Knowledge extraction method and system based on memory neural network and device

Abstract

The invention relates to the field of knowledge extraction and particularly relates to a knowledge extraction method and system based on a memory neural network and a device. The invention aims to solve the problem of information redundancy existing in the prior art. The method comprises a step of obtaining a possible relationship type in an input text and a semantic coding vector by using a convolutional neural network under the premise of giving a predefined relationship type, a step of carrying out semantic encoding by using a two-way long and short time memory neural network and obtaininga semantic vector, a step of taking the relationship type as an initial value of the two-way long and short time memory network and a first label in a decoding module and thus fusing the relationshiptype information into encoding information and label information of the decoding module, and a step of obtaining a label sequence by using a decoding module of a single-way long and short time memorynetwork and then obtaining structured information by parsing the label sequence. According to the knowledge extraction method and system and the device, the efficiency of structured information extraction is greatly improved, and the problem of information redundancy existing in the prior art is solved.

Description

Technical field [0001] The present invention relates to the field of knowledge extraction, in particular to a knowledge extraction method, system and equipment based on a memory neural network. Background technique [0002] The rapid development of the Internet has caused an explosive growth of text data on the Internet. Massive text data contains a large amount of knowledge and also has the problem of information redundancy. On the one hand, from the perspective of users, how to quickly obtain the key information that is really needed from the increasing mass of text data has become an urgent need in people’s daily life and work; on the other hand, from the perspective of intelligent applications , All kinds of intelligent applications, such as automatic question and answer, intelligent search, personalized recommendation, etc., all need the support of knowledge resources, and a large number of knowledge resources can only be used by machines if they are organized together in th...

AI Technical Summary

Problems solved by technology

[0006] The existing triplet extraction algorithms for predefined relation types can be simply divided into two categories: one is the series extraction method, which has the advantage of being convenient for optimizing entity recognition tasks and relation extraction tasks separately, but the disadvantage is that they are used to obtain triplets. The intermediate product (entity or relationship type) of the group is the target, and the result of entity recognition will further affect the result of relationship extraction, resulting in the accumulation of errors; the other is the association extraction algorithm of entities and relationships. and relational relevance to improve the effect of triplet extraction

Although these methods have their own advantages, their basic idea is to first obtain entity tuples and relational tuples, and then obtain the basic knowledge unit "triple", which will eventually cause information redundancy.

Images