Latent Dirichlet Allocation

← Back to document page

Journal of Machine Learning Research 3 (2003) 993-1022Submitted 2/02; Published 1/03Latent Dirichlet AllocationDavid M. Science DivisionUniversity of CaliforniaBerkeley, CA 94720, USAAndrew Y. Science DepartmentStanford UniversityStanford, CA 94305, USAMichael I. Science Division and Department of StatisticsUniversity of CaliforniaBerkeley, CA 94720, USAEditor:John LaffertyAbstractWe describelatent Dirichlet Allocation (LDA), a generative probabilistic model for collections ofdiscrete data such as text corpora. LDA is a three-level hierarchical Bayesian model, in which eachitem of a collection is modeled as a finite mixture over an underlying set of topics. Each topic is, inturn, modeled as an infinite mixture over an underlying set of topic probabilities.

LATENT DIRICHLET ALLOCATION This line of thinking leads to the latent Dirichlet allocation (LDA) model that we present in the current paper. It is important to emphasize that an assumption of exchangeability is not equivalent to an as-

  Talent, Allocation, Latent dirichlet allocation, Dirichlet