贝叶斯分类（二）

今天我们来说明下如何构建贝叶斯的训练器，先附上代码,这篇续接上一篇贝叶斯分类(一)http://blog.csdn.net/xueyunf/article/details/9243481，依然先附上代码：

def trainNB0(trainMatrix,trainCategory):
    numTrainDocs = len(trainMatrix)
    numWords = len(trainMatrix[0])
    pAbusive = sum(trainCategory)/float(numTrainDocs)
    p0Num = zeros(numWords)
    p1Num = zeros(numWords)      #change to ones() 
    #print(p0Num,p1Num )
    p0Denom = 0.0
    p1Denom = 0.0                        #change to 0.0
    for i in range(numTrainDocs):
        if trainCategory[i] == 1:
            p1Num += trainMatrix[i]
            p1Denom += sum(trainMatrix[i])
        else:
            p0Num += trainMatrix[i]
            p0Denom += sum(trainMatrix[i])
    p1Vect = p1Num/p1Denom         #change to log()
    p0Vect = p0Num/p0Denom     #change to log()
    return p0Vect,p1Vect,pAbusive

首先输入的是文本和词的对应矩阵，求出每个词在对应文本中的条件概率。然后返回概率矩阵。当然这个是非常简单的。不需要我做太多的解释。

下面来编写分类的代码：

def classifyNB(vec2Classify, p0Vec, p1Vec, pClass1):
    p1 = sum(vec2Classify*p1Vec)+log(pClass1)
    p0 = sum(vec2Classify*p0Vec)+log(1.0-pClass1)
    if p1>p0:
        return 1
    else:
        return 0

当然也非常简单，其实就是计算和每一类的匹配度，返回高匹配度的分类标签。