Li, JunALANAZI, MOHAMMED SAAD2025-08-202025APAs298908https://hdl.handle.net/20.500.14154/76217This thesis examines airport service quality by analyzing passenger feedback collected from multiple online platforms, including TripAdvisor, Google Maps, Twitter, and airlinequality.com. Using a dataset of about 300,000 reviews, the research applied advanced Artificial Intelligence techniques, such as Machine Learning (ML) and Deep Learning (DL), to evaluate passenger opinions on specific aspects of airport services, such as security, check-in, facilities, staff behavior, and wayfinding. A key method used was Aspect-Based Sentiment Analysis (ABSA), which allows identifying strengths and weaknesses in each service area rather than giving only a general opinion. Several predictive models, including SVM, Random Forest, CNN, and LSTM, were developed and achieved high accuracy in classifying passenger sentiments. The study also proposed a standardized evaluation framework to make results consistent across different review sources. Overall, the research provides both theoretical contributions, by advancing sentiment analysis methods and integrating diverse datasets, and practical contributions, by offering airport managers actionable insights to improve service quality and enhance passenger satisfaction.This thesis presents a comprehensive analysis of airport service quality through the lens of passenger feedback, utilizing advanced computational techniques such as machine learning (ML) and deep learning (DL). The primary objectives of the research include developing and validating predictive models, employing Aspect-Based Sentiment Analysis (ABSA), standardizing evaluation aspects across different platforms, and conducting a comparative analysis of feedback from various online sources. To achieve these objectives, a substantial dataset of approximately 300,000 reviews was collected from platforms such as airlinequality.com, TripAdvisor, Google Maps, and Twitter. This diverse and extensive dataset enabled a detailed examination of passenger sentiments related to key aspects of airport services, including facilities, terminal, security, check-in processes, wayfinding, and staff behaviour. The application of ABSA highlights specific areas of service excellence and those requiring improvement. The research introduced methodological innovations by integrating multiple data sources and employing advanced predictive models, including Support Vector Machines (SVM), Decision Trees, Random Forests, Convolutional Neural Networks (CNNs), and Long Short-Term Memory (LSTM) networks. These models demonstrated high accuracy and reliability in predicting passenger sentiments, offering robust tools for real-time monitoring and proactive management of airport service quality. A standardized evaluation framework was developed to ensure consistent assessments across different platforms, enhancing the reliability and comparability of the findings. The comparative analysis revealed distinct trends and characteristics in passenger feedback across various platforms, providing a comprehensive understanding of passenger experiences and expectations. The theoretical contributions of this thesis include the advancement of sentiment analysis methodologies, the integration of diverse online data sources, and the development of a standardized framework for service quality assessment. Practically, the research provides actionable insights for airport management to enhance service quality and improve passenger satisfaction. The findings also have broader applicability across various service-oriented industries, demonstrating the versatility and impact of the methodologies developed. In conclusion, this thesis sets a new standard for the analysis of airport service quality, providing a detailed, reliable, and actionable framework that can be applied universally across different review platforms and service contexts. The findings offer valuable insights for enhancing passenger experience and driving service improvements in the aviation industry and beyond.192enAirport Service qualitypassenger feedbackonline feedback platformaspect-base sentimental analysisThesis