Abstract Twitter Sentiment Analysis is the way of identifying sentiments and opinions in tweets. The main computational steps in this process are determining the polarity or sentiment of the tweet and then categorizing them into the positive tweet or negative tweet. The primary issue with Twitter sentiment analysis is the identification of the most suitable sentiment classifier that can correctly classify the tweets. Generally, base classification technique like Naive Bayes classifier, Random Forest classifier, SVMs and Logistic Regression are being used. In this paper, an ensemble classifier has been proposed that combines the base learning classifier to form a single classifier, with an aim of improving the performance and accuracy of sentiment classification technique. The results show that the proposed ensemble classifier performs better than stand-alone classifiers and majority voting ensemble classifier. In addition, the role of data pre-processing and feature representation in sentiment classification technique is also explored as part of this work.