对发帖情感进行分析。
字段说明: Announce_ID字段代表用户ID,User_Name字段代表用户名,topic字段代表发帖主题,body字段代表发帖内容,post_type字段代表发帖话题是否与工作相关,sentiment字段表明发帖情感色彩,IP字段代表用户IP地址。
关于classify.xls文件 提取码:fkwq
stopwords.txt 文件 提取码:05z1
相关代码
import pandas as pd
import jieba
from sklearn.feature_extraction.text import CountVectorizer
from sklearn.decomposition import TruncatedSVD
from sklearn import model_selection, metrics
from sklearn.naive_bayes import BernoulliNB
# 读入classify文件,Sheetname设为classify
df = pd.read_excel('classify.xls', sheet_name='classify')
# print(df)
# 对post_type特征做频数统计
# print(df['post_type'].value_counts())
# 删去User_Name、post_type、IP特征
df.drop(columns=['User_Name', 'post_type', 'IP'], inplace=True)
# print(df)
# 对Body特征进行中文分词处理
df['Body'] = df['Body'].map(lambda x:jieba.lcut(x))
# print(df['Body'].head())
# 读取停用词表,剔除停用词
s = ''
with open('stopwords.txt', 'r', encoding='utf8') as r:
for i in r.readlines():
s += i.strip()
df['Body'] = df['Body'].map(lambda x:[i for i in x if(i not in s) and (len(i) > 1)])
# print(df['Body'].head())
# 将剔除停用词后的Body特征转换为矩阵向量
list1 = df['Body'].map(lambda line:' '.join(line))
cv = CountVectorizer()
w = cv.fit_transform(list1)
# 降维处理得到特征X
svd = TruncatedSVD(50)
X = svd.fit_transform(w)
# 切片sentiment特征作为目标标签Y
Y = df['sentiment']
# 按照8:2划分数据集
train_x, test_x, train_y, test_y = model_selection.train_test_split(X, Y, test_size=0.2, random_state=0)
# 建立朴素贝叶斯分类模型
model = BernoulliNB()
model.fit(train_x, train_y)
# 进行模型训练和预测
h = model.predict(test_x)
# 打印混淆矩阵和分类报告
print('classification_report:\n', metrics.classification_report(test_y, h))
print('confusion_matrix:\n', metrics.confusion_matrix(test_y, h))