Knowledge graph-based multi-path training scene construction method

Abstract

The invention discloses a knowledge graph-based multi-path training scene construction method. The method comprises the steps: building a main knowledge graph according to the intention motivation ofa verbal skill problem, covering all intentions in a training scene, and carrying out the classification and correlation according to a service mode and a link; establishing a corresponding verbal problem pair associated with the intention on each entity of the intention map; establishing a knowledge reasoning path with a strong relationship and a weak relationship in each question pair graph, wherein the knowledge reasoning path comprises semantic analysis and various knowledge reasoning models; and generating one or more training scenes according to the selected conditions, wherein the mainconditions comprise links of the training scenes, evaluation point keywords and question pair quantity indexes. According to the invention, a user can select one of the training scenes automatically generated by tens of millions of paths for training, so that the convenience of creating the training scene and the intelligent degree of the scene are greatly improved.

Description

technical field [0001] The present invention relates to the technical field of knowledge graphs, in particular to a method for constructing multi-path training scenarios based on knowledge graphs. Background technique [0002] At present, the training scenarios set for man-machine pair training still need to be manually created through the configuration process. The scenarios are single, rigid, fixed, low in intelligence, time-consuming and laborious, and cannot meet the real needs of customers to create flexible and changeable training scenarios. . Contents of the invention [0003] The purpose of the present invention is to provide a method for constructing a multi-path training scene based on a knowledge graph. The multi-path training scene allows users to freely select a certain training scene according to multiple paths for man-machine pair training, which is more convenient and flexible. [0004] To achieve the above object, the present invention provides the follow...

AI Technical Summary

Problems solved by technology

[0002] At present, the training scenarios set for man-machine pair training still need to be manually created through the configuration process. The scenarios are single, rigid, fixed, low in intelligence, time-consuming and laborious, and cannot meet the real needs of customers to create flexible and changeable training scenarios.

Images