大作业-中文文本分类

import os
import numpy as np
import sys
from datetime import datetime
import gc
path  =r'C:\Users\Administrator\Desktop\0369'
path
import jieba
with open(r'C:\Users\Administrator\Desktop\stopsCN.txt', encoding='utf-8') as f:
    stops = f.read().split('\n')
def processing (tokens):
    #去掉停用词
    tokens = [token for token in tokens if token not in stops]
    #去掉非字母汉字的字符
    tokens = "".join([token for token in tokens if token.isalpha()])
    #结巴分词，去掉小于2的词
    tokens =" ".join([token for token in jieba.cut(tokens,cut_all=True) if len(token)>=2])
    
    return tokens
conList = []
targetList = []

for root,dirs,files in os.walk(path): # 用os.walk获取需要的变量，并拼接文件路径再打开每一个文件
#     print(root)
#     print(dirs)
#     print(files)
    for filetext in files:
        Filetext = os.path.join(root,filetext)
        size = os.path.getsize(Filetext)
        print(Filetext,size)
        with open(Filetext,'r',encoding='utf-8') as f:
            content = f.read() # 获取文本
        target = Filetext.split('\\')[-2] # 获取类别标签
        targetList.append(target)
        conList.append(processing(content))
targetList[0:10]
conList[0:10]
#### 划分训练集测试集
from sklearn.feature_extraction.text import TfidfVectorizer  #Feature extraction文本特征提取
from sklearn.model_selection import train_test_split
from sklearn.naive_bayes import GaussianNB,MultinomialNB
from sklearn.model_selection import cross_val_score
from sklearn.metrics import classification_report
x_train,x_test,y_train,y_test  = train_test_split(conList,targetList,test_size=0.2,stratify=targetList)

vector = TfidfVectorizer()
X_train = vector.fit_transform(x_train)
X_test = vector.transform(x_test)
# 建立模型,这里用多项式朴素贝叶斯，因为样本特征的a分布大部分是多元离散值
bys = MultinomialNB()
MX = bys.fit(X_train,y_train)
#预测
y_predict = MX.predict(X_test)

scores = cross_val_score(bys,X_test,y_test,cv=3)
print("Accuracy:%.3f"%scores.mean())
print("classification_report:\n",classification_report(y_predict,y_test))

　　

 

2、boston房价预测

from sklearn.datasets import load_boston
from sklearn.model_selection import train_test_split
data = load_boston()
data.keys()
#划分数据集
x_train,x_test,y_train,y_test = train_test_split(data.data,data.target,test_size=0.3)
#建立多元线性回归模型
from sklearn.linear_model import LinearRegression
mlr = LinearRegression()
mlr.fit(x_train,y_train) #训练模型
print('系数',mlr.coef_,"\n截距",mlr.intercept_)
#判断模型的好坏
from sklearn.metrics import regression
y_predict = mlr.predict(x_test) #预测
print("预测的均方误差：", regression.mean_squared_error(y_test,y_predict))
print("预测的平均绝对误差：", regression.mean_absolute_error(y_test,y_predict))
# 模型的分数
print("多元线性回归模型的分数：",mlr.score(x_test, y_test))
#多元多项式回归模型
from sklearn.preprocessing import PolynomialFeatures
poly = PolynomialFeatures(degree=2)
x_poly_train = poly.fit_transform(x_train) #创建多项式训练集
x_poly_test = poly.transform(x_test) #创建多项式数据集
#建立模型
mlrp = LinearRegression()
mlrp.fit(x_poly_train,y_train)
y_predict1 = mlrp.predict(x_poly_test)

print("预测的均方误差：", regression.mean_squared_error(y_test,y_predict1))
print("预测的平均绝对误差：", regression.mean_absolute_error(y_test,y_predict1))
# 打印模型的分数
print("多项式回归模型的分数：",mlrp.score(x_poly_test, y_test))

　　结果：