决策树的应用

最后更新： 2017-10-22

一、Python机器学习的库：scikit-learn

• 简单高效的数据挖掘和机器学习分析
• 对所有用户开放，根据不同需求高度可重用性
• 基于Numpy, SciPy和matplotlib
• 开源，商用级别：获得 BSD许可

安装 scikit-learn 自行 Google

二、实战

参考: http://scikit-learn.org/stable/modules/tree.html

源码:

import csv
from sklearn.feature_extraction import DictVectorizer
from sklearn import preprocessing

from sklearn import tree

from sklearn.externals.six import StringIO

# Read in the csv file and put features into list of dict and list of class label
allElectronicsData = open('AllElectronics.csv', 'r')
reader = csv.reader(allElectronicsData)


featureList = []
labelList = []

with open('AllElectronics.csv', 'r') as csvfile:
    reader = csv.reader(csvfile)

    # 获得所有的头部属性
    headers = next(reader)

    for row in reader:
        # 将每一行的结果取出来
        labelList.append(row[len(row)-1])

        # 将对应的属性生成字典存起来, 第一列 是 RID, 不需要
        rowDict = {}
        for i in range(1, len(row)-1):
            rowDict[headers[i]] = row[i]
        featureList.append(rowDict)

print(featureList)

# Vetorize features
vec = DictVectorizer()
dummyX = vec.fit_transform(featureList).toarray()

# 按照 get_feature_names 获取的属性去生成对应的 矩阵
# ['age=middle_aged', 'age=senior', 'age=youth', 'credit_rating=excellent', 'credit_rating=fair', 'income=high', 'income=low', 'income=medium', 'student=no', 'student=yes']
print(vec.get_feature_names())
print(dummyX)

print("labelList: " + str(labelList))

# # vectorize class labels

# 将结果集
lb = preprocessing.LabelBinarizer()
dummyY = lb.fit_transform(labelList)
print("dummyY: " + str(dummyY))

# Using decision tree for classification
# clf = tree.DecisionTreeClassifier()

# 按照什么属性来设置对对应的算法
# 可以参考: http://scikit-learn.org/stable/modules/tree.html
clf = tree.DecisionTreeClassifier(criterion='entropy')
clf = clf.fit(dummyX, dummyY)
print("clf: " + str(clf))


# # Visualize model
with open("allElectronicInformationGainOri.dot", 'w') as f:
    f = tree.export_graphviz(clf, feature_names=vec.get_feature_names(), out_file=f)


# 修改一个值 来预测
oneRowX = dummyX[0, :]
print("oneRowX: " + str(oneRowX))

newRowX = oneRowX
newRowX[0] = 1
newRowX[2] = 0
print("newRowX: " + str(newRowX))

predictedY = clf.predict([newRowX])
print("predictedY: " + str(predictedY))