cmcdh集群管理課程bayes分類

1、問題引樣本：10000 封郵件，每個郵件被標(biāo)記記已有公問題分判為 還是與及假設(shè)文檔單詞出現(xiàn)獨是指的:問題引樣本：10000 封郵件，每個郵件被標(biāo)記記已有公問題分判為 還是與及假設(shè)文檔單詞出現(xiàn)獨是指的:c1類別的出現(xiàn)的單詞總數(shù)Px|c)*P(cPx|)*P(x|c).)P(c1N12P(x|c)*P(c實例說0101011 my,dog,has,flea,problems,實例說0101011 my,dog,has,flea,problems,help,maybe,not,take,him,to,dog,park,my,dalmation,is,so,cute,I,love,stop,ting,

2、worthless,mr,licks,ate,my,steak,how,to,stop,quit,buying,worthless,dog,food,性，0 代表文本形形式利于判斷。通常將文本轉(zhuǎn)換為詞向量向量空間模型) my, dog, has, flea, problems, help, please, maybe, not, take, him, to, park, ,dalmation, is, so, cute, I, love, stop, how, quit, buying, foodting, worthless, garbage, mr, licks, ate, my,dog,

3、has,flea,problems,help,1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 詞集模型只標(biāo)識了單詞在一篇文檔中是否出現(xiàn)代碼實代碼實先驗概3性，類條件概 及性中，單詞dog出現(xiàn)的概率000my,dog,has,flea,先驗概3性，類條件概 及性中，單詞dog出現(xiàn)的概率000my,dog,has,flea,problems,help,my,dalmation,is,so,cute,I,love,mr,licks,ate,my,steak,how

4、,to,stop,9中單詞dog出現(xiàn)的次數(shù)111stop, ting, worthless, 56quit,buying,worthless,dog,food,dogg| 辱常) 性1侮常辱性1詞詞總總2數(shù)數(shù)1maybe,not,take,him,to,dog,park,8性所有單詞總數(shù)：5+6+8=19 dog2 性與存在問01maybe,not,take,him,to,dog,park,8性所有單詞總數(shù)：5+6+8=19 dog2 性與存在問0 x1 沒有出現(xiàn)過。P(x|c)*P(x|c).P(x ( |)lo(x(x|P c 11 11 21xNN1|c2111)211logP(x2 代

5、碼實 代碼實n1p(x|c11 n2類條件概率及先驗概類條件概率及先驗概-2.56494936 -2.56494936 -2.56494936 -3.25809654 -3.25809654 -2.56494936 -2.56494936 -3.25809654 -2.56494936 -2.56494936 -2.56494936 -3.25809654 -3.25809654 -2.15948425 -3.25809654 -2.56494936 -3.25809654 -2.56494936 -2.56494936 -3.25809654 -2.56494936 -2.56494936 -3.25809654 -2.56494936 -3.25809654 -3.04452244 -3.04452244 -3.04452244 -2.35137526 -2.35137526 -3.04452244 -3.04452244 -2.35137526 -2.35137526 -3.04452244 -3.04452244 -2.35137526 -2.35137526 -2.35137526 -2.35137526 -3.04452244 -1.94591015 -3.0445