Knowledge base optimization method and system

Abstract

The embodiment of the invention provides a knowledge base optimization method. The method comprises the steps: querying a topic tree which is hit by a user and contains potential context associated information based on a knowledge base; selecting a heuristic question corresponding to at least one knowledge point from a plurality of knowledge points of the topic tree to feed back to a user, and guiding the user to select; determining the next round of questions of the user as heuristic labels for the user; and optimizing the knowledge base at least based on the heuristic annotation. The embodiment of the invention further provides a knowledge base optimization system. According to the knowledge base optimization method for the heuristic dialogue, potential information in the user dialogue process can be mined, knowledge base optimization can be efficiently completed, association in user problems can be fully mined, and more accurate and rapid knowledge base optimization is achieved with low human input. In addition, for ever-increasing customer service composite requirements and more complex service scenes, more accurate and higher-quality optimization is carried out, and the intelligent service level can be improved more quickly.

Description

technical field [0001] The invention relates to the field of natural language understanding, in particular to a knowledge base optimization method and system. Background technique [0002] The knowledge base is the core of the Q&A service provided by intelligent customer service products. Improving the accuracy of questions and answers in the intelligent customer service knowledge base can improve the overall service level of intelligent customer service and improve customer satisfaction. Knowledge base technology is a technology that belongs to the category of natural language understanding (NLP). The effect of knowledge base question answering is related to factors such as application scenarios, topic range, and training data. Without training and optimizing the question method, most of the knowledge base will return blind answers or similar recommendations, frequently triggering blind answers, or frequent recommendations without direct answers, which is not good for cust...

AI Technical Summary

Problems solved by technology

[0007] In order to at least solve the problem that the existing knowledge base optimization does not take advantage of whether the user's question is related to the next question, there is less information available for optimization, and the improvement effect is limited

Images