European Journal of Scientific Research ISSN 1450-216X / 1450-202X Vol. 146 No 1 June, 2017, pp.93 - 101 http://www. europeanjournalofscientificresearch.com Twitter Messages Sentiment Analysis Model based on Deep and Machine Learning Marina Essam Badeaa Modern Academy for Computer Science and Technology E-mail: marina_essam91@yahoo.com Wael H.Gomaa Faculty of Computers & Information Beni-Suef University E-mail: wael.goma@fcis.bsu.edu.eg Mohamed H.Haggag Faculty of Computers & Information Helwan University E-mail: mohamed.haggag@fci.helwan.edu.eg Abstract This research aims at introducing a new hybrid model for Twitter sentiment analysis, which categorizes a tweet’s sentiment polarity into positive and negative. Depending on the lexicon-based, regular machine learning-based and deep learning-based, these three methodologies were trialed, concluding that the out-turn polarity of each methodologies has become an input data for a majority voting algorithm. Through a simple rule, the voting algorithm reached the final polarity. Whereas the hybrid model promotes the accuracy of every approach individually on two data sets according to the experimental outcomes. Keyword: Twitter, Sentiment Classification, Voting Algorithm, Hybrid Model, Deep Learning 1. Introduction 500 million tweets every day by 1.3 billion users have made the social networking site “Twitter” the most commonly used microblogging service web, and a very interesting platform as well. According to the account privacy, a tweet on Twitter (similar to a post on Facebook)is shared publicly or among defined followers through a number of characters restricted to 140; distinguishing Twitter from the other social networking websites. However, it is challenging to users to express their brainswith never-ending sentences, since they are restricted by limited characters. On the other side, Twitter is considered a great reflection of the events globally occur, as it is widely used by all strata. From the companies evaluation, Twitter is among the main platforms because of the latest trends appear on Twitter. These trends are used to achieve marketing ends such advertising for products on trends go viral. Also, tweets analysis is an advantage for the companies to create competitive marketing ideas.