Text similarity identification method and device based on information interaction, equipment and medium

Abstract

The invention discloses a text similarity recognition method and device based on information interaction, equipment and a medium, and the method comprises the steps: respectively converting first text information and second text information to obtain two character sequences, and respectively extracting two keyword sequences from the two pieces of text information, performing mapping conversion on the two character sequences to obtain character coding information, inputting the character coding information into a neural network model to obtain a character vector, respectively obtaining keyword vectors of the two keyword sequences according to the character vector, and inputting the keyword vectors into a self-attention neural network to obtain two keyword representation vectors; and performing identification through the similarity identification model to obtain an identification result. The invention belongs to the technical field of semantic analysis and further relates to a block chain technology, two keyword representation vectors corresponding to two pieces of text information respectively are further obtained through character vectors obtained through vector information interaction, and similarity identification is performed based on vector information interaction, so that the time consumption of an identification process can be reduced; and the accuracy of similarity identification is greatly improved.

Description

technical field [0001] The present invention relates to the technical field of semantic analysis, and belongs to the application scenario of intelligently analyzing text information in a smart city to realize text similarity recognition based on information interaction, in particular to a text similarity recognition method, device, and equipment based on information interaction. medium. Background technique [0002] The identification of text similarity is mainly based on natural language processing (NLP) technology, and a semantic matching model can be built based on natural language processing (NLP) technology to identify text similarity. However, the semantic matching model based on traditional technical methods usually encodes the two texts to be matched to obtain two text vectors, and calculates the cosine similarity or Euclidean distance between the two text vectors to obtain the matching The similarity of two texts is matched and identified. However, this traditiona...

AI Technical Summary

Problems solved by technology

[0003] The embodiment of the present invention provides a text similarity recognition method, device, device and medium based on information interaction, aiming to solve the problem of low recognition efficiency in the process of recognizing text similarity in the prior art method

Images