Twitter Sentiment Analysis using Deep Neural Network

摘要

People usually try as certainly online opinions about products travel plans or movies. Thus, opinions and sentiment analysis become a daily process. Therefore, the need for text sentiment analysis automation has arisen. It is helpful to access views and sentimental views about specific subject effectively, rather than surfing reviews. In this paper, two main approaches of sentiment analysis are applied, machine learning approaches support vector machine, naive Bayes, decision tree, and K-nearest neighbor and the second approach is the deep neural network, a recurrent neural network using Long Short-Term Memory (LSTM). Also, we used three twitter datasets which are IMDB, Amazon, and Airline to apply two approaches. Also, we illustrate a comparison between different algorithms and the experiment results are shown the recurrent neural network using LSTM that is scored the highest accuracy of 88%, 87%, and 93%.