A data normalization method, device and medium for identity recognition

Abstract

The present invention provides a data normalization method, device and medium for identity recognition. The method first constructs a corresponding identity attribute data set; then judges whether there is a matching rule for identifying the identity attribute data set, and if so, uses the rule matching method to match Identify the identity attribute data set, if not, use the path matching method to identify the identity attribute data set; then calculate the credibility of at least two data records in the obtained identification results, if the credibility reaches a certain threshold, then The at least two data records are stored in the database after being normalized. According to the different characteristics of the data records, the present invention adaptively selects whether to use the rule matching algorithm or the path matching algorithm, can quickly normalize the identities conforming to the rule characteristics, and can more comprehensively normalize the identities without obvious consistent characteristics , this method will greatly improve the accuracy of identity normalization, and a rule matching algorithm and a path matching algorithm are proposed.

Description

technical field [0001] The invention relates to the technical field of computer data processing, in particular to a data normalization method, device and storage medium for identification. Background technique [0002] In the era of mobile Internet, massive amounts of data are generated every day, such as accommodation, driving, travel in real life, instant messaging in the virtual world, third-party payment, etc.; these data are large in volume and have no unified identification characteristics, resulting in various types of data Scattered and irrelevant, how to automatically analyze and normalize the identities of related data has become a difficulty in improving the ability and efficiency of massive data analysis. [0003] Since the data is constantly increasing with the increase of various application types, and there is no unified identification feature, the existing identity normalization methods currently on the market mainly use manual configuration to judge the rela...

AI Technical Summary

Problems solved by technology

[0002] In the era of mobile Internet, massive amounts of data are generated every day, such as accommodation, driving, travel in real life, instant messaging in the virtual world, third-party payment, etc.; these data are large in volume and have no unified identification characteristics, resulting in various types of data Scattered and irrelevant, how to automatically analyze and normalize the identities of relevant data has become a difficulty in improving the analysis ability and efficiency of massive data

[0003] Since the data is constantly increasing with the increase of various application types, and there is no unified identification feature, the existing identity normalization methods currently on the market mainly use manual configuration to judge the relationship between data sources one by one. These technologies cannot meet the complex analysis needs in reality, and their technical defects are as follows:

[0004] 1) The manual configuration method requires a lot of business research time, and is prone to errors and omissions, which greatly affects the efficiency and quality of data analysis work;

[0005] 2) Single matching rule: In many cases, the data cannot be related only by a single rule, resulting in insufficient coverage of data that can be related in the end, which seriously affects the use effect and user experience of the system

Images