Database indexing method and system

Abstract

The invention discloses a database index method and system. The method comprises the steps that index keywords are obtained firstly; according to the keyword, the database is searched to acquire query information or a query cycle ; then the query information or the query cycle is compared with predefined thresholds, afterwards, previous service information is obtained according to the comparative result; according to the previous service information, a plurality of probability index keywords are acquired, afterwards, the database is searched for obtaining target data according to the probability index keywords and the index keywords. By the means of counting the occurrence probability of every service, unknown variate of services with a high occurrence probability can be known, which can help access path match with more index entries and increase access efficiency of this kind of services, thereby improving the treatment efficiency of the whole services.

Description

technical field [0001] The invention relates to the field of data indexing, in particular to a database indexing method and system. Background technique [0002] In the existing data query, most of the known conditions are used to filter the data that meets the known conditions in the database through the massive data for users to choose. The more known conditions, the more accurate the target data obtained by screening; if the known conditions If there are less data and there are many data in the database that meet the known conditions, the user cannot effectively and quickly locate the target data they need, and at the same time, the matching of massive data also causes a certain load on the hardware and software; for this reason In the current field of data indexing, when there are few known conditions, that is, there are many unknown conditions, the amount of data satisfied by the matching part of the index is large, resulting in a long time to access the database; espec...

AI Technical Summary

Problems solved by technology

[0002] In the existing data query, most of the known conditions are used to filter the data that meets the known conditions in the database through the massive data for users to choose. The more known conditions, the more accurate the target data obtained by screening; if the known conditions When there are fewer data in the database and there are many data that meet the known conditions in the database, the user cannot effectively and quickly locate the target data they need, and at the same time, the matching of massive data also causes a certain load on the hardware and software; for this reason In the current field of data indexing, when there are few known conditions, that is, there are many unknown conditions, the amount of data satisfied by the matching part of the index is large, resulting in a long time to access the database; especially when processing such data in batches, it takes Sometimes the time window is very large, which cannot meet the business processing requirements

[0003] The existing common solution is to use the intersection of its characteristics to access the database for all business types, so that there are few database index fields that can be matched, so the returned query results are large, resulting in slow subsequent processing

Images