Dynamic Relational Topic Models

Abstract

This paper presents the Dynamic Relational Topic Model, a new dynamic topic model that incorporates both document text and relationships for discovering the underlying topics in document collections and their evolution over time. We derive an approximate variational inference algorithm for our model and demonstrate its effectiveness over previous approaches by analyzing papers in Computer Science from CiteSeerX [5] from 1993 to 2009.

Extracted Key Phrases

5 Figures and Tables

Cite this paper

@inproceedings{Lee2015DynamicRT, title={Dynamic Relational Topic Models}, author={Albert Lee and Andrea S. LaPaugh}, year={2015} }