V = len(dictionary)
countC = sum(dictionary.values())

sumOfProbs = 0

for word in newTweet:
if (word in dictionary):
x = (dictionary.get(word)+1) / (countC + V)

sumOfProbs = sumOfProbs + math.log(x)

return sumOfProbs

def classification(tweet):

totalNegTweets =0
for line in trainNegData:
totalNegTweets += 1

totalPosTweets =0
for line in trainPosData:
totalPosTweets += 1


 totalNumOfTweets = totalNegTweets + totalPosTweets
 positiveOverTotal = totalPosTweets / totalNumOfTweets
 negativeOverTotal = totalNegTweets / totalNumOfTweets

 positive = (math.log(positiveOverTotal)) +
 (calcWordProbability(tweet,posDict))
 negative = (math.log(negativeOverTotal)) +
 (calcWordProbability(tweet,negDict))

 if(positive < negative):
 prediction = 0
 else:
 prediction = 1

 return prediction

positive = (math.log(positiveOverTotal))+(calcWordProbability(tweet,posDict))
negative = (math.log(negativeOverTotal))+(calcWordProbability(tweet,negDict))