Turkish PoS Tagging by Reducing Sparsity with Morpheme Tags in Small Datasets
This addresses sparsity issues in agglutinating languages like Turkish for NLP applications, but it is incremental as it adapts existing methods to a specific domain.
The paper tackles sparsity in Turkish part-of-speech tagging by using morphological features, achieving 94.1% accuracy in morpheme tagging and 89.2% in PoS tagging on a small 5K training dataset.
Sparsity is one of the major problems in natural language processing. The problem becomes even more severe in agglutinating languages that are highly prone to be inflected. We deal with sparsity in Turkish by adopting morphological features for part-of-speech tagging. We learn inflectional and derivational morpheme tags in Turkish by using conditional random fields (CRF) and we employ the morpheme tags in part-of-speech (PoS) tagging by using hidden Markov models (HMMs) to mitigate sparsity. Results show that using morpheme tags in PoS tagging helps alleviate the sparsity in emission probabilities. Our model outperforms other hidden Markov model based PoS tagging models for small training datasets in Turkish. We obtain an accuracy of 94.1% in morpheme tagging and 89.2% in PoS tagging on a 5K training dataset.