Approximate query method for predicting future queries based on offline learning historical queries

Abstract

The invention discloses an approximate query method for predicting future queries based on offline learning historical queries, and belongs to the field of database technology application. The methodcomprises the following steps of: 1, modeling underlying data distribution by offline learning historical queries: 1.1, splitting complex SQL queries into simple queries; 1.2, extracting features contained in the SQL query and an approximate result; 1.3, modeling underlying data distribution by utilizing historical queries, an approximate result and a real result; 2, predicting a result of newly-arrived queries on line: 2.1, splitting the newly-arrived queries into simple queries; 2.2, predicting a query result on line for each simple query; 2.3, combining the prediction results of the simplequery, and outputting a final prediction result. The complexity of constructing the underlying data distribution model can be reduced, the accuracy of the underlying data distribution model can be improved, and query accuracy can be improved without excessive time consumption.

Description

technical field [0001] The invention belongs to the technical field of databases, and in particular relates to an approximate query method for predicting future queries based on off-line learning historical queries. Background technique [0002] Social media, mobile devices, and wireless sensors are generating large spatiotemporal data at an unprecedented rate, and aggregated queries on big data have become the basis of many decision support systems. Traditional databases process data queries in a blocking manner, and it takes a long time to return an accurate result after the user submits the query, so the query efficiency is low. On the one hand, the query time required to return an accurate result is unacceptable to the user. On the other hand, a decision support system does not need a completely accurate answer in many scenarios, as long as the approximate result and the real result can make the same decision. Users are more inclined to choose an answer that is not com...

AI Technical Summary

Problems solved by technology

[0004]The purpose of this invention is to solve the problems of complex process and low prediction accuracy brought by the existing technology of approximate query by learning the underlying data distribution, and proposes a An Approximate Query Method Based on Offline Learning Historical Query Prediction for Future Query

Images