Sentiment Analysis using VADER: Unveiling Customer Sentiment and Predicting Buying Behavior in the Airline Industry

This study focuses on leveraging sentiment analysis using VADER (Valence Aware Dictionary and Sentiment Reasoner) to analyze customer reviews and predict buying behavior in the airline industry, with a specific focus on British Airways. The study aims to enhance customer experience and provide valuable insights for the airline company. The study begins by applying web scraping techniques to collect customer reviews from the Airline Quality website. The collected unstructured data is then processed and cleaned for further analysis. VADER sentiment analysis is utilized to determine the sentiment of each customer review, classifying them as positive, negative, or neutral. This analysis provides valuable insights into customer sentiment towards British Airways, enabling the airline to identify areas of improvement and address customer concerns. Furthermore, the study involves creating a classification model to predict customer buying behavior. Various features such as flight routes, booking origins, trip types, number of passengers, and customer preferences are considered to build the model. Random Forest Classifier, XGBoost Classifier, Logistic Regression and KNN algorithms are employed to train and evaluate the models. By accurately predicting buying behavior, British Airways can tailor their services and offerings to meet customer expectations, resulting in improved customer satisfaction and increased bookings. The findings and insights from this study provide British Airways with valuable information to optimize their operations, enhance customer experience, and stay competitive in the airline industry. By understanding customer sentiment and predicting buying behavior, the airline can make data-driven decisions to attract and retain customers, ultimately leading to improved business performance and success.