Unsupervised Neural Hidden Markov Models
arXiv:1609.09007v162 citations
Originality Incremental advance
AI Analysis
This work addresses unsupervised part-of-speech tagging for natural language processing, representing an incremental advancement in generative models.
The authors tackled the problem of unsupervised tag induction by neuralizing a Hidden Markov Model, achieving competitive performance with state-of-the-art methods using a simpler model that is easily extendable.
In this work, we present the first results for neuralizing an Unsupervised Hidden Markov Model. We evaluate our approach on tag in- duction. Our approach outperforms existing generative models and is competitive with the state-of-the-art though with a simpler model easily extended to include additional context.