Man-machine dialog method, training method of pre-trained language model, man-machine dialog system, and electronic device

A man-machine dialog method, comprising:

• receiving, through an interaction interface, a question input by a user, generating a question request according to the question and sending the question request to a could server;
• and
• receiving answer content fed back from the cloud server for the question request, and exhibiting and displaying the answer content;
• wherein the answer content is generated by the could server according to a document knowledge graph and the question corresponding to the question request, the document knowledge graph is generated based on hierarchical structure information and content information of a plurality of documents.

An electronic device, comprising:

• a processor, a memory, a communications interface and a communication bus, wherein the processor, the memory and the communications interface complete communication with each other through the communication bus;
• the memory is configured to store at least one executable instruction, and the executable instruction enables the processor to execute following operations:
• obtaining a question vector corresponding to a question;
• determining, from a plurality of document nodes, a target document node that is matched with the question vector, according to feature representations corresponding to the plurality of document nodes in a pre-generated document knowledge graph, wherein the document knowledge graph is generated based on hierarchical structure information and content information of a plurality of documents;
• and
• generating an answer for the question according to node content corresponding to the target document node that is matched with the question vector.