情感分析

对一句话进行分析，给出其是正向还是负面。

`from sklearn.naive_bayes import MultinomialNB
from snownlp import SnowNLP
import pandas as pd
import jieba
from sklearn.feature_extraction.text import TfidfVectorizer

def a1():
a = “环境优美” # 一条正向句子
_b = “垃圾遍地” # 一条负面句子
_m = SnowNLP(a)
n = SnowNLP(b)
print(“正面：”,m.sentiments)
print(“负面”,n.sentiments)

def a2():
# 读取文件
_data = pd.read_csv(“酒店评论.csv”)
remarks = data[“review”].values.tolist() # 评论
labels = data[“label”].values.tolist() # 标签
content = []
for line in remarks:
line = str(line)
line = line.strip(“\r\t\n”)
temp = jieba.lcut(line)
content.append(“ “.join(temp))
# print(content)

_tfidf = TfidfVectorizer()<br />    tfidf.fit(content)<br />    train = tfidf.transform(content)
_# 朴素贝叶斯<br />    _model = MultinomialNB()<br />    _# 训练模型<br />    _model.fit(train,labels)
_# 使用模型进行测试<br />    _test = "垃圾"<br />    test = jieba.lcut(test)<br />    test = tfidf.transform(test)<br />    _# 识别情感<br />    _rs = model.predict(test)<br />    em = "正面"<br />    if rs == 0:<br />        em = "负面"<br />    print("测试语句：",test,"的情感预测为：",em)
_# 38 - 46行代码只能针对test是单个词进行分析<br />    # 现在对其优化，使其能够真正对一句话进行分析<br />    _testNew = "真他妈垃圾"<br />    testNew = jieba.lcut(testNew)<br />    print("testNew = ",testNew)<br />    testNew = tfidf.transform(testNew)<br />    rs = model.predict(testNew)<br />    print("rs = ",rs)<br />    emnum = sum(rs)<br />    print("emnum = ",emnum)<br />    em = "正面"<br />    if emnum <= len(rs)/2:<br />        em = "负面"<br />    print("测试语句：", testNew, "的情感预测为：", em)`<br />![image.png](https://cdn.nlark.com/yuque/0/2022/png/26766835/1653056692795-525d03e3-bad5-442f-a8e9-cc9846bce102.png#clientId=u12aee0c2-38ef-4&crop=0&crop=0&crop=1&crop=1&from=paste&height=282&id=ud0fc4b1f&margin=%5Bobject%20Object%5D&name=image.png&originHeight=282&originWidth=718&originalType=binary&ratio=1&rotation=0&showTitle=false&size=19181&status=done&style=none&taskId=u927ea659-a8a7-4ddc-ba83-4804320d42a&title=&width=718)

训练数据：酒店评论.csv