Unsupervised learning of word sense disambiguation rules by estimating an optimum iteration number in the EM algorithm.
Hiroyuki ShinnouMinoru SasakiPublished in: CoNLL (2003)
Keyphrases
- em algorithm
- word sense disambiguation
- expectation maximization
- unsupervised learning
- maximum likelihood
- mixture model
- parameter estimation
- gaussian mixture model
- generative model
- maximum likelihood estimation
- natural language processing
- gaussian mixture
- maximum a posteriori
- semantic similarity
- hyperparameters
- supervised learning
- mixture modeling