High-Reproducibility and High-Accuracy Method for Automated Topic Classification
Open Access
- 29 January 2015
- journal article
- research article
- Published by American Physical Society (APS) in Physical Review X
- Vol. 5 (1) , 011007
Abstract
Much of human knowledge sits in large databases of unstructured text. Leveraging this knowledge requires algorithms that extract and record metadata on unstructured text documents. Assigning topics to documents will enable intelligent searching, statistical characterization, and meaningful classification. Latent Dirichlet allocation (LDA) is the state of the art in topic modeling. Here, we perform a systematic theoretical and numerical analysis that demonstrates that current optimization techniques for LDA often yield results that are not accurate in inferring the most suitable model parameters. Adapting approaches from community detection in networks, we propose a new algorithm that displays high reproducibility and high accuracy and also has high computational efficiency. We apply it to a large set of documents in the English Wikipedia and reveal its hierarchical structure.Keywords
All Related Versions
Funding Information
- Army Research Office (W911NF-14-1-0259)
This publication has 24 references indexed in Scilit:
- Efficiently inferring community structure in bipartite networksPhysical Review E, 2014
- Model selection for degree-corrected block modelsJournal of Statistical Mechanics: Theory and Experiment, 2014
- Hierarchical Block Structures and High-Resolution Model Selection in Large NetworksPhysical Review X, 2014
- Stochastic blockmodels and community structure in networksPhysical Review E, 2011
- Identifying functional miRNA–mRNA regulatory modules with correspondence latent dirichlet allocationBioinformatics, 2010
- Unsupervised Learning of Human Action Categories Using Spatial-Temporal WordsInternational Journal of Computer Vision, 2008
- Statistical mechanics methods and phase transitions in optimization problemsTheoretical Computer Science, 2001
- Learning the parts of objects by non-negative matrix factorizationNature, 1999
- Fractals and Disordered Systems1991
- Indexing by latent semantic analysisJournal of the American Society for Information Science, 1990