| Experiments | SEG |
| Experiments | 8- ....... _<>~~~~ °’ SEG <> 3- ’-—’A ‘ ' ' ‘ ' ' ‘ ' - e - - - - “A v- o, g,"’ 0 L 00 r,—A, l0_ [\ E’— O\CAP 0 |
| Experiments | SEG F1 MQA |
| Joint Query Annotation | 2Q 2 {CAP, TAG, SEG }. |
| Models 2.1 Baseline Models | When tested against a human-annotated gold standard of linguistic morpheme segmentations for Finnish, this algorithm outperforms competing unsupervised methods, achieving an F—score of 67.0% on a 3 million sentence corpus (Creutz and Lagus, 2006). |
| Models 2.1 Baseline Models | In order to get robust, common segmentations , we trained the segmenter on the 5000 most frequent words2; we then used this to segment the entire data set. |
| Models 2.1 Baseline Models | Of the phrases that included segmentations (‘Morph’ in Table 1), roughly a third were ‘productive’, i.e. |