CN116910729B

CN116910729B - Nuclear body processing method and system applied to multi-organization architecture

Info

Publication number: CN116910729B
Application number: CN202311170222.3A
Authority: CN
Inventors: 张磊
Original assignee: Hangzhou Yiyi Network Technology Co ltd
Current assignee: Hangzhou Yiyi Network Technology Co ltd
Priority date: 2023-09-12
Filing date: 2023-09-12
Publication date: 2023-12-08
Anticipated expiration: 2043-09-12
Also published as: CN116910729A

Abstract

The invention provides a core body processing method and a system applied to a multi-organization architecture, and relates to the technical field of artificial intelligence. In the invention, the preset network operation record data corresponding to the network operation object is determined; constructing an object relation knowledge network based on preset network operation record data; determining the identified knowledge net members with the identification information and the non-identified knowledge net members without the identification information; carrying out the evaluation operation of the identification information on the knowledge network members, and outputting the evaluation result of the identification information; analyzing target identification information of the non-identified knowledge net members based on the identification information and the identification information evaluation result; and extracting the result of the matching operation of the identity information of the network operation object corresponding to the non-identification knowledge network member by each organization structure, and determining whether the identity information passes the matching operation or not based on the result. Based on the above, the efficiency of the core processing can be improved to some extent.

Description

Nuclear body processing method and system applied to multi-organization architecture

Technical Field

The invention relates to the technical field of artificial intelligence, in particular to a nuclear body processing method and system applied to a multi-organization architecture.

Background

And the identity information is checked and matched through a plurality of organization structures, and then whether the identity information passes the check and match operation or not can be determined based on the result, so that the reliability of the identity information check can be improved. However, in the prior art, since the verification operation of the identity information is generally performed for each object, there is a problem in that the efficiency of the verification process is not high.

Disclosure of Invention

Accordingly, the present invention is directed to a method and a system for processing a core for a multi-organization architecture, so as to improve the efficiency of the core processing to a certain extent.

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

a method for processing a core body applied to a multi-organization architecture, comprising:

determining a plurality of network operation objects corresponding to preset network operation, and determining preset network operation record data corresponding to each network operation object, wherein the preset network operation record data records the process of the preset network operation by the network operation object in a text form;

performing distribution state analysis processing of designated network operations based on preset network operation record data corresponding to each network operation object, and constructing a corresponding object relation knowledge network based on analyzed distribution state characterization data, wherein the designated network operations are preset network operations matched with configured target conditions, and the two knowledge network members associated with a connecting line segment in the object relation knowledge network have the same designated network operations between the network operation objects corresponding to the two knowledge network members;

Performing knowledge network member identification operation on the object relation knowledge network to determine an identification knowledge network member with identification information and a non-identification knowledge network member without identification information, wherein the identification information is used for reflecting whether a corresponding network operation object performs abnormal network operation;

based on the data corresponding to each knowledge net member in the object relation knowledge net, carrying out the evaluation operation of the identification information on the knowledge net member so as to output the evaluation result of the identification information corresponding to each knowledge net member;

analyzing the target identification information of each non-identified knowledge net member based on the identification information of each identified knowledge net member and the identification information evaluation result corresponding to each knowledge net member;

and for each non-identified knowledge net member with the target identification information representing abnormal network operation, extracting the result of checking and matching operation on the identity information of the network operation object corresponding to the non-identified knowledge net member by each organization architecture, and determining whether the identity information of the network operation object passes the checking and matching operation or not based on the result.

In some preferred embodiments, in the above-mentioned core processing method applied to a multi-organization architecture, the step of performing an evaluation operation of identification information on the knowledge network members based on the data corresponding to each knowledge network member in the object relationship knowledge network to output an evaluation result of the identification information corresponding to each knowledge network member includes:

performing key information mining operation on data corresponding to each knowledge net member in the object relation knowledge net to output a knowledge net member description vector corresponding to each knowledge net member, wherein the data corresponding to the knowledge net member is attribute data of the knowledge net member, and the attribute data comprises object identity information and object action information of a network operation object corresponding to the knowledge net member;

for each knowledge net member in the object relation knowledge net, carrying out fusion operation on a knowledge net member description vector corresponding to the knowledge net member and a knowledge net member description vector corresponding to an associated knowledge net member of the knowledge net member to form a fusion member description vector corresponding to the knowledge net member;

performing evaluation operation of identification information based on fusion member description vectors corresponding to the knowledge network members so as to output an identification information evaluation result corresponding to each knowledge network member;

The step of performing a fusion operation on the knowledge net member description vector corresponding to the knowledge net member and the knowledge net member description vector corresponding to the knowledge net member associated with the knowledge net member for each knowledge net member in the object relationship knowledge net to form a fusion member description vector corresponding to the knowledge net member includes:

for any one of the knowledge net members in the object relation knowledge net, performing linear mapping operation on knowledge net member description vectors corresponding to the knowledge net members based on target mapping parameter distribution to form first mapping description vectors corresponding to the knowledge net members;

based on the target mapping parameter distribution, performing linear mapping operation on the knowledge network member description vectors corresponding to each associated knowledge network member of the knowledge network members respectively to form each second mapping description vector corresponding to the knowledge network member;

performing a averaging operation on each second mapping description vector corresponding to the knowledge network member to form a mean mapping description vector corresponding to the knowledge network member;

and performing cascading combination operation on the first mapping description vector corresponding to the knowledge network member and the mean mapping description vector corresponding to the knowledge network member to form a fusion member description vector corresponding to the knowledge network member.

In some preferred embodiments, in the above-mentioned core processing method applied to a multi-organization architecture, the step of analyzing the target identification information of each non-identified knowledge net member based on the identification information of each identified knowledge net member and the identification information evaluation result corresponding to each knowledge net member includes:

constructing a corresponding identification information evaluation result cluster based on the identification information evaluation results corresponding to the knowledge network members;

constructing an identification information knowledge network corresponding to the identification information evaluation result cluster based on the identification information evaluation result cluster, wherein the identification information knowledge network comprises identification information evaluation results in the identification information evaluation result cluster and has the same architecture as the object relation knowledge network;

and carrying out the operation of transmitting and updating the identification information on the identification information knowledge net based on the identification information of each identification knowledge net member so as to output the identification information of each non-identification knowledge net member.

In some preferred embodiments, in the above-mentioned core processing method applied to a multi-organization architecture, the core processing method applied to a multi-organization architecture further includes:

Extracting a typical object relation knowledge network, wherein the typical object relation knowledge network comprises typical identification knowledge network members with identification information and typical non-identification knowledge network members without identification information; and performing the following steps by using the candidate identification information evaluation model:

based on the description vector fusion operation and the identification information transfer updating operation, the identification information evaluation results corresponding to the typical knowledge network members are output;

performing a lottery operation of identification information of each typical identification knowledge net member to form a preferred typical knowledge net member and an non-preferred typical knowledge net member, wherein the preferred typical knowledge net member is a typical knowledge net member with the lottery identification information, and the non-preferred typical knowledge net member is a typical identification knowledge net member other than the preferred typical knowledge net member;

determining a typical identification information evaluation result cluster according to the identification information of the preferred typical knowledge network member and the identification information evaluation result corresponding to each typical knowledge network member;

analyzing the evaluation identification information of each non-preferred typical knowledge net member based on the identification information of each preferred typical knowledge net member and the typical identification information evaluation result cluster;

And carrying out network optimization operation on the candidate identification information evaluation models based on the identification information of the non-preferred typical knowledge network members and the distinguishing information between the evaluation identification information of the non-preferred typical knowledge network members so as to form target identification information evaluation models corresponding to the candidate identification information evaluation models, wherein the target identification information evaluation models are used for carrying out the evaluation operation of the identification information.

In some preferred embodiments, in the above-mentioned core processing method applied to a multi-organization architecture, the step of determining a typical identification information evaluation result cluster according to identification information of the preferred typical knowledge network member and identification information evaluation results corresponding to each typical knowledge network member includes:

constructing a corresponding typical identification information evaluation result cluster based on the identification information evaluation results corresponding to each typical knowledge network member;

and updating the typical identification information evaluation result cluster according to the identification information of the preferred typical knowledge network member to form a new typical identification information evaluation result cluster, wherein the identification information evaluation result corresponding to the preferred typical knowledge network member is included in any one of the preferred typical knowledge network members in the typical identification information evaluation result cluster, and the identification information of the preferred typical knowledge network member is included in the new typical identification information evaluation result cluster.

In some preferred embodiments, in the above-mentioned core processing method applied to a multi-organization architecture, the step of analyzing the estimated identification information of each non-preferred canonical knowledge net member based on the identification information of each preferred canonical knowledge net member and the canonical identification information estimation result cluster includes:

maintaining the identification information of each preferred typical knowledge net member in the typical identification information evaluation result cluster;

and carrying out identification information transmission updating operation on the constructed typical identification information knowledge network according to the identification information of each preferred typical knowledge network member so as to output the evaluation identification information of each non-preferred typical knowledge network member, wherein the typical identification information knowledge network is formed based on the typical identification information evaluation result cluster, and the typical identification information knowledge network belongs to a knowledge network with the same architecture as the typical object relation knowledge network.

extracting a typical object relation knowledge network, wherein the typical object relation knowledge network comprises typical identification knowledge network members and typical non-identification knowledge network members, and the typical identification knowledge network members and the typical non-identification knowledge network members have identification information;

The following steps are performed using the candidate identification information evaluation model:

constructing a corresponding typical identification information evaluation result cluster according to the identification information of the typical identification knowledge network members and the identification information evaluation results corresponding to the typical knowledge network members;

analyzing the estimated identification information of each typical non-identified knowledge net member based on the identification information of each typical identified knowledge net member and the typical identification information estimated result cluster;

performing network optimization operation on the candidate identification information evaluation model based on the identification information of the typical non-identification knowledge network member and the distinguishing information between the evaluation identification information of the typical non-identification knowledge network member so as to form a target identification information evaluation model corresponding to the candidate identification information evaluation model, wherein the target identification information evaluation model is used for performing the evaluation operation of the identification information, and performing the network optimization operation on the candidate identification information evaluation model comprises determining a network optimization index corresponding to the candidate identification information evaluation model based on the distinguishing information;

The calculation process of the network optimization index comprises the following steps:

constructing a first identification information distribution matrix based on the identification information of the typical non-identification knowledge network member, and constructing a second identification information distribution matrix based on the evaluation identification information of the typical non-identification knowledge network member; performing matrix parameter negative correlation mapping operation on the first identification information distribution matrix to form a corresponding third identification information distribution matrix, and performing matrix parameter negative correlation mapping operation on the second identification information distribution matrix to form a corresponding fourth identification information distribution matrix; performing a target operation of matrix parameters on the second identification information distribution matrix to form a corresponding fifth identification information distribution matrix, and performing a target operation of matrix parameters on the fourth identification information distribution matrix to form a corresponding sixth identification information distribution matrix; and performing differential information analysis operation based on the first identification information distribution matrix, the third identification information distribution matrix, the fifth identification information distribution matrix and the sixth identification information distribution matrix to form a network optimization index corresponding to the candidate identification information evaluation model.

In some preferred embodiments, in the above-mentioned core processing method applied to a multi-organization architecture, the step of constructing a corresponding typical identification information evaluation result cluster according to the identification information of the typical identification knowledge net member and the identification information evaluation result corresponding to each typical knowledge net member includes;

and updating the typical identification information evaluation result cluster according to the identification information of the typical identification information member so as to form a new typical identification information evaluation result cluster, wherein the identification information evaluation result corresponding to the typical identification information member is included in any typical identification information evaluation result cluster, and the identification information of the typical identification information member is included in the new typical identification information evaluation result cluster.

In some preferred embodiments, in the above-mentioned core processing method applied to a multi-organization architecture, the step of analyzing the estimated identification information of each of the typical non-identified knowledge net members based on the identification information of each of the typical identified knowledge net members and the typical identification information estimation result cluster includes:

Maintaining the identification information of each identification knowledge net member in the typical identification information evaluation result cluster;

and carrying out identification information transmission updating operation on the constructed typical identification information knowledge net according to the identification information of each identification knowledge net member so as to output the evaluation identification information of each typical non-identification knowledge net member, wherein the typical identification information knowledge net is formed based on the typical identification information evaluation result cluster, and the typical identification information knowledge net belongs to a knowledge net with the same architecture as the typical object relation knowledge net.

The embodiment of the invention also provides a core body processing system applied to the multi-organization architecture, 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 core body processing method applied to the multi-organization architecture.

The method and the system for processing the nuclear bodies applied to the multi-organization architecture can determine the preset network operation record data corresponding to the network operation object; constructing an object relation knowledge network based on preset network operation record data; determining the identified knowledge net members with the identification information and the non-identified knowledge net members without the identification information; carrying out the evaluation operation of the identification information on the knowledge network members, and outputting the evaluation result of the identification information; analyzing target identification information of the non-identified knowledge net members based on the identification information and the identification information evaluation result; and extracting the result of the matching operation of the identity information of the network operation object corresponding to the non-identification knowledge network member by each organization structure, and determining whether the identity information passes the matching operation or not based on the result. Based on the foregoing, the identification information can be checked and matched only for the non-identified knowledge net members, so that the efficiency of the kernel processing can be improved to a certain extent, and the problem of low efficiency in the prior art is solved.

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 a core processing system applied to a multi-organization architecture according to an embodiment of the present invention.

Fig. 2 is a flowchart illustrating steps included in a method for processing a core body applied to a multi-organization architecture according to an embodiment of the present invention.

Fig. 3 is a schematic diagram of each module included in a core processing apparatus applied to a multi-organization architecture according to an 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 a core processing system applied to a multi-organization architecture. The core processing system applied to the multi-organization architecture can comprise 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, so as to implement the core processing method applied to the multi-organization architecture provided by the embodiment of the present invention.

Alternatively, 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.

Alternatively, 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.

Alternatively, in some possible embodiments, the core processing system applied to the multi-organization architecture may be a server with data processing capability.

With reference to fig. 2, an embodiment of the present invention further provides a core processing method applied to a multi-organization architecture, which can be applied to the above-mentioned core processing system applied to the multi-organization architecture. The method steps defined by the flow related to the core processing method applied to the multi-organization architecture can be realized by the core processing system applied to the multi-organization architecture.

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

Step S110, a plurality of network operation objects corresponding to the preset network operation are determined, and preset network operation record data corresponding to each network operation object is determined.

In the embodiment of the invention, the core processing system applied to the multi-organization architecture can determine a plurality of network operation objects corresponding to preset network operation and determine preset network operation record data corresponding to each network operation object. The preset network operation record data records the process of the network operation object for performing the preset network operation in the form of text, for example, the time, the position and the like of the network operation object for performing the preset network operation can be recorded. In addition, the preset network operation may be a data access operation performed on the database, or the like.

Step S120, performing distribution state analysis processing of the specified network operation based on the preset network operation record data corresponding to each network operation object, and constructing a corresponding object relationship knowledge network based on the analyzed distribution state characterization data.

In the embodiment of the invention, the core processing system applied to the multi-organization architecture can perform distribution state analysis processing of the designated network operation based on the preset network operation record data corresponding to each network operation object, and construct a corresponding object relationship knowledge network based on the analyzed distribution state characterization data. The specified network operation is preset network operation matched with the configured target condition, and the two knowledge network members associated with the connecting line segments in the object relation knowledge network have the same specified network operation between the network operation objects corresponding to the two knowledge network members. For example, when data access operation is performed on the database in the same time period between the network operation objects corresponding to the two knowledge net members, a connection line segment is configured between the two knowledge net members to represent the association. And if the data access operation is not performed on the database in the same time period between the network operation objects corresponding to the two knowledge net members, the connection line segments are not configured between the two knowledge net members, and the connection line segments are not associated.

Step S130, carrying out knowledge net member identification operation on the object relation knowledge net to determine the identified knowledge net members with identification information and the non-identified knowledge net members without identification information.

In the embodiment of the invention, the core processing system applied to the multi-organization architecture can perform knowledge net member identification operation on the object relation knowledge net so as to determine the identified knowledge net members with identification information and the non-identified knowledge net members without identification information. The identification information is used for reflecting whether the corresponding network operation object performs abnormal network operation. That is, a knowledge net member having identification information may be marked as an identified knowledge net member, and a knowledge net member not having identification information may be marked as a non-identified knowledge net member. The specific type of the abnormal network operation is not limited, such as illegal data access and other operations.

Step S140, based on the data corresponding to each knowledge network member in the object relationship knowledge network, performing an evaluation operation of the identification information on the knowledge network member, so as to output an evaluation result of the identification information corresponding to each knowledge network member.

In the embodiment of the present invention, the core processing system applied to the multi-organization architecture may perform the evaluation operation of the identification information on the knowledge network members based on the data corresponding to each knowledge network member in the object relationship knowledge network, so as to output the evaluation result of the identification information corresponding to each knowledge network member, that is, whether the evaluation is performed on the abnormal network operation.

Step S150, analyzing the target identification information of each non-identified knowledge net member based on the identification information of each identified knowledge net member and the identification information evaluation result corresponding to each knowledge net member.

In the embodiment of the present invention, the core processing system applied to the multi-organization architecture may analyze the target identification information of each non-identified knowledge net member, that is, whether the final result of the abnormal network operation is performed, based on the identification information of each identified knowledge net member and the identification information evaluation result corresponding to each knowledge net member.

Step S160, for each non-identified knowledge net member whose target identification information characterizes the abnormal network operation, extracting the result of the matching operation of the identity information of the network operation object corresponding to the non-identified knowledge net member by each of the organization structures, and determining whether the identity information of the network operation object passes the matching operation based on the result.

In the embodiment of the invention, the core processing system applied to the multi-organization architecture can characterize each non-identified knowledge net member subjected to abnormal network operation for the target identification information, extract the result of the verification matching operation of the identity information of the network operation object corresponding to the non-identified knowledge net member by the multi-organization architecture, and determine whether the identity information of the network operation object passes the verification matching operation or not based on the result. The organization architecture can be different personnel, namely, the identity information is checked and matched through different personnel, and can also be different equipment, namely, the identity information is checked and matched through different equipment. For example, it is determined that the identity information of the network operation object does not pass the checkmatching operation as long as the result of one organization structure indicates that the checkmatching operation is not passed, or it is determined that the identity information of the network operation object passes the checkmatching operation as long as the result of a plurality of organization structures indicates that the checkmatching operation is passed.

Based on the foregoing (i.e., the above steps S110 to S160), the determination of the non-identified knowledge net member is performed first, so that the verification and matching operation of the identity information can be performed only for the non-identified knowledge net member, which can improve the efficiency of the kernel processing to a certain extent, thereby improving the problem of low efficiency in the prior art.

Optionally, in some possible embodiments, step S140 in the foregoing description, that is, the step of performing, based on the data corresponding to each knowledge network member in the object relationship knowledge network, an evaluation operation of identification information on the knowledge network member to output an evaluation result of identification information corresponding to each knowledge network member, further includes the following specific implementation contents:

performing key information mining operation on data corresponding to each knowledge net member in the object relation knowledge net to output a knowledge net member description vector corresponding to each knowledge net member, wherein the data corresponding to the knowledge net member is attribute data of the knowledge net member, the attribute data comprises object identity information and object action information of a network operation object corresponding to the knowledge net member, the object identity information can refer to registration information of a corresponding network platform, the object action information can refer to network operation performed on the corresponding network platform, and the registration information can be recorded in a text form, so that the key information mining operation can be performed on the data to form a corresponding knowledge net member description vector, the key information mining operation can be a coding operation, such as convolution operation;

For each knowledge net member in the object relation knowledge net, carrying out fusion operation on a knowledge net member description vector corresponding to the knowledge net member and a knowledge net member description vector corresponding to an associated knowledge net member of the knowledge net member to form a fusion member description vector corresponding to the knowledge net member, so that the representation capability of the fusion member description vector can be enhanced;

the evaluation operation of the identification information is performed based on the fusion member description vector corresponding to each knowledge network member, so as to output an identification information evaluation result corresponding to each knowledge network member, for example, the classification output may be performed based on the fusion member description vector so as to obtain the identification information evaluation result corresponding to the knowledge network member, for example, the fusion member description vector may be subjected to full connection processing first so as to form a corresponding full connection description vector, then the full connection description vector may be subjected to activation processing so as to form a corresponding identification information evaluation result, the full connection processing may be implemented based on a full connection network, and the activation processing may be implemented based on a softmax function or the like.

Optionally, in some possible embodiments, the step of performing, for each of the knowledge net members in the object relationship knowledge net, a fusion operation on a knowledge net member description vector corresponding to the knowledge net member and a knowledge net member description vector corresponding to an associated knowledge net member of the knowledge net member to form a fused member description vector corresponding to the knowledge net member may further include the following specific implementation content:

for any one of the knowledge net members in the object relation knowledge net, performing linear mapping operation on knowledge net member description vectors corresponding to the knowledge net members based on target mapping parameter distribution to form first mapping description vectors corresponding to the knowledge net members, wherein the target mapping parameter distribution can be used as network parameters of a corresponding neural network to be formed in a network optimization process, and the linear mapping operation can be to multiply the target mapping parameter distribution with the knowledge net member description vectors to obtain the corresponding first mapping description vectors;

based on the target mapping parameter distribution, performing linear mapping operation on the knowledge net member description vector corresponding to each associated knowledge net member of the knowledge net members to form each second mapping description vector corresponding to the knowledge net member, for example, the knowledge net member description vector corresponding to the associated knowledge net member and the target mapping parameter distribution may be multiplied, in addition, the associated knowledge net members may refer to other knowledge net members associated with the knowledge net members through connection line segments, for example, the associated knowledge net members are associated only through one connection line segment, or are associated through connection line segments with the number smaller than the preset number, for example, the connection line segments are associated between the knowledge net member 1 and the knowledge net member 2, the connection line segments are associated between the knowledge net member 2 and the knowledge net member 3, and the connection line segments are not associated between the knowledge net member 1 and the knowledge net member 3, so that the knowledge net member 1 and the knowledge net member 2 may interact as the associated knowledge net member, the knowledge net member 2 and the knowledge net member 3 may interact as the associated knowledge net member, if the preset number is equal to 2, the knowledge net member 1 and the knowledge net member 3 cannot be associated with the knowledge net member 2 as the knowledge net member 2 if the preset number is equal to 2, and the knowledge net member 3 cannot be associated with the knowledge net member 1;

and performing cascading combination operation on the first mapping description vector corresponding to the knowledge network member and the mean mapping description vector corresponding to the knowledge network member to form a fusion member description vector corresponding to the knowledge network member, wherein the fusion member description vector can be { the first mapping description vector, the mean mapping description vector }, or in other embodiments, focus feature analysis operation can be performed on the first mapping description vector corresponding to the knowledge network member based on the mean mapping description vector to form a corresponding focus description vector, and then cascading combination operation can be performed on the first mapping description vector and the focus description vector.

Optionally, in some possible embodiments, step S150 in the foregoing description, that is, the step of analyzing the target identification information of each non-identified knowledge net member based on the identification information possessed by each identified knowledge net member and the identification information evaluation result corresponding to each knowledge net member, may further include the following specific implementation matters:

Constructing a corresponding identification information evaluation result cluster based on the identification information evaluation results corresponding to the knowledge network members, that is, the identification information evaluation result cluster may include the identification information evaluation results corresponding to the knowledge network members;

constructing an identification information knowledge network corresponding to the identification information evaluation result cluster based on the identification information evaluation result cluster, wherein the identification information knowledge network comprises identification information evaluation results in the identification information evaluation result cluster and has the same architecture as the object relationship knowledge network, for example, knowledge network members of the identification information knowledge network are identification information evaluation results corresponding to the knowledge network members, and whether connection line segments are associated with the knowledge network members or not can be determined based on whether connection line segments are associated with the corresponding knowledge network members in the object relationship knowledge network or not, so that the knowledge network has consistency;

Optionally, in some possible embodiments, the step of performing an update operation of transferring the identification information on the identification information knowledge network based on the identification information of each of the identified knowledge network members to output the identification information of each of the non-identified knowledge network members may further include the following specific implementation matters:

Performing matrixing operation on the identification information knowledge network to form a corresponding first matrix, wherein each matrix parameter is used for reflecting whether a connecting line segment is associated between knowledge network members in the identification information knowledge network, if the matrix parameter is equal to 1, the connecting line segment is represented, and if the matrix parameter is equal to 0, the connecting line segment is represented; in addition, the first matrix may be a symmetric matrix, and each matrix parameter on a diagonal is used to reflect whether a knowledge net member is associated with a connecting line segment between itself, for example, configured to be associated with a connecting line segment;

determining a second matrix corresponding to the first matrix, wherein each matrix parameter on a diagonal line in the second matrix is used for reflecting the number of other knowledge net members, which are associated with the connecting line segments, of the knowledge net members at the corresponding positions in the first matrix;

multiplying the-0.5 th power of the second matrix, the first matrix and the-0.5 th power of the second matrix to output a corresponding third matrix;

performing matrixing operation on the identification information of each identified knowledge net member to form a corresponding fourth matrix, and configuring a fifth matrix identical to the fourth matrix;

For the first iterative computation, performing multiplication operation on the third matrix and the fourth matrix to form a corresponding sixth matrix, and performing weighted summation computation on the sixth matrix and the fifth matrix to form a target matrix corresponding to the first iterative computation, wherein the weighted coefficient of the weighted summation computation can be used as a parameter of a corresponding neural network to be formed in the process of network optimization;

for each iteration calculation except the first iteration calculation, performing multiplication operation on the third matrix and a target matrix corresponding to the previous iteration calculation to form a corresponding sixth matrix, performing weighted summation calculation on the sixth matrix and the fifth matrix to form a target matrix corresponding to the iteration calculation, and determining identification information of each non-identification knowledge net member based on the target matrix corresponding to the last iteration calculation; in addition, the number of iterative calculations performed is not limited, and may be positively correlated with the number of connecting line segments in the identification information knowledge network, for example.

Optionally, in some possible embodiments, each step may be implemented based on a target identification information evaluation model, based on which the core processing method applied to the multi-organization architecture may further include the following specific implementation matters:

Extracting a typical object relation knowledge network, wherein the typical object relation knowledge network comprises typical identification knowledge network members with identification information and typical non-identification knowledge network members without identification information, the typical object relation knowledge network can refer to a typical object relation knowledge network, namely an object relation knowledge network which is used as a basis for network optimization, the typical identification knowledge network members can refer to typical identification knowledge network members, namely identification knowledge network members which are used as a basis for network optimization, and the typical non-identification knowledge network members can refer to typical non-identification knowledge network members, namely non-identification knowledge network members which are used as a basis for network optimization;

based on the description vector fusion operation and the identification information transfer updating operation (as described in the previous related description), the identification information evaluation result corresponding to each typical knowledge network member is output;

performing a lottery operation of identification information of each typical identification knowledge net member to form a preferred typical knowledge net member and an non-preferred typical knowledge net member, wherein the preferred typical knowledge net member is a typical knowledge net member with the selected identification information (random lottery, random lottery or lottery according to a configuration rule can be performed), and the non-preferred typical knowledge net member is a typical identification knowledge net member other than the preferred typical knowledge net member;

based on the identification information of the non-preferred typical knowledge network member and the distinguishing information between the evaluation identification information of the non-preferred typical knowledge network member, performing network optimization operation on the candidate identification information evaluation model to form a target identification information evaluation model corresponding to the candidate identification information evaluation model, wherein the target identification information evaluation model is used for performing the evaluation operation of the identification information; that is, the candidate identification information evaluation model may be subjected to a network optimization operation based on the direction in which the discrimination information is lowered to form the target identification information evaluation model.

Optionally, in some possible embodiments, the step of determining a cluster of typical identification information evaluation results according to the identification information possessed by the preferred typical knowledge network member and the identification information evaluation results corresponding to each typical knowledge network member may further include the following specific implementation matters:

Constructing a corresponding typical identification information evaluation result cluster based on the identification information evaluation results corresponding to each typical knowledge network member, that is, the typical identification information evaluation result cluster comprises; the identification information evaluation results corresponding to the typical knowledge network members;

and updating the typical identification information evaluation result cluster according to the identification information of the preferred typical knowledge network member to form a new typical identification information evaluation result cluster, wherein any one of the preferred typical knowledge network members comprises an identification information evaluation result corresponding to the preferred typical knowledge network member, and the new typical identification information evaluation result cluster comprises identification information of the preferred typical knowledge network member, namely, the identification information evaluation result corresponding to the preferred typical knowledge network member is replaced and the like based on the identification information of the preferred typical knowledge network member.

Optionally, in some possible embodiments, the step of analyzing the evaluation identification information of each non-preferred exemplary knowledge net member based on the identification information of each preferred exemplary knowledge net member and the exemplary identification information evaluation result cluster may further include the following specific implementation matters:

and carrying out identification information transmission updating operation on the constructed typical identification information knowledge network according to the identification information of each preferred typical knowledge network member so as to output the evaluation identification information of each non-preferred typical knowledge network member, wherein the typical identification information knowledge network is formed based on the typical identification information evaluation result clusters, and the typical identification information knowledge network belongs to a knowledge network with the same architecture as the typical object relation knowledge network, as described in the previous related description.

based on the description vector fusion operation and the identification information transfer updating operation, the identification information evaluation results corresponding to the typical knowledge network members are output, as described in the related description;

based on the distinguishing information between the identification information of the typical non-identification knowledge network member and the evaluation identification information of the typical non-identification knowledge network member, performing network optimization operation on the candidate identification information evaluation model to form a target identification information evaluation model corresponding to the candidate identification information evaluation model, wherein the target identification information evaluation model is used for performing the identification information evaluation operation, and performing network optimization operation on the candidate identification information evaluation model comprises determining a network optimization index corresponding to the candidate identification information evaluation model based on the distinguishing information, for example, performing optimization adjustment of model parameters along the direction of reducing the network optimization index.

Optionally, in some possible embodiments, the step of constructing a corresponding typical identification information evaluation result cluster according to the identification information possessed by the typical identification knowledge network member and the identification information evaluation result corresponding to each typical knowledge network member may further include the following specific implementation contents:

constructing a corresponding typical identification information evaluation result cluster based on the identification information evaluation results corresponding to the typical knowledge network members, that is, the typical identification information evaluation result cluster comprises the identification information evaluation results corresponding to the typical knowledge network members;

and updating the typical identification information evaluation result cluster according to the identification information of the typical identification information member to form a new typical identification information evaluation result cluster, wherein the identification information evaluation result corresponding to the typical identification information member is included in any one of the typical identification information evaluation result clusters, and the identification information of the typical identification information member is included in the new typical identification information evaluation result cluster, that is, the identification information evaluation result corresponding to the typical identification information member is replaced based on the identification information of the typical identification information member.

Optionally, in some possible embodiments, the step of analyzing the estimated identification information of each of the typical non-identified knowledge net members based on the identification information of each of the typical identified knowledge net members and the typical identification information estimation result cluster may further include the following specific implementation matters:

maintaining the identification information of each identified knowledge net member in the typical identification information evaluation result cluster, that is, in the process of carrying out the transmission updating operation of the identification information, the identification information of each identified knowledge net member is not updated;

and carrying out identification information transmission updating operation on the constructed typical identification information knowledge net according to the identification information of each identification knowledge net member so as to output the evaluation identification information of each typical non-identification knowledge net member, wherein the typical identification information knowledge net is formed based on the typical identification information evaluation result cluster, and the typical identification information knowledge net belongs to a knowledge net with the same architecture as the typical object relation knowledge net, as described in the previous related description.

Optionally, in some possible embodiments, the calculation process of the network optimization index may further include the following specific implementation matters:

constructing a first identification information distribution matrix based on the identification information of the typical non-identification knowledge network member, and constructing a second identification information distribution matrix based on the evaluation identification information of the typical non-identification knowledge network member;

performing matrix parameter negative correlation mapping operation on the first identification information distribution matrix to form a corresponding third identification information distribution matrix, and performing matrix parameter negative correlation mapping operation on the second identification information distribution matrix to form a corresponding fourth identification information distribution matrix; for example, for performing a matrix parameter negative correlation mapping operation, the sum of the matrix parameters after the negative correlation mapping operation and the matrix parameters before the negative correlation mapping operation may be equal to a fixed parameter, such as 1;

performing a target operation of matrix parameters on the second identification information distribution matrix to form a corresponding fifth identification information distribution matrix, and performing a target operation of matrix parameters on the fourth identification information distribution matrix to form a corresponding sixth identification information distribution matrix, wherein the target operation may be, for example, a logarithmic operation;

And performing differential information analysis operation based on the first identification information distribution matrix, the third identification information distribution matrix, the fifth identification information distribution matrix and the sixth identification information distribution matrix to form a network optimization index corresponding to the candidate identification information evaluation model.

Optionally, in some possible embodiments, the step of performing a differential information analysis operation based on the first identifier information distribution matrix, the third identifier information distribution matrix, the fifth identifier information distribution matrix, and the sixth identifier information distribution matrix to form a network optimization index corresponding to the candidate identifier information evaluation model may further include the following specific implementation matters:

performing multiplication operation on the first identification information distribution matrix and the fifth identification information distribution matrix to output a corresponding first parameter matrix, and performing multiplication operation on the third identification information distribution matrix and the sixth identification information distribution matrix to output a corresponding second parameter matrix;

and finally, determining a network optimization index corresponding to the candidate identification information evaluation model based on the target sum value, for example, the network optimization index and the target sum value can have a negative correlation relationship, for example, the sum value between the network optimization index and the target sum value is equal to a fixed value, such as 1, 0 and the like.

With reference to fig. 3, an embodiment of the present invention further provides a core processing device applied to a multi-organization architecture, which is applicable to the above-mentioned core processing system applied to the multi-organization architecture. The core body processing device applied to the multi-organization architecture can comprise:

the object data determining module is used for determining a plurality of network operation objects corresponding to preset network operation, and determining preset network operation record data corresponding to each network operation object, wherein the preset network operation record data records the process of the preset network operation by the network operation objects in a text form;

the knowledge network construction module is used for carrying out distribution state analysis processing of designated network operation based on preset network operation record data corresponding to each network operation object, and constructing a corresponding object relation knowledge network based on analyzed distribution state characterization data, wherein the designated network operation is preset network operation matched with a configured target condition, and two knowledge network members associated with a connecting line segment in the object relation knowledge network have the same designated network operation between network operation objects corresponding to each other;

The knowledge network member identification module is used for carrying out knowledge network member identification operation on the object relation knowledge network so as to determine an identified knowledge network member with identification information and a non-identified knowledge network member without identification information, wherein the identification information is used for reflecting whether a corresponding network operation object carries out abnormal network operation or not;

the evaluation module of the identification information is used for carrying out the evaluation operation of the identification information on the knowledge network members based on the data corresponding to each knowledge network member in the object relation knowledge network so as to output the evaluation result of the identification information corresponding to each knowledge network member;

the identification information determining module is used for analyzing the target identification information of each non-identified knowledge net member based on the identification information of each identified knowledge net member and the identification information evaluation result corresponding to each knowledge net member;

the verification matching module is used for representing each non-identified knowledge net member subjected to abnormal network operation for the target identification information, extracting the result of the verification matching operation for the identity information of the network operation object corresponding to the non-identified knowledge net member by the organization architecture, and determining whether the identity information of the network operation object passes the verification matching operation or not based on the result.

In summary, the method and the system for processing the core body applied to the multi-organization architecture provided by the invention can determine the preset network operation record data corresponding to the network operation object; constructing an object relation knowledge network based on preset network operation record data; determining the identified knowledge net members with the identification information and the non-identified knowledge net members without the identification information; carrying out the evaluation operation of the identification information on the knowledge network members, and outputting the evaluation result of the identification information; analyzing target identification information of the non-identified knowledge net members based on the identification information and the identification information evaluation result; and extracting the result of the matching operation of the identity information of the network operation object corresponding to the non-identification knowledge network member by each organization structure, and determining whether the identity information passes the matching operation or not based on the result. Based on the foregoing, the identification information can be checked and matched only for the non-identified knowledge net members, so that the efficiency of the kernel processing can be improved to a certain extent, and the problem of low efficiency in the prior art is solved.

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. A method for processing a core body applied to a multi-organization architecture, comprising:

2. The method for processing a core body applied to a multi-organization architecture according to claim 1, wherein the step of performing an evaluation operation of identification information on the knowledge network members based on the data corresponding to each of the knowledge network members in the object relation knowledge network to output an evaluation result of the identification information corresponding to each of the knowledge network members comprises:

3. The method for processing a core body applied to a multi-organization architecture according to claim 1, wherein the step of analyzing the target identification information of each of the non-identified knowledge net members based on the identification information possessed by each of the identified knowledge net members and the identification information evaluation result corresponding to each of the knowledge net members comprises:

4. The method for processing a core body applied to a multi-organization architecture according to claim 1, wherein the method for processing a core body applied to a multi-organization architecture further comprises:

5. The method for processing a core body applied to a multi-organization architecture according to claim 4, wherein the step of determining a cluster of typical identification information evaluation results according to the identification information of the preferred typical knowledge network member and the identification information evaluation results corresponding to each typical knowledge network member comprises:

6. The method for processing a core body applied to a multi-organization architecture according to claim 4, wherein the step of analyzing the evaluation identification information of each of the non-preferred canonical knowledge network members based on the identification information possessed by each of the preferred canonical knowledge network members and the canonical identification information evaluation result cluster comprises:

7. The method for processing a core body applied to a multi-organization architecture according to claim 1, wherein the method for processing a core body applied to a multi-organization architecture further comprises:

8. The method for processing a core body applied to a multi-organization architecture according to claim 7, wherein the step of constructing a corresponding typical identification information evaluation result cluster according to the identification information of the typical identification knowledge net member and the identification information evaluation result corresponding to each typical knowledge net member comprises the steps of;

9. The method for processing a core body applied to a multi-organization architecture according to claim 7, wherein the step of analyzing the evaluation identification information of each of the typical non-identification knowledge net members based on the identification information possessed by each of the typical identification knowledge net members and the typical identification information evaluation result cluster comprises:

10. A core processing system for application in a multi-organization architecture, comprising a processor and a memory, the memory for storing a computer program, the processor for executing the computer program to implement the method of any of claims 1-9.