Joint Inference of Multiple Label Types in Large Networks

TL;DR

This paper introduces EdgeExplain, a scalable method for jointly inferring multiple label types in large social networks, outperforming traditional label propagation significantly in accuracy.

Contribution

The paper presents EdgeExplain, a novel approach explicitly modeling label type interactions for scalable joint inference in large networks.

Findings

EdgeExplain outperforms label propagation with up to 120% recall@1 improvement.

Scalable inference achieved on a billion-node Facebook subgraph.

Joint inference improves accuracy for multiple label types.

Abstract

We tackle the problem of inferring node labels in a partially labeled graph where each node in the graph has multiple label types and each label type has a large number of possible labels. Our primary example, and the focus of this paper, is the joint inference of label types such as hometown, current city, and employers, for users connected by a social network. Standard label propagation fails to consider the properties of the label types and the interactions between them. Our proposed method, called EdgeExplain, explicitly models these, while still enabling scalable inference under a distributed message-passing architecture. On a billion-node subset of the Facebook social network, EdgeExplain significantly outperforms label propagation for several label types, with lifts of up to 120% for recall@1 and 60% for recall@3.