Variational Framework for Information Diffusion Prediction

TL;DR

DyDiff-VAE is a novel dynamic variational model that predicts information diffusion on social media by modeling evolving user interests and integrating cascade content with user forwarding sequences, outperforming existing methods.

Contribution

The paper introduces a dynamic encoder for evolving user interests and a dual attentive decoder for better propagation likelihood estimation, advancing diffusion prediction models.

Findings

Achieves 43.3% relative gain over baselines

Demonstrates superior performance on Twitter and Youtube datasets

Offers lower run-time compared to RNN-based models

Abstract

This paper describes a novel diffusion model, DyDiff-VAE, for information diffusion prediction on social media. Given the initial content and a sequence of forwarding users, DyDiff-VAE aims to estimate the propagation likelihood for other potential users and predict the corresponding user rankings. Inferring user interests from diffusion data lies the foundation of diffusion prediction, because users often forward the information in which they are interested or the information from those who share similar interests. Their interests also evolve over time as the result of the dynamic social influence from neighbors and the time-sensitive information gained inside/outside the social media. Existing works fail to model users' intrinsic interests from the diffusion data and assume user interests remain static along the time. DyDiff-VAE advances the state of the art in two directions: (i) We propose a dynamic encoder to infer the evolution of user interests from observed diffusion data. (ii) We propose a dual attentive decoder to estimate the propagation likelihood by integrating information from both the initial cascade content and the forwarding user sequence. Extensive experiments on four real-world datasets from Twitter and Youtube demonstrate the advantages of the proposed model; we show that it achieves 43.3% relative gains over the best baseline on average. Moreover, it has the lowest run-time compared with recurrent neural network based models.