TnT - A Statistical Part-of-Speech Tagger
About
Trigrams'n'Tags (TnT) is an efficient statistical part-of-speech tagger. Contrary to claims found elsewhere in the literature, we argue that a tagger based on Markov models performs at least as well as other current approaches, including the Maximum Entropy framework. A recent comparison has even shown that TnT performs significantly better for the tested corpora. We describe the basic model of TnT, the techniques used for smoothing and for handling unknown words. Furthermore, we present evaluations on two corpora.
Thorsten Brants• 2000
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Part-of-Speech Tagging | UD Average 1.2 (test) | Accuracy94.61 | 22 | |
| Part-of-Speech Tagging | UD Indo-European 1.2 (test) | Accuracy94.7 | 8 | |
| Part-of-Speech Tagging | UD non-Indo-European 1.2 (test) | Accuracy94.57 | 8 | |
| Part-of-Speech Tagging | UD Bulgarian 1.2 (test) | Accuracy96.84 | 4 | |
| Part-of-Speech Tagging | UD English 1.2 (test) | Accuracy92.66 | 4 | |
| Part-of-Speech Tagging | UD Arabic 1.2 (test) | Accuracy97.82 | 3 |
Showing 6 of 6 rows