RESUMEN
In some developing countries, particularly China, a significant number of individual farmers manage small field scale of cultivated land. However, the existing research on cultivated land quality assessment mainly focuses on large-scale regions, establishing comprehensive index systems from a macro perspective, while lacking evaluations customized to individual farmers, who constitute a crucial component in agricultural production, and a demand-driven field-scale assessment of cultivated land quality. Therefore, we developed a field-scale index system that meets the needs of individual farmers in the black soil region of Northeast China. Additionally, we proposed a machine learning model for field-scale cultivated land quality assessment. The experimental results showed that our model achieved an [Formula: see text] value of 0.9660 and an [Formula: see text] of [Formula: see text] under fourfold cross-validation, which represents an improvement of 5.19% and a reduction of 1.13%, respectively, relative to the XGBoost model. Ultimately, we conducted obstacle factor diagnosis, aiming to assist individual farmers in identifying the existing issues in their cultivated land fields. This study not only provides guidance to individual farmers but also addresses the research gap in cultivated land quality assessment by offering an individual farmer demand-driven index system for field-scale studies.