Vertical domain knowledge graph construction method based on transfer learning

The invention provides a vertical domain knowledge graph construction method based on transfer learning, and the method is characterized in that the method comprises the following steps: S1, inputting a knowledge text into a pre-training natural language model A to obtain lexical elements; s2, inputting the natural language tag into a pre-trained natural language model B to obtain a feature representation set; s3, performing dot product calculation on the lexical elements and the feature representation set to obtain a classification result; step S4, filling existing relation words into a Prompt template and then inputting the existing relation words into the pre-trained natural language model C to obtain vector representation; s5, inputting the classified sentences into a pre-training natural language model D to obtain a coding result; s6, performing similarity calculation on the coding result and the vector representation to obtain a relationship classification result; and S7, constructing tuples according t