PGM

生成模型与判别模型：http://www.cnblogs.com/kaituorensheng/p/3379170.html

PGM：
http://www.cnblogs.com/sea-wind/p/4324394.html

http://tripleday.cn/2016/07/14/hmm-memm-crf/

http://homepages.inf.ed.ac.uk/csutton/publications/crftut-fnt.pdf

http://www.cnblogs.com/xulb597/archive/2012/07/31/2616108.html
lstm + crf
http://x-algo.cn/index.php/2017/01/16/1639/

crf
http://blog.163.com/facteur@126/blog/static/2320803020154794459829/

http://leon.bottou.org/projects/sgd

https://github.com/taozhijiang/chinese_nlp

http://blog.csdn.net/felomeng/article/details/4367250

navie bayes
http://scikit-learn.org/stable/modules/naive_bayes.html

PGM blogs :
http://blog.csdn.net/ycheng_sjtu/article/details/25865057\
PGM course:
https://www.coursera.org/specializations/probabilistic-graphical-models

---

probability theory

1.Basic concept
<1>Joint probability distribution
<2>Marginal probability distribution
<3>Conditional probability distribution
2.Bayes and naive bayes
Naive bayes derives from the independence of random variables.
In the application scenario of text classification:
(1) set x=a1,a2,...,am , each a is a attributive character of x.
(2)category set C=y1,y2,...,ym