High polymer material research and development guidance generation method and system based on large language model

Abstract

The invention discloses a high polymer material research and development guidance generation method and system based on a large language model, and the method comprises the steps: collecting a to-be-labeled industrial standard text in a vertical field, and labeling the industrial standard text; cleaning and preprocessing the marked industrial standard text, and converting the marked industrial standard text into a structured data set; selecting a large language model, performing knowledge injection training by using LoRA fine tuning, injecting specific high polymer material knowledge, updating parameters of an attention layer and a feed-forward network layer in the large language model, and storing as a first large language model; performing depth fine tuning on the first large language model by using a structured data set, and storing the first large language model as a second large language model; and generating industry standard data by using the second large language model. According to the hybrid field large model constructed by the method, on the premise of original functions and performance, the application effect and accuracy of the hybrid field large model in a specific vertical field are enhanced, and the research and development efficiency and innovation capability of materials can be effectively improved.

Images