Machine Learning and Friends Lunch

Group and Topic Discovery from Relations and Their Attributes

Abstract

We present a probabilistic generative model of entity relationships and their attributes that simultaneously discovers groups among the entities and topics among the corresponding textual attributes. Block-models of relationship data have been studied in social network analysis for some time. Here we simultaneously cluster in several modalities at once, incorporating the attributes (words in this paper) associated with certain relationships. Significantly, joint inference allows the discovery of groups to be guided by the emerging topics, and vice-versa. We present experimental results on two large data sets: sixteen years of bills put before the U.S. Senate, comprising their corresponding text and voting records, and fourty-three years of similar data from the United Nations. We show that in comparison with traditional, separate latent-variable models for words or Blockstructures for votes, the Group-Topic model's joint inference improves both the groups and topics discovered. Joint work with Natasha Mohanty and Andrew McCallum.

Back to ML Lunch home