富饶 发表于 2025-3-30 11:57:53

Prepositional Phrase Attachment Through a Backed-off Model,is applicable. Results on Wall Street Journal data of 84.5% accuracy are obtained using this method. A surprising result is the importance of low-count events — ignoring events which occur less than 5 times in training data reduces performance to 81.6%.

KEGEL 发表于 2025-3-30 15:40:00

http://reply.papertrans.cn/67/6618/661800/661800_52.png

META 发表于 2025-3-30 20:05:32

http://reply.papertrans.cn/67/6618/661800/661800_53.png

Obloquy 发表于 2025-3-30 20:52:39

Trainable Coarse Bilingual Grammars for Parallel Text Bracketing,owledge of one language’s constraints to the task of bracketing the texts in both languages. The second approach generalizes the inside-outside algorithm to adjust the grammar parameters so as to improve the likelihood of a training corpus. Preliminary experiments on parallel English-Chinese text are supportive of these strategies.

GENRE 发表于 2025-3-31 02:05:41

http://reply.papertrans.cn/67/6618/661800/661800_55.png

compassion 发表于 2025-3-31 05:42:16

http://reply.papertrans.cn/67/6618/661800/661800_56.png

climax 发表于 2025-3-31 11:06:44

http://reply.papertrans.cn/67/6618/661800/661800_57.png

RECUR 发表于 2025-3-31 17:05:05

Statistical Augmentation of a Chinese Machine-Readable Dictionary,man evaluators and against a previously available dictionary. We also evaluated performance improvement in automatic Chinese tokenization. Results show that our method outputs legitimate words, acronymic constructions, idioms, names and titles, as well as technical compounds, many of which were lacking from the original dictionary.
页: 1 2 3 4 5 [6]
查看完整版本: Titlebook: Natural Language Processing Using Very Large Corpora; Susan Armstrong,Kenneth Church,David Yarowsky Book 1999 Springer Science+Business Me