Abstract: In June 2019, with the announcement of the fifth batch of Chinese traditional villages, a total of 6819 villages with important protection values have been included in the list of traditional villages in China. The quantitative and qualitative evaluation indicators are mixed in the identified evaluation index system. There is no quantifiable evaluation index system, and the lack of intelligent evaluation methods. In this paper, the author has collected and sorted out relevant data of more than 6,000 outstanding villages. The dimensions include the village's basic characteristics, village history, natural environment, site selection pattern, traditional architecture, and folk culture. Based on this data set, the decision tree, which is a supervised learning method in machine learning, is used to study and build a traditional village intelligent evaluation model to predict whether the village objects meet the evaluation requirements of Chinese traditional villages. At the same time, according to the key measurement values of each variable in the model, the quantifiable characteristic attributes, which are in a key position in the traditional village identification evaluation, are described, and a quantifiable evaluation index system is explored and established. This intelligent evaluation model can be a powerful supplement to expert review, improve the efficiency of identification work, and increase the timeliness and accuracy of identification. The evaluation of provincial and municipal traditional villages can also draw on the decision tree algorithm used in this research. Based on the sample data of provincial and municipal traditional villages, an intelligent assessment model of provincial and municipal traditional villages can be built as an intelligent technical support.