CN114548820B

CN114548820B - Big data wind control method and server for distance education service

Info

Publication number: CN114548820B
Application number: CN202210216943.2A
Authority: CN
Inventors: 窦跃飞; 闫继平
Original assignee: Geekbang Holdings Beijing Co ltd
Current assignee: Geekbang Holdings Beijing Co ltd
Priority date: 2022-03-07
Filing date: 2022-03-07
Publication date: 2022-11-01
Anticipated expiration: 2042-03-07
Also published as: CN115293650A; CN114548820A

Abstract

The invention provides a big data wind control method and a server for remote education service, wherein initial service operation and maintenance risk description is mined firstly through a machine learning model, and then staged operation and maintenance risk items, staged risk description contents and integrated risk description contents can be determined in parallel according to the mined initial service operation and maintenance risk description, so that the staged operation and maintenance risk items, the staged risk description contents and the integrated risk description contents are prevented from being independently mined, the mining timeliness can be ensured, the initial service operation and maintenance risk description contents which need to be referred to by the three types of key data are mined, the three types of key data are determined through the integrally mined initial service operation and maintenance risk description, the initial service operation and maintenance risk description is not required to be mined when each type of key data is determined, the initial service operation and maintenance risk description is frequently mined, and the resource overhead of wind control analysis is reduced to a certain extent.

Description

Big data wind control method and server for distance education service

Technical Field

The invention relates to the technical field of big data, in particular to a big data wind control method and a server for remote education service.

Background

Distance education (network education) refers to a teaching mode using transmission media such as televisions, the internet and the like, the distance education can break through the boundary line of time and space, and compared with the traditional education, the distance education has the advantages that firstly, a user does not need to attend a class in a designated place; 2. the study can be mutually assisted through various different pipelines such as television broadcasting, the Internet, a tutoring special line, a lesson and research society and the like.

With the continuous development of distance education, when a teaching end interacts with a client in the distance education process, interaction of a large amount of data is involved, so that the problem of interaction safety in the interaction process is a great concern for users. Therefore, risk prevention and control for remote education service processes are indispensable, and related risk prevention and control technologies may occupy more resources when performing feature analysis, which is not favorable for the efficiency of wind control analysis.

Disclosure of Invention

The invention provides a big data wind control method and a server for remote education service, and adopts the following technical scheme in order to achieve the technical purpose.

The first aspect is a big data wind control method for remote education service, which is applied to a big data wind control server, and the method at least comprises the following steps: determining a remote education service operation report to be subjected to wind control analysis; excavating the initial service operation and maintenance risk description of the remote education service operation report to be subjected to the wind control analysis; and determining staged operation and maintenance risk items of the remote education service operation report to be subjected to the wind control analysis, staged risk description contents corresponding to the staged operation and maintenance risk items and integrated risk description contents of the remote education service operation report to be subjected to the wind control analysis according to the initial service operation and maintenance risk description.

According to the design, initial service operation and maintenance risk description is mined firstly through a machine learning model, and then according to the mined initial service operation and maintenance risk description, staged operation and maintenance risk items, staged risk description contents and integrated risk description contents can be determined in parallel, so that the staged operation and maintenance risk items, staged risk description contents and integrated risk description contents are prevented from being mined independently, the mining timeliness can be guaranteed, the initial service operation and maintenance risk description contents are risk description contents which need to be referred to when the three types of key data are mined, the three types of key data are determined through the integrally mined initial service operation and maintenance risk description, frequent mining of the initial service operation and maintenance risk description caused by mining the initial service operation and maintenance risk description when each type of key data is determined is not needed, the resource overhead of wind control analysis is reduced to a certain extent, and the high efficiency of wind control analysis is facilitated.

Under some design concepts that can be independently implemented, the determining, according to the initial service operation and maintenance risk description, staged operation and maintenance risk items of the remote education service operation report to be subjected to the wind control analysis, staged risk description contents corresponding to the staged operation and maintenance risk items, and integrated risk description contents of the remote education service operation report to be subjected to the wind control analysis includes: and obtaining staged operation and maintenance risk items of the remote education service operation report to be subjected to the wind control analysis, staged risk description contents corresponding to the staged operation and maintenance risk items and integrated risk description contents of the remote education service operation report to be subjected to the wind control analysis through an operation and maintenance risk identification model according to the initial service operation and maintenance risk description.

Therefore, the excavation of the staged operation and maintenance risk items of the remote education service operation report to be subjected to the wind control analysis, the staged risk description content and the integrated risk description content corresponding to the staged operation and maintenance risk items can be realized through one operation and maintenance risk identification model without being respectively realized through a plurality of machine learning models, so that not only can the model configuration of the operation and maintenance risk identification model be adjusted, but also the parallel determination of the three types of key data can be realized, and the quality of the wind control analysis can be further ensured.

Under some design ideas which can be independently implemented, configuring the operation and maintenance risk identification model, wherein the configuration comprises the following steps: determining a plurality of sets of authenticated remote education service operation reports; for each group of authenticated remote education service operation reports, determining at least one group of authenticated associated service operation reports with transfer characteristics of the authenticated remote education service operation reports, and binding the authenticated remote education service operation reports with each group of authenticated associated service operation reports respectively to obtain at least one authenticated report binary group; and configuring the operation and maintenance risk identification model through the authenticated report binary group.

Therefore, the transfer characteristics can reflect the attention change of the acquisition thread when the two remote education service operation reports are acquired, the authenticated report binary group comprising the two remote education service operation reports with the transfer characteristics is used as authenticated information for configuring the operation and maintenance risk identification model, the performance that the configured operation and maintenance risk identification model is compatible with the attention change of the acquisition thread when the key data is identified can be improved, and the accuracy and the reliability of the configured operation and maintenance risk identification model for exporting the three types of key data can be improved.

Under some independently implementable design considerations, the authenticated remote education service execution report includes a first authenticated service session report, the authenticated report duplet encompassing a first authenticated session report duplet; the method for determining at least one authenticated associated service operation report with the transfer characteristics of the authenticated remote education service operation reports for each authenticated remote education service operation report group, and binding the authenticated remote education service operation reports with each authenticated associated service operation report group respectively to obtain at least one authenticated report binary group, comprises the following steps: determining a plurality of sets of first authenticated service session reports under a target service session; determining dynamic service session data corresponding to the target service session according to a plurality of groups of the first authenticated service session reports; and determining at least one pair of first authenticated service session reports with transfer characteristics from the first authenticated service session reports according to the dynamic service session data to obtain at least one first authenticated session report binary group.

In this way, the transfer characteristics among the sets of first authenticated service session reports can be accurately determined through the determined dynamic service session data, and further, a plurality of first authenticated session report duplets with the transfer characteristics can be accurately determined. And the remote education service operation reports under the target service session of the two remote education service operation reports included in the first authenticated session report binary group are configured for the operation and maintenance risk identification model through the plurality of first authenticated session report binary groups, so that the performance that the configured operation and maintenance risk identification model is compatible with the attention change of the acquisition thread when key data is identified can be improved, the compatibility of the configured operation and maintenance risk identification model under a certain target service session can be improved, and further the wind control analysis reliability under a certain target service session is improved.

In some independently implementable design considerations, the authenticated remote education service run report includes a second authenticated service session report, the authenticated report duplet encompassing a second authenticated session report duplet; the determining, for each set of authenticated teleeducation service operation reports, at least one set of authenticated association service operation reports having transfer characteristics with the authenticated teleeducation service operation reports, and binding the authenticated teleeducation service operation reports with each set of authenticated association service operation reports, respectively, to obtain at least one authenticated report binary, including: determining a plurality of sets of second authenticated service session reports under a non-target service session; and for each group of second authenticated service session reports, performing first preprocessing on the second authenticated service session reports to obtain at least one group of target associated service operation reports with transfer characteristics of the second authenticated service session reports, and binding the second authenticated service session reports with each group of target associated service operation reports respectively to obtain at least one second authenticated session report binary group.

In this way, through the first preprocessing, a target associated service operation report with transfer characteristics existing in the second authenticated service session report can be obtained, and the second authenticated service session report and the target associated service operation report are both remote education service operation reports under non-target service sessions, so that the operation and maintenance risk identification model is configured through the second authenticated session report binary group formed by binding the second authenticated service session report and the target associated service operation report, the performance of compatibility of the configured operation and maintenance risk identification model with attention change of a collection thread during key data identification can be improved, the compatibility of the configured operation and maintenance risk identification model under various service sessions can be improved, and the wind control analysis reliability of the operation and maintenance risk identification model under various service sessions can be improved.

Under some design ideas which can be independently implemented, configuring the operation and maintenance risk identification model through the authenticated report binary group, including: performing second preprocessing on a part of the authenticated report binary group to obtain at least one authenticated report binary group which is subjected to the second preprocessing; the authenticated reporting duplet encompasses a first authenticated reporting duplet and/or a second authenticated reporting duplet; and configuring the operation and maintenance risk identification model through the part of the authenticated report duplet and the authenticated report duplet which completes the preprocessing.

Thus, the remote education service operation report in the pre-processed authenticated report binary group through the second pre-processing is different from the remote education service operation report in the basic authenticated report binary group, such as: and the information such as the identification degree and the disturbance of the difference exists, so that the compatibility of the configured operation and maintenance risk identification model to the slight error of the remote education service operation report can be improved by taking the authenticated report binary group and the authenticated report binary group which completes the preprocessing as the authenticated information, such as: the compatibility of the configured operation and maintenance risk identification model to the situations of identification degree change, disturbance change and the like of the remote education service operation report can be improved, and the accuracy of the risk identification result derived by the operation and maintenance risk identification model is improved.

In some independently implementable design concepts, the second pre-processing includes one or more of: non-limiting disturbance processing and report feature recognition degree adjustment processing.

In this way, the non-restrictive disturbance processing can change disturbance information of the remote education service operation report in the authenticated report binary group, the report feature recognition degree adjustment processing can change recognition degree information of the remote education service operation report in the authenticated report binary group, model configuration is performed through the remote education service operation report in the authenticated report binary group which is subjected to the pre-processing by the second pre-processing, compatibility of the configured operation and maintenance risk identification model to situations of change of recognition degree, disturbance change and the like of the remote education service operation report can be improved, and accuracy of a risk identification result derived by the operation and maintenance risk identification model can be improved.

Under some design ideas which can be independently implemented, configuring the operation and maintenance risk identification model through the authenticated report binary group, including: respectively determining staged guiding operation and maintenance risk items of each authenticated remote education service operation report in the authenticated report binary group; determining active example operation and maintenance risk items and passive example operation and maintenance risk items corresponding to the staged guiding operation and maintenance risk items from the authenticated associated service operation reports in the authenticated report duplet; and configuring the operation and maintenance risk identification model through the staged operation and maintenance guiding risk items, the active example operation and maintenance risk items and the passive example operation and maintenance risk items corresponding to the staged operation and maintenance guiding risk items.

In this way, the operation and maintenance risk identification model is configured through the active examples and the passive examples, so that the identification reliability of the configured operation and maintenance risk identification model in the process of identifying the staged risk description content of the remote education service operation report can be improved.

Under some independently implementable design ideas, the determining, from the authenticated associated service operation report in the authenticated report binary, an active example operation and maintenance risk item and a passive example operation and maintenance risk item corresponding to the staged guided operation and maintenance risk item includes: taking the operation and maintenance risk items corresponding to the staged guiding operation and maintenance risk items in the authenticated associated service operation report as active example operation and maintenance risk items; and determining the operation and maintenance risk item with the maximum common index with the staged guiding operation and maintenance risk item from the operation and maintenance risk items except the active example operation and maintenance risk item in the authenticated associated service operation report, and taking the operation and maintenance risk item with the maximum common index as the passive example operation and maintenance risk item.

In this way, the configuration ideas corresponding to the operation and maintenance risk items of the active example and the passive example are different, the configuration idea of the operation and maintenance risk item of the active example is a common index approaching to the basic example and the active example, and the active example which is repeated with the basic example exists in the authenticated related service operation report which has the transfer characteristics with the authenticated remote education service operation report, so that the staged risk description content of the operation and maintenance risk item of the active example screened from the authenticated related service operation report is most similar to the staged risk description content corresponding to the staged guidance operation and maintenance risk item, and the configuration effect of the operation and maintenance risk item of the active example can be better satisfied by the active example screened from the authenticated related service operation report; the configuration thought of the passive example operation and maintenance risk items is to avoid the common index of the basic example and the passive example, so the operation and maintenance risk item with the maximum common index of the staged guiding operation and maintenance risk items is screened as the passive example operation and maintenance risk item, the staged risk description content of the screened passive example operation and maintenance risk item is most similar to the staged risk description content corresponding to the staged guiding operation and maintenance risk item, the configuration effect of the passive example operation and maintenance risk item can be better met by avoiding the common index of the basic example and the passive example, and the identification reliability of the staged risk description content of the configured operation and maintenance risk identification model can be improved.

Under some design ideas which can be independently implemented, the configuring the operation and maintenance risk identification model through the staged operation and maintenance guiding risk item, the active example operation and maintenance risk item corresponding to the staged operation and maintenance guiding risk item, and the passive example operation and maintenance risk item includes: loading the authenticated report binary group into the operation and maintenance risk identification model to obtain a staged test operation and maintenance risk item, a first test risk description content corresponding to the staged guide operation and maintenance risk item, a second test risk description content corresponding to the active example operation and maintenance risk item, and a third test risk description content corresponding to the passive example operation and maintenance risk item; determining a first model quality evaluation according to the staged test operation and maintenance risk items and the staged guide operation and maintenance risk items; determining a second model quality evaluation through a first test risk description content corresponding to the staged guiding operation and maintenance risk item, a second test risk description content corresponding to the active example operation and maintenance risk item and a third test risk description content corresponding to the passive example operation and maintenance risk item; and configuring the operation and maintenance risk identification model through the first model quality evaluation and the second model quality evaluation.

The first model quality evaluation is the model quality evaluation between the staged test operation and maintenance risk items and the staged guide operation and maintenance risk items, the operation and maintenance risk identification model is configured through the first model quality evaluation, the identification reliability of the staged operation and maintenance risk items identified by the configured operation and maintenance risk identification model can be improved, the second model quality evaluation is the model quality evaluation among the staged first test risk description content, the staged second test risk description content and the staged third test risk description content, and the operation and maintenance risk identification model is configured through the second model quality evaluation, so that the accuracy of the staged operation and maintenance risk description contents identified by the configured operation and maintenance risk identification model can be improved.

Under some design ideas which can be independently implemented, loading the authenticated report binary into the operation and maintenance risk identification model to obtain a staged test operation and maintenance risk item, a first test risk description content corresponding to the staged guide operation and maintenance risk item, a second test risk description content corresponding to the active example operation and maintenance risk item, and a third test risk description content corresponding to the passive example operation and maintenance risk item, including: loading the authenticated report binary group into the operation and maintenance risk identification model, and mining a sub-model according to the initial description in the operation and maintenance risk identification model to obtain a test type initial service operation and maintenance risk description corresponding to the authenticated report binary group; and identifying a sub-model by the description in the operation and maintenance risk identification model, and obtaining staged test operation and maintenance risk items, first test risk description contents corresponding to the staged guide operation and maintenance risk items, second test risk description contents corresponding to the active example operation and maintenance risk items and third test risk description contents corresponding to the passive example operation and maintenance risk items according to the test model initial service operation and maintenance risk description.

Therefore, through the configuration of the initial description mining submodel, the accuracy of the test type initial service operation and maintenance risk description derived from the initial description mining submodel can be improved, the description identification submodel can derive the staged test operation and maintenance risk item, the first test risk description content, the second test risk description content, the third test risk description content and the like with higher accuracy through the test type initial service operation and maintenance risk description with higher accuracy, the description identification submodel is configured through the staged test operation and maintenance risk item, the first test risk description content, the second test risk description content, the third test risk description content and the like, and finally, the identification reliability of the configured operation and maintenance risk identification submodel can be improved.

Under some design ideas which can be independently implemented, the configuring the operation and maintenance risk identification model through the staged operation and maintenance guiding risk item, the active example operation and maintenance risk item corresponding to the staged operation and maintenance guiding risk item, and the passive example operation and maintenance risk item includes: loading a plurality of first authenticated report binary groups to the operation and maintenance risk identification model, respectively obtaining integrated fourth test risk description content corresponding to each group of remote education service operation reports in each first authenticated report binary group, and adding the integrated fourth test risk description content to a designated space; and configuring the operation and maintenance risk identification model according to a plurality of second authenticated report duplets and fourth test risk description contents in the designated space.

In this way, the fourth test risk description content obtained through the first authenticated reporting binary group is recorded in the designated space to serve as a negative example, so that the number of the negative examples for configuration can be expanded, and the accuracy of identifying the integrated risk description content by the configured operation and maintenance risk identification model is improved.

Under some design ideas which can be independently implemented, configuring the operation and maintenance risk identification model according to a plurality of second authenticated reporting duplets and fourth test risk description contents in the designated space, including: loading a plurality of second authenticated report binary groups to the operation and maintenance risk identification model, and respectively obtaining an integrated fifth test risk description content of each remote education service operation report in each second authenticated report binary group; and determining third model quality evaluation according to the fourth test risk description content and the fifth test risk description content in the designated space, and configuring the operation and maintenance risk identification model according to the third model quality evaluation.

The third model quality evaluation is the model quality evaluation between the fifth test risk description content and the fourth test risk description content, the model quality evaluation corresponding to the integrated risk description content is determined, and the operation and maintenance risk identification model is configured through the third model quality evaluation, so that the accuracy of identifying the integrated risk description content by the configured operation and maintenance risk identification model can be improved.

Under some design ideas which can be independently implemented, the fifth test risk description content comprises a sixth test risk description content and a seventh test risk description content; determining a third model quality evaluation according to a fourth test risk description and the fifth test risk description in the designated space, including: for each second authenticated report binary group, determining a sixth test risk description content of an authenticated remote education service operation report in the second authenticated report binary group, taking an authenticated association service operation report in the second authenticated report binary group as an active authenticated remote education service operation report corresponding to the authenticated report, and determining a seventh test risk description content of the active authenticated remote education service operation report; determining a fourth test risk description content with the largest commonality index with a sixth test risk description content of the certified remote education service operation report in the second certified report binary group from the designated space, and taking the determined fourth test risk description content as an eighth test risk description content corresponding to a negative certified report; and determining a third model quality evaluation according to the sixth test risk description content, the seventh test risk description content and the eighth test risk description content.

In this way, by the sixth test risk description and the seventh test risk description, a model quality evaluation corresponding to an active example between the integrated risk description contents can be determined, by the fourth test risk description having the largest commonality index with the sixth test risk description of the certified remote education service operation report, a model quality evaluation corresponding to a passive example between the integrated risk description contents can be determined, and then by the determined model quality evaluation corresponding to the active example and the model quality evaluation corresponding to the passive example, the credibility of the generated third model quality evaluation can be improved.

Under some independently implementable design ideas, after obtaining the integrated fifth test risk description content of each remote education service operation report in each second authenticated report binary group respectively, the method further comprises the following steps: optimizing the designated space according to the integrated fifth test risk description content of each remote education service operation report.

Therefore, optimization of the electrode example can be achieved through the fifth test risk description content optimization set determined in real time, the integrated fifth test risk description content generated by each configuration is higher in precision, and the reliability of the obtained third model quality evaluation can be improved through the high-precision negative example.

The second aspect is a big data wind control server, comprising a memory and a processor; the memory and the processor are coupled; the memory for storing computer program code, the computer program code comprising computer instructions; wherein the computer instructions, when executed by the processor, cause the big data wind control server to perform the method of the first aspect.

Drawings

Fig. 1 is a schematic flow chart of a big data wind control method for distance education service according to an embodiment of the present invention.

Fig. 2 is a block diagram of a big data wind control device for remote education service according to an embodiment of the present invention.

Detailed Description

In the following, the terms "first", "second" and "third", etc. are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or to imply that the number of indicated technical features is significant. Thus, features defined as "first", "second" or "third", etc., may explicitly or implicitly include one or more of the features.

Fig. 1 is a flowchart illustrating a big data wind control method for a distance education service according to an embodiment of the present invention, where the big data wind control method for the distance education service may be implemented by a big data wind control server, and the big data wind control server may include a memory and a processor; the memory and the processor are coupled; the memory for storing computer program code, the computer program code comprising computer instructions; when the processor executes the computer instructions, the big data wind control server is made to execute the technical scheme described in the following steps.

And step 101, determining a remote education service operation report to be subjected to wind control analysis.

In the embodiment of the invention, the remote education service operation report to be subjected to the wind control analysis can be a remote education service operation report acquired by a thread under a target service session, or a remote education service operation report under a non-target service session, and when staged operation and maintenance risk items exist in the remote education service operation report to be subjected to the wind control analysis, the staged operation and maintenance risk items can be the item characteristics with the highest risk information attention degree in a set service set corresponding to the service items corresponding to the remote education service operation report to be subjected to the wind control analysis.

It can be understood that the staged operation and maintenance risk items can reflect staged detailed descriptions or core descriptions of remote education service operation reports to be subjected to the wind control analysis, can be used for marking target abnormal events in the remote education service operation reports to be subjected to the wind control analysis, and can realize the pairing of the remote education service operation reports to be subjected to the wind control analysis and other remote education service operation reports through the pairing of the staged operation and maintenance risk items of the remote education service operation reports to be subjected to the wind control analysis and the staged operation and maintenance risk items of the other remote education service operation reports.

For example, the remote education service operation report to be subjected to the wind control analysis can be understood as a remote education service operation report to be processed, and the remote education service operation report can be an operation report generated in the interaction process of an online education service terminal and a client terminal, and mainly focuses on a data information security and protection layer. Further, different service sessions may be understood as different online education scenarios. The staged operation and maintenance risk items can be understood as local characteristic items.

And 102, mining the initial service operation and maintenance risk description of a remote education service operation report to be subjected to wind control analysis.

In the embodiment of the invention, according to the determined remote education service operation report to be subjected to the wind control analysis, the initial service operation and maintenance risk description of the remote education service operation report to be subjected to the wind control analysis can be mined, wherein the initial service operation and maintenance risk description can be a service operation and maintenance risk description (refer to operation and maintenance risk description) which needs to be referred to when the staged operation and maintenance risk item, the staged risk description content corresponding to the staged operation and maintenance risk item and the integrated risk description content of the remote education service operation report to be subjected to the wind control analysis are determined.

And 103, determining staged operation and maintenance risk items of the remote education service operation report to be subjected to the wind control analysis, staged risk description contents corresponding to the staged operation and maintenance risk items and integrated risk description contents of the remote education service operation report to be subjected to the wind control analysis according to the initial service operation and maintenance risk description.

In the embodiment of the present invention, the staged risk description content may be understood as a risk description content for describing a global description in a set service set corresponding to a service item corresponding to the staged operation and maintenance risk item, and each staged operation and maintenance risk item may have a staged risk description content. The integrated risk description can be understood as the risk description for describing the global description of the remote education service operation report to be subjected to the wind control analysis, and a single integrated service operation and maintenance risk description exists in one remote education service operation report to be subjected to the wind control analysis. Wherein, the staged risk description content and the integrated risk description content can be recorded by rich description vectors.

In the embodiment of the present invention, after determining the initial service operation and maintenance risk description, each staged operation and maintenance risk item of the remote education service operation report to be subjected to the wind control analysis and the staged risk description content corresponding to each staged operation and maintenance risk item may be determined at the same time on the premise of determining that staged operation and maintenance risk items exist in the remote education service operation report to be subjected to the wind control analysis, where on the premise of determining that staged operation and maintenance risk items exist in the remote education service operation report to be subjected to the wind control analysis, not less than one staged operation and maintenance risk item may exist in one remote education service operation report to be subjected to the wind control analysis, in other words, not less than one staged risk description content may exist in one remote education service operation report to be subjected to the wind control analysis. In addition, according to the initial service operation and maintenance risk description and the remote education service operation report to be subjected to the wind control analysis, the integrated risk description content of the remote education service operation report to be subjected to the wind control analysis can be determined at the same time.

In practical implementation, the determination of each staged risk description content needs not only to be performed by means of the initial service operation and maintenance risk description and the remote education service operation report to be subjected to the wind control analysis, but also needs to be performed by means of the staged operation and maintenance risk item corresponding to the staged operation and maintenance risk description. It is to be appreciated that the staged risk description can be further understood as a staged visual array, and the integrated risk description can be further referred to as an integrated visual array.

Therefore, initial service operation and maintenance risk description is mined firstly through a machine learning model, and then staged operation and maintenance risk items, staged risk description contents and integrated risk description contents can be determined in parallel according to the mined initial service operation and maintenance risk description, so that the staged operation and maintenance risk items, staged risk description contents and integrated risk description contents are prevented from being mined independently, the mining timeliness can be guaranteed, the initial service operation and maintenance risk description contents are risk description contents which need to be referred to by mining the three types of key data, the three types of key data are determined through the integrally mined initial service operation and maintenance risk description, the initial service operation and maintenance risk description is not required to be mined when each type of key data is determined, the initial service operation and maintenance risk description is frequently mined, and the resource overhead of wind control analysis is reduced to a certain extent.

In some examples, for step 103, a staged operation and maintenance risk item of the remote education service operation report to be subjected to the wind control analysis, staged risk description content corresponding to the staged operation and maintenance risk item, and an integrated risk description content of the remote education service operation report to be subjected to the wind control analysis may be derived from the initial service operation and maintenance risk description through the operation and maintenance risk identification model.

In the embodiment of the present invention, the big data wind control method for distance education service provided by the embodiment of the present invention may be applied to an operation and maintenance risk identification model, and steps 101 to 103 may be implemented by the operation and maintenance risk identification model. It can be understood that the determined remote education service operation report to be subjected to the wind control analysis can be processed through the operation and maintenance risk identification model, an initial service operation and maintenance risk description of the remote education service operation report to be subjected to the wind control analysis is derived, and then staged operation and maintenance risk items of the remote education service operation report to be subjected to the wind control analysis, staged risk description contents corresponding to the staged operation and maintenance risk items and integrated risk description contents of the remote education service operation report to be subjected to the wind control analysis are derived through the initial service operation and maintenance risk description and the remote education service operation report to be subjected to the wind control analysis.

In practical implementation, the operation and maintenance risk identification model may include an initial description mining sub-model (a feature extraction layer) and a description identification sub-model (a feature identification layer), and the description identification sub-model may include a staged operation and maintenance risk item identification unit, a staged risk description content identification unit, and an integrated risk description content identification unit. The operation and maintenance risk identification model provided by the embodiment of the invention derives the staged operation and maintenance risk items of the remote education service operation report to be subjected to the wind control analysis, the staged risk description content corresponding to the staged operation and maintenance risk items and the design thought of the integrated risk description content. The system comprises an initial description mining submodel, a staged operation and maintenance risk item identification unit, a staged risk description content identification unit and an integrated risk description content identification unit, wherein the initial description mining submodel is used for mining initial service operation and maintenance risk description of a remote education service operation report to be subjected to wind control analysis, the staged operation and maintenance risk item identification unit is used for deriving staged operation and maintenance risk items of the remote education service operation report to be subjected to wind control analysis, the staged risk description content identification unit is used for deriving staged risk description content corresponding to the staged operation and maintenance risk items, and the integrated risk description content identification unit is used for deriving integrated risk description content of the remote education service operation report to be subjected to wind control analysis.

In addition, the operation and maintenance risk identification model can derive identification information with high reliability and accuracy only by configuration, so that the embodiment of the invention also provides a design idea for configuring the operation and maintenance risk identification model, which exemplarily comprises the following steps.

Step 301, sets of authenticated remote education service operation reports are determined.

And 302, for each group of authenticated remote education service operation reports, determining at least one group of authenticated associated service operation reports with the transfer characteristics of the authenticated remote education service operation reports, and binding the authenticated remote education service operation reports with each group of authenticated associated service operation reports respectively to obtain at least one authenticated report binary group. Wherein the authenticated reporting duplet with the delivery characteristics covers the remote education service operation reporting duplet with the partial repeated service items.

And 303, configuring the operation and maintenance risk identification model through the authenticated report binary group.

In the embodiment of the present invention, the authenticated remote education service operation report may be a remote education service operation report in an arbitrary service session, and the delivery characteristics (which may be understood as a common view relationship) may reflect the change in attention of the collection thread when two remote education service operation reports are collected, and the number and distribution of repetitive service events between the two remote education service operation reports. The configured operation and maintenance risk identification model can derive staged operation and maintenance risk items of the authenticated remote education service operation report with higher reliability and accuracy, staged risk description contents corresponding to the staged operation and maintenance risk items and integrated risk description contents of the remote education service operation report to be subjected to the wind control analysis according to the processing of the authenticated remote education service operation report.

After the authenticated report binary group is determined, each pair of authenticated remote education service operation reports in the authenticated report binary group can be loaded to the operation and maintenance risk identification model, then, according to the derivation result of the operation and maintenance risk identification model, model quality evaluation for cyclic configuration of the operation and maintenance risk identification model is determined, and the operation and maintenance risk identification model is cyclically configured through the determined model quality evaluation to obtain the configured operation and maintenance risk identification model.

In addition, after obtaining a plurality of authenticated reporting duplets, a second pre-processing may be performed on a part of the authenticated reporting duplets, so as to obtain no less than one authenticated reporting duplet which completes the pre-processing. Wherein the second pre-processing (which may be understood as pre-processing) may include one or more of: non-limiting disturbance processing and report feature recognition degree adjustment processing.

In practical implementation, a part of authenticated reporting duplets can be determined from the multiple authenticated reporting duplets, then, for the determined part of authenticated reporting duplets, non-restrictive disturbance processing and reporting feature recognition degree adjustment processing can be performed on each authenticated reporting duplet in the part of authenticated reporting duplets, non-restrictive disturbance processing can also be performed on part of the authenticated reporting duplets in the part of authenticated reporting duplets, and reporting feature recognition degree adjustment processing can be performed on part of the authenticated reporting duplets in the part of authenticated reporting duplets. And then circularly configuring the operation and maintenance risk identification model through the obtained multiple authenticated report duplets and the authenticated report duplets which are subjected to preprocessing.

Thus, the remote education service operation report in the pre-processed authenticated report binary group through the second pre-processing is different from the remote education service operation report in the basic authenticated report binary group, such as: the authenticated report binary group and the pre-processed authenticated report binary group are used as authenticated information, so that the compatibility of the configured operation and maintenance risk identification model with slight errors of the remote education service operation report can be improved, for example, the compatibility of the configured operation and maintenance risk identification model with the conditions of change of the recognition degree, change of disturbance and the like of the remote education service operation report can be improved, and the accuracy of the risk identification result derived by the operation and maintenance risk identification model is improved.

In some examples, the authenticated remote education service execution report may include a first authenticated service session report, the authenticated report duplet encompassing the first authenticated session report duplet.

For step 302, the authenticated reporting doublet may be determined according to the following steps.

Step A, determining a plurality of groups of first authenticated service session reports under the target service session.

And step B, determining dynamic service session data corresponding to the target service session according to a plurality of groups of first authenticated service session reports.

And step C, determining at least one pair of first authenticated service session reports with transfer characteristics from the first authenticated service session reports according to the dynamic service session data to obtain at least one first authenticated session report binary group.

In the embodiment of the invention, in order to improve the identification reliability of the configured operation and maintenance risk identification model for the remote education service operation report under the target service session, the operation and maintenance risk identification model can be configured through the remote education service operation report under the target service session.

The first authenticated service session report may be a remote education service execution report determined under the target service session, the first authenticated session report doublet including two sets of first authenticated service session reports having a transitive feature, filtered from the sets of first authenticated service session reports.

In practical implementation, after determining a plurality of different sets of authenticated service session reports, according to a plurality of sets of authenticated service session reports, performing dynamic feature mapping on a target service session corresponding to a couple of authenticated session reports through a feature mapping idea to determine dynamic service session data corresponding to the target service session, for each first authenticated service session report in a plurality of sets of first authenticated service session reports, determining, through the dynamic service session data, other first authenticated service session reports having a transfer feature with the first authenticated service session report from the plurality of sets of first authenticated service session reports, binding each first authenticated service session report in the other first authenticated service session reports having a transfer feature with the first authenticated service session report to obtain a first authenticated session report couple corresponding to the couple of first authenticated session reports, and then determining that each first authenticated service session report couple corresponding to each first authenticated service session report in the plurality of sets of first authenticated service session reports is not less than a couple of first authenticated session reports, and determining that a couple of authenticated service session reports is not less than a couple of authenticated session reports in this embodiment.

And finally, circularly configuring the operation and maintenance risk identification model through the determined first authenticated session report binary group to obtain the configured operation and maintenance risk identification model.

In other examples, the authenticated remote education service execution report may further include a second authenticated service session report, and similarly, the authenticated report duplet may further include a second authenticated session report duplet. In practical implementation, the second authenticated service session report is a remote education service operation report under a non-target service session, after a plurality of groups of second authenticated service session reports are determined, for each group of second authenticated service session reports, a first preprocessing can be performed on the second authenticated service session reports, so that not less than one group of target associated service operation reports having transfer characteristics with the second authenticated service session report can be determined, and then the second authenticated service session report is bound with each target associated service operation report in the determined not less than one group of target associated service operation reports, so that not less than one second authenticated session report binary corresponding to the second authenticated session report binary is obtained. Further, according to the first pre-processing, a second authenticated session report doublet corresponding to each second authenticated session report doublet may be determined.

Wherein the first pre-processing may be a feature adjustment operation upon which it may be determined that there is not less than a set of target associated service operation reports of the delivery feature with the second authenticated service session report.

After determining the second authenticated session reporting duplet, the operation and maintenance risk identification model may be configured cyclically through the first authenticated session reporting duplet and the second authenticated session reporting duplet simultaneously. In practical implementation, the operation and maintenance risk identification model may be configured cyclically by determining a first number of first authenticated session reporting tuples and a second number of second authenticated session reporting tuples for configuration by setting a criterion (such as setting a ratio), and then filtering the first number of first authenticated session reporting tuples and the second number of second authenticated session reporting tuples together.

In some examples, the set criteria may be that the number of first authenticated session reporting tuples and the number of second authenticated session reporting tuples is 7:3 in percentage by weight.

Therefore, the operation and maintenance risk identification model is configured through the first authenticated session report binary group and the second authenticated session report binary group together, the identification reliability of the configured operation and maintenance risk identification model under a certain target service session can be improved, the compatibility of the configured operation and maintenance risk identification model under other service sessions can be improved, and the identification reliability under other service sessions can be improved.

In addition, for the screened first number of first authenticated session report tuples and the screened second number of second authenticated session report tuples, a part of the first authenticated session report tuples and/or the second authenticated session report tuples can be screened, second preprocessing is performed on the first authenticated session report tuples and/or the second authenticated session report tuples, and the first authenticated session report tuples and/or the second authenticated session report tuples which are subjected to preprocessing are obtained, and the operation and maintenance risk identification model can be circularly configured through the first authenticated session report tuples and/or the second authenticated session report tuples, the first authenticated session report tuples which are subjected to preprocessing and/or the second authenticated session report tuples which are subjected to preprocessing at the same time.

Therefore, the configured operation and maintenance risk identification model can adapt to attention change of the acquisition thread, identification degree change and disturbance change of the remote education service operation report, and can have reliable identification reliability under certain target service sessions and other non-target service sessions.

For step 303, after the authenticated reporting duplet is determined, the staged operation and maintenance risk item included in the remote education service operation report in each authenticated reporting duplet may be determined through a correlation algorithm (e.g., an adaptive transaction identification algorithm), the staged operation and maintenance risk item included in the authenticated remote education service operation report in the determined authenticated reporting duplet may be used as the staged guided operation and maintenance risk item corresponding to the authenticated report, then the positive example operation and maintenance risk item and the negative example operation and maintenance risk item corresponding to the determined staged guided operation and maintenance risk item may be determined from the authenticated correlated service operation report in the authenticated reporting duplet, and finally the operation and maintenance risk identification model may be configured circularly through the staged guided operation and maintenance risk item, the positive example operation and maintenance risk item and the negative example operation and maintenance risk item, so as to obtain a configured operation and maintenance risk identification model.

Further, the authenticated remote education service operation report in each authenticated report binary may include a plurality of staged operation and maintenance risk items, in other words, an active example operation and maintenance risk item and a passive example operation and maintenance risk item corresponding to each staged guiding operation and maintenance risk item in the plurality of staged operation and maintenance risk items may be determined from the authenticated associated service operation report corresponding to the authenticated report. The operation and maintenance risk identification model can be configured circularly through each determined staged guiding operation and maintenance risk item and the corresponding active example operation and maintenance risk item and the corresponding passive example operation and maintenance risk item.

In some examples, regarding the determination of the active example operation and maintenance risk item and the passive example operation and maintenance risk item corresponding to any staged operation and maintenance guidance risk item included in the authenticated remote education service operation report in the authenticated report binary, the staged operation and maintenance risk item corresponding to the staged operation and maintenance guidance risk item may be determined from the authenticated associated service operation report and used as the active example operation and maintenance risk item corresponding to the staged operation and maintenance guidance risk item. From staged operation and maintenance risk items included in the authenticated association service operation report in addition to the active example operation and maintenance risk items, a negative example operation and maintenance risk item corresponding to the staged leading operation and maintenance risk item can then be determined. In actual implementation, for each staged operation and maintenance risk item in the staged operation and maintenance risk items included in the authenticated associated service operation report except for the active example operation and maintenance risk item, a common index between each staged operation and maintenance risk item and the staged guiding operation and maintenance risk item may be determined, and then the staged operation and maintenance risk item with the largest common index is screened as the passive example operation and maintenance risk item.

It is to be appreciated that after determining the staged guiding operation and maintenance risk item, the active example operation and maintenance risk item, and the passive example operation and maintenance risk item included in the authenticated reporting duplet, the authenticated reporting duplet may be loaded into an operation and maintenance risk identification model, and for each staged guiding operation and maintenance risk item included in the authenticated reporting duplet and its corresponding active example operation and maintenance risk item and passive example operation and maintenance risk item, the operation and maintenance risk identification model may derive a staged testing operation and maintenance risk item corresponding to the staged guiding operation and maintenance risk item, a first testing risk description content corresponding to the staged guiding operation and maintenance risk item, a second testing risk description content corresponding to the staged guiding operation and maintenance risk item, and a third testing risk description content corresponding to the staged guiding operation and maintenance risk item and its corresponding passive example operation and maintenance risk item according to the processing of the remote education service operation report in the authenticated reporting duplet. And the first test risk description content, the second test risk description content and the third test risk description content are all derived staged risk description contents.

Then, a first model quality evaluation about staged operation and maintenance risk item identification can be generated by using the staged guiding operation and maintenance risk item and the staged testing operation and maintenance risk item corresponding to the staged guiding operation and maintenance risk item as actual values, and then the staged operation and maintenance risk item identification unit can be configured circularly by the first model quality evaluation, so that the staged operation and maintenance risk item identification unit in the configured operation and maintenance risk identification model can derive the staged testing operation and maintenance risk item with higher precision.

It can be understood that the second model quality evaluation may be determined by the first test risk description content, the second test risk description content, and the third test risk description content corresponding to the staged guiding operation and maintenance risk item, where the second model quality evaluation may be cross entropy loss.

In this way, the staged risk description content identification unit can be configured in a circulating manner through the quality evaluation of the second model, and the staged risk description content identification unit in the configured operation and maintenance risk identification model can derive the staged test risk description content with higher precision.

In the embodiment of the invention, the first model quality evaluation and the second model quality evaluation can be determined simultaneously, and then the operation and maintenance risk identification model can be configured circularly through the first model quality evaluation and the second model quality evaluation, so as to obtain the configured operation and maintenance risk identification model.

In addition, the big data wind control method for the remote education service provided by the embodiment of the invention can also realize the configuration of the integrated risk description content identification unit.

In practical implementation, a plurality of first authenticated reporting duplets can be loaded into the operation and maintenance risk identification model at the same time, wherein the first authenticated reporting duplet may be a remote education service operation reporting duplet filtered from the determined plurality of authenticated reporting duplets, for each first authenticated reporting duplet of the imported plurality of first authenticated reporting duplets, an initial description mining submodel in the operation and maintenance risk identification model, the identifying initial descriptive content of the remote education service operation report included in the first authenticated report tuple can be mined, then, the staged operation and maintenance risk item identification unit in the operation and maintenance risk identification model can determine staged test operation and maintenance risk items corresponding to each group of remote education service operation reports in the first authenticated report binary group and the staged risk description content identification unit in the operation and maintenance risk identification model according to the determined identification initial description content and the remote education service operation reports included in the first authenticated report binary group, the first test risk description content, the second test risk description content and the third test risk description content corresponding to each group of remote education service operation reports in the first authenticated report binary group and the integrated risk description content identification unit in the operation and maintenance risk identification model can be derived according to the staged test operation and maintenance risk matters and the identification initial description content corresponding to the first authenticated report binary group and the remote education service operation reports included in the first authenticated report binary group, and deriving an integrated fourth test risk description corresponding to each group of remote education service operation reports in the first authenticated report binary group according to the identification initial description content and the remote education service operation reports included in the first authenticated report binary group. Wherein the fourth test risk description is an exported integrated risk description.

In this way, a second model quality evaluation and a third model quality evaluation can be constructed through the staged test operation and maintenance risk item, the first test risk description content, the second test risk description content and the third test risk description content, the staged operation and maintenance risk item identification unit and the staged risk description content identification unit are configured in a circulating mode, the third model quality evaluation can be generated through the fourth test risk description content, and meanwhile, the integrated risk description content identification unit is configured in a circulating mode.

It is to be understood that, in the cyclic configuration of the integrated risk description content identification unit by the certified reporting binary group, for one certified remote education service operation report, the corresponding active certified remote education service operation report may be a remote education service operation report having a transfer characteristic with the certified remote education service operation report, and the passive certified remote education service operation report may be all remote education service operation reports having no transfer characteristic with the current remote education service operation report, so that for each imported certified report binary, the certified remote education service operation report and the certified related service operation report included therein have a transfer characteristic, and thus for each certified remote education service operation report included in the imported certified report binary, there is only an active certified remote education service operation report, and there is no passive remote certified remote education service operation report, but in order to increase the recognition reliability of the integrated test risk description content derived by the configured integrated risk description content identification unit, an embodiment of the present invention proposes a concept according to the assigned space, and stores the determined integrated test risk description content of the configured global risk description in the global risk description unit as a circular configuration, and the integrated test risk description of the integrated test risk description of the integrated risk description.

In practical implementation, after the fourth test risk description content is determined, the fourth test risk description content may be recorded in a designated space as an integrated test risk description content corresponding to a negative authenticated report configured in a next cycle, and a determination time period of each fourth test risk description content is recorded, but at this time, the third model quality evaluation is not determined by the fourth test risk description content, in other words, the integrated risk description content identification unit is not configured, so that the number of integrated test risk description contents corresponding to the negative authenticated report at this time is small, and the configuration effect is poor. After the configuration of the current round is completed, a new first authenticated report binary group screened from the determined multiple authenticated report binary groups is loaded into the operation and maintenance risk identification model again, the integrated risk description content identification unit derives a fourth test risk description content corresponding to the new first authenticated report binary group, records the fourth test risk description content in a designated space, records a determined time period of the fourth test risk description content determined in real time, calculates the fourth test risk description content one by one, and completes the reset of the designated space (storage sequence) when the configuration times of the operation and maintenance risk identification model meet the set times, at the moment, the designated space already records a certain number of fourth test risk description contents, and the configuration is performed through the recorded fourth test risk description contents, so that a better configuration effect can be met, and therefore, the authenticated report binary group imported from the current round is used for the configuration of the operation and maintenance risk identification model.

In practical implementation, second authenticated report duplets different from the first authenticated report duplets and selected from the determined multiple authenticated report duplets may be loaded into the operation and maintenance risk identification model, so that an integrated fifth test risk description content of each remote education service operation report in each second authenticated report duplet may be obtained, and then a third model quality evaluation for configuring the operation and maintenance risk identification model may be determined through the fifth test risk description content and a fourth test risk description content in a designated space.

In some examples, the fifth test risk description may include a sixth test risk description corresponding to an authenticated report in the second authenticated reporting duplet and a seventh test risk description corresponding to an authenticated correlation service run report in the second authenticated reporting duplet.

In practical implementation, for each second certified reporting doublet of the plurality of second certified reporting doublets loaded into the operation and maintenance risk identification model, a sixth test risk description content corresponding to a certified report in each second certified reporting doublet and a seventh test risk description content corresponding to a certified related service operation report in the second certified reporting doublet may be determined, and then a fourth test risk description content having the largest commonality index with the sixth test risk description content may be determined from fourth test risk description contents recorded in a designated space as integrated test risk description contents corresponding to negative certified reports, and the fourth test risk description content may be used as integrated test risk description contents corresponding to negative certified reports for determining a third model quality evaluation, and the eighth test risk description content may be used for determining a third model quality evaluation through the sixth test risk description content, the seventh test risk description content and the eighth test risk description content, wherein the third model quality evaluation may be performed on the integrated test risk description contents, and the current configuration risk loss of the operation risk identification model may be a hinge quality evaluation.

In addition, after the eighth test risk description content corresponding to the negative type authenticated report of the current round is determined, the fifth test risk description content obtained by the current round can be recorded into the designated space, and according to the determination time period of the fourth test risk description content recorded in the designated space and the third number of the fifth test risk description content, the third number of the fourth test risk description content before the determination time period in the designated space is eliminated, so that the optimization of the designated space is completed. And then, the next configuration of the operation and maintenance risk identification model can be completed according to the steps through a new second authenticated remote education service operation report, and finally, the configured operation and maintenance risk identification model can be obtained.

It can be understood that the design idea of circularly configuring the operation and maintenance risk identification model according to the specified space is provided by the embodiment of the present invention.

In addition, when the operation and maintenance risk identification model is configured circularly each time, the configuration can also be understood as the configuration of the initial description mining submodel, and when the configured operation and maintenance risk identification model is obtained, the configuration of the initial description mining submodel is also completed, so that the identification initial description content with better precision can be derived.

Therefore, the configured operation and maintenance risk identification model obtained through the simultaneously configured ideas can parallelly determine the staged operation and maintenance risk items of the remote education service operation report to be subjected to the wind control analysis with high precision, the staged risk description contents corresponding to the staged operation and maintenance risk items and the integrated risk description contents of the remote education service operation report to be subjected to the wind control analysis according to the mined initial service operation and maintenance risk description, and the quality of the wind control analysis of the remote education service operation report to be subjected to the wind control analysis is improved.

In some examples, the big data wind control method for the remote education service provided in the embodiment of the present invention may further include, first, circularly configuring, by the authenticated reporting binary group, the initial description mining sub-model, the staged operation and maintenance risk item identification unit, and the staged risk description content identification unit in the operation and maintenance risk identification model, after the configuration is completed, circularly configuring, by the authenticated reporting binary group, the integrated risk description content identification unit in the operation and maintenance risk identification model, and finally obtaining the configured operation and maintenance risk identification model, where an embodiment of the present invention is not limited herein.

Based on the same inventive concept, fig. 2 illustrates a block diagram of a big data wind control device for distance education service provided by an embodiment of the present invention, and the big data wind control device for distance education service may include the following modules implementing the related method steps illustrated in fig. 1.

And the report determining module 21 is used for determining the running report of the remote education service to be subjected to the wind control analysis.

And the risk mining module 22 is used for mining the initial service operation and maintenance risk description of the remote education service operation report to be subjected to the wind control analysis.

And the description analysis module 23 is configured to determine, according to the initial service operation and maintenance risk description, staged operation and maintenance risk items of the remote education service operation report to be subjected to the wind control analysis, staged risk description contents corresponding to the staged operation and maintenance risk items, and integrated risk description contents of the remote education service operation report to be subjected to the wind control analysis.

The related embodiment applied to the invention can achieve the following technical effects: the method comprises the steps of firstly mining initial service operation and maintenance risk description through a machine learning model, then determining staged operation and maintenance risk items, staged risk description contents and integrated risk description contents in parallel according to the mined initial service operation and maintenance risk description, avoiding independently mining the staged operation and maintenance risk items, staged risk description contents and integrated risk description contents, ensuring mining timeliness, determining the three types of key data through the integrally mined initial service operation and maintenance risk description because the initial service operation and maintenance risk description is the risk description contents which need to be referred to when the three types of key data are mined, not needing to mine the initial service operation and maintenance risk description when each type of key data is determined to cause frequent mining of the initial service operation and maintenance risk description, and reducing resource overhead of wind control analysis to a certain extent.

The foregoing is only illustrative of the present invention. Those skilled in the art can conceive of changes or substitutions based on the specific embodiments provided by the present invention, and all such changes or substitutions are intended to be included within the scope of the present invention.

Claims

1. A big data wind control method for remote education service, which is characterized in that the method is applied to a big data wind control server and at least comprises the following steps:

determining a remote education service operation report to be subjected to wind control analysis, and mining initial service operation and maintenance risk description of the remote education service operation report to be subjected to wind control analysis;

according to the initial service operation and maintenance risk description, determining staged operation and maintenance risk items of the remote education service operation report to be subjected to the wind control analysis, staged risk description contents corresponding to the staged operation and maintenance risk items and integrated risk description contents of the remote education service operation report to be subjected to the wind control analysis;

the step operation and maintenance risk items of the remote education service operation report to be subjected to the wind control analysis, the step risk description content corresponding to the step operation and maintenance risk items and the integrated risk description content of the remote education service operation report to be subjected to the wind control analysis are determined according to the initial service operation and maintenance risk description, and the step operation and maintenance risk description content comprises the following steps: obtaining staged operation and maintenance risk items of the remote education service operation report to be subjected to the wind control analysis, staged risk description contents corresponding to the staged operation and maintenance risk items and integrated risk description contents of the remote education service operation report to be subjected to the wind control analysis according to the initial service operation and maintenance risk description through an operation and maintenance risk identification model; configuring the operation and maintenance risk identification model, including: determining a plurality of sets of authenticated remote education service operation reports; for each group of authenticated remote education service operation reports, determining at least one group of authenticated association service operation reports with the authenticated remote education service operation reports having transfer characteristics, and binding the authenticated remote education service operation reports with each group of authenticated association service operation reports respectively to obtain at least one authenticated report binary group; configuring the operation and maintenance risk identification model through the authenticated report binary group;

wherein the configuring the operation and maintenance risk identification model through the authenticated report duplet comprises: respectively determining staged guiding operation and maintenance risk items of each authenticated remote education service operation report in the authenticated report binary group; determining active example operation and maintenance risk items and passive example operation and maintenance risk items corresponding to the staged guiding operation and maintenance risk items from the authenticated associated service operation reports in the authenticated report duplet; configuring the operation and maintenance risk identification model through the staged guiding operation and maintenance risk items, and active example operation and maintenance risk items and passive example operation and maintenance risk items corresponding to the staged guiding operation and maintenance risk items; wherein, the determining the active example operation and maintenance risk items and the passive example operation and maintenance risk items corresponding to the staged guiding operation and maintenance risk items from the authenticated associated service operation reports in the authenticated report duplet includes: taking the operation and maintenance risk items corresponding to the staged guiding operation and maintenance risk items in the authenticated associated service operation report as active example operation and maintenance risk items; and determining the operation and maintenance risk item with the maximum common index with the staged guiding operation and maintenance risk item from the operation and maintenance risk items except the active example operation and maintenance risk item in the authenticated associated service operation report, and taking the operation and maintenance risk item with the maximum common index as the passive example operation and maintenance risk item.

2. The method of claim 1, wherein the authenticated remote education service execution report includes a first authenticated service session report, and wherein the authenticated report duplet encompasses a first authenticated session report duplet; the determining, for each set of authenticated teleeducation service operation reports, at least one set of authenticated association service operation reports having transfer characteristics with the authenticated teleeducation service operation reports, and binding the authenticated teleeducation service operation reports with each set of authenticated association service operation reports, respectively, to obtain at least one authenticated report binary, including:

determining a plurality of sets of first authenticated service session reports under a target service session;

determining dynamic service session data corresponding to the target service session according to a plurality of groups of the first authenticated service session reports;

and determining at least one pair of first authenticated service session reports with transfer characteristics from the first authenticated service session reports according to the dynamic service session data to obtain at least one first authenticated session report tuple.

3. The method of claim 2, wherein the authenticated remote education service execution report includes a second authenticated service session report, and wherein the second authenticated report tuple encompasses a second authenticated session report tuple; the determining, for each set of authenticated teleeducation service operation reports, at least one set of authenticated association service operation reports having transfer characteristics with the authenticated teleeducation service operation reports, and binding the authenticated teleeducation service operation reports with each set of authenticated association service operation reports, respectively, to obtain at least one authenticated report binary, including:

determining a plurality of sets of second authenticated service session reports under a non-target service session;

and for each group of second authenticated service session reports, performing first preprocessing on the second authenticated service session reports to obtain at least one group of target associated service operation reports with transfer characteristics of the second authenticated service session reports, and binding the second authenticated service session reports with each group of target associated service operation reports respectively to obtain at least one second authenticated session report binary group.

4. The method of claim 3, wherein the configuring the operation and maintenance risk identification model through the authenticated reporting duplet comprises:

performing second preprocessing on a part of the authenticated report binary group to obtain at least one authenticated report binary group which is subjected to the second preprocessing; the authenticated reporting duplet encompasses a first authenticated reporting duplet and/or a second authenticated reporting duplet;

configuring the operation and maintenance risk identification model through the part of the authenticated report duplet and the authenticated report duplet which completes the preprocessing; wherein the second pre-processing comprises one or more of: non-limiting disturbance processing and report feature recognition degree adjustment processing.

5. The method of claim 1, wherein the configuring the operation and maintenance risk identification model through the staged guiding operation and maintenance risk item, the active example operation and maintenance risk item and the passive example operation and maintenance risk item corresponding to the staged guiding operation and maintenance risk item comprises:

loading the authenticated report binary group into the operation and maintenance risk identification model to obtain a staged test operation and maintenance risk item, a first test risk description content corresponding to the staged guide operation and maintenance risk item, a second test risk description content corresponding to the active example operation and maintenance risk item, and a third test risk description content corresponding to the passive example operation and maintenance risk item;

determining a first model quality evaluation according to the staged test operation and maintenance risk items and the staged guide operation and maintenance risk items;

determining a second model quality evaluation through a first test risk description content corresponding to the staged guiding operation and maintenance risk item, a second test risk description content corresponding to the active example operation and maintenance risk item and a third test risk description content corresponding to the passive example operation and maintenance risk item;

configuring the operation and maintenance risk identification model through the first model quality evaluation and the second model quality evaluation;

the loading the authenticated report binary group into the operation and maintenance risk identification model to obtain a staged test operation and maintenance risk item, a first test risk description content corresponding to the staged guide operation and maintenance risk item, a second test risk description content corresponding to the active example operation and maintenance risk item, and a third test risk description content corresponding to the passive example operation and maintenance risk item includes: loading the authenticated report binary group into the operation and maintenance risk identification model, and mining a sub-model according to the initial description in the operation and maintenance risk identification model to obtain a test type initial service operation and maintenance risk description corresponding to the authenticated report binary group; and identifying a sub-model by the description in the operation and maintenance risk identification model, and obtaining staged test operation and maintenance risk items, first test risk description contents corresponding to the staged guide operation and maintenance risk items, second test risk description contents corresponding to the active example operation and maintenance risk items and third test risk description contents corresponding to the passive example operation and maintenance risk items according to the test model initial service operation and maintenance risk description.

6. The method of claim 1, wherein the configuring the operation and maintenance risk identification model through the staged guiding operation and maintenance risk item, the active example operation and maintenance risk item and the passive example operation and maintenance risk item corresponding to the staged guiding operation and maintenance risk item comprises:

loading a plurality of first authenticated report binary groups to the operation and maintenance risk identification model, respectively obtaining integrated fourth test risk description contents corresponding to each group of remote education service operation reports in each first authenticated report binary group, and adding the integrated fourth test risk description contents to a designated space;

configuring the operation and maintenance risk identification model according to a plurality of second authenticated report duplets and fourth test risk description contents in the designated space;

wherein the configuring the operation and maintenance risk identification model according to the plurality of second authenticated reporting duplets and the fourth test risk description content in the designated space includes: loading a plurality of second authenticated report duplets to the operation and maintenance risk identification model, and respectively obtaining integrated fifth test risk description contents of each remote education service operation report in each second authenticated report duplet; determining a third model quality evaluation according to a fourth test risk description content and the fifth test risk description content in the designated space, and configuring the operation and maintenance risk identification model through the third model quality evaluation;

wherein, after obtaining the integrated fifth test risk description content of each remote education service operation report in each second authenticated report binary group, the method further comprises: optimizing the designated space according to the integrated fifth test risk description of each remote education service operation report.

7. The method of claim 6, wherein the fifth test risk description comprises a sixth test risk description and a seventh test risk description; determining a third model quality evaluation according to a fourth test risk description and the fifth test risk description in the designated space, including:

for each second authenticated report binary group, determining a sixth test risk description content of an authenticated remote education service operation report in the second authenticated report binary group, taking an authenticated association service operation report in the second authenticated report binary group as an active authenticated remote education service operation report corresponding to the authenticated report, and determining a seventh test risk description content of the active authenticated remote education service operation report;

determining a fourth test risk description content with the largest commonality index with a sixth test risk description content of the certified remote education service operation report in the second certified report binary group from the designated space, and taking the determined fourth test risk description content as an eighth test risk description content corresponding to a negative certified report;

and determining third model quality evaluation according to the sixth test risk description content, the seventh test risk description content and the eighth test risk description content.

8. A big data wind control server, comprising: a memory and a processor; the memory and the processor are coupled; the memory for storing computer program code, the computer program code comprising computer instructions; wherein the computer instructions, when executed by the processor, cause the big data wind server to perform the method of any of claims 1-7.