Abstract
In the competitive luxury hotel industry, continual improvement and responsiveness to guest feedback are crucial. Negative reviews, while seemingly detrimental, provide valuable operational insights. This paper introduces the Aspect-Based Sentiment Summarization (ABSS) framework, enhanced with Generative Pretrained Transformer (GPT) models, as an effective tool for converting negative reviews into actionable intelligence. Focusing on a single luxury hotel's negative feedback from Google Maps, the ABSS-GPT approach utilizes natural language processing to identify the root causes of guest dissatisfaction and translate sentiments into measurable patterns. This analysis informs decision-making processes, helping luxury hotels improve service quality and operational efficiency. The research demonstrates that the ABSS-GPT methodology plays a pivotal role in leveraging negative feedback to enhance guest satisfaction and maintain competitiveness in the luxury hospitality sector. Ultimately, this framework transforms adverse reviews into strategic advantages for luxury hotels.