Variable Selection for Latent Dirichlet Allocation
This is an incremental improvement for researchers and practitioners using topic modeling to enhance topic quality by reducing noise from irrelevant words.
The paper tackles the problem of irrelevant vocabulary words in latent Dirichlet allocation (LDA) by proposing vsLDA, a variable selection method that makes topics multinomial distributions over a subset of the vocabulary, resulting in more robust and discriminative topics. The result shows vsLDA outperforms symmetric LDA in likelihood and classification, and asymmetric LDA in consistency and classification, with similar performance in other comparisons.
In latent Dirichlet allocation (LDA), topics are multinomial distributions over the entire vocabulary. However, the vocabulary usually contains many words that are not relevant in forming the topics. We adopt a variable selection method widely used in statistical modeling as a dimension reduction tool and combine it with LDA. In this variable selection model for LDA (vsLDA), topics are multinomial distributions over a subset of the vocabulary, and by excluding words that are not informative for finding the latent topic structure of the corpus, vsLDA finds topics that are more robust and discriminative. We compare three models, vsLDA, LDA with symmetric priors, and LDA with asymmetric priors, on heldout likelihood, MCMC chain consistency, and document classification. The performance of vsLDA is better than symmetric LDA for likelihood and classification, better than asymmetric LDA for consistency and classification, and about the same in the other comparisons.