Deep Learning Models to Analyze Sentiments of People Regarding New Vaccines

In recent years, sentiment analysis has been one of the most active research areas in natural language processing. The wide range of its applications is the main reason for its popularity. The purpose of this work is to analyze the sentiments of Twitter users towards new vaccines using deep learning algorithms. The goal of this studey is to develop an effective model for classifying sentiments expressed in tweets. We use three deep learning algorithms that are CNN, LSTM and Bi-LSTM. To train and create a classifier from textual data, text should be transformed into numerical data. We use GloVe to transform text and create the word embedding for the DL models. We conduct comparison on the three developed algorithms. Our experiments demonstrate that the best model for the classification is the Bi-LSTM since it achieved the highest accuracy value of 0.82 on the test data.