Python中的自然语言处理实例：情感分析

pip install nltkpip install TwitterAPI

import refrom nltk.corpus import stopwordsdef clean_text(text):    text = text.lower() # 将文本转换成小写字母    text = re.sub(r'[^ws]', '', text) # 去除标点符号    text = re.sub(r'd+', '', text) # 去除数字    stop_words = set(stopwords.words('english'))    words = text.split()    words = [w for w in words if w not in stop_words] # 去除停用词    text = ' '.join(words)    return text

import nltkfrom nltk.corpus import movie_reviewsimport randomdocuments = [(list(movie_reviews.words(fileid)), category)             for category in movie_reviews.categories()             for fileid in movie_reviews.fileids(category)]random.shuffle(documents)

all_words = nltk.FreqDist(w.lower() for w in movie_reviews.words())word_features = list(all_words.keys())[:2000]def document_features(document):    document_words = set(document)    features = {}    for word in word_features:        features['contains({})'.format(word)] = (word in document_words)    return featuresfeaturesets = [(document_features(d), c) for (d,c) in documents]train_set, test_set = featuresets[200:], featuresets[:200]classifier = nltk.NaiveBayesClassifier.train(train_set)

from TwitterAPI import TwitterAPIimport jsonconsumer_key = 'your consumer key'consumer_secret = 'your consumer secret'access_token_key = 'your access token key'access_token_secret = 'your access token secret'api = TwitterAPI(consumer_key,                 consumer_secret,                 access_token_key,                 access_token_secret)def analyze_tweet(tweet):    tweet_text = tweet['text']    tweet_clean = clean_text(tweet_text)    tweet_features = document_features(tweet_clean.split())    sentiment = classifier.classify(tweet_features)    return sentimentkeywords = 'Trump'for tweet in api.request('search/tweets', {'q': keywords, 'lang': 'en', 'count': 10}):    sentiment = analyze_tweet(tweet)    print(tweet['text'])    print(sentiment)    print('')