1. A data resource sharing method applied to big data service is characterized by comprising the following steps:

extracting resource content by combining historical service data resources through a data resource sharing thread to be updated to obtain target resource content distribution of the historical service data resources;

determining dynamic sharing state information corresponding to the historical service data resources through the data resource sharing thread to be updated in combination with the target resource content distribution;

and updating the thread parameters of the data resource sharing thread by combining the effective sharing state information and the dynamic sharing state information of the historical service data resources to obtain the updated data resource sharing thread.

2. The data resource sharing method applied to big data services according to claim 1, wherein the obtaining of the target resource content distribution of the historical service data resources by performing resource content extraction through a data resource sharing thread to be updated in combination with the historical service data resources comprises:

acquiring historical service data resources and effective sharing state information corresponding to the historical service data resources, wherein the effective sharing state information corresponding to the historical service data resources comprises effective item classification information of each resource item in the historical service data resources;

inputting the historical service data resources into a resource preprocessing thread in the data resource sharing threads to be updated, and extracting resource contents of the historical service data resources through a resource content extraction sub-thread of the resource preprocessing thread to obtain target resource content distribution of the historical service data resources;

the determining, by the data resource sharing thread to be updated, dynamic sharing state information corresponding to the historical service data resource in combination with the target resource content distribution includes:

determining a sub-thread through the sharing state of the resource preprocessing thread, and determining dynamic sharing state information corresponding to the historical service data resource based on the content distribution of the target resource, wherein the dynamic sharing state information corresponding to the historical service data resource comprises dynamic item classification information of each resource item in the historical service data resource;

the updating the thread parameters of the data resource sharing thread by combining the effective sharing state information and the dynamic sharing state information of the historical service data resources to obtain the updated data resource sharing thread comprises:

determining, by a sharing state classification sub-thread in the data resource sharing thread to be updated, a first determination result that the dynamic sharing state information belongs to idle sharing state information of the historical service data resource and a second determination result that the effective sharing state information belongs to idle sharing state information of the historical service data resource based on effective sharing state information and dynamic sharing state information of the historical service data resource;

and updating the thread parameters of the data resource sharing thread based on the first judgment result and the second judgment result to obtain the updated data resource sharing thread.

3. The method as claimed in claim 2, wherein the obtaining of the target resource content distribution of the historical service data resource by the resource content extraction sub-thread of the resource preprocessing thread extracting the resource content of the historical service data resource comprises:

and extracting the resource content of the historical service data resource through a resource content extraction sub-thread of the resource preprocessing thread to obtain the resource content distribution of a plurality of service scenes of the historical service data resource, and integrating the resource content distribution of the plurality of service scenes to obtain the target resource content distribution of the historical service data resource.

4. The data resource sharing method applied to big data services according to claim 3, wherein the resource content extraction sub-thread comprises a resource content integration unit and at least two resource content extraction units connected in sequence;

the resource content extraction sub-thread of the resource preprocessing thread extracts the resource content of the historical service data resource to obtain resource content distribution of a plurality of service scenes of the historical service data resource, and integrates the resource content distribution of the plurality of service scenes to obtain target resource content distribution of the historical service data resource, and the resource content extraction sub-thread comprises the following steps:

extracting resource contents of the historical service data resources through the resource content extraction units which are connected in sequence to obtain resource content distribution of different service scenes output by different resource content extraction units;

and integrating the resource content distribution of different service scenes according to the sequence from the last resource content extraction unit to the first resource content extraction unit by the resource content integration unit to obtain the target resource content distribution of the historical service data resources.

5. The data resource sharing method applied to big data services according to claim 4, wherein the number of the resource content integration units is one less than that of the resource content extraction units;

the integrating, by the resource content integrating unit, resource content distribution of the different service scenes according to a sequence from a last resource content extracting unit to a first resource content extracting unit to obtain target resource content distribution of the historical service data resource includes:

performing service scene adjustment processing on the resource content distribution input into the current resource content integration unit to obtain adjusted resource content distribution, wherein the adjusted resource content distribution is the same as the service scene of the resource content distribution extracted by the last resource content extraction unit in the resource content distribution which does not participate in the integration processing; if the current resource content integration unit is the last resource content integration unit, inputting the resource content distribution of the current integration unit as the resource content distribution extracted by the last resource content extraction unit;

and performing resource content distribution integration on the adjusted resource content distribution and the resource content distribution extracted by the last resource content extraction unit in the resource content distribution which does not participate in integration processing through the current resource content integration unit, and inputting the integrated resource content distribution into the previous resource content integration unit, wherein if the current resource content integration unit is the first resource content integration unit, the integrated resource content distribution obtained by the current resource content integration unit is the target resource content distribution.

6. The method for sharing data resources applied to big data services according to any one of claims 2 to 5, wherein the obtaining historical service data resources and the effective sharing state information corresponding to the historical service data resources comprises:

obtaining a sample service data resource of a data resource sharing thread to be updated, wherein the resource characteristics of the sample service data resource comprise: effective sample sharing state information of sample business data resources, wherein the effective sample sharing state information comprises effective item classification information of all resource items in the sample business data resources;

acquiring at least one preset size of service data resource from sample service data resources, and taking the acquired service data resource as a historical service data resource of the data resource sharing thread to be updated;

and acquiring the effective sharing state information of the historical service data resources from the effective sample sharing state information of the sample service data resources based on the resource distribution condition of the historical service data resources in the corresponding sample service data resources.

7. The method for sharing data resources applied to big data services according to any of claims 2 to 5, wherein the effective sharing state information is the service data resources with an effective sharing state, and the dynamic sharing state information is the service data resources with a dynamic sharing state;

correspondingly, the determining, by a sharing state classification sub-thread in the data resource sharing thread to be updated, a first determination result that the dynamic sharing state information belongs to the idle sharing state information of the historical service data resource and a second determination result that the effective sharing state information belongs to the idle sharing state information of the historical service data resource based on the effective sharing state information and the dynamic sharing state information of the historical service data resource includes:

the historical service data resources and the corresponding service data resources with the effective sharing state are correlated to obtain the service data resources with the effective sharing state after correlation, and the historical service data resources and the corresponding service data resources with the dynamic sharing state are correlated to obtain the service data resources with the dynamic sharing state after correlation;

acquiring sub-threads through the resource content of the sharing state classification sub-threads, and acquiring resource content information of a first service data resource from the associated service data resource with the dynamic sharing state;

determining, by the state discrimination unit of the shared state classification sub-thread, a service data resource having a dynamic shared state corresponding to the service data resource having the dynamic shared state after association based on the resource content information of the first service data resource, and a first discrimination result of the service data resource belonging to the idle shared state of the historical service data resource;

acquiring the sub-thread through the resource content of the sharing state classification sub-thread, and acquiring resource content information of a second service data resource from the service data resource which is associated and has the effective sharing state;

and determining, by the state discrimination unit of the shared state classification sub-thread, a service data resource having an effective shared state corresponding to the service data resource having an effective shared state after association based on the resource content information of the second service data resource, and a second discrimination result of the service data resource belonging to the idle shared state of the historical service data resource.

8. The method according to any one of claims 2 to 5, wherein the updating the thread parameter of the data resource sharing thread based on the first and second determination results to obtain an updated data resource sharing thread comprises:

determining a first sharing state identification deviation of the resource preprocessing thread based on the first judgment result;

determining a second sharing state identification deviation of the resource preprocessing thread based on an information comparison result between the dynamic item classification information and the effective item classification information of the same resource item in the effective sharing state information and the dynamic sharing state information of the historical service data resource;

updating the thread parameters of the resource preprocessing thread based on the first sharing state identification deviation and the second sharing state identification deviation to obtain an updated resource preprocessing thread;

determining a state classification deviation of the shared state classification sub-thread based on the first judgment result and the second judgment result;

updating thread parameters of the shared state classification sub-thread based on the state classification bias;

correspondingly, the dynamic item classification information comprises a dynamic item classification and a dynamic judgment result in the dynamic item classification;

correspondingly, the determining a second sharing state identification deviation of the resource preprocessing thread based on the information comparison result between the dynamic item classification information and the valid item classification information of the same resource item in the valid sharing state information and the dynamic sharing state information of the historical service data resource includes:

determining effective item classification of each resource item in the historical service data resource and dynamic discrimination results of each resource item in the dynamic sharing state information in the corresponding effective item classification based on the effective sharing state information and the dynamic sharing state information of the historical service data resource;

and determining a second sharing state identification deviation of the resource preprocessing thread based on the effective item classification of the resource items of the historical service data resources and the dynamic judgment result in the effective item classification.

9. The method according to claim 7, wherein the number of the resource sharing requirement conditions of the service data resources in the effective sharing state and the service data resources in the dynamic sharing state is the same, and the local service data resources corresponding to the resource sharing requirement conditions of the service data resources in the effective sharing state include: whether each resource item of the historical service data resource is effective item classification information corresponding to the resource sharing requirement condition or not; the local service data resources corresponding to the resource sharing requirement conditions of the service data resources with the dynamic sharing state comprise: each resource item of the historical service data resource is a dynamic judgment result of a dynamic item classification corresponding to a resource sharing requirement condition;

correspondingly, the associating the historical service data resource with the corresponding service data resource with the effective sharing state to obtain the service data resource with the effective sharing state after association, and associating the historical service data resource with the corresponding service data resource with the dynamic sharing state to obtain the service data resource with the dynamic sharing state after association, includes:

the historical service data resources and the local service data resources of the service data resources with the effective sharing state sharing requirement conditions are used as the local service data resources of the service data resources with the effective sharing state sharing requirement conditions after association, and the historical service data resources and the service data resources with the effective sharing state are associated to obtain the service data resources with the effective sharing state after association;

and associating the historical service data resources and the service data resources with the dynamic sharing state to obtain the service data resources with the dynamic sharing state after association.

10. A big data server is characterized by comprising a processing engine, a network module and a memory; the processing engine and the memory communicate through the network module, the processing engine reading a computer program from the memory and operating to perform the method of any of claims 1-9.

Background

Shared economy (Shared economy) refers to an owner of idle business resources, and the idle business resources are Shared, so that other business participants create value through the idle business resources Shared by the owner. In the shared economy, idle traffic resources are the first and most critical elements. It is the basis for resource owner and resource user to share resource. The idle traffic resources under the shared economic concept can be understood as: the service resource is originally used by an individual or an organization, and is an idle service resource when not in a use state or an occupied state.

At present, with the continuous development of big data technology, many online services begin to adopt a shared economic mode to perform service processing, so that resources of different service ends can be effectively integrated and intercommunicated, and the service processing intelligence degree is improved.

In general, the sharing economy based on big data traffic mostly involves data resource sharing. However, in practical applications, the inventor finds that the related data resource sharing technology still has some problems, such as easily causing interference to the traffic in the transaction state during the data resource sharing process.

Disclosure of Invention

In view of the foregoing, the present application provides the following.

One embodiment of the present application provides a data resource sharing method applied to a big data service, including: extracting resource content by combining historical service data resources through a data resource sharing thread to be updated to obtain target resource content distribution of the historical service data resources; determining dynamic sharing state information corresponding to the historical service data resources through the data resource sharing thread to be updated in combination with the target resource content distribution; and updating the thread parameters of the data resource sharing thread by combining the effective sharing state information and the dynamic sharing state information of the historical service data resources to obtain the updated data resource sharing thread.

The scheme of one embodiment of the application provides a big data server, which comprises a processing engine, a network module and a memory; the processing engine and the memory communicate through the network module, and the processing engine reads the computer program from the memory and operates to perform the above-described method.

In the description that follows, additional features will be set forth, in part, in the description. These features will be in part apparent to those skilled in the art upon examination of the following and the accompanying drawings, or may be learned by production or use. The features of the present application may be realized and attained by practice or use of various aspects of the methodologies, instrumentalities and combinations particularly pointed out in the detailed examples that follow.

Drawings

The present application will be further explained by way of exemplary embodiments, which will be described in detail by way of the accompanying drawings. These embodiments are not intended to be limiting, and in these embodiments like numerals are used to indicate like structures, wherein:

FIG. 1 is a flow diagram illustrating an exemplary data resource sharing method and/or process for big data traffic according to some embodiments of the present application;

FIG. 2 is a block diagram illustrating an exemplary data resource sharing arrangement for big data traffic according to some embodiments of the present application;

FIG. 3 is a block diagram illustrating an exemplary data resource sharing system for big data traffic, according to some embodiments of the present application, an

FIG. 4 is a diagram illustrating hardware and software components in an exemplary big data server, according to some embodiments of the present application.

Detailed Description

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings used in the description of the embodiments will be briefly introduced below. It is obvious that the drawings in the following description are only examples or embodiments of the application, from which the application can also be applied to other similar scenarios without inventive effort for a person skilled in the art. Unless otherwise apparent from the context, or otherwise indicated, like reference numbers in the figures refer to the same structure or operation.

It should be understood that "system", "device", "unit" and/or "module" as used herein is a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, other words may be substituted by other expressions if they accomplish the same purpose.

As used in this application and the appended claims, the terms "a," "an," "the," and/or "the" are not intended to be inclusive in the singular, but rather are intended to be inclusive in the plural unless the context clearly dictates otherwise. In general, the terms "comprises" and "comprising" merely indicate that steps and elements are included which are explicitly identified, that the steps and elements do not form an exclusive list, and that a method or apparatus may include other steps or elements.

Flow charts are used herein to illustrate operations performed by systems according to embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in the exact order in which they are performed. Rather, the various steps may be processed in reverse order or simultaneously. Meanwhile, other operations may be added to the processes, or a certain step or several steps of operations may be removed from the processes.

In order to better understand the technical solutions of the present invention, the following detailed descriptions of the technical solutions of the present invention are provided with the accompanying drawings and the specific embodiments, and it should be understood that the specific features in the embodiments and the examples of the present invention are the detailed descriptions of the technical solutions of the present invention, and are not limitations of the technical solutions of the present invention, and the technical features in the embodiments and the examples of the present invention may be combined with each other without conflict.

The application aims to provide a data resource sharing method and a big data server applied to big data services, firstly, a data resource sharing thread to be updated is adopted to extract resource content of historical service data resources to obtain target resource content distribution, secondly, dynamic sharing state information corresponding to the historical service data resources is determined through the target resource content distribution, and therefore thread parameters of the data resource sharing thread can be updated by combining effective sharing state information and the dynamic sharing state information to obtain the updated data resource sharing thread.

It can be understood that, since the effective sharing state information and the dynamic sharing state information are simultaneously considered when updating the thread parameters of the data resource sharing thread, the prior business state and the current business state can be simultaneously considered, and since the dynamic sharing state information is determined according to the target resource content distribution, the business relevance and continuity between the effective sharing state information and the dynamic sharing state information can be ensured. Therefore, the thread parameters of the data resource sharing thread can be updated based on the service relevance layer and the service continuity layer, so that the actual service condition can be matched when the updated data resource sharing thread is adopted for data resource sharing, and the influence on the service in the handling state in the data resource sharing process is minimized. Therefore, the technical problem that the related data resource sharing technology easily causes interference to the business in the transaction state can be solved.

First, an exemplary data resource sharing method applied to a big data service is described, please refer to fig. 1, which is a flowchart illustrating an exemplary data resource sharing method and/or process applied to a big data service according to some embodiments of the present application, and the data resource sharing method applied to a big data service may include the technical solutions described in the following steps 100 to 300.

Step 100, combining historical service data resources, and extracting resource contents through a data resource sharing thread to be updated to obtain target resource content distribution of the historical service data resources.

For example, the historical traffic data resource may be a traffic data resource that has been shared or partially shared. The data resource sharing thread to be updated has a data resource sharing function, and the thread can be understood as a neural network model or a module unit with a data sharing function formed in other manners. The data resource sharing thread is deployed in the big data server and used for executing related data resource sharing operation. Further, the target resource content distribution may be the same as the detailed information describing the historical service data resource, such as the data resource type, the data resource storage address, the data resource sharing condition, and the like. In general, the target resource content distribution may be expressed in the form of a graph.

In some possible embodiments, in combination with the historical service data resource, the resource content extraction is performed by the data resource sharing thread to be updated to obtain the target resource content distribution of the historical service data resource, which is described in step 100, and the following steps may be performed: acquiring historical service data resources and effective sharing state information corresponding to the historical service data resources, wherein the effective sharing state information corresponding to the historical service data resources comprises effective item classification information of each resource item in the historical service data resources; and inputting the historical service data resource into a resource preprocessing thread in the data resource sharing threads to be updated, and extracting the resource content of the historical service data resource through a resource content extraction sub-thread of the resource preprocessing thread to obtain the target resource content distribution of the historical service data resource.

For example, the effective sharing state information may be understood as prior sharing state information, such as the sharing condition (sharing period, sharing object, sharing manner, etc.) of the prior data resource, and furthermore, the historical business data resource includes a plurality of resource items, which may relate to a plurality of fields, such as blockchain finance, blockchain medicine, blockchain office, blockchain factory, etc. Accordingly, valid item classification information may be used to differentiate resource items. Further, the resource preprocessing thread may be one of the data resource sharing threads, and is used for performing extraction of the target resource content distribution.

In some optional embodiments, the obtaining of the historical service data resources and the effective shared state information corresponding to the historical service data resources described in the above steps may be implemented by the following implementation manners: acquiring sample service data resources of a data resource sharing thread to be updated; acquiring at least one preset size of service data resource from sample service data resources, and taking the acquired service data resource as a historical service data resource of the data resource sharing thread to be updated; and acquiring the effective sharing state information of the historical service data resources from the effective sample sharing state information of the sample service data resources based on the resource distribution condition of the historical service data resources in the corresponding sample service data resources.

In a related embodiment, the resource characteristics of the sample traffic data resources include: effective sample sharing state information of the sample business data resources, wherein the effective sample sharing state information comprises effective item classification information of all resource items in the sample business data resources. Further, the preset size may be understood as a data amount of the service data resource, such as xxxMB or xxxGB. The preset size can be adjusted according to actual requirements, so that the data processing pressure of the data resource sharing thread can be reduced as much as possible on the premise of ensuring the use efficiency of historical service data resources.

In other possible embodiments, the above-described steps of performing resource content extraction on the historical service data resource by using the resource content extraction sub-thread of the resource preprocessing thread to obtain the target resource content distribution of the historical service data resource may be implemented by the following implementation manners: and extracting the resource content of the historical service data resource through a resource content extraction sub-thread of the resource preprocessing thread to obtain the resource content distribution of a plurality of service scenes of the historical service data resource, and integrating the resource content distribution of the plurality of service scenes to obtain the target resource content distribution of the historical service data resource.

For example, the resource content distributions of different service scenarios may be different, and by determining and integrating the resource content distributions of a plurality of service scenarios, extraction of the resource content distribution of each service scenario can be ensured, thereby ensuring the integrity of the target resource content distribution and avoiding omission of the resource content distribution of some service scenarios.

In some examples, the resource content extraction sub-thread may include a resource content integration unit and at least two resource content extraction units connected in sequence, and the resource content integration unit and the resource content extraction unit may be corresponding functional unit modules. Based on this, the resource content extraction sub-thread through the resource content extraction sub-thread of the resource preprocessing thread described in the above steps performs resource content extraction on the historical service data resource to obtain resource content distribution of a plurality of service scenes of the historical service data resource, and integrates the resource content distribution of the plurality of service scenes to obtain target resource content distribution of the historical service data resource, which can be implemented by the following implementation manners: extracting resource contents of the historical service data resources through the resource content extraction units which are connected in sequence to obtain resource content distribution of different service scenes output by different resource content extraction units; and integrating the resource content distribution of different service scenes according to the sequence from the last resource content extraction unit to the first resource content extraction unit by the resource content integration unit to obtain the target resource content distribution of the historical service data resources.

It can be understood that the order of the resource content extraction units can be determined by a pre-configured priority rule, and thus, the integrity of the distribution of the target resource content can be ensured and the order of the distribution of the target resource content can be improved as much as possible by determining the distribution of the target resource content by the resource content integration unit and the at least two resource content extraction units connected in sequence.

In some possible examples, the number of the resource content integration units is one less than that of the resource content extraction units, and based on this, the resource content distributions of the different service scenes are integrated by the resource content integration units according to the sequence from the last resource content extraction unit to the first resource content extraction unit to obtain the target resource content distribution of the historical service data resources, which may be implemented by the following implementation manners: performing service scene adjustment processing on the resource content distribution input into the current resource content integration unit to obtain adjusted resource content distribution; and performing resource content distribution integration on the adjusted resource content distribution and the resource content distribution extracted by the last resource content extraction unit in the resource content distribution which does not participate in integration processing through the current resource content integration unit, and inputting the integrated resource content distribution into the previous resource content integration unit.

In a related embodiment, the adjusted resource content distribution is the same as the service scenario of the resource content distribution extracted by the last resource content extraction unit in the resource content distributions not participating in the integration processing. And if the current resource content integration unit is the last resource content integration unit, inputting the resource content distribution of the current integration unit into the resource content distribution extracted by the last resource content extraction unit. Further, if the current resource content integration unit is the first resource content integration unit, the integrated resource content distribution obtained by the current resource content integration unit is the target resource content distribution.

In the related embodiment, the service scenario adjustment process may be understood as an update or a swap of a service scenario, so as to implement an update process on resource content distribution. It can be understood that, by analyzing the sequence of the resource content integration units, integration omission of individual resource content distribution can be avoided as much as possible, thereby ensuring the integrity of the target resource content distribution.

And 200, determining dynamic sharing state information corresponding to the historical service data resources through the data resource sharing thread to be updated by combining the target resource content distribution.

For example, the dynamic sharing state information may be understood as current/real-time sharing state information, such as the sharing condition (sharing period, sharing object, sharing mode, etc.) of the data resource of the historical service data resource in the current period or current service scene. It can be understood that by determining the dynamic sharing state information, the sharing condition of the data resources of the historical service data resources in the current time period or the current service scene can be analyzed, so as to locate other services in the transaction state, and thus, the influence of subsequent data resource sharing on other services in the transaction state can be avoided.

In some possible embodiments, the determining, by the data resource sharing thread to be updated, the dynamic sharing state information corresponding to the historical service data resource according to the distribution of the target resource content described in step 200 may include the following: determining a sub-thread through the sharing state of the resource preprocessing thread, and determining dynamic sharing state information corresponding to the historical service data resource based on the content distribution of the target resource, wherein the dynamic sharing state information corresponding to the historical service data resource comprises dynamic item classification information of each resource item in the historical service data resource.

For example, the sharing status determining sub-thread may be used to determine dynamic sharing status information, and accordingly, the dynamic item classification information may be used to distinguish real-time classification situations of resource items. This facilitates analysis of prior and current sharing of different resource items.

And step 300, updating the thread parameters of the data resource sharing thread by combining the effective sharing state information and the dynamic sharing state information of the historical service data resource to obtain the updated data resource sharing thread.

For example, by updating the thread parameters of the data resource sharing thread, the data resource updating policy of the data resource sharing thread can be adjusted, so that the updated data resource sharing thread can be ensured to match the actual service condition and the data resource use condition when the data resource is shared, and the influence on the service in the transaction state in the data resource sharing process is further minimized. Further, the thread parameters include a sharing call mode for the data resource, a sharing call screening rule, and the like.

In some possible embodiments, the updating the thread parameter of the data resource sharing thread in combination with the effective sharing state information and the dynamic sharing state information of the historical service data resource, which are described in the above step 300, to obtain an updated data resource sharing thread may include the following: determining, by a sharing state classification sub-thread in the data resource sharing thread to be updated, a first determination result that the dynamic sharing state information belongs to idle sharing state information of the historical service data resource and a second determination result that the effective sharing state information belongs to idle sharing state information of the historical service data resource based on effective sharing state information and dynamic sharing state information of the historical service data resource; and updating the thread parameters of the data resource sharing thread based on the first judgment result and the second judgment result to obtain the updated data resource sharing thread.

For example, a shared state classification child thread may be used to classify the associated shared state, such as the shared state classification child thread may be a classifier, but is not limited to a secondary. Further, the first determination result that the dynamic shared state information belongs to the idle shared state information of the historical service data resource may be understood as a first possibility (for example, a first probability value) that the dynamic shared state information belongs to the idle shared state information of the historical service data resource, and the second determination result that the effective shared state information belongs to the idle shared state information of the historical service data resource may be understood as a second possibility (for example, a second probability value) that the effective shared state information belongs to the idle shared state information of the historical service data resource.

Therefore, the thread parameters of the data resource sharing thread can be updated based on the judgment result that different sharing state information is possibly in an idle state, and the updated data resource sharing thread is obtained. By the design, the thread parameters of the data resource sharing thread can be updated based on the service relevance layer and the service continuity layer, so that the actual service condition can be matched when the updated data resource sharing thread is adopted for data resource sharing, and the influence on the service in a handling state in the data resource sharing process is minimized. Therefore, the technical problem that the related data resource sharing technology easily causes interference to the business in the transaction state can be solved.

In some other embodiments, the effective shared state information is a service data resource having an effective shared state, and the dynamic shared state information is a service data resource having a dynamic shared state. Based on this, the classifying sub-thread through the sharing state in the data resource sharing thread to be updated, which is described in the above steps, determines that the dynamic sharing state information belongs to the first determination result of the idle sharing state information of the historical service data resource and the second determination result of the idle sharing state information of the historical service data resource based on the effective sharing state information and the dynamic sharing state information of the historical service data resource, and may be implemented through the following contents described in steps 310 to 350.

And 310, associating the historical service data resources with the corresponding service data resources with the effective sharing state to obtain the service data resources with the effective sharing state after association, and associating the historical service data resources with the corresponding service data resources with the dynamic sharing state to obtain the service data resources with the dynamic sharing state after association.

It can be understood that by associating the service data resources in the effective sharing state and associating the service data resources in the dynamic sharing state, the relevance and continuity of the service data resources in different sharing states can be ensured as much as possible, so that the reliability of the determination is improved when the idle sharing state is determined.

In some possible examples, the number of resource sharing requirement conditions of the service data resource with the effective sharing state and the service data resource with the dynamic sharing state is the same. Further, the local service data resource corresponding to each resource sharing requirement condition of the service data resource having the effective sharing state includes: and whether each resource item of the historical service data resource is the information of the effective item classification corresponding to the resource sharing requirement condition. In addition, the local service data resources corresponding to the resource sharing requirement conditions of the service data resources with the dynamic sharing state include: and each resource item of the historical service data resource is a dynamic judgment result of a dynamic item classification corresponding to the resource sharing requirement condition.

Further, the associating the historical service data resource with the corresponding service data resource having the effective sharing state to obtain the service data resource having the effective sharing state after the associating, and associating the historical service data resource with the corresponding service data resource having the dynamic sharing state to obtain the service data resource having the dynamic sharing state after the associating, which is described in the above steps, may include the following contents: the historical service data resources and the local service data resources of the service data resources with the effective sharing state sharing requirement conditions are used as the local service data resources of the service data resources with the effective sharing state sharing requirement conditions after association, and the historical service data resources and the service data resources with the effective sharing state are associated to obtain the service data resources with the effective sharing state after association; and associating the historical service data resources and the service data resources with the dynamic sharing state to obtain the service data resources with the dynamic sharing state after association.

By the design, different resource sharing requirement conditions can be considered when the business data resources are associated, so that the association of the business data resources can be matched with the corresponding resource sharing requirement conditions, and the resource sharing adaptability of the business data resources in an effective sharing state after the association and the business data resources in a dynamic sharing state after the association can be ensured.

And 320, acquiring the sub-thread through the resource content of the sharing state classification sub-thread, and acquiring the resource content information of the first service data resource from the associated service data resource with the dynamic sharing state.

It will be appreciated that the resource content information is used to record the relevant data resource content in the first service data resource.

Step 330, determining, by the state determination unit of the shared state classification sub-thread, a service data resource having a dynamic shared state corresponding to the service data resource having a dynamic shared state after association based on the resource content information of the first service data resource, and a first determination result of the service data resource having an idle shared state belonging to the historical service data resource.

It can be understood that the state determination unit may be a related determination decision layer, and by performing decision analysis on the service data resource with the dynamic sharing state corresponding to the service data resource with the dynamic sharing state after being associated, it is possible to accurately determine the service data resource with the dynamic sharing state corresponding to the service data resource with the dynamic sharing state after being associated, and a first determination result of the service data resource with the idle sharing state belonging to the historical service data resource.

And 340, acquiring the sub-thread through the resource content of the sharing state classification sub-thread, and acquiring resource content information of a second service data resource from the associated service data resource with the effective sharing state.

Step 350, determining, by the state determination unit of the shared state classification sub-thread, a service data resource with an effective shared state corresponding to the service data resource with an effective shared state after association based on the resource content information of the second service data resource, and a second determination result of the service data resource with an idle shared state belonging to the historical service data resource.

It is understood that the description of step 340 and step 350 is similar to that of step 320 and step 330, and is not repeated here.

In some possible embodiments, the updating the thread parameter of the data resource sharing thread based on the first and second determination results to obtain the updated data resource sharing thread described in the above steps may be implemented by: determining a first shared state identification deviation (which may be expressed by similarity or accuracy, for example) of the resource preprocessing thread based on the first discrimination result, where the first shared state identification deviation is used to characterize a local identification deviation; determining a second shared state identification deviation (the second shared state identification deviation is used for representing a global identification deviation) of the resource preprocessing thread based on an information comparison result (for example, may be expressed by a cosine distance between classification labels corresponding to classification information) between the dynamic item classification information and the valid item classification information of the same resource item in the valid shared state information and the dynamic shared state information of the historical service data resource; updating the thread parameters of the resource preprocessing thread based on the first sharing state identification deviation and the second sharing state identification deviation to obtain an updated resource preprocessing thread; determining a state classification deviation of the shared state classification sub-thread based on the first judgment result and the second judgment result; updating thread parameters of the shared state classification child thread based on the state classification bias.

It can be understood that, when the above-mentioned contents are implemented, not only the thread parameter of the resource preprocessing thread may be updated based on the first shared state identification deviation and the second shared state identification deviation, but also the state classification deviation of the shared state classification sub-thread may be determined based on the first determination result and the second determination result, so that the update of the thread parameter of the data resource sharing thread may be refined, the update of the thread parameter of the related sub-thread may be ensured, and the update efficiency and reliability of the global thread parameter of the data resource sharing thread may be ensured.

In some further examples, the dynamic item classification information includes a dynamic item classification and a dynamic discrimination result in the dynamic item classification, where the dynamic discrimination result may be a discrimination result that varies over time. Correspondingly, the determining of the second sharing status identification deviation of the resource preprocessing thread based on the information comparison result between the dynamic item classification information and the valid item classification information of the same resource item in the valid sharing status information and the dynamic sharing status information of the historical service data resource described in the above steps may include the following: determining effective item classification of each resource item in the historical service data resource and dynamic discrimination results of each resource item in the dynamic sharing state information in the corresponding effective item classification based on the effective sharing state information and the dynamic sharing state information of the historical service data resource; and determining a second sharing state identification deviation of the resource preprocessing thread based on the effective item classification of the resource items of the historical service data resources and the dynamic judgment result in the effective item classification.

In the related embodiment, the second sharing state identification deviation can be ensured to have a time sequence change characteristic by determining the valid item classification of each resource item in the historical service data resource and the dynamic judgment result of each resource item in the corresponding valid item classification in the dynamic sharing state information, so that the timeliness of the second sharing state identification deviation is ensured, and an accurate and reliable decision basis is provided for updating the thread parameter.

In some optional embodiments, after obtaining the updated data resource sharing thread, the updated data resource sharing thread may be utilized to share the relevant data service resource. After the sharing of the related data service resources is completed, the behavior detection can be performed on the service terminal which performs service interaction by using the shared data service resources. Based on this, on the basis of the above step 300, the method may further include the following: detecting object content and intention content of the first cloud service interaction data by using a pre-called behavior detection model to obtain first detection object content and first detection intention content; optimizing a first model parameter of the pre-called behavior detection model based on the first detection object content and the first detection intention content, and optimizing a second model parameter of the pre-called behavior detection model through the first cloud service interaction data and the second cloud service interaction data to obtain an optimized behavior detection model; determining reference object content, reference intention content, second detection object content and second detection intention content by using the optimized behavior detection model and the second cloud service interaction data; training the optimized behavior detection model through the reference object content, the reference intention content, the second detection object content and the second detection intention content to obtain a trained behavior detection model, and performing behavior detection on object content and intention content of target business interaction operation behavior in cloud business interaction data through the trained behavior detection model.

For the related technical solutions of the above behavior detection, the following embodiments may be referred to.

Step S1: and detecting object content and intention content of the first cloud service interaction data by using a pre-called behavior detection model to obtain first detection object content and first detection intention content.

In this embodiment, the pre-called behavior detection model may be an artificial intelligence network (NN) based behavior detection model, which may be understood as an initial behavior detection model, for example, an initial behavior detection model called by a big data server from another database or another server. Generally speaking, if the behavior detection model called in advance is directly used for user behavior detection, problems of service scene misadaptation, model parameter deviation and the like may exist, and thus a larger error may occur in a user behavior detection result. Therefore, the optimization and the secondary training of the behavior detection model called in advance are realized based on the loop iteration of model use and model training, so that the learning capacity of the behavior detection model for different cloud service interaction data is enhanced, the detection sensitivity of the behavior detection model for service interaction operation behaviors is improved, the behavior detection efficiency and the reliability of the behavior detection model are improved, and the adaptive capacity of the behavior detection model under different cloud service scenes is ensured.

In this embodiment, the cloud service interaction data may be service interaction data between the service terminal and the big data server, or service interaction data between different service terminals. Further, the cloud service interaction data may be payment service interaction data, game service interaction data, government and enterprise service interaction data, smart medical interaction data, smart city interaction data, smart factory interaction data, and visual service interaction data, and accordingly, the application fields of the service terminal and the big data server may also include the fields corresponding to the service interaction data.

In addition, object content detection and intention content detection can be understood as part of behavior detection, the object content detection is used for analyzing the identity of an interactive object of cloud business interaction data and related business behaviors, and the intention content detection can dig out intention trends of the related business behaviors, so that related countermeasures can be timely executed when abnormal intention content is mined/predicted.

In some possible embodiments, the performing, by using the behavior detection model called in advance, object content and intention content detection on the first cloud service interaction data, which is described in step S1, to obtain first detection object content and first detection intention content may include the following contents described in steps S11 and S12.

Step S11: the method comprises the steps of obtaining first cloud service interaction data and second cloud service interaction data which comprise target service interaction operation behaviors, wherein the first cloud service interaction data are cloud service interaction data added with target object contents and target intention contents of the target service interaction operation behaviors.

Step S12: and detecting object content and intention content of the target service interaction operation behavior in the first cloud service interaction data through a pre-called behavior detection model to obtain first detection object content and first detection intention content.

In this embodiment, the target service interaction operation behavior may be a pre-marked operation behavior, taking a payment service as an example, the target service interaction operation behavior may be a cross-border payment or cloud game service as an example, and the target service interaction operation behavior may be a game assistant starting behavior. The target object content may be understood as direct object content and indirect object content related to the target business interaction operation behavior, and the target intention content may be understood as behavior intention content related to the target business interaction operation behavior, such as business requirements or behavior trends corresponding to the target business interaction operation behavior.

It is understood that, when step S12 is executed, the resulting first detection-object content and first detection-intention content are different from the target-object content and target-intention content, in other words, the target-object content and target-intention content may correspond to the real-object content and the real-intention content. If the subsequent behavior detection is directly performed in step S12, there may be a large error, and for this reason, it is necessary to continuously perform iterative optimization on the behavior detection model called in advance based on the first detection object content, the first detection intention content, the target object content, and the target intention content.

In some possible embodiments, the "performing object content and intention content detection on the target business interaction operation behavior in the first cloud business interaction data through a behavior detection model called in advance to obtain first detection object content and first detection intention content" described in the above step S12 may be implemented by the following steps S121 and S122.

Step S121: and performing service interaction feature extraction on the first cloud service interaction data through the pre-called behavior detection model to obtain first service interaction feature information corresponding to the first cloud service interaction data.

In step S121, the service interaction feature may summarize and summarize the service interaction condition of the first cloud service interaction data, so as to reduce the data volume on the premise of ensuring complete description of the service interaction condition, thereby improving the processing efficiency of the behavior detection model invoked in advance.

In some embodiments, the pre-invoked behavior detection model includes a cold business feature processing unit that includes a classifier comprised of a plurality of dynamic recognition units. For example, the cold business feature processing unit can identify and capture some low-heat business features, so that the integrity of business interaction feature extraction is ensured, and model training deviation caused by omission of the low-heat business features is avoided. Based on this, the "performing, by using the pre-called behavior detection model, service interaction feature extraction on the first cloud service interaction data to obtain first service interaction feature information corresponding to the first cloud service interaction data" described in step S121 may include the content described in the following step S1210.

Step 1210: and sequentially carrying out identification operation on the first cloud service interaction data through a plurality of dynamic identification units of the classifier so as to extract first service interaction characteristic information corresponding to the first cloud service interaction data. It can be understood that, because the classifier includes a plurality of dynamic identification units, each dynamic identification unit has different capturing capabilities and identification capabilities for the cold service features, different cold service features can be captured and identified based on actual service conditions, thereby ensuring the integrity of the first service interaction feature information corresponding to the extracted first cloud service interaction data and avoiding omission of some cold service features.

Step S122: and performing object content and intention content detection on the target service interaction operation behavior in the first cloud service interaction data based on the first service interaction characteristic information through the pre-called behavior detection model to obtain first detection object content and first detection intention content.

In step S122, the first service interaction feature information may be input into the behavior detection model called in advance, and then the first detection object content and the first detection intention content output by the behavior detection model called in advance are obtained.

In some optional embodiments, the behavior detection model called in advance includes an operation behavior classification unit, and the operation behavior classification unit includes a decision tree classification execution network layer, an intention content detection execution network layer, and an object content classification execution network layer. On this basis, the "performing, by the behavior detection model called in advance, object content and intention content detection on the target service interaction operation behavior in the first cloud service interaction data based on the first service interaction feature information to obtain first detection object content and first detection intention content" described in the above step S122 may include the following contents described in steps S1221 to S1223.

Step S1221: and carrying out decision tree classification on the target service interaction operation behavior in the first cloud service interaction data based on the first service interaction characteristic information through the decision tree classification execution network layer to obtain a decision tree classification result.

In practical implementation, the first service interaction feature information may be input into the decision tree classification execution network layer, and the decision tree classification execution network layer performs decision tree-based classification on the first service interaction feature information, so as to obtain a decision tree classification result.

Step S1222: and detecting the difference between the non-hot behavior of the target service interaction operation behavior in the first cloud service interaction data and the real-time behavior portrait and the delayed behavior portrait of the hot behavior of the target service interaction operation behavior by the intention content detection execution network layer based on the first service interaction feature information and the decision tree classification result, so as to obtain first detection intention content of the target service interaction operation behavior in the first cloud service interaction data.

In practical implementation, the first service interaction feature information and the decision tree classification result may be input into the intention content detection execution network layer, and the intention content detection execution network layer detects a real-time behavior sketch and a delayed behavior sketch difference between a non-hotness behavior of the target service interaction operation behavior and a hotness behavior of the target service interaction operation behavior in the first cloud service interaction data.

Further, the intention content detection execution network layer may perform real-time behavior and image difference detection on the non-hot behavior of the target service interaction operation behavior and the hot behavior of the target service interaction operation behavior in the first cloud service interaction data, and then perform delay behavior and image difference detection on the non-hot behavior of the target service interaction operation behavior and the hot behavior of the target service interaction operation behavior in the first cloud service interaction data. Or performing delay behavior portrait difference detection on the non-hot behavior of the target service interaction operation behavior and the hot behavior of the target service interaction operation behavior in the first cloud service interaction data, and then performing real-time behavior portrait difference detection on the non-hot behavior of the target service interaction operation behavior and the hot behavior of the target service interaction operation behavior in the first cloud service interaction data.

In this embodiment, the hot behavior may be understood as a hot behavior, the non-hot behavior may be understood as a cold behavior, the real-time behavior image may be understood as an explicit behavior image of the current time period, and the delayed behavior image may be understood as a potential behavior image of the current time period.

It can be understood that, after the first service interaction feature information and the decision tree classification result are input into the intention content detection execution network layer, the intention content detection execution network layer may adjust a behavior portrait difference detection index based on the decision tree classification result, and then process the first service interaction feature information, thereby implementing real-time/delayed behavior portrait difference detection of hot behaviors and non-hot behaviors.

Step S1223: and performing object content detection on the target service interaction operation behavior in the first cloud service interaction data through the object content classification execution network layer based on the first service interaction feature information and the decision tree classification result to obtain first detection object content of the target service interaction operation behavior.

In practical implementation, the first service interaction feature information and the decision tree classification result may be input to the object content classification execution network layer, and the object content classification execution network layer performs object content detection on the target service interaction operation behavior in the first cloud service interaction data based on the decision tree classification result and the first service interaction feature information in sequence to obtain a first detection object content of the target service interaction operation behavior.

For example, the object content classification execution network layer may update network layer parameters based on the decision tree classification result, and then perform object content detection on the target service interaction operation behavior in the first cloud service interaction data to obtain first detection object content of the target service interaction operation behavior.

Therefore, by executing the steps S1221 to S1223, the first service interaction feature information can be processed by using the decision tree classification execution network layer, the intention content detection execution network layer, and the object content classification execution network layer included in the operation behavior classification unit, and since the decision tree classification execution network layer is located before the intention content detection execution network layer and the object content classification execution network layer, when the processing is performed based on the intention content detection execution network layer and the object content classification execution network layer, the decision tree classification result can be fully considered, thereby effectively reducing the errors of the first detection intention content and the first detection object content.

Step S2: optimizing a first model parameter of the pre-called behavior detection model based on the first detection object content and the first detection intention content, and optimizing a second model parameter of the pre-called behavior detection model through the first cloud service interaction data and the second cloud service interaction data to obtain an optimized behavior detection model.

In this embodiment, the first model parameter and the second model parameter may correspond to different model performances of the behavior detection model invoked in advance, for example, the first model parameter may correspond to a model prediction accuracy, and the second model parameter may correspond to a model generalization capability. Based on this, the "optimizing the first model parameters of the behavior detection model called in advance based on the first detection object content and the first detection intention content, and optimizing the second model parameters of the behavior detection model called in advance by the first cloud service interaction data and the second cloud service interaction data to obtain the optimized behavior detection model" described in the above step S2 may include the following steps S20.

Step S20: comparing the first detection object content with the target object content, comparing the first detection intention content with the target intention content to optimize a first model parameter of the pre-called behavior detection model, and performing deep learning on the first cloud service interaction data and the second cloud service interaction data through the pre-called behavior detection model to optimize a second model parameter of the pre-called behavior detection model to obtain the optimized behavior detection model.

It is understood that by comparing the first detection object content with the target object content and comparing the first detection intention content with the target intention content, the difference between the first detection object content and the target object content and the difference between the first detection intention content and the target intention content can be obtained, so that the optimization of the first model parameter of the behavior detection model called in advance can be realized based on the differences, such as the optimization of the first model parameter according to the set parameter adjustment step size.

Furthermore, the first cloud service interaction data and the second cloud service interaction data are deeply learned through the pre-called behavior detection model to be understood as counterlearning, so that the second model parameters of the pre-called behavior detection model can be optimized through counterlearning results, and the generalization capability of the model is improved.

As can be seen from this, by implementing step S20, the model parameters can be optimized and adjusted based on the model prediction accuracy and the model generalization capability level, and an optimized behavior detection model can be obtained.

In some examples, the first cloud service interaction data may be historical cloud service interaction data, the second cloud service interaction data may be cloud service interaction data to be processed, the historical cloud service interaction data is cloud service interaction data to which target object content and target intention content of the target service interaction operation behavior are added, and the cloud service interaction data to be processed is cloud service interaction data to which the target object content and the target intention content of the target service interaction operation behavior are not added. Further, the pre-invoked behavior detection model includes an auto-supervised countermeasure training unit that includes a countermeasure outcomes adjustment layer.

On this basis, the "comparing the first detection object content with the target object content and comparing the first detection intention content with the target intention content to optimize the first model parameters of the pre-called behavior detection model, and performing deep learning on the first cloud business interaction data and the second cloud business interaction data through the pre-called behavior detection model to optimize the second model parameters of the pre-called behavior detection model to obtain the optimized behavior detection model" described in the above step S20 may include the following steps S21 to S24.

Step S21: and performing service interaction feature extraction on the second cloud service interaction data through the pre-called behavior detection model to obtain second service interaction feature information corresponding to the second cloud service interaction data.

For further explanation of step S21, reference may be made to the above description of step S121.

Step S22: and constructing a global model performance index based on the decision tree classification result, the first detection intention content and the first detection object content.

In this embodiment, the global model performance index may reflect the model performance as a whole, for example, the global model performance index may be determined according to model state weights corresponding to the decision tree classification result, the first detection intention content, and the first detection object content, where a higher model state weight indicates that the corresponding decision tree classification result, the first detection intention content, or the first detection object content is more important than the model, so that it may be ensured that the global model performance index matches with the actual service condition.

Step S23: comparing the first detection object content with the target object content by the global model performance indicator, and comparing the first detection intent content with the target intent content to optimize a first model parameter of the pre-invoked behavior detection model.

In this embodiment, a difference threshold of the first detection object content and the target object content may be determined according to a global model performance index, and a difference threshold of the first detection intention content and the target intention content may be determined according to a global model performance index, and then optimization of the first model parameter of the behavior detection model invoked in advance may be achieved by continuously converging the first detection object content and the target object content and continuously converging the first detection intention content and the target intention content.

Step S24: the historical cloud service interaction data or the cloud service interaction data to be processed to which the first service interaction feature information and the second service interaction feature information belong are detected through the behavior of the self-supervision countermeasure training unit to obtain a behavior detection result, countermeasure result adjustment is conducted on the behavior detection result through the countermeasure result adjustment layer to analyze the service interaction feature value without time sequence change, deep learning is conducted on the types of the first cloud service interaction data and the second cloud service interaction data based on the service interaction feature value without time sequence change to optimize second model parameters of the pre-called behavior detection model, and the optimized behavior detection model is obtained.

In this embodiment, the first service interaction feature information and the second service interaction feature information may be respectively input into the self-supervision countermeasure training unit, behavior detection is performed by the self-supervision countermeasure training unit to obtain behavior detection results corresponding to the historical cloud service interaction data or the cloud service interaction data to be processed, and then a countermeasure result adjustment layer is used to adjust the behavior detection results, so that a service interaction feature value without time sequence change can be analyzed and a service interaction feature value without time sequence change can be determined. The service interaction characteristic value can quantitatively express different service interaction characteristics, so that the behavior detection model can be assisted to deeply learn the types of the first cloud service interaction data and the second cloud service interaction data, and the generalization capability of the behavior detection model is improved by optimizing the second model parameter of the pre-called behavior detection model.

Step S3: and determining reference object content, reference intention content, second detection object content and second detection intention content by using the optimized behavior detection model and the second cloud service interaction data.

It can be understood that by determining the reference object content, the reference intention content, the second detection object content, and the second detection intention content, further training of the optimized behavior detection model can be achieved, thereby ensuring model stability of the optimized behavior detection model. Based on this, the "determining reference object content, reference intention content, second detection object content, and second detection intention content using the optimized behavior detection model and the second cloud service interaction data" described in the above step S3 may include the following steps S31 and S32.

Step S31: and acquiring the object content and the intention content with the maximum business interaction characteristic tag value corresponding to the target business interaction operation behavior in the second cloud business interaction data as reference object content and reference intention content respectively through the optimized behavior detection model.

In step S31, the service interaction feature tag value can be used to distinguish the object content from the intention content, and can also be used to quantify the feature recognition degree of the object content and the intention content. For example, the larger the value of the feature tag of the service interaction is, the higher the feature recognition degree of the corresponding object content and intention content is, and by selecting the object content and the intention content with the maximum value of the feature tag of the service interaction as the reference object content and the reference intention content, respectively, the reference object content and the reference intention content can be ensured to have the higher feature recognition degree, thereby facilitating the subsequent model training.

In some possible embodiments, the "obtaining, by the optimized behavior detection model, the object content and the intention content with the largest business interaction feature tag value corresponding to the target business interaction operation behavior in the second cloud business interaction data as the reference object content and the reference intention content respectively" described in the above step S31 may include the following contents described in steps S311 and S313.

Step S311: and performing service interaction feature extraction on the second cloud service interaction data through the optimized behavior detection model to obtain third service interaction feature information.

It is understood that the description regarding step S311 may be with respect to similar steps as described above.

Step S312: and detecting object content and intention content of the target service interaction operation behavior in the second cloud service interaction data based on the third service interaction feature information to obtain at least one optimized detection object content and a corresponding service interaction feature tag value thereof and at least one optimized detection intention content and a corresponding service interaction feature tag value thereof.

In this embodiment, the corresponding behavior detection may be performed based on the optimized behavior detection model, so as to obtain the optimized detection object content and the optimized detection intention content, and meanwhile, the optimized behavior detection model may be used to determine respective service interaction feature tag values corresponding to the optimized detection object content and the optimized detection intention content, so as to implement the binding of the feature recognition degrees of the optimized detection object content and the optimized detection intention content.

Step S313: and screening out object content with the maximum service interaction characteristic tag value from the optimized detection object content as reference object content corresponding to the target service interaction operation behavior in the second cloud service interaction data, and screening out intention content with the maximum service interaction characteristic tag value from the optimized detection intention content as reference intention content corresponding to the target service interaction operation behavior in the second cloud service interaction data.

On the basis, the reference object content and the reference intention content meeting the requirements can be screened out through the determined service interaction characteristic label value.

Step S32: and inputting the second cloud service interaction data into the optimized behavior detection model to detect object content and intention content to obtain second detection object content and second detection intention content.

In some possible embodiments, the step S32 of inputting the second cloud service interaction data into the optimized behavior detection model for object content and intention content detection to obtain second detection object content and second detection intention content may include the following steps S321 and S322.

Step S321: and performing service interaction feature extraction on the second cloud service interaction data through the optimized behavior detection model to obtain fourth service interaction feature information corresponding to the second cloud service interaction data.

Step S322: and performing decision tree classification and object content and intention content detection on the target service interaction operation behavior in the second cloud service interaction data based on the fourth service interaction feature information through the optimized behavior detection model to obtain second detection object content and second detection intention content of the target service interaction operation behavior.

By implementing the steps S31 and S32, the object content and the intention content with the largest service interaction feature tag value can be selected as the reference object content and the reference intention content, respectively, so as to ensure that the reference object content and the reference intention content have high feature recognition degree, thereby facilitating subsequent model training.

Step S4: training the optimized behavior detection model through the reference object content, the reference intention content, the second detection object content and the second detection intention content to obtain a trained behavior detection model, and performing behavior detection on object content and intention content of target business interaction operation behavior in cloud business interaction data through the trained behavior detection model.

In this embodiment, the optimized behavior detection model is trained through the reference object content, the reference intention content, the second detection object content and the second detection intention content, so that the learning capability of the behavior detection model for different cloud service interaction data can be enhanced, the detection sensitivity of the behavior detection model for service interaction operation behaviors is improved, the behavior detection efficiency and the reliability of the behavior detection model are improved, and the adaptability of the behavior detection model in different cloud service scenes is ensured. Based on this, the "training the optimized behavior detection model through the reference object content, the reference intention content, the second detection object content and the second detection intention content to obtain a trained behavior detection model, so as to perform behavior detection on the object content and the intention content of the target business interaction operation behavior in the cloud business interaction data through the trained behavior detection model" described in the above step S4 may include the following steps S40.

Step S40: and comparing the second detection object content with the reference object content, and comparing the second detection intention content with the reference intention content to train the optimized behavior detection model to obtain a trained behavior detection model, so as to perform behavior detection on the object content and the intention content of the target service interaction operation behavior in the cloud service interaction data through the trained behavior detection model.

In some possible embodiments, the step S40 of comparing the second test object content with the reference object content and comparing the second test intention content with the reference intention content to train the optimized behavior test model to obtain a trained behavior test model may include the following steps S41-S43.

Step S41: and comparing the second detection object content with the reference object content through the first model performance index to obtain first model performance data.

Step S42: and comparing the second detection intention content with the reference intention content through a second model performance index to obtain second model performance data.

Step S43: and constructing a target global model performance index based on the first model performance data and the second model performance data, optimizing the model parameters of the optimized behavior detection model through the target global model performance index, taking the optimized behavior detection model after optimizing the model parameters as a pre-called behavior detection model, returning to execute the pre-called behavior detection model to perform object content and intention content detection on the target business interaction operation behavior in the first cloud business interaction data, and obtaining the operation of first detection object content and first detection intention content until the performance deviation value corresponding to the model performance data of the target global model performance index is minimum, thereby obtaining the trained behavior detection model.

In some embodiments, the first model performance indicator and the second model performance indicator may be different loss functions, and accordingly, the first model performance data and the second model performance data may be different loss function values. Based on the above, a target global model performance index can be accurately constructed through the first model performance data and the second model performance data, and then the model parameters of the optimized behavior detection model are continuously optimized and adjusted based on the target global model performance index. And then, feeding back and executing the previous steps, and continuously realizing the iterative training of the model.

Accordingly, the performance deviation value corresponding to the model performance data of the target global model performance index may be a global loss value reflecting the overall performance of the model. Therefore, the learning capacity of the behavior detection model for different cloud service interaction data can be enhanced, so that the detection sensitivity of the behavior detection model for service interaction operation behaviors is improved, the behavior detection efficiency and the reliability of the behavior detection model are improved, and the adaptive capacity of the behavior detection model in different cloud service scenes is ensured.

On the basis of the above, the method may further include the following contents described in step S51 to step S53.

Step S51: and acquiring the cloud service interaction data to be detected, which comprises the target service interaction operation behavior.

Step S52: and performing service interaction feature extraction on the cloud service interaction data to be detected through the trained behavior detection model to obtain target service interaction feature information.

Step S53: and identifying the object content and the intention content of the target service interaction operation behavior in the cloud service interaction data to be detected based on the target service interaction characteristic information through the trained behavior detection model.

It can be understood that, by implementing the steps S51-S53, the accuracy and reliability of the object content and the intention content of the target business interaction operation behavior in the cloud business interaction data to be detected can be ensured. Thereby facilitating subsequent operational behavior analysis and business service adjustment.

In some optional embodiments, after identifying, by the trained behavior detection model, the object content and the intention content of the target business interaction operation behavior in the cloud business interaction data to be detected based on the target business interaction feature information, the method may further include: and analyzing abnormal operation behaviors through the object content and the intention content of the target service interaction operation behaviors in the cloud service interaction data to be detected, and generating a data information security strategy according to the analysis result of the abnormal operation behaviors. Therefore, the corresponding data information security protection strategy can be customized through abnormal operation behavior analysis, so that the protection on the related data information in the big data server is realized, and the illegal use of the data information by the abnormal operation behavior is avoided.

In some alternative embodiments, the data information security policy may focus on data anonymization processing, and for this reason, the "generating data information security policy according to abnormal operation behavior analysis result" described in the above may further include the following steps S61-S6: 5.

Step S61: and responding to a data anonymity processing instruction, and acquiring the data state reference content without attributes matched with the target stored user data according to the abnormal operation behavior analysis result.

In this embodiment, the data anonymization processing indication may be sent to the big data server by another service terminal, the target stored user data exists in the big data server, no attribute may be understood as no attribute tag, and the data state reference content may be understood as a data state used for characterizing the user data.

Step S62: and determining different data demand characteristics respectively corresponding to the data state reference content without the attribute matched with the target stored user data.

In this embodiment, the data requirement characteristic is used to characterize the service requirement condition of the data state reference content.

Step S63: determining, based on the different data demand characteristics, a data segment with an update that matches the different data demand characteristics; and dynamically screening the data state reference content without the attribute matched with the target stored user data based on the updated data segment to form a reference content set without the attribute matched with the corresponding data demand characteristics.

In this embodiment, the data segment may be obtained by data splitting the target stored user data. An attribute-free reference content set may be understood as a set corresponding to data state reference content without tags.

Step S64: and calling a data state conversion thread dynamically configured by a multidimensional characteristic training sample, and carrying out anonymous processing on the attribute-free reference content set through the data state conversion thread to obtain a data anonymous processing result of the target stored user data.

In this embodiment, the data state transition threads include a content extraction sub-thread, a sample classification sub-thread, and a feature classification sub-thread. The data state transition thread can be understood as a neural network model with related functions.

Step S65: and applying the data anonymity processing result of the target stored user data.

In this embodiment, the anonymization processing may be K anonymization processing, which can ensure data security of the target stored user data.

It can be understood that, by implementing the above steps S61-S65, anonymization processing can be performed on the non-attribute reference content set based on the data state transition thread, so as to obtain the data anonymization processing result of the target stored user data, which can protect the relevant data information in the big data server and avoid illegal use of the data information by abnormal operation behaviors.

On the basis of the above-described steps S61 to S65, the method may further include the following steps S661 to S664.

Step S661: and acquiring a historical anonymization processing record of the content and the data characteristics of the target stored user data set.

Step S662: and obtaining a reference content set matched with the data state conversion thread according to the historical anonymization processing record of the content and the data characteristics of the target stored user data set, wherein the reference content set comprises reference contents with different multidimensional characteristics.

Step S663: determining different time step lengths, processing the reference content set according to the corresponding time step lengths, and determining training sample sets with different multidimensional characteristics matched with the data state conversion thread, wherein the training sample sets comprise at least one group of training samples.

Step S664: training the data state conversion thread according to training sample sets of different multidimensional characteristics matched with the data state conversion thread, and determining thread configuration data matched with the data state conversion thread so as to realize anonymous processing on target stored user data corresponding to the data characteristics through the data state conversion thread.

It can be understood that, by implementing the above steps S661 to S664, the reference content set can be processed based on different time steps, and training sample sets of different multidimensional features matching the data state transition thread are determined, so that the data state transition thread is trained according to the training sample sets of different multidimensional features matching the data state transition thread, and thread configuration data adapted to the data state transition thread is determined, so that the data state transition thread can be adjusted based on the thread configuration data as appropriate as possible, thereby ensuring the anti-attack performance of anonymization processing on target stored user data corresponding to the data features by the data state transition thread, and avoiding cracking of anonymized processing data by abnormal operation behavior.

When the contents described in the above steps S1 to S4 are implemented, model parameters of the behavior detection model can be optimized based on the behavior detection result of the behavior detection model, and training of the behavior detection model is implemented, so that the learning ability of the behavior detection model for different cloud service interaction data can be enhanced, the detection sensitivity of the behavior detection model for service interaction operation behaviors in the cloud service interaction data is improved, and meanwhile, the behavior detection accuracy of the behavior detection model can be improved based on the optimization of different model parameters, so that the behavior detection efficiency and the reliability of the behavior detection model are improved. In addition, the adaptability of the behavior detection model under different cloud service scenes can be ensured based on deep learning of different cloud service interaction data. Therefore, the trained behavior detection model obtained by optimizing and training the pre-called behavior detection model can be used for effectively detecting and analyzing the behavior of the cloud service interaction data.

Next, for the data resource sharing method applied to the big data service, an exemplary data resource sharing device applied to the big data service is further provided in the embodiment of the present invention, as shown in fig. 2, the data resource sharing device 200 applied to the big data service may include the following functional modules.

And a content distribution extraction module 210, configured to extract resource content through a data resource sharing thread to be updated in combination with the historical service data resource, so as to obtain target resource content distribution of the historical service data resource.

And a sharing state determining module 220, configured to determine, by combining with the content distribution of the target resource, dynamic sharing state information corresponding to the historical service data resource through the data resource sharing thread to be updated.

And a shared thread updating module 230, configured to update thread parameters of the data resource sharing thread in combination with the effective shared state information of the historical service data resource and the dynamic shared state information, to obtain an updated data resource sharing thread.

Then, based on the above method embodiment and apparatus embodiment, the embodiment of the present invention further provides a system embodiment, that is, a data resource sharing system applied to a big data service, please refer to fig. 3, where the data resource sharing system 30 applied to the big data service may include a big data server 10 and a service terminal 20. Wherein the big data server 10 and the service terminal 20 communicate to implement the above method, and further, the functionality of the data resource sharing system 30 applied to the big data service is described as follows. The big data server 10 extracts resource contents through a data resource sharing thread to be updated by combining historical service data resources to obtain target resource content distribution of the historical service data resources; determining dynamic sharing state information corresponding to the historical service data resources through the data resource sharing thread to be updated in combination with the target resource content distribution; and updating the thread parameters of the data resource sharing thread by combining the effective sharing state information and the dynamic sharing state information of the historical service data resources to obtain the updated data resource sharing thread.

Further, referring to fig. 4 in conjunction, the big data server 10 may include a processing engine 110, a network module 120, and a memory 130, the processing engine 110 and the memory 130 communicating through the network module 120.

Processing engine 110 may process the relevant information and/or data to perform one or more of the functions described herein. For example, in some embodiments, processing engine 110 may include at least one processing engine (e.g., a single core processing engine or a multi-core processor). By way of example only, the Processing engine 110 may include a Central Processing Unit (CPU), an Application-Specific Integrated Circuit (ASIC), an Application-Specific Instruction Set Processor (ASIP), a Graphics Processing Unit (GPU), a Physical Processing Unit (PPU), a Digital Signal Processor (DSP), a Field Programmable Gate Array (FPGA), a Programmable Logic Device (PLD), a controller, a microcontroller Unit, a Reduced Instruction Set Computer (RISC), a microprocessor, or the like, or any combination thereof.

Network module 120 may facilitate the exchange of information and/or data. In some embodiments, the network module 120 may be any type of wired or wireless network or combination thereof. Merely by way of example, the Network module 120 may include a cable Network, a wired Network, a fiber optic Network, a telecommunications Network, an intranet, the internet, a Local Area Network (LAN), a Wide Area Network (WAN), a Wireless Local Area Network (WLAN), a Metropolitan Area Network (MAN), a Public Switched Telephone Network (PSTN), a bluetooth Network, a Wireless personal Area Network, a Near Field Communication (NFC) Network, and the like, or any combination thereof. In some embodiments, the network module 120 may include at least one network access point. For example, the network module 120 may include wired or wireless network access points, such as base stations and/or network access points.

The Memory 130 may be, but is not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Read-Only Memory (EPROM), an electrically Erasable Read-Only Memory (EEPROM), and the like. The memory 130 is used for storing a program, and the processing engine 110 executes the program after receiving the execution instruction.

It will be appreciated that the configuration shown in fig. 4 is merely illustrative and that the big data server 10 may also include more or fewer components than shown in fig. 4, or have a different configuration than shown in fig. 4. The components shown in fig. 4 may be implemented in hardware, software, or a combination thereof.

It should be understood that, for the above, a person skilled in the art can deduce from the above disclosure to determine the meaning of the related technical term without doubt, for example, for some values, coefficients, weights, indexes, factors, and other terms, a person skilled in the art can deduce and determine from the logical relationship between the above and the following, and the value range of these values can be selected according to the actual situation, for example, 0 to 1, for example, 1 to 10, and for example, 50 to 100, which are not limited herein.

The skilled person can unambiguously determine some preset, reference, predetermined, set and target technical features/terms, such as threshold values, threshold intervals, threshold ranges, etc., from the above disclosure. For some technical characteristic terms which are not explained, the technical solution can be clearly and completely implemented by those skilled in the art by reasonably and unambiguously deriving the technical solution based on the logical relations in the previous and following paragraphs. Prefixes of unexplained technical feature terms, such as "first", "second", "previous", "next", "current", "history", "latest", "best", "target", "specified", and "real-time", etc., can be unambiguously derived and determined from the context. Suffixes of technical feature terms not to be explained, such as "list", "feature", "sequence", "set", "matrix", "unit", "element", "track", and "list", etc., can also be derived and determined unambiguously from the foregoing and the following.

The foregoing disclosure of embodiments of the present invention will be apparent to those skilled in the art. It should be understood that the process of deriving and analyzing technical terms, which are not explained, by those skilled in the art based on the above disclosure is based on the contents described in the present application, and thus the above contents are not an inventive judgment of the overall scheme.

It should be appreciated that the system and its modules shown above may be implemented in a variety of ways. For example, in some embodiments, the system and its modules may be implemented in hardware, software, or a combination of software and hardware. Wherein the hardware portion may be implemented using dedicated logic; the software portions may be stored in a memory for execution by a suitable instruction execution system, such as a microprocessor or specially designed hardware. Those skilled in the art will appreciate that the methods and systems described above may be implemented using computer executable instructions and/or embodied in processor control code, such code being provided, for example, on a carrier medium such as a diskette, CD-or DVD-ROM, a programmable memory such as read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The system and its modules of the present application may be implemented not only by hardware circuits such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., but also by software executed by various types of processors, for example, or by a combination of the above hardware circuits and software (e.g., firmware).

It is to be noted that different embodiments may produce different advantages, and in different embodiments, any one or combination of the above advantages may be produced, or any other advantages may be obtained.

Having thus described the basic concept, it will be apparent to those skilled in the art that the foregoing detailed disclosure is to be considered merely illustrative and not restrictive of the broad application. Various modifications, improvements and adaptations to the present application may occur to those skilled in the art, although not explicitly described herein. Such modifications, improvements and adaptations are proposed in the present application and thus fall within the spirit and scope of the exemplary embodiments of the present application.

Also, this application uses specific language to describe embodiments of the application. Reference throughout this specification to "one embodiment," "an embodiment," and/or "some embodiments" means that a particular feature, structure, or characteristic described in connection with at least one embodiment of the present application is included in at least one embodiment of the present application. Therefore, it is emphasized and should be appreciated that two or more references to "an embodiment" or "one embodiment" or "an alternative embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, some features, structures, or characteristics of one or more embodiments of the present application may be combined as appropriate.

Moreover, those skilled in the art will appreciate that aspects of the present application may be illustrated and described in terms of several patentable species or situations, including any new and useful combination of processes, machines, manufacture, or materials, or any new and useful improvement thereon. Accordingly, various aspects of the present application may be embodied entirely in hardware, entirely in software (including firmware, resident software, micro-code, etc.) or in a combination of hardware and software. The above hardware or software may be referred to as "data block," module, "" engine, "" unit, "" component, "or" system. Furthermore, aspects of the present application may be represented as a computer product, including computer readable program code, embodied in one or more computer readable media.

The computer storage medium may comprise a propagated data signal with the computer program code embodied therewith, for example, on baseband or as part of a carrier wave. The propagated signal may take any of a variety of forms, including electromagnetic, optical, etc., or any suitable combination. A computer storage medium may be any computer-readable medium that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code located on a computer storage medium may be propagated over any suitable medium, including radio, cable, fiber optic cable, RF, or the like, or any combination of the preceding.

Computer program code required for the operation of various portions of the present application may be written in any one or more programming languages, including an object oriented programming language such as Java, Scala, Smalltalk, Eiffel, JADE, Emerald, C + +, C #, VB.NET, Python, and the like, a conventional programming language such as C, Visual Basic, Fortran 2003, Perl, COBOL 2002, PHP, ABAP, a dynamic programming language such as Python, Ruby, and Groovy, or other programming languages, and the like. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any network format, such as a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet), or in a cloud computing environment, or as a service, such as a software as a service (SaaS).

Additionally, the order in which elements and sequences of the processes described herein are processed, the use of alphanumeric characters, or the use of other designations, is not intended to limit the order of the processes and methods described herein, unless explicitly claimed. While various presently contemplated embodiments of the invention have been discussed in the foregoing disclosure by way of example, it is to be understood that such detail is solely for that purpose and that the appended claims are not limited to the disclosed embodiments, but, on the contrary, are intended to cover all modifications and equivalent arrangements that are within the spirit and scope of the embodiments herein. For example, although the system components described above may be implemented by hardware devices, they may also be implemented by software-only solutions, such as installing the described system on an existing server or mobile device.

Similarly, it should be noted that in the preceding description of embodiments of the application, various features are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure aiding in the understanding of one or more of the embodiments. This method of disclosure, however, is not intended to require more features than are expressly recited in the claims. Indeed, the embodiments may be characterized as having less than all of the features of a single embodiment disclosed above.

Numerals describing the number of components, attributes, etc. are used in some embodiments, it being understood that such numerals used in the description of the embodiments are modified in some instances by the use of the modifier "about", "approximately" or "substantially". Unless otherwise indicated, "about", "approximately" or "substantially" indicates that the numbers allow for adaptive variation. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximations that may vary depending upon the desired properties of the individual embodiments. In some embodiments, the numerical parameter should take into account the specified significant digits and employ a general digit preserving approach. Notwithstanding that the numerical ranges and parameters setting forth the broad scope of the range are approximations, in the specific examples, such numerical values are set forth as precisely as possible within the scope of the application.

The entire contents of each patent, patent application publication, and other material cited in this application, such as articles, books, specifications, publications, documents, and the like, are hereby incorporated by reference into this application. Except where the application is filed in a manner inconsistent or contrary to the present disclosure, and except where the claim is filed in its broadest scope (whether present or later appended to the application) as well. It is noted that the descriptions, definitions and/or use of terms in this application shall control if they are inconsistent or contrary to the statements and/or uses of the present application in the material attached to this application.

Finally, it should be understood that the embodiments described herein are merely illustrative of the principles of the embodiments of the present application. Other variations are also possible within the scope of the present application. Thus, by way of example, and not limitation, alternative configurations of the embodiments of the present application can be viewed as being consistent with the teachings of the present application. Accordingly, the embodiments of the present application are not limited to only those embodiments explicitly described and depicted herein.