Artificial Intelligence Agents in Spatial Analysis teaching

• Grant number: 25/09029-9
• Support Opportunities: Regular Research Grants
• Start date: February 01, 2026
• End date: January 31, 2029
• Field of knowledge: Humanities - Geography
• Principal Investigator: Alfredo Pereira de Queiroz Filho
• Grantee: Alfredo Pereira de Queiroz Filho
• Host Institution: Faculdade de Filosofia, Letras e Ciências Humanas (FFLCH). Universidade de São Paulo (USP). São Paulo , SP, Brazil
• City of the host institution: São Paulo

Abstract

The incorporation of technology into Geography teaching activities in recent decades can be considered growing, although heterogeneous. There are many incentives to innovate classroom activities; however, many teachers have superficial experience with technology and serious time constraints to develop new skills. The objective of this research is to contribute to the construction of a methodological framework on Artificial Intelligence (AI) agents in Spatial Analysis (SA) teaching activities. It aims to explore the interdependence between spatial questions, data, methods and technological resources of SA. The proposal is to insert content on topological relationships (adjacency, connectivity and containment) between vector representation forms (point, line and polygon) into the AI ¿¿agent, which will be trained by students and teachers to create a digital assistant. The tests will be carried out in the Spatial Analysis and Geoprocessing discipline of the Geography course - FFLCH - USP. The assumptions are that: it is essential to insert technologies into a coherent pedagogical context linked to a progression of questions, in order to relate the information obtained to an integrated network of geographic content; and that the greatest challenge is to select simple, free or low-cost teaching resources, and that their incorporation requires a small number of hours. As a result, we hope to contribute methodologically to the use of AI agents in activities involving spatial reasoning, generate codes to execute spatial queries and reflect on the benefits and problems that this technological incorporation could cause in the teaching and learning process. (AU)