Analysis of the Urban Development Technologies of Ukek using Neural Networks (preliminary research results)

The application of neural networks to analyze urban planning technologies in the Golden Horde city of Ukek, focusing on building materials such as bricks and ceramic water pipes, is explored. This study presents the main approaches, algorithms, and data processing methods, including geoinformation systems (GIS), clustering, optimization algorithms, and various types of neural networks such as convolutional, recurrent, and deep networks. The research is based on the analysis of historical and archaeological materials from the Uvek settlement. Complex patterns and associations between archaeological, historical, and geological data from different years were identified. The software tools included convolutional neural networks for image analysis, recurrent neural networks for time sequence analysis, and deep neural networks for complex classification, modeling, and verification tasks. Key aspects of the construction and infrastructure of the city of Ukek were identified through this analysis. These include the localization of raw material extraction sites, the number of construction sites, number of bricks produced, water lines, and the territorial delimitation of the urban area based on a digital elevation model (DEM). The chronology of landslides and their impact on the city's development was also examined. The research results confirm the effectiveness of mixed and multi-level neural networks for analyzing archaeological data related to the characteristics of medieval technologies, the process of concentrating productive forces, and means of production.