Markov Random Topic Fields

Most approaches to topic modeling assume an independence between documents that is frequently violated. We present an topic model that makes use of one or more user-specified graphs describing relationships between documents. These graph are encoded in the form of a Markov random field over topics and serve to encourage related documents to have similar topic structures. Experiments on show upwards of a 10% improvement in modeling performance.

• Type: Journal
• Year: 2009
• Where: ACL
• Authors: Hal Daume III
• Keywords: ACL 2009, Computational Linguistics, Markov Random Field, Similar Topic Structures, Topic Modeling Assume