Query sorting method and device under multiple space-time conditions, equipment and storage medium

Abstract

The invention discloses a query sorting method under multiple space-time conditions, which comprehensively considers spatial similarity and spatial diversity, combines a self-information amount index,a similarity index and a spatial diversity index, and uses a Z-score evaluation method to nondimensionalize the index value of different target eigenvalues to obtainthe query value corresponding to each target eigenvalue, and achieves the query sorting of target features in multi-time-space state. That is to say, a target characteristic value meeting a space-time query condition is retrieved through sorting, further the most matched query result is determined after the dimensionless processing of the index values under different evaluation indexes. The method can be widely used for a varietyof target feature value queries, such as terminal information, intelligent IC card information, identity card ID information, and the like, can integrate the spatial similarity and the spatial diversity, and can obtain an accurate matching result.

Description

technical field [0001] The invention relates to the field of query and sorting of time series data, in particular to a query and sorting method, device, equipment and storage medium under multi-temporal and spatial conditions. Background technique [0002] At present, there are many time-series data calculation models, including TSDB time-series database, etc. Many monitoring systems use TSDB as a database system to store a large amount of various index data that is strictly time-increasing and has a very simple structure to a certain extent. , the stored data structure simply means that a metric will only have one value at a certain point in time and space, without complex structures (nesting, hierarchy, etc.) and relationships (association, primary and foreign keys, etc.), the existing time-series based The target query of the database is to query and sort the data of a single time and space, such as finding the time and space where a specified Mac address appears, or sort...

AI Technical Summary

Problems solved by technology

[0002] At present, there are many time-series data calculation models, including TSDB time-series database, etc. Many monitoring systems use TSDB as a database system to store a large amount of various index data that is strictly time-increasing and has a very simple structure to a certain extent. , the stored data structure simply means that a metric will only have one value at a certain point in time and space, without complex structures (nesting, hierarchy, etc.) and relationships (association, primary and foreign keys, etc.), the existing time-series based The target query of the database is to query and sort the data of a single time and space, such as finding the time and space where a specified Mac address appears, or sorting its indicators, and does not involve the query of uncertain target eigenvalues ​​in a multi-time and space state , and the query process does not comprehensively consider the spatial similarity and spatial diversity, so it is necessary to propose a comprehensive evaluation method to achieve the sorting of the query results, so as to obtain the target eigenvalue data information of the uncertain target

Images