Accounting for Burstiness in Topic Models

Abstract

Many different topic models have been used successfully for a variety of applications. However, even state-of-the-art topic models suffer from the important flaw that they do not capture the tendency of words to appear in bursts; it is a fundamental property of language that if a word is used once in a document, it is more likely to be used again. We introduce a topic model that uses Dirichlet compound multinomial (DCM) distributions to model this burstiness phenomenon. On both text and non-text datasets, the new model achieves better held-out likelihood than standard latent Dirichlet allocation (LDA). It is straightforward to incorporate the DCM extension into topic models that are more complex than LDA

Cite

Text

Doyle and Elkan. "Accounting for Burstiness in Topic Models." International Conference on Machine Learning, 2009. doi:10.1145/1553374.1553410

Markdown

[Doyle and Elkan. "Accounting for Burstiness in Topic Models." International Conference on Machine Learning, 2009.](https://mlanthology.org/icml/2009/doyle2009icml-accounting/) doi:10.1145/1553374.1553410

BibTeX

@inproceedings{doyle2009icml-accounting,
  title     = {{Accounting for Burstiness in Topic Models}},
  author    = {Doyle, Gabriel and Elkan, Charles},
  booktitle = {International Conference on Machine Learning},
  year      = {2009},
  pages     = {281-288},
  doi       = {10.1145/1553374.1553410},
  url       = {https://mlanthology.org/icml/2009/doyle2009icml-accounting/}
}