[Purpose/Significance] Digital countryside is the strategic direction of rural revitalization and an important part of the construction of digital China. It is of great significance to quantitatively analyze the policy texts related to digital countryside from the perspective of text analysis and to explore the fields of policy action, the law of evolution, and the logic of development planning, in order to improve the development of the policy. [Method/Process] Firstly, text preprocessing was carried out on the policy text corpus. Second, TF-IDF was used to extract the feature words of the policy text, and the set of feature words was screened. Then, based on the FastText model, the feature words set was trained with a word vector, K-means cluster analysis was carried out and the co-occurrence matrix of feature words was constructed. Finally, the Gephi tool was used to draw the semantic network diagram. [Result/Conclusion] The cluster analysis results show that the previous digital rural policies mainly affect the construction and development of digital rural from five aspects: policy receptors, policy perspectives, policy fields, supporting measures, and policy functions, which shows the important leading role of digital rural policies in promoting the comprehensive construction and development of digital rural. The semantic network diagram reveals that the previous policies radiated various fields of digital rural construction and development with the agricultural products and modern agriculture as important nodes. The new policies have strengthened and supplemented based on the old policies, but the evolution of digital rural policies has a certain lag and exposes some relatively weak areas in the past, such as management systems, information security, and environmental governance. This study provides an important basis for the policy effect analysis and policy direction of the digital countryside.