2008年4月24日星期四

n-Gram splitter

   While we process chinese, we need to split chinese sentences into chinaese words,a  statistical algorithm is N-Gram split algorithm, which needn't dictionary. 2-gram is very easy to implemente,but the algorithm becomes complex while n > 2. Now, I have implemented the prototype of 2-gram, I will implement the algorithm n-gram while n >2 and HMM algorithm.These algorithms are very useful on finding new words.
   Oyeah!

没有评论: