An Approach Based on Artificial Intelligence and Spatio- Temporal Data Mining for the Prevention of Territorial: Application to Territory Planning

Territories are progressively undergoing socialization through spatial dynamics and the swift pace of change, resulting in tangible spatial mutations and intricate spatio-temporal interactions. Territorial planning involves developing public spaces, structures, facilities, services, and transportation systems whose form and function adapts to temporalities, geo-graphic information, and diverse applications. A novel approach, ZIT-IME, focuses on artificial intelligence and spatio-temporal data mining to enhance decision support in territorial planning. The approach identifies association rules and affinity analyses to extract correlations and suitable variables and manipulate event-based data which are responsible for the changes observed at the level of spatio-temporal entities. The solution involves supervised learning by artificial neural networks for prediction models and knowledge extraction in the field of territory planning for the building of primary schools and identification of suitable areas in Tunisia.