Proceedings of the Annual Meeting of the Cognitive Science Society

This paper presents a big data urban visualization platform for Guangdong Province, aimed at enhancing the efficiency and intelligence of urban planning. The platform utilizes Python to collect city-specific data, with data storage implemented using a MySQL database, complemented by NoSQL technologies to support the integration of unstructured data. The Flask backend employs deep learning and data mining algorithms to identify complex relationships among urban data, and we have also integrated graph neural network methods to capture spatial dependencies across different geographic regions within the city. Meanwhile, the ECharts frontend generates dynamic charts to present diverse information. Through a front-end and back-end separation architecture, the system ensures real-time updates, enhancing user experience. This research further emphasizes the need to explore challenges in real-time data integration, expanding data sources, optimizing user interactions, and protecting data privacy, providing important directions for future AI-driven urban planning.