Question

我试图在python中使用TextBlob在推文上实现朴素贝叶斯分类器。我已经能够训练数据集，并且可以使用以下方法成功地对单个推文进行分类：

print cl.classify("text")

现在我想打开一个csv文件并对该文件中的所有推文进行分类。有关如何实现这一目标的任何建议？我的代码如下：

import csv
from textblob import TextBlob

with open(test_path, 'rU') as csvfile:
    lineReader = csv.reader(csvfile,delimiter=',',quotechar="\"")
    lineReader = csv.reader(csvfile,delimiter=',')

    test = []
    for row in lineReader:
      blob = (row[0]) 
      blob = TextBlob(blob)
      test.append([blob])

      print (test.classify())

AttributeError：＆＃39; list＆＃39;对象没有属性＆＃39;分类＆＃39;

Answer 1

你也需要先训练（不清楚你是否做过这个？），

train = [] 
# then repeat your above lines, appending each tweet to train set
# but for a separate training set (or slice up the rows)

# do your test append loop -----

# 1. Now train a model
my_classifier = NaiveBayesClassifier(train)

# 2. test given to the model to get accuracy
accu = my_classifier.accuracy(test)

TextBlob朴素贝叶斯文本分类

1 个答案: