Precise Recommendation Method of Suitable Planting Areas of Maize Varieties Based on Knowledge Graph

The rapid increase in the number of new maize varieties and intensification market competition have raised need to precisely promote suitable planting areas fully exploit variety potential win competition. This paper proposes a precise recommendation method for based on knowledge graph. meteorology graph ecological regions is constructed at county-scale RippleNet model used mine spatial correlation suitability different meteorological environments. then realized. In total, 331 agricultural data 59 experimental Huang-Huai-Hai region are training testing (accuracy 76.3%). Through comparison, accuracy this 24.3% higher than that six traditional machine learning methods, 11.2% attention networks, 5.8% convolution neural networks. study provides data-driven solution positioning varieties, enhances scientificity helps their potential.