Analyzing user perceptions and adoption of urban public transport applications through natural language processing

An enhanced understanding of urban public transport app adoption has been achieved by applying innovative data mining and sentiment analysis techniques to large user review datasets from Google Play Store and Apple App Store, covering major European metropolitan areas. Natural language processing (NLP) methods were employed to perform sentiment analysis, emotion analysis, and topic modeling (LDA), extracting valuable insights from unstructured text. The results highlight the importance of citizen science and citizen participation in app development. This is in line with the growing trend of actively involving users in the development of digital services. Intermodal integration, accessibility, improving the user experience, environmental engagement, integrating digital payments seamlessly, and providing real-time information have been identified as key user priorities. The perceived usefulness of reviews was impacted by the relationship between sentiment and review features. This offered suggestions for ways to improve app development. The findings present useful recommendations for creating more inclusive, sustainable, and user-aligned services in smart city settings, as well as a conceptual framework for technology adoption in urban mobility applications.