Address text similarity calculation method and device

Abstract

The invention provides an address text similarity calculation method. The method comprises the steps that address structure analysis is performed on a first address text and a second address text respectively to split the first address text and the second address text into a plurality of address fields in one-to-one correspondence; text similarities between the address fields in the first addresstext and the corresponding address fields in the second address text are sequentially calculated, and summation is performed on the text similarities corresponding to all the address fields in the first address text to obtain a first similarity score; maximum values of the text similarities between the address fields in the first address text and all the address fields in the second address text are sequentially calculated, and summation is performed on the maximum values corresponding to all the address fields in the first address text to obtain a second similarity score; and the maximum value in the first similarity score and the second similarity score is calculated, and the maximum value is set as a text similarity score of the first address text and the second address text.

Description

technical field [0001] The present application relates to the field of computer applications, in particular to a method and device for calculating similarity of address text. Background technique [0002] In related technologies, address similarity calculation is used in many scenarios. For example, in the process of business application in the financial field, it is necessary to match the address entered by the user with the historically used addresses stored in the system to perform similarity calculations, and then use the similarity calculation results to verify the user submitted Whether the address has ever been used. [0003] However, in practical applications, the addresses entered by users are usually diverse, and different people may express the same address in different ways; It is of great significance to improve the accuracy of the address similarity calculation results based on the influence of the address similarity calculation results. Contents of the inv...

AI Technical Summary

Problems solved by technology

[0003] However, in practical applications, the addresses entered by users are usually diverse, and different people may express the same address in different ways; It is of great significance to improve the accuracy of the address similarity calculation results due to the influence of the address similarity calculation results.

Images