Graph-Sparse LDA: A topic model with structured sparsity

Finale Doshi-Velez, Byron C. Wallace, Ryan Adams

Research output: Chapter in Book/Report/Conference proceedingConference contribution

14 Scopus citations

Abstract

Topic modeling is a powerful tool for uncovering latent structure in many domains, including medicine, finance, and vision. The goals for the model vary depending on the application: sometimes the discovered topics are used for prediction or another downstream task. In other cases, the content of the topic may be of intrinsic scientific interest. Unfortunately, even when one uses modem sparse techniques, discovered topics are often difficult to interpret due to the high dimensionality of the underlying space. To improve topic interpretability, we introduce Graph-Sparse LDA, a hierarchical topic model that uses knowledge of relationships between words (e.g., as encoded by an ontology). In our model, topics are summarized by a few latent concept-words from the underlying graph that explain the observed words. Graph-Sparse LDA recovers sparse, interpretable summaries on two real-world biomedical datasets while matching state-of-the-art prediction performance.

Original languageEnglish (US)
Title of host publicationProceedings of the 29th AAAI Conference on Artificial Intelligence, AAAI 2015 and the 27th Innovative Applications of Artificial Intelligence Conference, IAAI 2015
PublisherAI Access Foundation
Pages2575-2581
Number of pages7
ISBN (Electronic)9781577357025
StatePublished - Jun 1 2015
Externally publishedYes
Event29th AAAI Conference on Artificial Intelligence, AAAI 2015 and the 27th Innovative Applications of Artificial Intelligence Conference, IAAI 2015 - Austin, United States
Duration: Jan 25 2015Jan 30 2015

Publication series

NameProceedings of the National Conference on Artificial Intelligence
Volume4

Other

Other29th AAAI Conference on Artificial Intelligence, AAAI 2015 and the 27th Innovative Applications of Artificial Intelligence Conference, IAAI 2015
CountryUnited States
CityAustin
Period1/25/151/30/15

All Science Journal Classification (ASJC) codes

  • Software
  • Artificial Intelligence

Fingerprint Dive into the research topics of 'Graph-Sparse LDA: A topic model with structured sparsity'. Together they form a unique fingerprint.

  • Cite this

    Doshi-Velez, F., Wallace, B. C., & Adams, R. (2015). Graph-Sparse LDA: A topic model with structured sparsity. In Proceedings of the 29th AAAI Conference on Artificial Intelligence, AAAI 2015 and the 27th Innovative Applications of Artificial Intelligence Conference, IAAI 2015 (pp. 2575-2581). (Proceedings of the National Conference on Artificial Intelligence; Vol. 4). AI Access Foundation.