Data processing method of enterprise informatization system

Abstract

The invention discloses a data processing method of an enterprise informatization system. The method comprises the steps of constructing index for the data and searching the date, wherein the index for the data is constructed according to the following steps: customizing attention aspects according to requirements; building a metadata model, at least comprising the following aspects such as keywords, types, type rating and associated words, aiming at the attention aspects; verifying the metadata model according to the standards set in advance, and converting the metadata model meeting the standards into a data base model; conducting dirty data processing; and enabling the data obtained by the last step to be persistent to be an index file. The data processing method does not concern a data base, the result obtained through searching can be traced, the data redundancy is avoided and the searching purpose is strong.

Description

technical field [0001] The present invention relates to a data processing method, in particular to a data processing method of an enterprise information system. The method includes indexing and retrieving data. Background technique [0002] With the deepening of enterprise information application, the amount of data in the enterprise information system is increasing day by day, it is inevitable that dirty data will appear, and the data quality will decline accordingly; on the other hand, with the increasingly complex data source and data dependency relationship, The efficiency with which enterprises can extract information through various systems is also gradually decreasing. [0003] In the existing enterprise informatization system, it is mainly a search informatization scheme based on full-text document search or an informatization scheme based on database full-text search. The full-text document search mainly includes the following steps: grabbing the content of the doc...

AI Technical Summary

Problems solved by technology

[0008] 3) There is a large amount of data redundancy in full-text document search. By comparing and filtering the redundant parts, the configuration of the index server or cluster is required to be very high and the cost is expensive;

[0010] 1) The data is not well targeted, and there will be a large amount of unnecessary dirty data;

[0011] 2) The full-text search and indexing of the database takes a long time and takes up a lot of system memory and space;

[0012] 3) There is no correlation in the full-text search of the database, and the relationship between information and information cannot be known. Information and information exist in an isolated manner

[0013] 4) The full-text search of the database must have a primary key for each table. The implication is that only the physical tables of the database can be indexed, and the virtual tables (views, stored procedures) in the system cannot be indexed.

[0014] 5) Database full-text search will rely too much on SQL statements, so there will be cross-database problems

Images