Privacy Policies and Measurement - Modelling and analyzing the massive policy discourse networks are of great importance in critical policy studies and have recently attracted increasing research interests. Yet, the effective representation scheme, quantitative policymaking metrics and the proper analysis methods remain largely unexplored. To address above challenges, with the Latent Dirichlet Allocation embedding, we proposed a government policy network, which models multiple entity types and complex relationships in between. Specifically, we have constructed the government policy network based on approximately 700 rural innovation and entrepreneurship policies released by the Chinese central government and eight provinces’ governments in the past eight years. We verified that the entity degree in the policy network is subject to the power-law distribution. Moreover, we propose a metric to evaluate the coordination between the central and local departments, namely coordination strength. And we find that this metric effectively reflects the coordination relationship between central and local departments. This study could be considered as a theoretical basis for applications such as policy discourse relationship prediction and departmental collaborative analysis.