Abstract: The gradual increase in the strength of the urban electromagnetic environment poses a potential threat to human health and the normal operation of electronic systems. In order to realize the rapid identification of the electromagnetic environment, this paper proposes a mining method for association rules among the electromagnetic environment, population heat and building density. Firstly, the electromagnetic environment, population heat, and building density data were discretely divided into three levels through two-step clustering. Then, the dynamic threshold frequent pattern growth (FP-growth) algorithm was used to conduct rule mining. Finally, the effective association rules were screened out. The proposed method is applied to 28424 sets of data sampled from roads in Beijing urban areas. The experimental results show that in areas with medium and low population thermal power or areas with low building density, the electromagnetic environment strength is usually lower than 1.52 V/m; in areas with high building density , the electromagnetic environment strength is usually in the range of 1.52−3.50 V/m; while the situation when the electromagnetic environment strength is higher than 3.50 V/m, mainly occurs in areas with high population heat. new ideas and references. The research work in this paper provides new ideas and references for rapid estimation, monitoring and early warning of urban electromagnetic environment.