[Objective] Conducting multi-crop land suitability analysis was carried out in order to provide an accurate and effective method for crop planning, land use planning, and management of agricultural production.[Methods] Based on data mining method, the comprehensive suitability of multi-crop land was quantitatively characterized by a random forest algorithm, a comprehensive index evaluation method, spatially constrained multivariate clustering, and spatial statistics.[Results] ① A random forest algorithm quantitatively and accurately produced the spatial layout of crop planting driven by multi-spatial elements, and identified factors that had a key influence on the selection of suitable planting land for maize and rapeseed. ② The suitable areas for maize and rapeseed planting exhibited significant spatial heterogeneity in the study area. The suitable areas for maize accounted for 91.23% of the total planting land area, but the suitable areas for rapeseed accounted for only 69.64%. The areas suitable for the planting of both crops was 13.08% of the total area, and were mainly located in the central and north parts of Fenggang County, the north part of Meitan County, and the northwest part of Yuqing County.[Conclusion] Data mining provides the possibility of selecting the optimal planting land use pattern. The method used in this study has good potential for evaluating the suitability of land in many locations for the production of various crops.