優點:計算複雜度不高,輸出結果易於理解,對中間再缺失不敏感,可以處理不相關特性。
缺點:可能產生過渡匹配問題
使用類型:數字型和標稱型
一,基本概念
1,信息熵
度量樣本集合純度的最常用指標之一。值越小,則純度越高。
公式爲:
其中,Pk爲第k類佔總數的比例。
例如,有10個樣本,4個爲好,6個爲壞,則信息熵爲:Ent(D)=-(4/10*log2(4/10)+6/10*log2(6/10))
2,信息增益
用於衡量分支影響力大小。信息增益越大,則影響力越大。ID3決策樹就是採用信息增益劃分屬性。
公式爲:
其中,D爲分支下的總數,Dv爲佔總數比例。
例如:數據如下:
1,青綠,蜷縮,濁響,清晰,凹陷,硬滑,是
2,烏黑,蜷縮,沉悶,清晰,凹陷,硬滑,是
3,烏黑,蜷縮,濁響,清晰,凹陷,硬滑,是
4,青綠,蜷縮,沉悶,清晰,凹陷,硬滑,是
5,淺白,蜷縮,濁響,清晰,凹陷,硬滑,是
6,青綠,稍蜷,濁響,清晰,稍凹,軟粘,是
7,烏黑,稍蜷,濁響,稍糊,稍凹,軟粘,是
8,烏黑,稍蜷,濁響,清晰,稍凹,硬滑,是
9,烏黑,稍蜷,沉悶,稍糊,稍凹,硬滑,否
10,青綠,硬挺,清脆,清晰,平坦,軟粘,否
11,淺白,硬挺,清脆,模糊,平坦,硬滑,否
12,淺白,蜷縮,濁響,模糊,平坦,軟粘,否
13,青綠,稍蜷,濁響,稍糊,凹陷,硬滑,否
14,淺白,稍蜷,沉悶,稍糊,凹陷,硬滑,否
15,烏黑,稍蜷,濁響,清晰,稍凹,軟粘,否
16,淺白,蜷縮,濁響,模糊,平坦,硬滑,否
17,青綠,蜷縮,沉悶,稍糊,稍凹,硬滑,否
信息熵有最後一列計算。ENT(D)=-((8/17)*log2(8/17)+(9/17)*log2(9/17))=0.998
第二列分類可知,青綠爲編號{1,4,6,10,13,16},則信息熵Ent(D1)=-((3/6)*log2(3/6)+(3/6)*log2(3/6))=1
烏黑和淺白計算類似信息熵爲0.918和0.722
則最終信息增益爲:Gain(D,色澤)=0.998-(6/17*1+6/17*0.918+5/17*0.722)=0.109
3,信息增益率
信息增益對屬性多有偏好,信息增益對屬性少的有偏好。是ID4.5決策樹算法。
公式如下:
4,基尼指數
基尼指數䦹一種衡量數據集純度指標。基尼指數越小,純度越高。
公式如下:
python代碼:
新建兩個Python文件:tree.py,用於決策樹分類;treePlotter.py用於繪製圖形
tree.py代碼:
from math import log
import operator
import treePlotter as tp
def createDataSet():
dataSet=[]
fr = open('watermelon1.txt')
for line in fr.readlines():
lineArr = line.strip().split(',')
dataSet.append(lineArr[:]) # 添加數據
labels = ['編號','色澤','根蒂','敲聲','紋理','頭部','觸感','好瓜']
return dataSet, labels
#計算信息熵 Ent(D)=-Σp*log2(p)
def calcShannonEnt(dataSet):
numEntries = len(dataSet) #數據總數
labelCounts = {}
for featVec in dataSet:
currentLabel = featVec[-1] #獲取類別
if currentLabel not in labelCounts.keys(): labelCounts[currentLabel] = 0 #新key加入字典賦值爲0
labelCounts[currentLabel] += 1 #已經存在的key,value+=1
shannonEnt = 0.0
for key in labelCounts:
prob = float(labelCounts[key])/numEntries
shannonEnt -= prob * log(prob,2) #計算信息熵
return shannonEnt
#獲取特徵值數據集
# dataSet --整個數據集
# axis --數據列
# value --類別
def splitSubDataSet(dataSet, axis, value):
retDataSet = []
for featVec in dataSet:
if featVec[axis] == value:
retDataSet.append([featVec[axis],featVec[-1]])
return retDataSet
#除去劃分完成的決策樹數據量
def splitDataSet(dataSet, axis, value):
retDataSet = []
for featVec in dataSet:
if featVec[axis] == value:
reducedFeatVec = featVec[:axis]
reducedFeatVec.extend(featVec[axis+1:])
retDataSet.append(reducedFeatVec)
return retDataSet
# 計算連續變量的分類點
# def calcconplot(subDataSet)
# 計算信息增益並返回信息增益最高的列
def chooseBestFeatureToSplit(dataSet):
numFeatures = len(dataSet[0]) - 1 #獲取所有特徵值數量(減1是除去最後一列分類)
baseEntropy = calcShannonEnt(dataSet) #計算基礎信息熵Ent(D)
bestInfoGain = 0.0; bestFeature = []
for i in range(1,numFeatures): #遍歷所有特徵值
featList = [example[i] for example in dataSet]#將特徵值保存在列表中
uniqueVals = set(featList) #獲取特徵值分類
newEntropy = 0.0 #特徵值不連續
for value in uniqueVals:
subDataSet = splitSubDataSet(dataSet, i, value)
prob = len(subDataSet)/float(len(dataSet))
newEntropy += prob * calcShannonEnt(subDataSet)
infoGain = baseEntropy - newEntropy #計算信息增益
if (infoGain > bestInfoGain): #保存信息增益最高的列
bestInfoGain = infoGain
bestFeature = i
return bestFeature #返回新增增益最高的列
def majorityCnt(classList):
classCount={}
for vote in classList:
if vote not in classCount.keys(): classCount[vote] = 0
classCount[vote] += 1
sortedClassCount = sorted(classCount.iteritems(), key=operator.itemgetter(1), reverse=True)
return sortedClassCount[0][0]
# 創建決策樹
def createTree(dataSet,labels):
classList = [example[-1] for example in dataSet]
if classList.count(classList[0]) == len(classList):
return classList[0]#當所有類都相同則不在分類
if len(dataSet[0]) == 1: #沒有更多特徵值時不再分類
return majorityCnt(classList)
bestFeat = chooseBestFeatureToSplit(dataSet) #選取信息增益最大的特徵值
bestFeatLabel = labels[bestFeat] #獲取特徵值列頭名
myTree = {bestFeatLabel:{}}
featValues = [example[bestFeat] for example in dataSet]
uniqueVals = set(featValues) # 獲取特徵值分類
del(labels[bestFeat]) # 刪除已經建立節點的特徵值
for value in uniqueVals:
subLabels = labels[:] # 複製出建立節點外的所有特徵值
myTree[bestFeatLabel][value] = createTree(splitDataSet(dataSet, bestFeat, value),subLabels) #建立子節點
return myTree
if __name__ == '__main__':
myData,label = createDataSet()
mytree = createTree(myData,label)
tp.createPlot(mytree)
treePlotter.py代碼:
import matplotlib.pyplot as plt
decisionNode = dict(boxstyle="sawtooth", fc="0.8") # 文本框圖形
leafNode = dict(boxstyle="round4", fc="0.8") # 線圖形
arrow_args = dict(arrowstyle="<-") # 箭頭圖形
# 獲取葉子數目
def getNumLeafs(myTree):
numLeafs = 0
firstStr = list(myTree.keys())[0] # 獲取第一個節點名
secondDict = myTree[firstStr] # 剩餘節點
for key in secondDict.keys():
if type(secondDict[key]).__name__=='dict':# 如果是節點繼續查詢
numLeafs += getNumLeafs(secondDict[key])
else: numLeafs +=1 #如果是葉子則累加
return numLeafs
# 獲取樹深度
def getTreeDepth(myTree):
maxDepth = 0
firstStr = list(myTree.keys())[0]
secondDict = myTree[firstStr]
for key in secondDict.keys():
if type(secondDict[key]).__name__=='dict': #如果是節點則深度加1
thisDepth = 1 + getTreeDepth(secondDict[key])
else: thisDepth = 1
if thisDepth > maxDepth: maxDepth = thisDepth
return maxDepth
def plotNode(nodeTxt, centerPt, parentPt, nodeType):
createPlot.ax1.annotate(nodeTxt, xy=parentPt, xycoords='axes fraction',
xytext=centerPt, textcoords='axes fraction',
va="center", ha="center", bbox=nodeType, arrowprops=arrow_args )
# 決策樹繪製
def createPlot(inTree):
fig = plt.figure(1, facecolor='white')
fig.clf()
axprops = dict(xticks=[], yticks=[])
createPlot.ax1 = plt.subplot(111, frameon=False, **axprops)
plotTree.totalW = float(getNumLeafs(inTree))
plotTree.totalD = float(getTreeDepth(inTree))
plotTree.xOff = -0.5/plotTree.totalW; plotTree.yOff = 1.0
plotTree(inTree, (0.5,1.0), '')
plt.show()
# 繪製節點文字
def plotMidText(cntrPt, parentPt, txtString):
xMid = (parentPt[0]-cntrPt[0])/2.0 + cntrPt[0]
yMid = (parentPt[1]-cntrPt[1])/2.0 + cntrPt[1]
createPlot.ax1.text(xMid, yMid, txtString, va="center", ha="center", rotation=30)
#計算節點位置
def plotTree(myTree, parentPt, nodeTxt):
numLeafs = getNumLeafs(myTree)
depth = getTreeDepth(myTree)
firstStr = list(myTree.keys())[0]
cntrPt = (plotTree.xOff + (1.0 + float(numLeafs))/2.0/plotTree.totalW, plotTree.yOff)
plotMidText(cntrPt, parentPt, nodeTxt)
plotNode(firstStr, cntrPt, parentPt, decisionNode)
secondDict = myTree[firstStr]
plotTree.yOff = plotTree.yOff - 1.0/plotTree.totalD
for key in secondDict.keys():
if type(secondDict[key]).__name__=='dict':
plotTree(secondDict[key],cntrPt,str(key))
else:
plotTree.xOff = plotTree.xOff + 1.0/plotTree.totalW
plotNode(secondDict[key], (plotTree.xOff, plotTree.yOff), cntrPt, leafNode)
plotMidText((plotTree.xOff, plotTree.yOff), cntrPt, str(key))
plotTree.yOff = plotTree.yOff + 1.0/plotTree.totalD
結果如下圖:
接下來,我們給出測試數據如下:
1,青綠,蜷縮,濁響,清晰,凹陷,軟粘
2,烏黑,稍蜷,沉悶,清晰,凹陷,硬滑
3,青綠,蜷縮,濁響,稍糊,平坦,硬滑
4,青綠,稍蜷,沉悶,清晰,凹陷,硬滑
5,淺白,蜷縮,濁響,稍糊,凹陷,硬滑
6,青綠,稍蜷,濁響,清晰,平坦,軟粘
7,烏黑,稍蜷,濁響,稍糊,稍凹,軟粘
8,青綠,稍蜷,濁響,清晰,稍凹,硬滑
9,烏黑,稍蜷,沉悶,稍糊,稍凹,硬滑
10,青綠,硬挺,濁響,清晰,平坦,軟粘
11,淺白,硬挺,清脆,模糊,平坦,硬滑
12,淺白,蜷縮,濁響,模糊,平坦,軟粘
13,青綠,稍蜷,濁響,稍糊,凹陷,硬滑
14,淺白,稍蜷,沉悶,稍糊,凹陷,硬滑
15,烏黑,稍蜷,濁響,清晰,稍凹,軟粘
16,淺白,蜷縮,濁響,模糊,平坦,硬滑
17,青綠,蜷縮,濁響,稍糊,稍凹,軟粘
然後在tree.py新增:
# 決策樹進行分類
def classify(inputTree,featLabels,testVec):
firstStr = list(inputTree.keys())[0]
secondDict = inputTree[firstStr]
featIndex = featLabels.index(firstStr)
key = testVec[featIndex]
valueOfFeat = secondDict[key]
if isinstance(valueOfFeat, dict):
classLabel = classify(valueOfFeat, featLabels, testVec)
else: classLabel = valueOfFeat
return classLabel
# 讀取測試數據
def createtestDataSet():
dataSet=[]
fr = open('testData.txt')
for line in fr.readlines():
lineArr = line.strip().split(',')
dataSet.append(lineArr[:]) # 添加數據
labels = ['編號','色澤','根蒂','敲聲','紋理','頭部','觸感']
return dataSet, labels
if __name__ == '__main__':
myData,label = createDataSet()
mytree = createTree(myData,label)
tp.createPlot(mytree)
testData,testlabel = createtestDataSet()
for data in testData:
cla = classify(mytree,testlabel,data)
print(data)
print(cla)
決策樹分類得到如下結果:
['1', '青綠', '蜷縮', '濁響', '清晰', '凹陷', '軟粘']
是
['2', '烏黑', '稍蜷', '沉悶', '清晰', '凹陷', '硬滑']
是
['3', '青綠', '蜷縮', '濁響', '稍糊', '平坦', '硬滑']
否
['4', '青綠', '稍蜷', '沉悶', '清晰', '凹陷', '硬滑']
是
['5', '淺白', '蜷縮', '濁響', '稍糊', '凹陷', '硬滑']
否
['6', '青綠', '稍蜷', '濁響', '清晰', '平坦', '軟粘']
是
['7', '烏黑', '稍蜷', '濁響', '稍糊', '稍凹', '軟粘']
是
['8', '青綠', '稍蜷', '濁響', '清晰', '稍凹', '硬滑']
是
['9', '烏黑', '稍蜷', '沉悶', '稍糊', '稍凹', '硬滑']
否
['10', '青綠', '硬挺', '濁響', '清晰', '平坦', '軟粘']
否
['11', '淺白', '硬挺', '清脆', '模糊', '平坦', '硬滑']
否
['12', '淺白', '蜷縮', '濁響', '模糊', '平坦', '軟粘']
否
['13', '青綠', '稍蜷', '濁響', '稍糊', '凹陷', '硬滑']
否
['14', '淺白', '稍蜷', '沉悶', '稍糊', '凹陷', '硬滑']
否
['15', '烏黑', '稍蜷', '濁響', '清晰', '稍凹', '軟粘']
否
['16', '淺白', '蜷縮', '濁響', '模糊', '平坦', '硬滑']
否
['17', '青綠', '蜷縮', '濁響', '稍糊', '稍凹', '軟粘']
是