Implement Kmeans algorithm.
Introduce the MaxEntropy simply.
1. Reduce memory employed: It is a word that is regarded as unit, rather than the sentence.
2. Reduce the number of the words changed: The random probability of a word need less when one should be changed.
3. Aim: Only several words are changed in a sentence.
4. Add common words influence (high frequency words)
5. Add MaxEntropy ========================================================================================
Look for friends that delete my wechat
A package for Word Segmentation using neural networks based on package LibN3L
Chinese to pinyin tool with Python
========================================================================================If you have some advices or suggestions, please contact me by jinyeqiong@hotmail.com