AnasAlmasri

import twitter

# initialize api instance
twitter_api = twitter.Api(consumer_key='YOUR_CONSUMER_KEY',
                        consumer_secret='YOUR_CONSUMER_SECRET',
                        access_token_key='YOUR_ACCESS_TOKEN_KEY',
                        access_token_secret='YOUR_ACCESS_TOKEN_SECRET')

# test authentication
print(twitter_api.VerifyCredentials())

AnasAlmasri

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)
        
        return [{"text":status.text, "label":None} for status in tweets_fetched]
    except:
        print("Unfortunately, something went wrong..")
        return None

AnasAlmasri

search_term = input("Enter a search keyword:")
testDataSet = buildTestSet(search_term)

print(testDataSet[0:4])

AnasAlmasri

def buidTrainingSet(corpusFile, tweetDataFile):
    import csv
    import time
    
    corpus = []
    
    with open(corpusFile,'rb') as csvfile:
        lineReader = csv.reader(csvfile,delimiter=',', quotechar="\"")
        for row in lineReader:
            corpus.append({"tweet_id":row[2], "label":row[1], "topic":row[0]})

AnasAlmasri

corpusFile = "YOUR_FILE_PATH/corpus.csv"
tweetDataFile = "YOUR_FILE_PATH/tweetDataFile.csv"

trainingData = buildTrainingSet(corpusFile, tweetDataFile)

AnasAlmasri

import re
from nltk.tokenize import word_tokenize
from string import punctuation 
from nltk.corpus import stopwords 

class PreProcessTweets:
    def __init__(self):
        self._stopwords = set(stopwords.words('english') + list(punctuation) + ['AT_USER','URL'])
        
    def processTweets(self, list_of_tweets):

AnasAlmasri

tweetProcessor = PreProcessTweets()
preprocessedTrainingSet = tweetProcessor.processTweets(trainingData)
preprocessedTestSet = tweetProcessor.processTweets(testDataSet)

AnasAlmasri

import nltk 

def buildVocabulary(preprocessedTrainingData):
    all_words = []
    
    for (words, sentiment) in preprocessedTrainingData:
        all_words.extend(words)

    wordlist = nltk.FreqDist(all_words)
    word_features = wordlist.keys()

AnasAlmasri

def extract_features(tweet):
    tweet_words = set(tweet)
    features = {}
    for word in word_features:
        features['contains(%s)' % word] = (word in tweet_words)
    return features 

AnasAlmasri

word_features = buildVocabulary(preprocessedTrainingData)
trainingFeatures = nltk.classify.apply_features(extract_features, preprocessedTrainingData)