[data mining] task 3: decision tree classification

requirement

requirement ： Weather factors include temperature 、 Humidity and wind , By giving data , Use decision tree algorithm to learn classification , Output a rule tree between exercise and inactivity and weather .

Training sets and test sets can be defined freely , In addition, temperature and humidity need to be generalized , Change the value into a general statement , For example, the temperature is hot , temperature , cool , Humidity becomes high , in .

from sklearn import tree
from sklearn.model_selection import train_test_split
import pandas as pd
import graphviz
import numpy as np

Data preprocessing

data fetch

df = pd.read_excel('data.xlsx', index_col=None)
df

	The weather	temperature	humidity	The wind	motion
0	Fine	85	85	nothing	Not suitable for
1	Fine	80	90	Yes	Not suitable for
2	cloudy	83	78	nothing	fit
3	There is rain	70	96	nothing	fit
4	There is rain	68	80	nothing	fit
5	There is rain	65	70	Yes	Not suitable for
6	cloudy	64	65	Yes	fit
7	Fine	72	95	nothing	Not suitable for
8	Fine	69	70	nothing	fit
9	There is rain	75	80	nothing	fit
10	Fine	75	70	Yes	fit
11	cloudy	72	90	Yes	fit
12	cloudy	81	75	nothing	fit
13	There is rain	71	80	Yes	Not suitable for

Quantitative text index

For the subsequent calculation of decision tree , We need to quantify the text index , Here's the conversion ：

The weather —— Fine -0, multivariate -1, There is rain -2

The wind —— nothing -0, Yes -1

motion —— Not suitable for -0, fit -1

df[' The weather '] = df[' The weather '].replace(" Fine ", 0)
df[' The weather '] = df[' The weather '].replace(" cloudy ", 1)
df[' The weather '] = df[' The weather '].replace(" There is rain ", 2)
df[' The wind '] = df[' The wind '].replace(" nothing ", 0)
df[' The wind '] = df[' The wind '].replace(" Yes ", 1)
df[' motion '] = df[' motion '].replace(" Not suitable for ", 0)
df[' motion '] = df[' motion '].replace(" fit ", 1)

Temperature and humidity generalization

Subject requirements , Change the temperature and humidity value into a general statement . Here, the temperature and humidity are summarized and converted into numerical values , The specific rules are as follows ：

temperature ：<70- cool -0,70~80- temperature -1,>80- heat -2

humidity ：>80- high -1,<=80- in -0

df[' temperature '] = np.where(df[' temperature '] < 70, 0, df[' temperature '])
df[' temperature '] = np.where((df[' temperature '] < 80) & (df[' temperature '] >= 70), 1, df[' temperature '])
df[' temperature '] = np.where(df[' temperature '] >= 80, 2, df[' temperature '])
df[' humidity '] = np.where(df[' humidity '] > 80, 1, 0)

The converted data is shown in the following table ：

df

	The weather	temperature	humidity	The wind	motion
0	0	2	1	0	0
1	0	2	1	1	0
2	1	2	0	0	1
3	2	1	1	0	1
4	2	0	0	0	1
5	2	0	0	1	0
6	1	0	0	1	1
7	0	1	1	0	0
8	0	0	0	0	1
9	2	1	0	0	1
10	0	1	0	1	1
11	1	1	1	1	1
12	1	2	0	0	1
13	2	1	0	1	0

Data set partitioning

according to 7/3 The training set and test set are divided in proportion

data = df[[' The weather ', ' temperature ', ' humidity ', ' The wind ']]
target = df[' motion ']
data = np.array(data)
target = np.array(target)
Xtrain, Xtest, Ytrain, Ytest = train_test_split(data, target, test_size=0.3)

Decision tree construction

Here, the standard of decision tree is Gini index , Finally, the classification accuracy is 60%

clf = tree.DecisionTreeClassifier(criterion="gini")
clf = clf.fit(Xtrain, Ytrain)
score = clf.score(Xtest, Ytest)
print(score)

0.6

Visualization results

feature_name = [' The weather ', ' temperature ', ' humidity ', ' The wind ']
dot_data = tree.export_graphviz(clf, feature_names=feature_name, class_names=[" Not suitable for ", " fit "], filled=True, rounded=True
                                )
graph = graphviz.Source(dot_data.replace(
    'helvetica', '"Microsoft YaHei"'), encoding='utf-8')
graph.view()