Efficient methods for topic model inference on streaming document collections

@inproceedings{Yao2009EfficientMF,
  title={Efficient methods for topic model inference on streaming document collections},
  author={Limin Yao and David Mimno and A. McCallum},
  booktitle={KDD},
  year={2009}
}
  • Limin Yao, David Mimno, A. McCallum
  • Published in KDD 2009
  • Computer Science
  • Topic models provide a powerful tool for analyzing large text collections by representing high dimensional data in a low dimensional subspace. Fitting a topic model given a set of training documents requires approximate inference techniques that are computationally expensive. With today's large-scale, constantly expanding document collections, it is useful to be able to infer topic distributions for new documents without retraining the model. In this paper, we empirically evaluate the… CONTINUE READING
    392 Citations
    Efficient Methods for Inferring Large Sparse Topic Hierarchies
    • 8
    • Highly Influenced
    • PDF
    Linear Time Samplers for Supervised Topic Models using Compositional Proposals
    • 6
    • PDF
    Online Belief Propagation for Topic Modeling
    • 12
    • Highly Influenced
    • PDF
    Sparse online topic models
    • 22
    • PDF
    Reducing the sampling complexity of topic models
    • 184
    • Highly Influenced
    • PDF
    A New Approach to Speeding Up Topic Modeling
    • 12
    • Highly Influenced
    • PDF

    References

    SHOWING 1-3 OF 3 REFERENCES
    Topic Models over Text Streams: A Study of Batch and Online Unsupervised Learning
    • 141
    • Highly Influential
    • PDF
    Fast collapsed gibbs sampling for latent dirichlet allocation
    • 509
    • Highly Influential
    • PDF
    Latent Dirichlet Allocation
    • 26,042
    • Highly Influential
    • PDF