CN116955451A

CN116955451A - Intelligent mapping data management method and system

Info

Publication number: CN116955451A
Application number: CN202310842002.4A
Authority: CN
Inventors: 高秋阳; 张艺馨; 徐明星
Original assignee: Jiangsu Zhuimeng Information Technology Co ltd
Current assignee: Jiangsu Zhuimeng Information Technology Co ltd
Priority date: 2023-07-11
Filing date: 2023-07-11
Publication date: 2023-10-27

Abstract

The invention provides an intelligent mapping data management method and system, and relates to the technical field of artificial intelligence. In the invention, the first data application action information and the second data application action information of a mapping data user to be processed are determined; determining a first ordered set of application actions and a second ordered set of application actions based on the first data application action information and the second data application action information; analyzing the feature representation to be analyzed of the target storage mapping data based on the first ordered set of application actions and the second ordered set of application actions; analyzing application correlation characterization parameters of a user of the mapping data to be processed on each target storage mapping data according to the to-be-analyzed characteristic representation of the target storage mapping data; and performing target data management operation on the target storage mapping data based on the size ordering among the application correlation characterization parameters. Based on the above, the reliability of data management can be improved to some extent.

Description

Intelligent mapping data management method and system

Technical Field

The invention relates to the technical field of artificial intelligence, in particular to an intelligent mapping data management method and system.

Background

The mapping data includes mapping text data, mapping image data, etc., wherein management of the mapping data generally includes management of compression, transmission, storage, etc. of the mapping data. However, in the prior art, mapping data is generally managed correspondingly based on information such as acquired time and manual random distribution, so that the reliability is not high.

Disclosure of Invention

In view of the above, the present invention is directed to providing an intelligent mapping data management method and system, so as to improve the reliability of data management to a certain extent.

In order to achieve the above purpose, the embodiment of the present invention adopts the following technical scheme:

an intelligent mapping data management method, comprising:

determining first data application action information of a user of mapping data to be processed in an action proceeding target interval, and determining second data application action information of the user of mapping data to be processed in an action proceeding sub-interval, wherein the interval width of the action proceeding target interval is larger than that of the action proceeding sub-interval, the action proceeding target interval comprises the action proceeding sub-interval, the action proceeding target interval and the action proceeding sub-interval both belong to a time interval, the first data application action information and the second data application action information are used for reflecting application actions of the user of mapping data to be processed on stored mapping data, the first data application action information and the second data application action information belong to text data, and the stored mapping data comprise text mapping data and/or image mapping data;

Determining a first ordered set of a first number of application actions and a second ordered set of a second number of application actions based on the first data application action information and the second data application action information, wherein the first ordered set of application actions comprises application action key information of the same data layer in the action proceeding target interval, and the second ordered set of application actions comprises application action key information of each data layer in the action proceeding subinterval;

analyzing to-be-analyzed feature representations of the to-be-processed mapping data user on a third number of targets storage mapping data based on the first ordered set of the first number of application actions and the second ordered set of the second number of application actions;

analyzing application correlation characterization parameters of the to-be-processed mapping data user on each target storage mapping data by utilizing a correlation analysis network according to-be-analyzed characteristic representation of the to-be-processed mapping data user on the third number of target storage mapping data;

and performing size sorting on the application relevance characterization parameters of the mapping data to be processed on each target storage mapping data by the user of the mapping data to be processed, and performing target data management operation on the third number of target storage mapping data based on the size sorting among the application relevance characterization parameters.

In some preferred embodiments, in the above intelligent mapping data management method, the step of determining the first ordered set of the first number of application actions and the second ordered set of the second number of application actions based on the first data application action information and the second data application action information includes:

according to the corresponding data layer, based on the first data application action information, a first ordered set of a first number of application actions is analyzed;

and analyzing a second ordered set of a second number of application actions based on the second data application action information according to the corresponding data layer.

In some preferred embodiments, in the above intelligent mapping data management method, the step of analyzing the first ordered set of the first number of application actions according to the corresponding data plane based on the first data application action information includes:

performing mining operation of application actions on the first data application action information so as to output first application actions corresponding to a first number of data layers in a target action execution interval, wherein the first application actions belong to actions corresponding to user application operations on the target stored mapping data by the mapping data to be processed;

And analyzing and forming a first ordered set of the first number of application actions according to the first application actions respectively corresponding to the first number of data layers in the action execution target interval, wherein the first ordered set of the application actions comprises application action first characteristic representations, the application action first characteristic representations are formed by carrying out key data mining on data of the corresponding data layers of the first application actions, and one application action first ordered set corresponds to one data layer.

In some preferred embodiments, in the above intelligent mapping data management method, the step of analyzing the second ordered set of the second number of application actions according to the corresponding data plane based on the second data application action information includes:

performing mining operation of application actions on the second data application action information so as to output a second number of second application actions in the action execution sub-interval, wherein the second application actions belong to actions corresponding to user application operations on the target stored mapping data by the mapping data to be processed;

and analyzing and forming a second ordered set of the second number of application actions according to the second number of second application actions in the action execution sub-interval, wherein the second ordered set of the application actions comprises application action second characteristic representations, the application action second characteristic representations are formed by carrying out key data mining on data of the second application actions in each data layer, and one second application action corresponds to one second application action in one second ordered set of the application actions.

In some preferred embodiments, in the above intelligent mapping data management method, the step of analyzing the to-be-processed mapping data user to the to-be-analyzed feature representation of the third number of targets stored mapping data based on the first ordered set of the first number of application actions and the second ordered set of the second number of application actions includes:

mining a first local feature representation of the mapping data to be processed for the third number of targets stored mapping data by the user based on the first ordered set of the first number of application actions;

mining a second local feature representation of the mapping data stored by the user of the mapping data to be processed for the third number of targets based on the second ordered set of the second number of application actions;

and performing feature representation aggregation operation on the first local feature representation and the second local feature representation to form corresponding feature representations to be analyzed.

In some preferred embodiments, in the above intelligent mapping data management method, the intelligent mapping data management method further includes:

determining user identity data of the user of the mapping data to be processed;

Performing mining operation of key data on the user identity data of the user of the mapping data to be processed so as to form user identity characteristic representation of the user of the mapping data to be processed;

the step of mining out a first local feature representation of the mapping data stored by the user of the mapping data to be processed for the third number of targets based on the first ordered set of the first number of application actions, comprises:

analyzing a first local feature representation of the mapping data user to be processed for the third number of targets based on the first ordered set of the first number of application actions and the user identity feature representation of the mapping data user to be processed;

the step of mining out a second local feature representation of the mapping data stored by the user of the mapping data to be processed for the third number of targets based on the second ordered set of the second number of application actions, comprises:

analyzing a second local feature representation of the mapping data user to be processed for the third number of targets based on the second ordered set of the second number of application actions and the user identity feature representation of the mapping data user to be processed;

The step of performing feature representation aggregation operation on the first local feature representation and the second local feature representation to form corresponding feature representations to be analyzed includes:

and performing feature representation aggregation operation on the first local feature representation, the second local feature representation and the user identity feature representation of the user of the mapping data to be processed to form a corresponding feature representation to be analyzed.

digging out a storage mapping data feature representation cluster, wherein the storage mapping data feature representation cluster comprises a fourth number of storage mapping data feature representation sub-clusters, the storage mapping data feature representation sub-clusters comprise storage mapping data feature representations of the same data layer of the third number of target storage mapping data, and the storage mapping data feature representation sub-clusters comprise different data layers corresponding to the storage mapping data feature representations;

the step of analyzing a first local feature representation of the mapping data user to be processed for the third number of targets based on the first ordered set of the first number of application actions and the user identity feature representation of the mapping data user to be processed, comprises:

Analyzing a first local feature representation of the to-be-processed mapping data user on the third number of targets according to the first ordered set of the first number of application actions, the user identity feature representation of the to-be-processed mapping data user and the stored mapping data feature representation cluster;

the step of analyzing a second local feature representation of the mapping data user to be processed for the third number of targets based on the second ordered set of the second number of application actions and the user identity feature representation of the mapping data user to be processed, comprises:

analyzing a second local feature representation of the to-be-processed mapping data user on the third number of targets according to the second ordered set of the second number of application actions, the user identity feature representation of the to-be-processed mapping data user and the stored mapping data feature representation cluster;

the step of analyzing, by using a correlation analysis network, the application correlation characterization parameters of the to-be-processed mapping data user to each target storage mapping data according to the to-be-analyzed feature representation of the to-be-processed mapping data user to the to-be-analyzed third number of target storage mapping data includes:

And analyzing application correlation characterization parameters of the user of the mapping data to be processed on each target storage mapping data by utilizing the correlation analysis network according to the feature representation to be analyzed and the stored mapping data feature representation cluster.

In some preferred embodiments, in the above intelligent mapping data management method, the step of sorting the size of the application relevance characterization parameters of each target stored mapping data by the user of the mapping data to be processed, and performing the target data management operation on the third number of target stored mapping data based on the sorting of the size between the application relevance characterization parameters includes:

the user of the mapping data to be processed performs size sorting on application correlation characterization parameters of each target storage mapping data, sorts the third number of target storage mapping data according to the size sorting result to form a target storage mapping data set, and in the target storage mapping data set, the application correlation characterization parameters corresponding to the target storage mapping data with the front sorting are larger than or equal to the application correlation characterization parameters corresponding to the target storage mapping data with the rear sorting;

Traversing the target storage mapping data set;

compressing the currently traversed target storage mapping data to form compressed target storage mapping data, wherein a compression ratio corresponding to the compression operation and a corresponding traversing stage have a negative correlation corresponding relation, and the compression ratio is equal to a data volume ratio between the compressed target storage mapping data and the currently traversed target storage mapping data;

and respectively carrying out storage operation on the compressed target storage mapping data corresponding to each target storage mapping data in the target storage mapping data set.

determining an example first application action information cluster, an example second application action information cluster, a target storage mapping data cluster and an actual correlation characterization parameter cluster of the to-be-processed mapping data user, wherein the example first application action information cluster comprises a plurality of example first application action information, the example second application action information cluster comprises a plurality of example second application action information, the application action information contained in the example first application action information and the example second application action information is used for reflecting the application action of the to-be-processed mapping data user on the stored mapping data, the application action information contained in the example first application action information cluster and the example second application action information cluster belong to different action intervals, the target storage mapping data cluster comprises the third number of target storage mapping data, and the actual correlation characterization parameter cluster comprises actual correlation characterization parameters corresponding to each target storage mapping data of the to-be-processed mapping data user;

Analyzing an example first ordered set cluster of actions and an example second ordered set cluster of actions according to the example first application action information cluster and the example second application action information cluster, wherein the example first ordered set cluster of actions comprises a plurality of example first ordered sets of actions, the example second ordered set cluster of actions comprises a plurality of example second ordered sets of actions, the example first ordered set of actions comprises application action key information of the same data layer, and the example second ordered set of actions comprises application action key information of each data layer;

analyzing to-be-analyzed feature representations of the to-be-processed mapping data user on the third number of target storage mapping data based on the example action first ordered set cluster and the example action second ordered set cluster;

analyzing application relevance characterization parameters of the to-be-processed mapping data user on each target storage mapping data by utilizing a candidate relevance analysis network according to-be-analyzed characteristic representations of the to-be-processed mapping data user on the third number of target storage mapping data and the target storage mapping data clusters;

and carrying out network optimization operation on the candidate correlation analysis network based on the actual correlation characterization parameter cluster and the application correlation characterization parameters of the mapping data stored by the mapping data to be processed on each target by the user, so as to form a corresponding correlation analysis network.

The embodiment of the invention also provides an intelligent mapping data management system, which comprises a processor and a memory, wherein the memory is used for storing a computer program, and the processor is used for executing the computer program so as to realize the intelligent mapping data management method.

The intelligent mapping data management method and system provided by the embodiment of the invention can determine the first data application action information and the second data application action information of the user of the mapping data to be processed; determining a first ordered set of application actions and a second ordered set of application actions based on the first data application action information and the second data application action information; analyzing the feature representation to be analyzed of the target storage mapping data based on the first ordered set of application actions and the second ordered set of application actions; analyzing application correlation characterization parameters of a user of the mapping data to be processed on each target storage mapping data according to the to-be-analyzed characteristic representation of the target storage mapping data; and performing target data management operation on the target storage mapping data based on the size ordering among the application correlation characterization parameters. Based on the foregoing, since the application action information of the data in the intervals with different interval widths is respectively determined, and the application action first ordered set and the application action second ordered set are included in the application action key information of the same data layer in the action proceeding target interval, and the application action second ordered set includes the application action key information of each data layer in the action proceeding sub-interval, the feature mining is performed on the application action first ordered set and the application action second ordered set of the user to be processed, so that the feature representation to be analyzed of the user to be processed on the plurality of target stored mapping data can be obtained, and the application action key information of the user to be processed on each data layer in different intervals can be accurately represented, so that the application relevance representation parameters of the user to be processed on the plurality of target stored mapping data output by the relevance analysis network are more matched with the actual conditions of the user to be processed, that is, the accuracy of the application relevance representation parameters of the plurality of the target stored mapping data is higher, and therefore, the reliability of the user to be processed can be improved to a certain extent, and the reliability of the prior art is not improved.

In order to make the above objects, features and advantages of the present invention more comprehensible, preferred embodiments accompanied with figures are described in detail below.

Drawings

Fig. 1 is a block diagram of an intelligent mapping data management system according to an embodiment of the present invention.

Fig. 2 is a flowchart illustrating steps included in the intelligent mapping data management method according to an embodiment of the present invention.

Fig. 3 is a schematic diagram of each module included in the intelligent mapping data management apparatus according to the embodiment of the present invention.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are only some embodiments of the present invention, but not all embodiments of the present invention. The components of the embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.

Thus, the following detailed description of the embodiments of the invention, as presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

As shown in FIG. 1, an embodiment of the present invention provides an intelligent mapping data management system. Wherein the intelligent mapping data management system may include a memory and a processor.

In detail, the memory and the processor are electrically connected directly or indirectly to realize transmission or interaction of data. For example, electrical connection may be made to each other via one or more communication buses or signal lines. The memory may store at least one software functional module (computer program) that may exist in the form of software or firmware. The processor may be configured to execute the executable computer program stored in the memory, thereby implementing the intelligent mapping data management method provided by the embodiment of the present invention.

It should be appreciated that in some possible embodiments, the Memory may be, but is not limited to, random access Memory (Random Access Memory, RAM), read Only Memory (ROM), programmable Read Only Memory (Programmable Read-Only Memory, PROM), erasable Read Only Memory (Erasable Programmable Read-Only Memory, EPROM), electrically erasable Read Only Memory (Electric Erasable Programmable Read-Only Memory, EEPROM), and the like.

It should be appreciated that in some possible embodiments, the processor may be a general purpose processor, including a central processing unit (Central Processing Unit, CPU), a network processor (Network Processor, NP), a System on Chip (SoC), etc.; but also Digital Signal Processors (DSPs), application Specific Integrated Circuits (ASICs), field Programmable Gate Arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

It should be appreciated that in some possible embodiments, the intelligent mapping data management system may be a server with data processing capabilities.

With reference to fig. 2, an embodiment of the present invention further provides an intelligent mapping data management method, which may be applied to the above intelligent mapping data management system. The method steps defined by the flow related to the intelligent mapping data management method can be realized by the intelligent mapping data management system.

The specific flow shown in fig. 2 will be described in detail.

Step S110, determining first data application action information of a mapping data user to be processed in an action proceeding target interval, and determining second data application action information of the mapping data user to be processed in an action proceeding sub-interval.

In the embodiment of the invention, the intelligent mapping data management system can determine the first data application action information of the mapping data user to be processed in the action proceeding target interval and determine the second data application action information of the mapping data user to be processed in the action proceeding sub-interval. The interval width of the action proceeding target interval is larger than the interval width of the action proceeding sub-interval, the action proceeding target interval comprises the action proceeding sub-interval, the action proceeding target interval and the action proceeding sub-interval both belong to time intervals, that is, the first data application action information belongs to data application action information in the latest larger time interval, the second data application action information belongs to data application action information in the latest smaller time interval, the first data application action information and the second data application action information are used for reflecting application actions, such as data modification actions, data reading actions and the like, of the to-be-processed mapping data user on stored mapping data, and the first data application action information and the second data application action information belong to text data, and the stored mapping data comprise text mapping data and/or image mapping data.

Step S120, determining a first ordered set of a first number of application actions and a second ordered set of a second number of application actions based on the first data application action information and the second data application action information.

In the embodiment of the invention, the intelligent mapping data management system can determine a first ordered set of a first number of application actions and a second ordered set of a second number of application actions based on the first data application action information and the second data application action information. The first ordered set of application actions comprises application action key information of the same data layer in the action proceeding target interval, and the second ordered set of application actions comprises application action key information of all data layers in the action proceeding subinterval. The data plane may refer to a time of a data application action, an object of the data application action, a type of the data application action, an amount of object data of the data application action, a time interval between a previous data application action, and the like.

Step S130, analyzing a to-be-analyzed feature representation of the to-be-processed mapping data user for a third number of targets storage mapping data based on the first ordered set of the first number of application actions and the second ordered set of the second number of application actions.

In the embodiment of the invention, the intelligent mapping data management system can analyze the to-be-analyzed characteristic representation of the mapping data stored by the to-be-processed mapping data user on the third number of targets based on the first ordered set of the first number of application actions and the second ordered set of the second number of application actions. The third number of objects may store all of the mapping data stored, or may be part of all of the mapping data stored.

Step S140, according to the to-be-analyzed feature representation of the to-be-processed mapping data user to the third number of objects, using the correlation analysis network to analyze the application correlation characterization parameters of the to-be-processed mapping data user to each object storage mapping data.

In the embodiment of the invention, the intelligent mapping data management system can analyze the application correlation characterization parameters of the mapping data to be processed for each target storage mapping data by utilizing a correlation analysis network according to the to-be-analyzed characteristic representation of the mapping data to be processed for the third number of target storage mapping data by the mapping data to be processed. The application relevance characterization parameter may refer to a probability of a requirement of the user of the mapping data to be processed for the target storage mapping data, and if the application relevance characterization parameter is larger, the probability that the user of the mapping data to be processed needs the target storage mapping data later is larger, that is, the probability of follow-up action is larger.

Step S150, performing size sorting on the application relevance characterization parameters of the mapping data to be processed on each target storage mapping data by the user of the mapping data to be processed, and performing target data management operation on the third number of target storage mapping data based on the size sorting between the application relevance characterization parameters.

In the embodiment of the invention, the intelligent mapping data management system may perform size sorting on the application relevance characterization parameters of each target storage mapping data by the mapping data user to be processed, and perform target data management operation on the third number of target storage mapping data based on the size sorting among the application relevance characterization parameters. Because the application correlation characterization parameter characterizes the probability of the subsequent action, the target data management operation is performed based on the application correlation characterization parameter, so that the management operation is more reliable, i.e. the subsequent action is performed.

Based on the foregoing (e.g., the foregoing steps S110-S150), since the application action information of the first ordered set of application actions and the second ordered set of application actions are determined based on the data application action information of the interval of different interval widths, respectively, and since the first ordered set of application actions includes the application action key information of the same data layer in the interval of action proceeding target, the second ordered set of application actions includes the application action key information of each data layer in the interval of action proceeding sub-interval, the feature mining is performed on the application action first ordered set and the application action second ordered set of the user of the mapping data to be processed, so that the feature representation to be analyzed can accurately represent the application action key information of the user of the mapping data to be processed in each data layer of different intervals, and therefore, the application correlation characterization parameters of the user of the mapping data to be processed output by the correlation analysis network on the plurality of target storage data are more matched with the actual conditions of the user of the mapping data to be processed, i.e., the application correlation characterization parameters of the plurality of target storage data are higher, and the reliability of the mapping data to be processed is not improved to a certain extent, and the reliability is not improved.

It should be appreciated that, in some possible embodiments, the step S120 above, that is, the step of determining the first ordered set of the first number of application actions and the second ordered set of the second number of application actions based on the first data application action information and the second data application action information may further include the following specific implementation procedure:

according to the corresponding data layer, based on the first data application action information, a first number of application action first ordered sets are analyzed, and for each of the application action first ordered sets, the application action first ordered sets can include application action key information in the same data layer, for example, the first application action first ordered set includes application action key information of each first data application action in the first data layer, the second application action first ordered set includes application action key information of each first data application action in the second data layer, the third application action first ordered set includes application action key information of each first data application action in the third data layer, the fourth application action first ordered set includes application action key information of each first data application action in the fourth data layer, and the like;

And according to the corresponding data layers, based on the second data application action information, analyzing a second ordered set of a second number of application actions, wherein for each second ordered set of application actions, the second ordered set of application actions comprises application action key information of each data layer, for example, a first second ordered set of application actions comprises application action key information of a first data application action at each data layer, a second ordered set of application actions comprises application action key information of a second data application action at each data layer, and the like.

It should be appreciated that, in some possible embodiments, the step of analyzing the first ordered set of the first number of application actions according to the corresponding data plane based on the first data application action information may further include the following specific implementation procedure:

According to the first application actions corresponding to the first number of data layers in the action execution target interval, a first number of application action first ordered sets are formed through analysis, the application action first ordered sets comprise application action first feature representations, the application action first feature representations are formed through key data mining on data of the corresponding data layers of the first application actions, one application action first ordered set corresponds to one data layer, one application action first feature representation included in the application action first ordered set is formed through mining data of one first application action on the corresponding data layer, and the application action first feature representations are formed through coding network implementation.

It should be appreciated that, in some possible embodiments, the step of analyzing the second ordered set of the second number of application actions according to the corresponding data plane based on the second data application action information may further include the following specific implementation procedure:

performing mining operation of application actions on the second data application action information so as to output a second number of second application actions in the action execution sub-interval, wherein the second application actions belong to actions corresponding to user application operations on the target stored mapping data (any one target stored mapping data) by the mapping data user to be processed;

And analyzing and forming a second ordered set of the second number of application actions according to the second number of second application actions in the action execution sub-interval, wherein the second ordered set of the application actions comprises application action second characteristic representations, the application action second characteristic representations are formed by carrying out key data mining on data of the second application actions on each data layer, one application action second ordered set corresponds to one second application action, and each application action second characteristic representation included in the application action second ordered set is formed by mining data of each data layer of one second application action, such as through a coding network.

It should be appreciated that, in some possible embodiments, step S130 above, that is, the step of analyzing the to-be-analyzed feature representation of the to-be-processed mapping data user for the third number of targets stored mapping data based on the first ordered set of the first number of application actions and the second ordered set of the second number of application actions, further includes the following specific implementation procedure:

mining a first local feature representation of the mapping data to be processed, namely depth feature mining, of the third number of target storage mapping data by the user based on the first ordered set of the first number of application actions to form a corresponding depth feature representation;

Mining a second local feature representation of the mapping data to be processed, namely depth feature mining, of the third number of target storage mapping data by the user based on the second ordered set of the second number of application actions to form a corresponding depth feature representation;

the first local feature representation and the second local feature representation are subjected to feature representation aggregation operation to form corresponding feature representations to be analyzed, and the first local feature representation and the second local feature representation can be subjected to weighted superposition to form feature representations to be analyzed, wherein corresponding weighting coefficients can be determined by corresponding neural networks in a network optimization process.

It should be appreciated that in some possible embodiments, the intelligent mapping data management method may further include the following specific implementation procedures:

determining user identity data (such as data related to the mapping field, such as occupation, working age and the like of the user) of the user of the mapping data to be processed; and carrying out key data mining operation on the user identity data of the user of the mapping data to be processed, for example, the key data mining operation can be realized through a coding network so as to form user identity characteristic representation of the user of the mapping data to be processed.

Based on the foregoing, the step of mining the first local feature representation of the third number of target stored mapping data by the user of the mapping data to be processed based on the first ordered set of the first number of application actions may further include the following specific implementation process:

and analyzing the first local characteristic representation of the mapping data to be processed for the third number of targets stored mapping data based on the first ordered set of the first number of application actions and the user identity characteristic representation of the mapping data to be processed, namely fusing the user identity characteristic representation of the mapping data to be processed in the first ordered set of the first number of application actions.

Based on the foregoing, the step of mining the second local feature representation of the third number of target stored mapping data by the user of the mapping data to be processed based on the second ordered set of the second number of application actions may further include the following specific implementation process:

and analyzing second local characteristic representations of the to-be-processed mapping data users for the third number of targets storage mapping data based on the second ordered set of the second number of application actions and the user identity characteristic representations of the to-be-processed mapping data users, namely fusing the user identity characteristic representations of the to-be-processed mapping data users in the second ordered set of the second number of application actions.

Based on the foregoing, the step of performing the aggregation operation of the feature representations on the first local feature representation and the second local feature representation to form the corresponding feature representation to be analyzed may further include the following specific implementation process:

and performing feature representation aggregation operation on the first local feature representation, the second local feature representation and the user identity feature representation of the user of the mapping data to be processed to form a corresponding feature representation to be analyzed, namely aggregating the information of the three aspects of the first local feature representation, the second local feature representation and the user identity feature representation of the user of the mapping data to be processed.

a stored mapping data feature representation cluster is mined, the stored mapping data feature representation cluster comprising a fourth number of stored mapping data feature representation sub-clusters, the stored mapping data feature representation sub-clusters comprising stored mapping data feature representations of the same data plane of the third number of target stored mapping data (e.g. the target stored mapping data may be feature mined to form a stored mapping data feature representation), the data planes corresponding to the stored mapping data feature representations included between the stored mapping data feature representation sub-clusters being different, the third number being greater than or equal to the first number, that is, the data plane of the target stored mapping data being greater than or equal to the data plane of the first data application action information, and in some applications may be equal.

Based on the foregoing, the step of analyzing the first local feature representation of the mapping data to be processed for the third number of objects stored by the mapping data user based on the first ordered set of the first number of application actions and the user identity feature representation of the mapping data to be processed may further include the following specific implementation procedures:

and analyzing a first local feature representation of the stored mapping data of the third number of targets by the user of the mapping data to be processed according to the first ordered set of the first number of application actions, the user identity feature representation of the user of the mapping data to be processed and the stored mapping data feature representation cluster, namely, fusing the information of the third aspect to form the first local feature representation.

Wherein it should be understood that, in some possible embodiments, the steps of analyzing the first local feature representation of the third number of target stored mapping data by the user of the mapping data to be processed according to the first ordered set of the first number of application actions, the user identity feature representation of the user of the mapping data to be processed, and the stored mapping data feature representation cluster may further include the following specific implementation procedures:

For each application action first ordered set in the first number of application action first ordered sets, determining a storage mapping data characteristic representation sub-cluster corresponding to a data layer corresponding to the application action first ordered set in the storage mapping data characteristic representation cluster, performing cascading combination operation on the storage mapping data characteristic representation included in the storage mapping data characteristic representation sub-cluster to form a corresponding storage cascading combination characteristic representation, calculating a number product between the storage cascading combination characteristic representation and the user identity characteristic representation of the mapping data user to be processed, and performing normalization operation on the number product to form a fusion coefficient corresponding to the application action first ordered set, and dividing the number product corresponding to one data layer by the sum value of the number products of each data layer to obtain a corresponding fusion coefficient when performing normalization operation;

and for each application action first ordered set in the first number of application action first ordered sets, performing cascading combination operation on each application action first feature representation included in the application action first ordered set to form a first cascading combination feature representation corresponding to the application action first ordered set, and performing weighted superposition operation on the first cascading combination feature representation corresponding to each application action first ordered set according to a fusion coefficient (serving as a weighting coefficient) corresponding to each application action first ordered set to form a first local feature representation of the mapping data to be processed for the third number of target storage mapping data by the mapping data user, or performing further linear mapping operation on a result of the weighted superposition operation to obtain a first local feature representation.

Based on the foregoing, the step of analyzing the second local feature representation of the mapping data user to be processed for the third number of objects stored mapping data based on the second ordered set of the second number of application actions and the user identity feature representation of the mapping data user to be processed may further include the following specific implementation procedures:

and analyzing a second local feature representation of the stored mapping data of the third number of targets by the user of the mapping data to be processed according to the second ordered set of the second number of application actions, the user identity feature representation of the user of the mapping data to be processed and the stored mapping data feature representation cluster, namely, fusing the information of the third aspect to form the second local feature representation.

Wherein it should be understood that, in some possible embodiments, the steps of analyzing the second local feature representation of the third number of target stored mapping data by the user of the to-be-processed mapping data according to the second ordered set of the second number of application actions, the user identity feature representation of the user of the to-be-processed mapping data, and the stored mapping data feature representation cluster may further include the following specific implementation procedures:

For one application action second ordered set in the second plurality of application action second ordered sets (each application action second ordered set can be sequentially or synchronously processed in the same way), performing target calculation operation on the application action second ordered set so as to output to-be-processed feature representation and importance characterization parameters corresponding to the application action second ordered set;

and carrying out weighted superposition operation on the to-be-processed feature representations corresponding to the second ordered sets of the application actions according to the importance characterization parameters (serving as weighting coefficients) corresponding to the second ordered sets of the application actions, so as to output second local feature representations of the to-be-processed mapping data users for the third number of target storage mapping data.

Wherein, it should be understood that, in some possible embodiments, the step of performing, for one application action second ordered set in the second plurality of application action second ordered sets, a target computing operation on the application action second ordered set to output a pending feature representation and an importance characterizing parameter corresponding to the application action second ordered set may include:

For a first calculation stage, performing a cascade combination operation on application action second feature representations included in the application action second ordered set to form a second cascade combination feature representation corresponding to the application action second ordered set, performing a cascade combination operation on stored mapping data feature representations included in the stored mapping data feature representation cluster to form a cascade combination stored data feature representation, performing a first weighted overlap-add operation on the second cascade combination feature representation and the cascade combination stored data feature representation, performing a first shift operation (e.g., adding a shift parameter) on a result of the weighted overlap-add operation, performing an activation operation on a result of the shift operation to output a first index parameter corresponding to the calculation stage, performing a second weighted overlap-add operation on a result of the weighted overlap-add operation (e.g., adding a shift parameter), performing an activation operation on a result of the shift operation to output a second index parameter corresponding to the calculation stage, performing a first shift operation on a result of the cascade combination feature representation and the cascade combination feature representation, performing a third weighted overlap-add operation on a result of the weighted overlap-add operation, performing a weighted overlap-add operation on a result of the cascade feature representation, performing a third weighted overlap-add operation on a result of the weighted overlap-add operation, performing a third index parameter corresponding to the cascade combination feature representation, performing a fourth shift operation (for example, adding a shift parameter) on the result of the weighted overlap-add operation, performing an activation operation on the result of the shift operation, multiplying the first index parameter by the result of the activation operation to obtain a first multiplied result, multiplying the second index parameter by the user identity characteristic representation of the user of the mapping data to be processed to obtain a second multiplied result, performing an overlap-add operation on the first multiplied result and the second multiplied result to output a fourth index parameter corresponding to the calculation stage, performing an activation operation on the fourth index parameter, performing a multiplication operation on the third index parameter and the result of the activation operation to output a fifth index parameter corresponding to the calculation stage, and calculating a number product between the data characteristic representation and the fifth index parameter stored by the cascade combination to obtain a sixth index parameter corresponding to the calculation stage, wherein the weighting coefficient of the first weighted overlap-add operation, the shift parameter of the first shift operation, the weighting coefficient of the second weighted overlap-add operation, the shift parameter of the second shift operation, the weighting coefficient of the third weighted overlap-add operation, the weighting operation and the fourth shift operation and the shift operation can form an optimized network in the shift operation;

For each calculation stage after the first calculation stage, performing a first weighted overlap-add operation on a fifth index parameter corresponding to a previous calculation stage of the calculation stage and the cascade combination stored data feature representation, performing a first shift operation (e.g., adding a shift parameter) on a result of the weighted overlap-add operation, performing an activation operation on a result of the shift operation to output a first index parameter corresponding to the calculation stage, performing a second weighted overlap-add operation on a fifth index parameter corresponding to a previous calculation stage of the calculation stage and the cascade combination stored data feature representation, performing a second weighted overlap-add operation (e.g., adding a shift parameter), performing an activation operation on a result of the shift operation to output a second index parameter corresponding to the calculation stage, performing a third weighted overlap-add operation on a fifth index parameter corresponding to a previous calculation stage of the calculation stage and the cascade combination stored data feature representation, performing a third weighted overlap-add operation on a result of the fifth index parameter corresponding to output a result of the calculation stage of the shift operation, performing a multiplication operation on a result of the first index parameter corresponding to the cascade combination stored data feature representation, performing a fourth weighted overlap-add operation on a result of the fifth index parameter corresponding to the calculation stage, multiplying a second index parameter of the calculation stage and a fourth index parameter of a previous calculation stage of the calculation stage to obtain a second multiplication result, performing superposition operation on the first multiplication result and the second multiplication result to output a fourth index parameter corresponding to the calculation stage, performing activation operation on the fourth index parameter, performing multiplication operation on the third index parameter and a result of the activation operation to output a fifth index parameter corresponding to the calculation stage, calculating a number product between the cascade combination storage data feature representation and the fifth index parameter, and performing normalization operation on the number product corresponding to the current calculation stage based on the number product corresponding to each calculation stage and the previous calculation stage to obtain a sixth index parameter corresponding to the calculation stage (current calculation stage);

And taking the fifth index parameter calculated in the last calculation stage as a to-be-processed feature representation corresponding to the second ordered set of the application actions, and taking the sixth index parameter calculated in the last calculation stage as an importance characterization parameter corresponding to the second ordered set of the application actions.

Wherein, it should be understood that, in some possible embodiments, the step of performing the aggregation operation of feature representations on the first local feature representation, the second local feature representation and the user identity feature representation of the user of the mapping data to be processed to form a corresponding feature representation to be analyzed may further include the following specific implementation procedures:

performing weighted superposition operation on the first local feature representation, the second local feature representation and the user identity feature representation of the mapping data user to be processed to form a corresponding weighted superposition feature representation, wherein a weighting coefficient can be formed in a corresponding network optimization process;

performing an activation operation on the weighted overlap feature representation to form a corresponding activation output parameter;

determining a first weighting coefficient and a second weighting coefficient respectively based on the activation output parameters;

And carrying out weighted superposition on the first local feature representation and the second local feature representation based on the first weighting coefficient and the second weighting coefficient to form a corresponding feature representation to be analyzed, wherein the first local feature representation has a corresponding relationship of negative correlation between the corresponding first weighting coefficient and the activation output parameter, and the second local feature representation has a corresponding relationship of positive correlation between the corresponding second weighting coefficient and the activation output parameter.

Wherein, it should be understood that, in some possible embodiments, the step of mining out the clusters of stored mapping data features may further include the following specific implementation procedures:

extracting a target storage mapping data cluster, wherein the target storage mapping data cluster comprises the third number of target storage mapping data, and the third number of target storage mapping data comprises key mapping data of a fourth number of data layers;

performing key data mining operation on the third number of target stored mapping data according to the fourth number of data layers to output stored mapping data characteristic representation sub-clusters corresponding to each data layer;

And constructing a corresponding storage mapping data characteristic representation cluster according to the storage mapping data characteristic representation sub-cluster corresponding to each data layer.

It should be understood that, in some possible embodiments, step S140 above, that is, the step of analyzing, using the correlation analysis network, the correlation characterization parameter applied to each target stored mapping data by the user of the mapping data to be processed according to the to-be-analyzed feature representation of the mapping data stored by the user of the mapping data to be processed for the third number of targets, may further include the following specific implementation procedures:

according to the feature representation to be analyzed and the stored mapping data feature representation cluster, using the correlation analysis network, analyzing application correlation characterization parameters of the user of the mapping data to be processed on each target stored mapping data, for example, the stored mapping data feature representations included in the stored mapping data feature representation cluster can be classified and combined according to corresponding target stored mapping data to form combined feature representations corresponding to each target stored mapping data, then calculating the distance, such as cosine distance, between the feature representation to be analyzed and each combined feature representation, and determining application correlation characterization parameters of the user of the mapping data to be processed on each target stored mapping data based on the distance, wherein the application correlation characterization parameters can be inversely correlated with the distance.

It should be appreciated that, in some possible embodiments, the step S150 above, that is, the step of sorting the application relevance characterization parameters of each target stored mapping data by the user of the mapping data to be processed, and performing the target data management operation on the third number of target stored mapping data based on the sorting of the application relevance characterization parameters, may further include the following specific implementation procedure:

traversing the target storage mapping data set;

compressing the currently traversed target storage mapping data to form compressed target storage mapping data, wherein a compression ratio corresponding to the compression operation and a corresponding traversing stage have a negative correlation corresponding relation, and the compression ratio is equal to a data volume ratio between the compressed target storage mapping data and the currently traversed target storage mapping data, namely, the larger the corresponding application correlation characterization parameter is, the larger the compression ratio of the compressed target storage mapping data is;

determining an example first application action information cluster, an example second application action information cluster, a target storage mapping data cluster and an actual correlation characterization parameter cluster of the to-be-processed mapping data user, wherein the example first application action information cluster comprises a plurality of example first application action information, the example second application action information cluster comprises a plurality of example second application action information, the example first application action information and the example second application action information comprise application action information used for reflecting the application action of the to-be-processed mapping data user on the stored mapping data, the application action information included in the example first application action information cluster and the example second application action information cluster belong to different action intervals, the target storage mapping data cluster comprises the third number of target storage mapping data, the actual correlation characterization parameter cluster comprises actual correlation characterization parameters corresponding to each target storage mapping data of the to-be-processed mapping data user, and the intervals corresponding to the example first application action information cluster and the example second application action information cluster are earlier than the target intervals;

Analyzing an example first ordered set cluster of actions and an example second ordered set cluster of actions according to the example first application action information cluster and the example second application action information cluster, wherein the example first ordered set cluster of actions comprises a plurality of example first ordered sets of actions, the example second ordered set cluster of actions comprises a plurality of example second ordered sets of actions, the example first ordered set of actions comprises application action key information of the same data layer, and the example second ordered set of actions comprises application action key information of each data layer as described in the previous related description;

analyzing to-be-analyzed feature representations of the to-be-processed mapping data user on the third number of target storage mapping data based on the example action first ordered set cluster and the example action second ordered set cluster, as previously described in relation thereto;

analyzing application relevance characterization parameters of the user of the to-be-processed mapping data on each target storage mapping data by utilizing a candidate relevance analysis network according to-be-analyzed characteristic representations of the user of the to-be-processed mapping data on the third number of target storage mapping data and the target storage mapping data cluster, as described in the previous relevance;

Based on the actual correlation characterization parameter cluster and the application correlation characterization parameters of the mapping data to be processed for each target storage mapping data, performing network optimization operation on the candidate correlation analysis network to form a corresponding correlation analysis network, and calculating a corresponding network optimization cost index based on the actual correlation characterization parameter cluster and the application correlation characterization parameters (difference between the actual correlation characterization parameters) of the mapping data to be processed for each target storage mapping data for the mapping data user; and carrying out optimization adjustment operation on the network parameters of the candidate correlation analysis network based on the network optimization cost index to form a corresponding correlation analysis network, wherein the optimization adjustment operation is carried out along the direction of reducing the network optimization cost index.

With reference to fig. 3, an embodiment of the present invention further provides an intelligent mapping data management apparatus, which is applicable to the above intelligent mapping data management system. Wherein, the intelligent mapping data management apparatus may include:

the action information determining module is used for determining first data application action information of a user of mapping data to be processed in an action proceeding target interval and determining second data application action information of the user of mapping data to be processed in an action proceeding sub-interval, wherein the interval width of the action proceeding target interval is larger than that of the action proceeding sub-interval, the action proceeding target interval comprises the action proceeding sub-interval, the action proceeding target interval and the action proceeding sub-interval both belong to a time interval, the first data application action information and the second data application action information are used for reflecting application actions of the user of mapping data to be processed on stored mapping data, the first data application action information and the second data application action information belong to text data, and the stored mapping data comprise text mapping data and/or image mapping data;

The action ordered set determining module is used for determining a first ordered set of a first number of application actions and a second ordered set of a second number of application actions based on the first data application action information and the second data application action information, wherein the first ordered set of the application actions comprises application action key information of the same data layer in the action execution target interval, and the second ordered set of the application actions comprises application action key information of each data layer in the action execution sub-interval;

the feature representation analysis module is used for analyzing to-be-analyzed feature representations of the to-be-processed mapping data user on a third number of targets stored mapping data based on the first ordered set of the first number of application actions and the second ordered set of the second number of application actions;

the application correlation analysis module is used for analyzing application correlation characterization parameters of the to-be-processed mapping data user on each target storage mapping data by utilizing a correlation analysis network according to-be-analyzed characteristic representation of the to-be-processed mapping data user on the third number of target storage mapping data;

and the target data management module is used for sequencing the magnitude of the application correlation characteristic parameters of the mapping data to be processed for each target storage mapping data by the user, and performing target data management operation on the third number of target storage mapping data based on the magnitude sequencing among the application correlation characteristic parameters.

In summary, the method and the system for intelligent mapping data management provided by the invention can determine the first data application action information and the second data application action information of the user of the mapping data to be processed; determining a first ordered set of application actions and a second ordered set of application actions based on the first data application action information and the second data application action information; analyzing the feature representation to be analyzed of the target storage mapping data based on the first ordered set of application actions and the second ordered set of application actions; analyzing application correlation characterization parameters of a user of the mapping data to be processed on each target storage mapping data according to the to-be-analyzed characteristic representation of the target storage mapping data; and performing target data management operation on the target storage mapping data based on the size ordering among the application correlation characterization parameters. Based on the foregoing, since the application action information of the data in the intervals with different interval widths is respectively determined, and the application action first ordered set and the application action second ordered set are included in the application action key information of the same data layer in the action proceeding target interval, and the application action second ordered set includes the application action key information of each data layer in the action proceeding sub-interval, the feature mining is performed on the application action first ordered set and the application action second ordered set of the user to be processed, so that the feature representation to be analyzed of the user to be processed on the plurality of target stored mapping data can be obtained, and the application action key information of the user to be processed on each data layer in different intervals can be accurately represented, so that the application relevance representation parameters of the user to be processed on the plurality of target stored mapping data output by the relevance analysis network are more matched with the actual conditions of the user to be processed, that is, the accuracy of the application relevance representation parameters of the plurality of the target stored mapping data is higher, and therefore, the reliability of the user to be processed can be improved to a certain extent, and the reliability of the prior art is not improved.

The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An intelligent mapping data management method, comprising:

2. The intelligent mapping data management method as set forth in claim 1, wherein the step of determining a first ordered set of a first number of application actions and a second ordered set of a second number of application actions based on the first data application action information and the second data application action information comprises:

3. The intelligent mapping data management method as set forth in claim 2, wherein the step of analyzing the first ordered set of the first number of application actions based on the first data application action information according to the corresponding data plane includes:

4. The intelligent mapping data management method as set forth in claim 2, wherein the step of analyzing a second ordered set of a second number of application actions based on the second data application action information according to the corresponding data plane includes:

5. The intelligent mapping data management method of claim 1, wherein the step of analyzing the representation of the to-be-processed mapping data user's to-be-analyzed features of the third number of targets stored mapping data based on the first ordered set of the first number of application actions and the second ordered set of the second number of application actions comprises:

6. The intelligent mapping data management method as set forth in claim 5, wherein the intelligent mapping data management method further comprises:

determining user identity data of the user of the mapping data to be processed;

7. The intelligent mapping data management method as set forth in claim 6, wherein the intelligent mapping data management method further comprises:

8. The intelligent mapping data management method of claim 1, wherein the step of ordering the application relevance characterization parameters of each target stored mapping data by the user of the mapping data to be processed, and performing target data management operations on the third number of target stored mapping data based on the ordering of the magnitudes between the application relevance characterization parameters, comprises:

Traversing the target storage mapping data set;

9. The intelligent mapping data management method as set forth in any one of claims 1 to 8, wherein the intelligent mapping data management method further includes:

10. An intelligent mapping data management system comprising a processor and a memory, the memory for storing a computer program, the processor for executing the computer program to implement the intelligent mapping data management method of any of claims 1-9.