Sentiment analysis on Malaysian airlines with BERT

Abstract

Sentiment analysis has been a popular research area in Natural Language Processing (NLP), where sentiments expressed through text data including positive, negative and neutral sentiments are analyzed and predicted. It is often performed to evaluate customer satisfaction and understand customer needs for businesses. In the airline industry, millions of people today use social networking sites such Twitter, Skytrax, TripAdvisor and more to express their emotions, opinions, reviews and share information about the aircraft service. This creates a treasure trove of information for the airline company, showcasing different points of views about the airline’s brand online and providing insightful information. Hence, this paper experiments with six different sentiment analysis models in order to determine and develop the best model to be used. The model with the best performance was then used to determine the social status, company reputation, and brand image of Malaysian airline companies. In conclusion, the BERT model was found to have the best performance out of the six models tested, scoring an accuracy of 86 percent.