A new energy policy text classification and recognition model based on convolutional neural networks

The sustainable development of new energy is the key to the low-carbon transformation of China's energy system, and the analysis and interpretation of new energy policies are the basis for the scientific layout and steady progress of the new energy business of enterprises. Based on this, this paper constructs a new energy policy text classification recognition model based on the CNN (convolutional neural networks) model, combs the policies released in the new energy field in China in the past 15 years, and uses expert consultation method and manual pre-processing method to extract, clean and classify the policy keywords. Among them, different from the previous policy classification methods, this paper classifies the policies into three types of policies, namely, guiding, incentive, and regulatory, according to the policy characteristics. Based on this, the constructed policy features are input into the CNN model for energy classification and identification. The results show that the accuracy of the model recognition is about 65.96%, which indicates that the CNN model applied in this paper can classify the new energy policies effectively.