Part of Speech Tagging with Backfreed Propagate Network
Zhi - hong TIAN · Journal of Shanxi University · 2001
A backfreed propagate (BP) network is used in this paper to resolve the problem of part of speech and this net has a copability that handing contextual information in text.Annotated sentences is used as training set and testing set,and giving a testing result of 3-7 word.The results show after the network is well trained with annotated sentences,it's validity coefficient is greater than 90% and disambiguation coefficient is greater than 87%,this result is the superior to the effect of POS tagging based on statistical method.