Causal relationship extraction method and device and electronic equipment

Abstract

The invention provides a causal relationship extraction method and device and electronic equipment, and relates to the field of artificial intelligence, in particular to the field of knowledge maps and deep learning. According to the specific implementation scheme, a to-be-processed text is obtained; the to-be-processed text and a first question are input into a first machine reading understanding model, a first round of answers are obtained, the first question is used for inquiring reasons in the to-be-processed text, and the first round of answers comprise at least one reason in the to-be-processed text; the to-be-processed text and a second question are input into a second machine reading understanding model, a second round of answers are obtained, the second question is used for inquiring the result of the reason in the to-be-processed text, and the second round of answers comprise at least one result of the reason. Therefore, the causal relationship extraction accuracy is improved.

Description

technical field [0001] The present disclosure relates to the technical fields of artificial intelligence such as knowledge graph and deep learning, and in particular to a causal relationship extraction method, device and electronic equipment. Background technique [0002] There is a large amount of causal-related knowledge in the event graph, and mining the causal relationship between events is of great value for downstream applications such as event prediction and event reasoning. [0003] At present, the mainstream causal relationship extraction method is still a rule-based method. By artificially summarizing causal-related conjunctions, such as "because", "resulting in", and "due to", etc., and then constructing a rule template, using the rule template to mine from the text cause, effect>. [0004] The limitation of this method is that not all causal relations are guided by causal conjunctions, and not all sentences containing causal conjunctions must have causal rela...

AI Technical Summary

Problems solved by technology

[0004]The limitation of this method is that not all causal relations are induced by causal conjunctions, nor are sentences containing causal conjunctions necessarily causal, and the rules cannot Exhaustive, so the accuracy of causality using rule templates is low

Images