Machine reading inference method based on graph neural network

Abstract

The invention provides a machine reading inference method based on a graph neural network. Overall process is as follows: a proposition judgment module, an entity identification module and an entity chain finger module are obtained through secondary training of a neural network; an information extraction module and a polarity discrimination module are combined respectively; a fact logic relation graph in a reading material and entity and polarity information in a to-be-inferred proposition are obtained, and then the fact logic relation graph, together with an environment knowledge graph, is input into a graph neural network subjected to secondary training together to obtain a final entity logic relation graph; and finally an inference conclusion and an inference route graph are obtained byusing a Bayesian network. According to the method, the graph neural network is applied to machine reading inference for the first time; on the basis of relation inference, the machine logic inferencecapacity is further given, and the automatic case inference process is achieved; and the method has important use value in the fields of criminal investigation, machine questioning and answering andthe like.

Description

technical field [0001] The invention relates to the field of artificial intelligence, in particular to the field of artificial intelligence machine reasoning. Background technique [0002] With the rapid development of society, the field of public security criminal investigation is facing new pressure. On the one hand, a large number of new forms of crime have emerged, leading to a high incidence of crimes in society; on the other hand, criminal investigations have higher requirements on the quality of investigators, and qualified criminals are in short supply. At present, there have been some researches on the application of artificial intelligence technology to assist criminal investigation, but they mainly focus on marginal work such as identification of criminal suspects, and rarely involve case reasoning, the core of criminal investigation. In addition, in the field of intelligent question-answering robots, current document question-answering robots can only answer som...

AI Technical Summary

Problems solved by technology

The patent "A Case Reasoning Method Based on Dynamic Knowledge Representation Learning" (public number CN110956254A) uses the relational reasoning model based on the recurrent neural network to find suspects on the constructed knowledge map, which has the ability of relational reasoning, but cannot perform logical reasoning

The patent "A Clinical Diagnosis Auxiliary Decision-Making System and Medical Knowledge Graph Accumulation Method" (publication number CN109686443A) uses logical formulas to perform logical reasoning on the medical knowledge graph to achieve clinical diagnosis, and cannot handle complex reasoning problems that include propositional logic and entity relationships at the same time

Images