A Method and Path for Predicting the Digital Transformation of New Agricultural Management Entities Based on Artificial Intelligence Algorithms

Due to the complex factors affecting the digital transformation, there is a significant deviation between the corresponding prediction results and the actual situation. Therefore, research on the digital transformation prediction method and path of new agricultural management entities based on artificial intelligence algorithms is proposed. After analyzing the driving factors of digital transformation from the perspectives of technology, market, policy, and talent, and considering that the prediction problem of digital transformation belongs to a linear and indivisible learning problem, based on adding relaxation variables to the linear support vector, the kinetic energy of digital transformation of new agricultural management entities in the feature space was calculated. And by propagating the calculation results forward and backward in the BP artificial neural network algorithm model, the final prediction result under error convergence state is determined. In the test results, the error between the predicted proportion of digital sales channels for 101 agricultural management entities and the actual situation using the design method is only 0.08%. Based on the predicted results, specific digital transformation paths have been proposed from the perspectives of talent and policy.