13-垃圾邮件分类2

1.读取

源代码：

#读取文件
file_path=r'D:PycharmProjects201706120186罗奕涛dataSMSSpamCollection'
sms=open(file_path,'r',encoding='utf-8')
sms_data=[]
sms_label=[]
csv_reader=csv.reader(sms,delimiter='	')
for line in csv_reader:
    sms_label.append(line[0])
    sms_data.append(preprocessing(line[1]))#对每封邮件做预处理
sms.close()

print(sms_label)
print(sms_data)

2.数据预处理

源代码：

import csv
import nltk
from nltk.corpus import stopwords
from nltk.stem import WordNetLemmatizer

print(nltk.__doc__)#输出版本号

def get_wordnet_pos(treebank_tag):#根据词性，生成还原参数pos
    if treebank_tag.startswith('J'):
        return nltk.corpus.wordnet.ADJ
    elif treebank_tag.startswith('V'):
        return nltk.corpus.wordnet.VERB
    elif treebank_tag.startswith('N'):
        return nltk.corpus.wordnet.NOUN
    elif treebank_tag.startswith('R'):
        return nltk.corpus.wordnet.ADV
    else:
        return nltk.corpus.wordnet.NOUN

#预处理
def preprocessing(text):
    tokens = [word for sent in nltk.sent_tokenize(text) for word in nltk.word_tokenize(sent)]#分词
    stops = stopwords.words("english")#停用词
    tokens = [token for token in tokens if token not in stops]#去掉停用词
    tokens = [token.lower() for token in tokens if len(token) >= 3]#将大写字母变为小写

    tag=nltk.pos_tag(tokens)#词性
    lmtzr = WordNetLemmatizer()
    tokens = [lmtzr.lemmatize(token,pos=get_wordnet_pos(tag[i][1])) for i,token in enumerate(tokens)]
    preprocessed_text = ''.join(tokens)
    return preprocessed_text

3.数据划分—训练集和测试集数据划分

from sklearn.model_selection import train_test_split

x_train,x_test, y_train, y_test = train_test_split(data, target, test_size=0.2, random_state=0, stratify=y_train)

源代码：

# 按0.8:0.2比例分为训练集和测试集
import numpy as np
from sklearn.model_selection import train_test_split

sms_data = np.array(sms_data)
sms_label = np.array(sms_label)
x_train, x_test, y_train, y_test = train_test_split(sms_data, sms_label, test_size=0.2, random_state=0,
                                                    stratify=sms_label)
print(len(sms_data),len(x_train),len(x_test))
print(x_train)

结果：

4.文本特征提取

sklearn.feature_extraction.text.CountVectorizer

https://scikit-learn.org/stable/modules/generated/sklearn.feature_extraction.text.CountVectorizer.html?highlight=sklearn%20feature_extraction%20text%20tfidfvectorizer

sklearn.feature_extraction.text.TfidfVectorizer

https://scikit-learn.org/stable/modules/generated/sklearn.feature_extraction.text.TfidfVectorizer.html?highlight=sklearn%20feature_extraction%20text%20tfidfvectorizer#sklearn.feature_extraction.text.TfidfVectorizer

from sklearn.feature_extraction.text import TfidfVectorizer

tfidf2 = TfidfVectorizer()

观察邮件与向量的关系

向量还原为邮件

源代码：

# 将其向量化
from sklearn.feature_extraction.text import TfidfVectorizer

vectorizer = TfidfVectorizer()
X_train = vectorizer.fit_transform(x_train)
X_test = vectorizer.transform(x_test)

print(X_train.toarray().shape)
print(X_test.toarray().shape)

结果：

4.模型选择

from sklearn.naive_bayes import MultinomialNB

说明为什么选择这个模型？

多项式朴素贝叶斯分类器适用于具有离散特征的分类(例如，用于文本分类的字数统计）

源代码：

from sklearn.naive_bayes import MultinomialNB

clf = MultinomialNB().fit(X_train, y_train)
y_nb_pred = clf.predict(X_test)
# x_test预测结果
print(y_nb_pred.shape,y_nb_pred)

结果：

5.模型评价：混淆矩阵，分类报告

from sklearn.metrics import confusion_matrix

confusion_matrix = confusion_matrix(y_test, y_predict)

说明混淆矩阵的含义

混淆矩阵是一个2 × 2的情形分析表，显示以下四组记录的数目：作出正确判断的肯定记录（真阳性）、作出错误判断的肯定记录（假阴性）、作出正确判断的否定记录（真阴性）以及作出错误判断的否定记录（假阳性）

from sklearn.metrics import classification_report

说明准确率、精确率、召回率、F值分别代表的意义

源代码：

from sklearn.metrics import confusion_matrix
from sklearn.metrics import classification_report

# 混淆矩阵
cm = confusion_matrix(y_test, y_nb_pred)
print('nb_confusion_matrix:')
print(cm)
# 主要分类指标的文本报告
cr = classification_report(y_test, y_nb_pred)
print('nb_classification_report:')
print(cr)

结果：

6.比较与总结

如果用CountVectorizer进行文本特征生成，与TfidfVectorizer相比，效果如何？

CountVectorizer只能转化英文的，不能转化中文的，因为是靠空格识别的。