def sentiment(comments): if not os.path.isfile('classifier.pickle'): tcl.training() fl = open('classifier.pickle', 'rb') classifier = pickle.load(fl) fl.close() pos = 0 neg = 0 for words in comments: comment = features(words) sentiment_value, confidence = VoteClassifier(classifier).classify( comment) if sentiment_value == 'positive': # and confidence * 100 >= 60: pos += 1 else: neg += 1 print("Positive sentiment : ", (pos * 100.0 / len(comments)), "%") print("Negative sentiment : ", (neg * 100.0 / len(comments)), "%") a = (pos * 100.0 / len(comments)) b = (neg * 100.0 / len(comments)) sys.stdout.flush() import matplotlib.pyplot as plt plt.figure(figsize=[5, 5]) labels = ["positive", "negative"] values = [a, b] plt.pie(values, labels=labels) plt.show()
def sentiment(comments): if not os.path.isfile('classifier.pickle'): tcl.training() fl = open('classifier.pickle', 'rb') classifier = pickle.load(fl) fl.close() pos = 0 neg = 0 for words in comments: comment = features(words) sentiment_value, confidence = VoteClassifier(classifier).classify( comment) if sentiment_value == 'positive': # and confidence * 100 >= 60: pos += 1 userpositive = open("userpositive.txt", "a", encoding="utf-8") x = words.split(":", 1) userpositive.write(x[0] + "\n") userpositive.close() else: neg += 1 usernegative = open("usernegative.txt", "a", encoding="utf-8") y = words.split(":", 1) usernegative.write(y[0] + "\n") usernegative.close() print("Positive sentiment : ", (pos * 100.0 / len(comments))) print("Negative sentiment : ", (neg * 100.0 / len(comments)))
def sentiment(comments): if not os.path.isfile('/home/faizollah/mysite/classifier.pickle'): tcl.training() fl = open('/home/faizollah/mysite/classifier.pickle', 'rb') classifier = pickle.load(fl) fl.close() pos = 0 neg = 0 for words in comments: comment = features(words) sentiment_value, confidence = VoteClassifier(classifier).classify( comment) if sentiment_value == 'positive': # and confidence * 100 >= 60: pos += 1 else: neg += 1 #print ("Positive sentiment : ", (pos * 100.0 /len(comments)) ) #print ("Negative sentiment : ", (neg * 100.0 /len(comments)) ) psent = (pos * 100.0 / len(comments)) nsent = (neg * 100.0 / len(comments)) return psent, nsent
def sentiment(comments): if not os.path.isfile('classifier.pickle'): tcl.training() fl = open('classifier.pickle','rb') classifier = pickle.load(fl) fl.close() pos = 0 neg = 0 for words in comments: comment = features(words) sentiment_value, confidence = VoteClassifier(classifier).classify(comment) df1['Sentiment']=sentiment_value if sentiment_value == 'positive':# and confidence * 100 >= 60: pos += 1 else: neg += 1
def sentiment(comments, language): if language == 0 and not os.path.isfile('chinese_classifier.pickle'): tcl.chinese_training() if language == 1 and not os.path.isfile('naive_bayes.pickle'): tcl.training() if language == 1: fl = open('naive_bayes.pickle', 'rb') else: fl = open('chinese_classifier.pickle', 'rb') naive_bayes_classifier = pickle.load(fl) fl.close() pos = 0 neg = 0 count = 0 for words in comments: count += 1 if language == 1: comment = features(words) else: comment = chinese_features(words) voted_classifier = VoteClassifier(naive_bayes_classifier) sentiment_value, confidence = voted_classifier.classify(comment) #if count <= 20: #print(count, sentiment_value, confidence) if sentiment_value == 'positive': # and confidence * 100 >= 60: pos += 1 else: neg += 1 output = [] output.append("Positive sentiment : " + str(int(pos * 100.0 / len(comments)))) output.append("Negative sentiment : " + str(int(neg * 100.0 / len(comments)))) return output
def sentiment(comments): if not os.path.isfile('classifier.pickle'): tcl.training() fl = open('classifier.pickle', 'rb') classifier = pickle.load(fl) fl.close() pos = 0 neg = 0 for words in comments: # print(words) comment = features(words) sentiment_value, confidence = VoteClassifier(classifier).classify( comment) if sentiment_value == 'positive': # and confidence * 100 >= 60: pos += 1 else: neg += 1 print("\nPositive sentiment : ", (pos * 100.0 / len(comments))) print("\nNegative sentiment : ", (neg * 100.0 / len(comments)))
def sentiment(comments): if not os.path.isfile('CommentSentimentData/classifier.pkl'): tcl.training() fl = open('CommentSentimentData/classifier.pkl', 'rb') classifier = pickle.load(fl) fl.close() with open('CommentSentimentData/bestwords.pkl', 'rb') as f: best_words = pickle.load(f) pos = 0 neg = 0 for words in comments: comment = features(words, best_words) sentiment_value, confidence = VoteClassifier(classifier).classify( comment) if sentiment_value == 'positive': # and confidence * 100 >= 60: pos += 1 else: neg += 1 print("Positive sentiment : ", (pos * 100.0 / len(comments))) print("Negative sentiment : ", (neg * 100.0 / len(comments)))