tc=args.tweet_count, st=args.search_type, ud=args.since_date, lan=args.language) opin_dict = tClient.opinion_mining(search_results['statuses']) ''' te_search_results = twitter.search(q=keyword, count=tweet_count, result_type='mixed', since_id=since, lang="te") trans = Translator() for result in te_search_results['statuses']: print result['text'] trans.translate(result['text']) ''' else: multi_search_results = tClient.multi_search_until(k=args.keyword, ud=args.since_date, lan=args.language) opin_dict = tClient.opinion_mining_multi(multi_search_results) while True: print "Print 1)Positive 2)Negative 3)Netral 4)Exit \n " option = input("Enter option(1/2/3/4)") opted_dict = {} if option == 1: opted_dict = opin_dict['positive'] elif option == 2: opted_dict = opin_dict['negative'] elif option == 3: opted_dict = opin_dict['neutral'] elif option == 4:
parser.add_argument("since_date", help = "The date from which tweets should be analyzed") parser.add_argument("-v", "--verbose", help = "Print traces for debugging", type=bool, default=False) ### End of arguments ### ####main#### args = parser.parse_args() tClient = TwitterClient() ReviewRules = [('VeryGood' , 90), ('Good' , 80), ('Watchable', 70), ('Average' , 60), ('Bad' , 30), ('VeryBad' , 0)] multi_search_results = tClient.multi_search_until(k=args.Movie, ud=args.since_date, verbose=args.verbose) total_tweets, opinion_dictionary = tClient.review_mining_multi(multi_search_results, ReviewRules) ''' opinion dictionary is a dictionary of dictionaries has the following structure {Sentiment : {tweet_id: tweet_text}} ex: {'VeryGood' : { 1 : "Must Watch", 2 : "Top Notch" } } ''' result_summary_dictionary = OrderedDict({int: [str,(float, int)]}) ''' {SentimentSNo:[Sentiment, (approved percetage(percentage of tweets who approved this rating),