Paper: Bayesian Document Generative Model with Explicit Multiple Topics

ACL ID D07-1044
Title Bayesian Document Generative Model with Explicit Multiple Topics
Venue Conference on Empirical Methods in Natural Language Processing
Session Main Conference
Year 2007
Authors

In this paper, we proposed a novel prob- abilistic generative model to deal with ex- plicit multiple-topic documents: Parametric Dirichlet Mixture Model(PDMM). PDMM is an expansion of an existing probabilis- tic generative model: Parametric Mixture Model(PMM) by hierarchical Bayes model. PMM models multiple-topic documents by mixing model parameters of each single topic with an equal mixture ratio. PDMM models multiple-topic documents by mix- ing model parameters of each single topic with mixture ratio following Dirichlet dis- tribution. We evaluate PDMM and PMM by comparing F-measures using MEDLINE corpus. The evaluation showed that PDMM is more effective than PMM.