Accurate assessment of soil fertility in agricultural land is beneficial for the development of precision agriculture, protecting the quality of agricultural land, and providing a decision-making basis for the sustainable use of soil resources. A total of 1 151 agricultural soil samples were collected in Changle County, Shandong Province. Based on 14 environmental characteristic factors, the spatial distribution of soil organic carbon was predicted by extreme gradient boosting(XGBoost)model. Using pH, organic carbon, alkaline nitrogen, available phosphorus and available potassium as evaluation indicators, the content and spatial distribution characteristics were analyzed, and the coefficient of variation method was applied to comprehensively evaluate soil fertility. The results show that the soil organic carbon content in the study area is 5.62~12.10 g/kg, which is at a relatively low level. The soil organic carbon content is influenced by nutrient factors, soil texture and biological factors. The soil pH in the study area is mainly neutral, and alkaline nitrogen, available phosphorus and available potassium have significant spatial correlations and are generally at moderate or above levels. The overall fertility of the soil in the research area is relatively poor to moderate, with a total of 91.84% of poor and moderate classes. The organic matter content is a limiting factor and the spatial distribution shows a spatial pattern of high, medium and low in the surrounding areas. It is necessary to increase the application of organic fertilizer and to optimize the cultivation method in order to improve the soil organic matter content and the soil fertility of the cultivated areas.