Address difference judgment method based on rules and statistical word segmentation

Abstract

The invention discloses an address difference judgment method based on rules and statistical word segmentation. The method comprises the following steps: obtaining national three-level administrativeregion address data; acquiring user address data input by a user, and dividing the user address data into three levels of administrative region addresses and detailed addresses of provinces and cities; complementing or correcting the default or nonstandard administrative region address by taking the national third-level administrative region address data as a standard; performing word segmentationprocessing on the detailed address; merging the administrative region address data after completion or correction with the detailed address data after word segmentation processing; judging whether the two addresses are the same address or not by comparing whether the address digital feature values are equal or not and calculating the address text similarity. According to the method, a large number of manual labeling tasks and computer storage space can be saved.

Description

technical field [0001] The invention relates to the technical field of geographic information processing, in particular to a method for judging address difference based on rules and statistical word segmentation. Background technique [0002] The application number is CN201910246155.6, and the name is a Chinese invention patent application for address standardization processing method and equipment. The solution includes an address standardization processing method and device, and the method includes: obtaining an initial address; performing preprocessing on the initial address to obtain a preprocessed address; extracting at least one piece of normalized address information recorded in the preprocessed address The address names in the database, the standardized address information database is pre-established, which records the standard names and level information of the national addresses at all levels and the alias names of the addresses at all levels, and the addresses at ...

AI Technical Summary

Problems solved by technology

The accuracy of the algorithm depends on the accuracy of the labeling, and labeling is a labor-intensive project

Word segmentation is an indispensable step in natural language processing in the field of artificial intelligence, but the large-scale address labeling data is not disclosed, which makes this technical difficulty mainly reflected in the acquisition of labeling data

Images