Abstract: The National Strategic Plan for Rural Revitalization (2018-2022) proposes to promote rural development under the classification: agglomeration for improvement, integration into cities and towns, characteristic protection, relocation and merging. China is a vast country with a large number of villages. However, the traditional classification method is less efficient and more arbitrary due to the interference of human factors. Taking the village layout planning of Wuxi County, Chongqing City as an example, we selected 13 learning indicators (input quantity) and 2 screening indicators from 6 aspects: ecological background, land bearing, traffic conditions, natural disasters, economy, history and humanities to form an indicator system in Wuxi County. The village classification is carried out by BP neural network, and the development priority measure of each type of village is calculated. The results show that the BP neural network has a high accuracy rate (98.6%) in classifying village types, and provides development timing guidance for different types of villages. The method could effectively improve the efficiency of village classification and the scientificity of planning.