Dirichlet-Survival Process: Scalable Inference of Topic-Dependent Diffusion Networks

TL;DR

This paper introduces Houston, a non-parametric, unsupervised model for inferring dynamic, topic-dependent diffusion networks from continuous-time data, improving over existing methods in clustering and subnetwork detection.

Contribution

It presents Houston, a scalable, online inference framework that jointly models content, timing, and network position for diffusion processes, advancing beyond prior parametric and limited-feature models.

Findings

Outperforms baselines in cluster recovery

Achieves better subnetwork inference

Scales linearly with dataset size

Abstract

Information spread on networks can be efficiently modeled by considering three features: documents' content, time of publication relative to other publications, and position of the spreader in the network. Most previous works model up to two of those jointly, or rely on heavily parametric approaches. Building on recent Dirichlet-Point processes literature, we introduce the Houston (Hidden Online User-Topic Network) model, that jointly considers all those features in a non-parametric unsupervised framework. It infers dynamic topic-dependent underlying diffusion networks in a continuous-time setting along with said topics. It is unsupervised; it considers an unlabeled stream of triplets shaped as \textit{(time of publication, information's content, spreading entity)} as input data. Online inference is conducted using a sequential Monte-Carlo algorithm that scales linearly with the size of the dataset. Our approach yields consequent improvements over existing baselines on both cluster recovery and subnetworks inference tasks.