Bayesian Checking for Topic Models
David Mimno, David M. Blei · 2011
Real document collections do not fit the independence assumptions asserted by most statistical topic models, but how badly do they violate them? We present a Bayesian method for measuring how well a topic model fits a corpus. Our approach is based on posterior predictive checking, a method for diagnosing Bayesian models in user-defined ways. Our method can identify where a topic model fits the data, where it falls short, and in which directions it might be improved. 1
Cited by 65 works.
Authors
References
- Latent dirichlet allocation
- Finding scientific topics
- Bayesianly Justifiable and Relevant Frequency Calculations for the Applied Statistician
- Dynamic topic models
- Reading Tea Leaves: How Humans Interpret Topic Models
- Online Learning for Latent Dirichlet Allocation
- POSTERIOR PREDICTIVE ASSESSMENT OF MODEL FITNESS VIA REALIZED DISCREPANCIES
- Evaluation methods for topic models
- Estimation in Parallel Randomized Experiments
- Automatic evaluation of topic coherence