Identifying and evaluating the impact of key service quality attributes on airline ratings using large language model (LLM): Case study Iberia

Online star ratings and sentiment classification are important to understand passenger perceptions regarding the quality of different airline service attributes. This study introduces a novel approach for classifying service quality attributes in airline reviews by leveraging question-based prompts using Large Language Model (LLMs). By classifying the sentiment polarity for each specific attribute, this method provides a more nuanced understanding and provides valuable insights into the key drivers of online reputation in the airline industry. The regression models highlighted that 70% of passengers evaluate at least two service quality attributes, focusing on what matters most to them. The analysis showed significant relations in 63 models, with ¨Check-in¨ and ¨Cleanliness¨ approaching as strong predictors of passenger satisfaction. The findings will offer practical recommendations for airlines seeking to improve their service and achieve a competitive position in the market.