Data matching method applied to data cleaning tool

Abstract

The invention relates to the technical field of data processing, and discloses a data matching method applied to a data cleaning tool, which comprises the following steps of: (1) importing a pluralityof differentiated samples and samples to generate model parameters according to data in an input table and an output table, and configuring the model parameters to a model table; (2) carrying out legality judgment on a query data primary key; (3) performing data type judgment on the non-null value and performing query to generate a record; (4) identifying data in all records according to a deep self-learning algorithm, and generating a universal matching model; and (5) using ETL to read, reading a model table to clean data.. According to the invention, on the basis of a big data sample, an intelligent learning algorithm is used to realize automatic matching of data in each independent system in a data extraction process; data scattered by patients in each system can be completely extracted to a hospital clinical data center, and compared with a method of manually judging data rules, the method is quick in data processing and saves a large amount of time and manpower.

Description

technical field [0001] The invention relates to the technical field of data processing, in particular to a data matching method applied to a data cleaning tool. Background technique [0002] At present, with the development of medical informatization, the construction of hospital clinical data center has been widely carried out, which is mainly used to establish patients' previous medical records, doctors' clinical decision-making support and doctor's scientific research decision-making support. The construction of a hospital clinical data center currently mainly uses data cleaning tools to disperse the hospital in various independent systems, such as: hospital information system, electronic medical record system, laboratory information system, medical imaging information system, nursing information system, etc. Due to the independent construction of each system, there is no unified patient identification specification among the systems. In the process of using data cleanin...

AI Technical Summary

Problems solved by technology

[0003] The present invention provides a data matching method applied to data cleaning tools in view of the shortcomings of insufficient manual energy and cumbersome data configuration in the prior art

Images