Count Vectorizer和朴素贝叶斯用于情感分析，TypeError：NoneType对象不可迭代

Question

我是机器学习的新手，并且想到了对Twitter数据进行情感分析的想法。我正在使用Tweepy获取100条推文，这将是我的测试集，并且有两列（文本，标签）。而我的训练集是一个csv文件，包含3列（文本，标签，主题）。到目前为止，我已经为训练集和测试集创建了一个数据框，并希望使用朴素贝叶斯分类器，但是有点卡在需要应用Count Vectorizer和Tfidf的地方。谢谢。

import pandas as pd
import twitter
import re
from nltk.tokenize import word_tokenize
from string import punctuation
from nltk.corpus import stopwords
import nltk
import string
import traceback
import pandas as pd

# initialize api instance
twitter_api = twitter.Api(consumer_key='123456789',
                          consumer_secret='123456789',
                          access_token_key='123456789',
                          access_token_secret='123456789')

# test authentication
print(twitter_api.VerifyCredentials())


def buildTestSet(search_keyword):
    try:
        tweets_fetched = twitter_api.GetSearch(search_keyword, count=100)
        print("Fetched " + str(len(tweets_fetched)) + " tweets for the term " + search_keyword)
        data = [{"text": status.text, "label": None} for status in tweets_fetched]
        df = pd.DataFrame(data)
        print(df)
        return df
    except Exception:
        print("Unfortunately, something went wrong..")
        traceback.print_exc()

def buildTrainSet(fileName):
    try:
        df2 = pd.read_csv(fileName, encoding='unicode_escape')
        print(df2)
        return df2
    except Exception:
        traceback.print_exc()

def cleaningSet(words):
    for text in words.iterrows() :
        lowerSet = text.__str__().lower()
        cleaned_text = lowerSet.__str__().translate(str.maketrans('', '', string.punctuation))
        cleaned_text = re.sub('((www\.[^\s]+)|(https?://[^\s]+))', 'URL', cleaned_text)  # remove URLs
        cleaned_text = re.sub('@[^\s]+', 'AT_USER', cleaned_text)  # remove usernames
        cleaned_text = re.sub(r'#([^\s]+)', r'\1', cleaned_text)  # remove the # in #hashtag

search_term = input("Enter a search keyword: ")
testDataSet= buildTestSet(search_term)
trainDataSet = buildTrainSet("TweetDataFile.csv")
testDataSet_clean = cleaningSet(testDataSet)
trainDataSet_clean = cleaningSet(trainDataSet)

categories = ['positive', 'negative', 'neutral']

try :
    from sklearn.feature_extraction.text import CountVectorizer
    count_vector = CountVectorizer()
    x_train_tf = count_vector.fit_transform(trainDataSet_clean)
    print(x_train_tf.shape)

    from sklearn.feature_extraction.text import TfidfTransformer
    tfidf_transformer = TfidfTransformer()
    x_train_tfidf = tfidf_transformer.fit_transform(x_train_tf)
    print(x_train_tfidf.shape)


    from sklearn.naive_bayes import MultinomialNB
    clf = MultinomialNB.fit(x_train_tfidf, categories)
    x_test_tf = count_vector.transform(testDataSet_clean)
    x_test_tfidf = tfidf_transformer.transform(x_test_tf)
    predicted = clf.predict(x_test_tfidf)
except Exception:
    traceback.print_exc()

编辑1：我对变量_testDataSet_clean_和_trainDataSet_clean_进行了注释，程序可以运行直到分类器再次给出另一个错误为止（参见图2）。

形状也被打印为（3x3）。

Answer 1

您可以看到错误消息。它说trainDataSet_clean是NoneType。我认为您在进行上述处理后，您的trainDataSet_clean就变成了NoneType，所以我可以看到您的trainDataSet_clean还是完整的代码？