| tech,9-5-P03-1051,ak | segmentation accuracy </term> , we use an <term> | unsupervised algorithm | </term> for automatically acquiring new <term> | #4717 To improve the segmentation accuracy, we use an unsupervised algorithm for automatically acquiring new stems from a 155 million word unsegmented corpus, and re-estimate the model parameters with the expanded vocabulary and training corpus. | |
| tech,17-2-P03-1051,ak | </term> and uses it to bootstrap an <term> | unsupervised algorithm | </term> to build the <term> Arabic word segmenter | #4657 Our method is seeded by a small manually segmented Arabic corpus and uses it to bootstrap an unsupervised algorithm to build the Arabic word segmenter from a large unsegmented Arabic corpus. |
