Análisis de algoritmos de inteligencia artificial para la determinación del avalúo catastral en inmuebles rurales agrícolas mayores a 1 hectárea del municipio de la Peña - Cundinamarca

Abstract

This article, aimed at obtaining the title of Specialist in Appraisals with a Rural Approach, focused on the investigation of cadastral valuation processes through the implementation of key variables and artificial intelligence algorithms. Through an exhaustive analysis, various variables that affect the determination of appraisals for rural real estate in the municipality of La Peña - Cundinamarca were explored. The results revealed that the inclusion of a broad set of variables, such as geographical location, land size, type of right over the property, among others, was essential to improve the accuracy of the prediction models. Particularly, the RANDOM FOREST model stood out as the most robust, yielding favorable results in the tests carried out. The ability of this algorithm to handle different situations and data sets suggests that it is a valuable tool for cadastral valuation. In addition, the importance of the implementation of artificial intelligence technologies in cadastral update processes was highlighted, highlighting the opportunity to optimize the automation of the valuation of rural real estate. These findings not only contribute as degree work, but can also have significant implications for the development of the country. The modernization of cadastral valuation processes through the integration of technologies can represent a before and after in the current work, it also lays the foundations for a more solid and updated framework in the valuation field. The research and application of these methods not only benefit the work of appraisal specialists, but also support the progress and competitiveness of the country in the economic and territorial context.