Knowledge graph inference method based on relation detection and reinforcement learning

Abstract

The invention discloses a knowledge graph inference method based on relation detection and reinforcement learning. The method comprises the steps that on the basis of character string fuzzy matching between a domain knowledge graph and an entity dictionary and a CNN-LSTM-CRF-based entity recognition model, an entity in a question input by a user is detected, and entity detection is completed; relation detection is completed by a neural network based semantic matching model, and the relation detection model is characterized in that low-dimension manifold expression is obtained through the neural network according to the input question, the relation related to the question and the relation not related to the question, on the basis of the low-dimension manifold expression, rank loss optimization model parameters are adopted, so that the question can search the relation set for the relation most similar to the semantics; according to knowledge graph inference based on reinforcement learning, for each time step, on the basis of a strategy function pi theta, under the current entity et, one out-going relation rt+1 is selected, the next entity et+1 is executed, the final entity eT is reached through a preset sequential decision with the maximum inference path length T, and the entity eT is adopted as an answer of the question to be output.

Description

technical field [0001] The invention relates to a knowledge map reasoning method based on relationship detection and reinforcement learning. Background technique [0002] At present, the question answering system based on knowledge graph is mainly based on SPARQL query statement and Multi-hop knowledge graph embedding method. [0003] The first is a knowledge graph question answering system based on SPARQL query statements. The system stores the knowledge graph in the graph database, and the system parses the natural language questions entered by the user into SPARQL statements, queries in the graph database, and returns the query results to the user. The system can give accurate answers, but it is a difficult task to generate SPARQL from natural language questions. Relevant experts need to summarize the question templates with high frequency according to the historical question and answer data, and write based on the question templates. It is difficult to expand the corre...

AI Technical Summary

Problems solved by technology

This method relies on the embedding representation of the knowledge graph. After migrating to a different domain, the embedding representation of the knowledge graph needs to be retrained, and it is difficult for TransE to model the situation where an entity corresponds to multiple relationships.

In addition, the reasoning model based on the memory network is difficult to deal with knowledge graphs where an entity contains multiple relationships, which limits its application scope.

Images