CN118133085A

CN118133085A - Control method and device for interaction task and computer equipment

Info

Publication number: CN118133085A
Application number: CN202410322176.2A
Authority: CN
Inventors: 杨林; 李邵杰
Original assignee: China Construction Bank Corp; CCB Finetech Co Ltd
Current assignee: China Construction Bank Corp; CCB Finetech Co Ltd
Priority date: 2024-03-20
Filing date: 2024-03-20
Publication date: 2024-06-04

Abstract

The disclosure relates to the technical field of big data, and particularly discloses a control method, a control device, computer equipment, a storage medium and a computer program product of an interaction task. The method comprises the following steps: responding to triggering operation of an account to a current interaction task, and acquiring a historical interaction task of the account, a corresponding risk category and interaction data of the current interaction task; determining a first risk detection result of the account according to the historical interaction task, the corresponding risk category and a preset first risk detection model; determining a second risk detection result of the account according to the interaction data of the current interaction task and a second risk detection model; and determining an interaction result of the current interaction task according to the first risk detection result, the second risk detection result and the risk strategy corresponding to the current interaction task. The method can improve the processing efficiency of the interactive task.

Description

Control method and device for interaction task and computer equipment

Technical Field

The present application relates to the field of big data technologies, and in particular, to a method, an apparatus, a computer device, a storage medium, and a computer program product for controlling an interaction task.

Background

With the continuous development of the internet, more and more application programs emerge. The application programs provide more benefits for users, so that the users are attracted to increase the application program usage amount, and the users can enjoy the benefits only by completing certain interaction tasks. However, as the number of users participating in an interaction task increases, so too does the act of taking advantage of activity vulnerabilities multiple times. Thus, risk control over interactive tasks becomes more important.

In the conventional risk control mode, when new interaction data are generated in the interaction process, various historical interaction data of an interaction account number are queried from a database, and are processed and analyzed in a combined mode. However, in this way, in the query process, all the interactive data needs to be queried every time, and under the condition of large interactive data volume, a large amount of computing resources are consumed, so that the memory is consumed, the risk control is affected, and the interaction task cannot normally operate.

Disclosure of Invention

In view of the foregoing, it is desirable to provide a method, an apparatus, a computer device, a computer-readable storage medium, and a computer program product for controlling an interactive task.

In a first aspect, the present application provides a method for controlling an interaction task. The method comprises the following steps:

responding to triggering operation of an account to a current interaction task, and acquiring a historical interaction task of the account, a corresponding risk category and interaction data of the current interaction task;

Determining a first risk detection result of the account according to the historical interaction task, the corresponding risk category and a preset first risk detection model;

Determining a second risk detection result of the account according to the interaction data of the current interaction task and a second risk detection model;

And determining an interaction result of the current interaction task according to the first risk detection result, the second risk detection result and the risk strategy corresponding to the current interaction task.

In one embodiment, the obtaining the interaction data of the historical interaction task and the corresponding risk category of the account number, and the current interaction task includes:

Under the condition that the account number does not trigger the current task for the first time, acquiring a first risk detection result of the account number and interaction data of the current interaction task;

And under the condition that the account carries out triggering operation on the current task for the first time, acquiring the historical interaction task of the account, the corresponding risk category and the interaction data of the current interaction task.

In one embodiment, the acquiring manner of the first risk detection model includes:

acquiring the risk category of the sample account in the history interaction task and the related history data of the sample account; the relevant historical data are used for representing risk categories of the sample account in non-interactive tasks;

and training an initial risk detection model according to the historical information and the related historical data to obtain the first risk detection model.

In one embodiment, the second risk detection model includes sub-models with different dimensions, and the determining, according to the interaction data of the current interaction task and the second risk detection model, a second risk detection result of the account includes:

Dividing the interaction data of the current interaction task into data with different dimensions according to different dimensions of the sub-model;

respectively inputting the data with different dimensions into corresponding sub-models to obtain risk detection results with different dimensions;

and determining a second risk detection result of the interaction data according to the risk detection results of the different dimensions.

In one embodiment, the training manner of the first risk detection model further includes:

Acquiring an interaction result of an account;

adding the interaction result of the account to the risk category of the account in the historical interaction task;

according to a preset rule, adjusting the weight of the risk category in the historical interaction task of the account;

and inputting the adjusted risk category and weight into the first risk detection model, and optimizing the first risk detection model.

In one embodiment, the method further comprises:

according to the interaction data and the current interaction task, determining the task completion degree of the account in sequence;

And determining a target account according to the task completion degree of the account and the interaction result.

In a second aspect, the application further provides a control device for the interaction task. The device comprises:

The data acquisition module is used for responding to the triggering operation of the account on the current interaction task and acquiring the historical interaction task of the account, the corresponding risk category and the interaction data of the current interaction task;

the risk prediction module is used for determining a first risk detection result of the account according to the historical interaction task, the corresponding risk category and a preset first risk detection model;

the risk prediction module is further configured to determine a second risk detection result of the account according to the interaction data of the current interaction task and a second risk detection model;

and the result determining module is used for determining the interaction result of the current interaction task according to the first risk detection result, the second risk detection result and the risk strategy corresponding to the current interaction task.

In one embodiment, the data acquisition module includes:

the data acquisition sub-module is used for acquiring a first risk detection result of the account and interaction data of a current interaction task under the condition that the account does not trigger the current task for the first time;

The data acquisition sub-module is further configured to acquire, when the account performs a triggering operation on a current task for the first time, a historical interaction task of the account, a corresponding risk category, and interaction data of the current interaction task.

In one embodiment, the risk prediction module includes:

the data acquisition sub-module is used for acquiring the risk category of the sample account in the history interaction task and the related history data of the sample account; the relevant historical data are used for representing risk categories of the sample account in non-interactive tasks;

And the model training sub-module is used for training the initial risk detection model according to the historical information and the related historical data to obtain the first risk detection model.

In one embodiment, the risk prediction module includes:

the data multidimensional sub-module is used for dividing the interactive data of the current interactive task into data with different dimensionalities according to the different dimensionalities of the sub-model;

The risk prediction sub-module is used for respectively inputting the data with different dimensions into the corresponding sub-model to obtain risk detection results with different dimensions;

And the risk determination submodule is used for determining a second risk detection result of the interaction data according to the risk detection results of the different dimensions.

In one embodiment, the model training sub-module further comprises:

the result acquisition unit is used for acquiring the interaction result of the account;

The data adding unit is used for adding the interaction result of the account to the risk category of the historical interaction task of the account;

The weight adjusting unit is used for adjusting the weight of the risk category in the historical interaction task of the account according to a preset rule;

And the model optimization unit is used for inputting the adjusted risk category and weight into the first risk detection model and optimizing the first risk detection model.

In one embodiment, the apparatus further comprises:

The condition confirmation module is used for sequentially determining the task completion degree of the account according to the interaction data and the current interaction task;

and the target confirmation module is used for determining a target account according to the task completion degree of the account and the interaction result.

In a third aspect, the present application also provides a computer device. The computer device comprises a memory and a processor, wherein the memory stores a computer program, and the processor realizes the control method of the interaction task according to any one of the embodiments of the disclosure when executing the computer program.

In a fourth aspect, the present application also provides a computer-readable storage medium. The computer readable storage medium has stored thereon a computer program which, when executed by a processor, implements a method of controlling an interactive task according to any of the embodiments of the present disclosure.

In a fifth aspect, the present application also provides a computer program product. The computer program product comprises a computer program which, when executed by a processor, implements a method of controlling an interactive task according to any one of the embodiments of the present disclosure.

According to the control method, the control device, the computer equipment, the storage medium and the computer program product of the interactive task, the risk detection model is utilized, the risk category of the account history interactive task and the interactive data of the current interactive task are used for predicting the account risk, and the risk is compared with the risk strategy of the current interactive task to obtain the interactive result of the current interactive task. By responding to the triggering operation of the account, the data of the current account can be obtained in real time, and the risk of the account is predicted according to the data in real time, so that the capability of controlling the risk of the interaction task is enhanced, and the safety of the interaction task system is remarkably improved. The first risk detection result is obtained by utilizing the historical interaction task and the corresponding risk category, and the potential risk account is identified by the historical risk category, so that the abnormal account is timely found, and the safety of the interaction task system is further improved; meanwhile, a second risk detection result is obtained by utilizing the interaction data of the current interaction task, the risk of the account can be predicted in time by the current interaction data, and the latest risk condition of the account is reflected. According to the risk strategy corresponding to the current interaction task, the first risk detection result and the second risk detection result are integrated, so that the interaction result of the account can be adjusted more flexibly and accurately, and the safety of the interaction task system is further improved. In addition, the method can automatically process the interaction task without manual intervention, greatly improves the processing efficiency of the interaction result, avoids subjectivity of manual processing, and improves the accuracy of the interaction result.

Drawings

FIG. 1 is an application environment diagram of a method of controlling interactive tasks in one embodiment;

FIG. 2 is a flow diagram of a method of controlling interactive tasks in one embodiment;

FIG. 3 is a schematic diagram of a data acquisition flow in one embodiment;

FIG. 4 is a flow chart of a first risk detection model training in one embodiment;

FIG. 5 is a flowchart illustrating determination of a second risk detection result according to an embodiment;

FIG. 6 is a flow diagram of a first risk detection model optimization in one embodiment;

FIG. 7 is a flow diagram of target account validation in one embodiment;

FIG. 8 is a first flow diagram of a control method implementation of an interactive task in one embodiment;

FIG. 9 is a second flow diagram of a control method implementation of an interactive task in one embodiment;

FIG. 10 is a block diagram of the control device of the interactive task in one embodiment;

FIG. 11 is a first internal block diagram of a computer device in one embodiment;

FIG. 12 is a second internal block diagram of a computer device in one embodiment.

Detailed Description

The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.

The technical scheme of the application obtains, stores, uses, processes and the like the data, which all meet the relevant regulations of national laws and regulations.

The control method of the interactive task provided by the embodiment of the application can be applied to an application environment shown in figure 1. Wherein the terminal 102 communicates with the server 104 via a network. The data storage system may store data that the server 104 needs to process. The data storage system may be integrated on the server 104 or may be located on a cloud or other network server. The account performs triggering operation on the current interaction task on the terminal 102, and the server 104 confirms the interaction result of the account after receiving the triggering operation of the account. The terminal 102 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, internet of things devices, and portable wearable devices, where the internet of things devices may be smart speakers, smart televisions, smart air conditioners, smart vehicle devices, and the like. The portable wearable device may be a smart watch, smart bracelet, headset, or the like. The server 104 may be implemented as a stand-alone server or as a server cluster of multiple servers.

In one embodiment, as shown in fig. 2, a method for controlling an interaction task is provided, and the method is applied to the server in fig. 1 for illustration, and includes the following steps:

Step S100, responding to triggering operation of an account to a current interaction task, and acquiring a history interaction task of the account, a corresponding risk category and interaction data of the current interaction task.

In an exemplary embodiment, the historical interaction tasks and the corresponding risk types of the account may include interaction results of interaction tasks that the account participated in and time of the corresponding interaction tasks; for example, account a is an abnormal interaction in the XX (task number) task of X years, X months and X days.

In an exemplary embodiment, the interaction operation of the current interaction task may include behavior data of the account in the current interaction task, such as clicking, browsing, purchasing, etc., and related data of task completion, task duration, etc. The server may obtain these interaction data via an interface, API, or other means of communication.

In an exemplary embodiment, the interactive operation of the current interactive task may include real behavior data, such as actions, expressions, sounds, etc. of the user acquired through a camera, a sensor, etc. device, and behavior features, emotional states, etc. of the user obtained by analyzing the data. The server may obtain the interaction data through an interface or other means of communication with the devices.

Step S200, determining a first risk detection result of the account according to the historical interaction task, the corresponding risk category and a preset first risk detection model.

In an exemplary embodiment, the first risk detection result of the account may include a category of the account, for example, the account is a high risk account, a long-term risk account, and the like; in another exemplary embodiment, the first risk detection result of the account may include a risk level of the account, for example, from 1 to 5, the risk level of the account is respectively indicated, and the risk level of the account a is 5, that is, the account a is a high risk account.

In an exemplary embodiment, the first risk detection model may include a support vector machine model; in another exemplary embodiment, the first risk detection model may include a Logistic regression model, e.g., a Logistic regression model, etc.

And step S300, determining a second risk detection result of the account according to the interaction data of the current interaction task and a second risk detection model.

In an exemplary embodiment, the second risk detection model may use a Support Vector Machine (SVM), a decision tree model, a random forest, a neural network, or the like.

In an exemplary embodiment, the second risk model may include detecting whether abnormal data exists; for example, a return to the order occurs immediately after purchase, a video is recorded with a camera to complete task data, and so on.

Step S400, determining an interaction result of the current interaction task according to the first risk detection result, the second risk detection result, and the risk policy corresponding to the current interaction task.

In an exemplary embodiment, the risk policy corresponding to the current interaction task may include setting an account number meeting the interaction result in the first risk detection result and the second risk detection result; for example, the first risk detection result includes a low risk account, a medium risk account, a high risk account, and the like, and the second risk result includes a low risk account, a medium risk account, a high risk account, and the like; the risk policy may include that the low risk account is an account that meets the interaction result when the first risk detection result and the second risk detection result are both low risk accounts; in another exemplary embodiment, the risk policy may also include determining that the account does not conform to the interaction result if the first risk detection result or the second risk detection result is a high risk account.

In an exemplary embodiment, when the first risk detection result and the second risk detection result are risk levels, a preset weight may be set for the first risk detection result and the second risk detection result, so as to determine a target risk result of the account; the risk policy may include a rank requirement of the target risk result, and the risk policy may further include a rank requirement of the target risk result and a requirement for the first risk detection result and the second risk detection result; for example, the risk levels of the first risk detection result and the second risk detection result are sequentially 1-5 from low to high, the risk policy may be that the target risk result of the account is from risk level 1 to risk level 3, and no risk level 5 exists in the first risk detection result and the second risk detection result.

In an exemplary embodiment, the interaction result of the current interaction task may include a compliance account under a risk policy; the system determines a target account through the task completion degree of the compliance account; in another exemplary embodiment, the interaction result of the current interaction task may include an out-of-compliance account under a risk policy; the system determines an initial target account number for the task completion degree of all the account numbers, and deletes the non-compliant account numbers from the initial target account number to obtain target account numbers; in another exemplary embodiment, the interaction result of the current interaction may include a compliance account number and a non-compliance account number under risk; the system determines the target account number through the compliant account number (or the non-compliant account number) and verifies the target account number by using the non-compliant account number (or the compliant account number).

In an exemplary embodiment, after the triggering operation of the account on the current interaction task is responded, relevant information of the account is obtained in real time, the information is pushed to the risk detection model in a streaming data mode in real time, an interaction result is generated, and the interaction result is pushed to the rights issuing system of the interaction task in a streaming data mode in real time to determine a target account.

In an exemplary embodiment, the control method of the interactive task may be applied to Complex Event Processing (CEP), where the account performs a triggering operation on the current task, and the complex event processing obtains a risk class of the historical interactive task and interactive data of the current interactive task; and matching the acquired data with a corresponding risk detection model, and comparing the output result with a configured task risk strategy to obtain an interaction result of the account.

In an exemplary embodiment, after the account performs a triggering operation on a current interaction task, the triggering operation and the account identifier may be pushed to a Complex Event Processing (CEP), the complex event processing obtains a historical risk category of the account by using the account identifier, and pushes the historical risk category to a first risk detection model, meanwhile, the complex event post-processing pushes the triggering operation to a second risk detection model, and finally, the first risk detection model and the second risk detection model push the detection result to the complex event processing, the complex time processing analyzes the detection result and a risk policy of the interaction task to obtain an interaction result, and pushes the interaction result to a rights issuing system to determine a target account, where the Complex Event Processing (CEP) may use a mainstream Flink CEP.

In the control method of the interactive task, the risk detection model is utilized to predict the account risk through checking the risk category of the account history interactive task and the interactive data of the current interactive task, and the risk detection model is compared with the risk strategy of the current interactive task to obtain the interactive result of the current interactive task. By responding to the triggering operation of the account, the data of the current account can be obtained in real time, and the risk of the account is predicted according to the data in real time, so that the capability of controlling the risk of the interaction task is enhanced, and the safety of the interaction task system is remarkably improved. The first risk detection result is obtained by utilizing the historical interaction task and the corresponding risk category, and the potential risk account is identified by the historical risk category, so that the abnormal account is timely found, and the safety of the interaction task system is further improved; meanwhile, a second risk detection result is obtained by utilizing the interaction data of the current interaction task, the risk of the account can be predicted in time by the current interaction data, and the latest risk condition of the account is reflected. According to the risk strategy corresponding to the current interaction task, the first risk detection result and the second risk detection result are integrated, so that the interaction result of the account can be adjusted more flexibly and accurately, and the safety of the interaction task system is further improved. In addition, the method can automatically process the interaction task without manual intervention, greatly improves the processing efficiency of the interaction result, avoids subjectivity of manual processing, and improves the accuracy of the interaction result.

In one embodiment, as shown in fig. 3, step S100 includes:

step S101, under the condition that the account number does not trigger the current task for the first time, a first risk detection result of the account number and interaction data of the current interaction task are obtained.

In an exemplary embodiment, when the triggering operation is not performed on the current task for the first time, a first risk detection result when the triggering operation is performed on the current task for the first time may be directly obtained; in another exemplary embodiment, if the difference between the first risk detection result and the second risk detection result is not within the preset range, re-acquiring the historical interaction task and the corresponding risk category of the account, and re-determining the first risk detection result of the account; for example, the first risk detection result and the second risk detection result sequentially represent that the risk is from low to high from risk level 1 to risk level 5, and when the difference between the first risk detection result and the second risk detection result is greater than 2, the historical interaction task and the corresponding risk category of the account are re-acquired, and the first risk detection result of the account is re-acquired.

In an exemplary embodiment, when the account is detected to reach the task completion degree of the interaction task, the first risk detection result of the account is detected again, and when the two first risk detection results are different, the first risk detection result may be detected again, or the target first risk detection result may be determined by using a special domestic discipline.

Step S102, under the condition that the account carries out triggering operation on the current task for the first time, acquiring the historical interaction task of the account, the corresponding risk category and the interaction data of the current interaction task.

In an exemplary embodiment, the situation that the account performs the triggering operation on the current task for the first time may include a part of the account completing the interaction task; the account participates in the interaction task; in another exemplary embodiment, the situation that the account performs the triggering operation on the current task for the first time may further include a situation that the interaction task is completed by the account and the completion degree of the acquisition right is reached.

In an exemplary embodiment, in a case that the account interacts with the interaction task, it may be confirmed that the account does not perform a triggering operation with the current task.

In this embodiment, when the account number does not trigger the interaction task for the first time, the first risk detection result predicted for the first time may be directly obtained, so that repeated detection of all data is avoided, calculation resources are saved, processing time is reduced, and processing efficiency of the interaction result is improved. Meanwhile, when the account triggers the interaction task for the first time, the historical interaction task and the corresponding risk category are acquired, and the risk information of the account in different tasks can be determined, so that the overall risk information of the account is judged more accurately, the potential risk account is predicted, and the accuracy of the interaction result is improved.

In one embodiment, as shown in fig. 4, step S200 includes:

step S201, acquiring risk categories of sample accounts in a history interaction task and relevant history data of the sample accounts; the relevant historical data are used for representing risk categories of the sample account in non-interactive tasks.

In an exemplary embodiment, the risk category of the sample account in the historical interaction task may include an interaction task in which the sample account participates historically, a risk category of the account, a time of the account, and the like; for example, account a is an abnormal interaction in the task XX (task number) of X years, X months and X days.

In an exemplary embodiment, the relevant historical data of the sample account may include scene-like behavior in other businesses within the ecology; such as credit information in the interactive resource, and specifically, may include loan offenders, credit card impression groups, and the like.

Step S202, training an initial risk detection model according to the historical information and the related historical data to obtain the first risk detection model.

In an exemplary embodiment, the first risk detection model may be trained according to a historical risk category of the sample account and related historical data by using a classification model to obtain a model capable of predicting the risk category of the account in the interaction task. Through the model, risk classification can be carried out on account numbers, and decision basis is provided for subsequent interaction task control.

In an exemplary embodiment, the first risk detection model may further predict a behavior of the account in the interaction task according to behavior data in the historical interaction task of the account and combine the relevant historical data of the account, and determine a risk category of the account according to a prediction result. Therefore, before the account performs the interaction task, risk early warning can be performed on the account, so that corresponding risk control measures can be timely taken, and the safety and smoothness of the interaction task are ensured.

In an exemplary embodiment, identification information of the account may be input to a first risk detection model, and the first risk prediction model obtains a first risk detection result based on a historical risk category and related historical data. In another exemplary embodiment, only the historical risk category of the account may be input into the first risk detection model, and the first risk detection result is predicted; in one exemplary embodiment, the classification model may be trained solely by risk categories of historical interaction tasks, resulting in a first risk detection model.

In this embodiment, training is performed on the classification model through the historical risk classification of the account and the related historical data to obtain a first risk detection model. The method and the device can determine potential risk information of the account by utilizing the historical risk category, predict risk behaviors of the account, and improve accuracy of a first risk detection result. Meanwhile, through the related historical data, the risk data of the account can be more completely determined, the risk control of the account is further improved, and the accuracy of the first risk detection result is improved.

In one embodiment, as shown in fig. 5, step S300 includes:

Step S301, dividing the interaction data of the current interaction task into data of different dimensions according to different dimensions of the sub-model.

In an exemplary embodiment, the different dimensions of the interaction task may include completion efficiency of the interaction task, frequency of the interaction task, and the like; for example, the interaction task completes 5% on average per active account per day, whereas the existing account completes 100% a day, then the account is a risk account; in the interaction task, each active account interacts ten times per minute on average, however, there are hundreds of thousands of interactions per minute for an account, and the account is a risk account.

In an exemplary embodiment, the different dimension sub-models may be trained using data of different dimensions, where the sub-models may use linear regression, polynomial regression, and the like; for example, support vector machines, random forests, neural networks, etc. are employed.

In an exemplary embodiment, the different dimension sub-model may include a plurality of, and the interaction data may include only data of a part of the different dimensions.

In an exemplary embodiment, according to the characteristics and risk characteristics of data in different dimensions, a suitable risk detection algorithm and model are selected for training, and a risk detection sub-model in a corresponding dimension is obtained. These sub-models may be run independently or may be used in combination to improve the accuracy and efficiency of risk detection.

Step S302, respectively inputting the data with different dimensions into corresponding sub-models to obtain risk detection results with different dimensions.

In an exemplary embodiment, the risk detection results of the different dimensions may include a behavior risk, a credit risk, and the like of the account in the interaction task. The risk detection results can provide decision basis for subsequent interactive task control, help enterprises to discover and control risks in time, and guarantee safety and smooth performance of interactive tasks.

Step S303, determining a second risk detection result of the interaction data according to the risk detection results of the different dimensions.

In an exemplary embodiment, the risk detection results of the different dimensions may be risk levels; the second risk detection result may include determining a risk level of the second risk detection result using a preset weight and risk levels of different dimensions; in another exemplary embodiment, the second risk detection result may include a highest level of the different dimensional risk levels; for example, the risk level of the account a in the a dimension is 3, and the risk level of the account a in the b dimension is 5, where it may be determined that the second risk detection result of the account a is 5.

In an exemplary embodiment, the second risk detection result of the interaction data may be comprehensively determined by using different dimension risk detection results, for example, the different dimension risk detection results only include risk existence and risk nonexistence, and the account may be determined to be a risk account when part of the data in different dimensions detects the risk existence; in another exemplary embodiment, if there is a risk in the risk monitoring results of different dimensions, then the second risk detection result is determined to be a risk.

In this embodiment, the second risk detection result is obtained comprehensively by dividing the interaction data of the current interaction task into data of different dimensions and inputting the data into sub-models of different dimensions. The method can predict the data of different dimensionalities of the interactive data, and improves the accuracy of integral prediction. The model is divided into a plurality of sub-models, so that the complexity of the model is reduced, the model is convenient to manage, and meanwhile, the model prediction efficiency is improved. In addition, the model is divided into a plurality of sub-models, so that the flexibility of model prediction is improved, prediction errors caused by missing data are avoided, and the accuracy of the second risk detection result is improved.

In one embodiment, as shown in fig. 6, the method further comprises:

Step S601, an interaction result of the account is obtained.

In an exemplary embodiment, the time for obtaining the interaction result of the account number may be at the end of the task, or the task may be performed in a middle.

Step S602, adding the interaction result of the account to the risk category of the account in the history interaction task.

In an exemplary embodiment, the adding the interaction result of the account to the risk category includes adding the interaction task code, the interaction task date, the interaction result, and the like of the account to the risk category in the history interaction task.

Step S603, adjusting the weight of the risk category in the historical interaction task of the account according to a preset rule.

In an exemplary embodiment, according to a preset rule, the weight of the risk category in the historical interaction task of the account number can be adjusted to more accurately reflect the risk condition of the account number. For example, historical interaction tasks that are farther from the current time may be set with lower weights for risk categories because the farther historical data may have less impact on the current risk prediction. In this way, the model can be ensured to pay more attention to recent interaction behaviors and data when predicting account risk.

Step S604, inputting the adjusted risk category and weight into the first risk detection model, and optimizing the first risk detection model.

In an exemplary embodiment, by continuously feeding back the interaction result of the account number to the risk category of the historical interaction task and adjusting the weight of the risk category according to a preset rule, the first risk detection model can be continuously learned and optimized, and accuracy and efficiency of risk detection are improved.

In one exemplary embodiment, the model may be training optimized when the interaction results are generated by adding the interaction results in the form of streaming data to risk categories of historical interaction tasks.

In the embodiment, the interaction result of the account is added to the risk category of the account in the historical interaction task, the weight of the historical interaction task is adjusted, and the first risk detection model is optimized, so that the model can be optimized through the interaction result, and the model can be more accurate. By adjusting the weight of the risk category, the model can use a plurality of problem scenes, and the generalization capability of the model is improved.

In one embodiment, as shown in fig. 7, the method includes:

step S701, determining the task completion degree of the account sequentially according to the interaction data and the current interaction task.

In an exemplary embodiment, the interaction result may be stored in the Redis cache, and a validity period is set for a certain period of time, and after the validity period, the interaction result of each account is recalculated, for example, the interaction result of the account a for the first time is that no abnormality is found, and after the validity period is recalculated, an abnormality may exist.

In an exemplary embodiment, the task completion degree may include the number of interactions of the task, etc., for example, the number of clicks, browses, purchases, etc.; in an exemplary embodiment, the task completion may include an interaction time, etc., e.g., a viewing duration, a movement duration, etc.; in another exemplary embodiment, the interaction completion degree may include determining the ranking of the respective accounts by using the interaction times or interaction times of the respective accounts, that is, ordering the respective accounts by using the interaction data of the respective accounts.

Step S702, determining a target account according to the task completion degree of the account and the interaction result.

In one exemplary embodiment, an interactive task system may be utilized to receive a trigger operation of an account to determine account information and interaction data; and the account information and the interaction data are transmitted to Complex Event Processing (CEP), the complex event processing uses the control method of the interaction task to determine the interaction result of the account, the complex event processing transmits the interaction result to the interaction task system again, the interaction task system determines the task completion degree according to the interaction data, and determines the target account according to the task completion degree and the interaction result, so that the decoupling between the complex event processing and the interaction task system is realized.

In the embodiment, the task completion degree of the account is determined through the interaction data, and the target account is determined by utilizing the interaction result and the task completion degree, so that the interaction result system and the operation system can be separated, decoupling of the two systems is realized, dependence between the two systems is reduced, complexity of the systems is reduced, and confirmation efficiency of the target object is improved.

In an exemplary embodiment, the method for controlling the interaction task may be implemented in an agricultural rural operation activity, as shown in fig. 8, and may specifically include:

step S881, obtaining sample data of historical activities and sample behavior information of a user of the historical activities;

Step S882, training the support vector machine model by using the sample data of the historical activity and the sample behavior information of the historical activity user to obtain a support vector machine model of the behavior information and a support vector machine model of the activity data with different dimensions;

Step S810, an agricultural rural operation system is established, and an activity flow, a participation mode, a specific limiting strategy and the like are formulated;

step S820, the operation system issues activity information to the outside;

Step S860, the active user participates in the activity;

Step S830, the active user completes the active main process and generates active data;

Step S840, determining the behavior information of the active user in the history related event by using the information of the active user, and pushing the behavior information and the active data to the CEP;

Step 870, dividing the received activity data into data with different dimensions corresponding to the support vector machine, pushing the behavior information and the data with different dimensions into the corresponding support vector machine, and pushing the risk data existing after calculation to the interaction system;

Step S840, the activity system receives the activity data and the risk data, determines the completion condition of the activity by using the activity configuration rule, and determines the rights and interests user by using the risk data;

step S850, issuing rights to the rights user, and training and optimizing the support vector machine model by taking the activity data generated by the activity as the sample data of the historical activity and the sample behavior information of the historical activity user.

In an exemplary embodiment, the method for controlling the interaction task may be implemented in the development of an enterprise activity, as shown in fig. 9, and may specifically include:

step S910, the activity official issues the activity outwards;

Step S920, the active user participates in the active main process published by the active official and generates active data;

Step S951, obtaining sample data of historical activities and sample behavior information of a user of the historical activities;

Step S952, an initial support vector machine model is established;

Step S953, training an initial support vector machine model by utilizing sample data of historical activities and sample behavior information of a user of the historical activities;

Step S954, testing the trained support vector machine model by using a test sample, and confirming a target data model under the condition that the model accuracy reaches a preset accuracy;

Step S921, determining behavior information of the active user in the history related event by utilizing the information of the active user, and pushing the behavior information and the active data to the CEP;

step S922, the CEP divides the activity data into data with different dimensions, and pushes the behavior information and the data with different dimensions to the corresponding SVM, to determine whether there is a risk;

step S923, the judged risk data is pushed to an active service system by using Kafka, and the data is stored in a Redis cache;

Step S930, receiving the activity data and the risk data in the Redis cache, determining the completion condition of the activity user by using the configured activity rule, and screening to obtain the rights object according to the risk data;

step S940, issuing rights to the rights object, and training and optimizing the support vector machine model by taking the activity data generated by the activity as the sample data of the historical activity and the sample behavior information of the historical activity user.

It should be understood that, although the steps in the flowcharts related to the embodiments described above are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.

Based on the same inventive concept, the embodiment of the application also provides a control device for realizing the interactive task of the control method of the interactive task. The implementation of the solution provided by the device is similar to the implementation described in the above method, so the specific limitation in the embodiments of the control device for one or more interactive tasks provided below may refer to the limitation of the control method for the interactive task hereinabove, and will not be described herein.

In one embodiment, as shown in fig. 10, there is provided a control device 100 for interaction tasks, including: a data acquisition module 101, a risk prediction module 102, and a result determination module 103, wherein:

In one embodiment, the data acquisition module includes: a data acquisition sub-module, wherein:

In one embodiment, the risk prediction module includes: the system comprises a data acquisition sub-module and a model training sub-module, wherein:

In one embodiment, the risk prediction module includes: the system comprises a data multidimensional sub-module, a risk prediction sub-module and a risk determination sub-module, wherein:

In one embodiment, the model training sub-module further comprises: the system comprises a result acquisition unit, a data addition unit, a weight adjustment unit and a model optimization unit, wherein:

In one embodiment, the apparatus further comprises: a situation confirmation module and a target confirmation module, wherein:

The modules in the control device for interaction tasks may be implemented in whole or in part by software, hardware, or a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.

In one embodiment, a computer device is provided, which may be a server, and the internal structure of which may be as shown in fig. 11. The computer device includes a processor, a memory, an Input/Output interface (I/O) and a communication interface. The processor, the memory and the input/output interface are connected through a system bus, and the communication interface is connected to the system bus through the input/output interface. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer equipment is used for storing interaction data of historical interaction tasks, risk categories and current interaction tasks. The input/output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used for communicating with an external terminal through a network connection. The computer program, when executed by a processor, implements a method of controlling an interactive task.

In one embodiment, a computer device is provided, which may be a terminal, and the internal structure thereof may be as shown in fig. 12. The computer device includes a processor, a memory, an input/output interface, a communication interface, a display unit, and an input means. The processor, the memory and the input/output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input/output interface. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input/output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless mode can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. The computer program, when executed by a processor, implements a method of controlling an interactive task. The display unit of the computer device is used for forming a visual picture, and can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, can also be a key, a track ball or a touch pad arranged on the shell of the computer equipment, and can also be an external keyboard, a touch pad or a mouse and the like.

It will be appreciated by persons skilled in the art that the structures shown in fig. 11/12 are block diagrams of only portions of structures associated with the present inventive arrangements and are not limiting of the computer device to which the present inventive arrangements are applied, and that a particular computer device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.

It should be noted that, the user information (including but not limited to user equipment information, user personal information, etc.) and the data (including but not limited to data for analysis, stored data, presented data, etc.) related to the present application are information and data authorized by the user or sufficiently authorized by each party, and the collection, use and processing of the related data need to comply with the related laws and regulations and standards of the related country and region.

Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, database, or other medium used in embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high density embedded nonvolatile Memory, resistive random access Memory (ReRAM), magneto-resistive random access Memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric Memory (Ferroelectric Random Access Memory, FRAM), phase change Memory (PHASE CHANGE Memory, PCM), graphene Memory, and the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in various forms such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), etc. The databases referred to in the embodiments provided herein may include at least one of a relational database and a non-relational database. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processor referred to in the embodiments provided in the present application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, or the like, but is not limited thereto.

The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.

The foregoing examples illustrate only a few embodiments of the application and are described in detail herein without thereby limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of the application should be assessed as that of the appended claims.

Claims

1. A method for controlling an interactive task, the method comprising:

2. The method according to claim 1, wherein the obtaining the interaction data of the historical interaction task and the corresponding risk category of the account number, the current interaction task, comprises:

3. The method of claim 1, wherein the acquiring the first risk detection model includes:

4. The method according to claim 1, wherein the second risk detection model comprises sub-models of different dimensions, and the determining the second risk detection result of the account according to the interaction data of the current interaction task and the second risk detection model comprises:

5. A method according to claim 3, wherein the training mode of the first risk detection model further comprises:

Acquiring an interaction result of an account;

6. The method according to claim 1, wherein the method further comprises:

7. A control device for an interactive task, the device comprising:

8. The apparatus of claim 7, wherein the data acquisition module comprises:

9. The apparatus of claim 7, wherein the risk prediction module comprises:

10. The apparatus of claim 7, wherein the risk prediction module comprises:

11. The apparatus of claim 9, wherein the model training sub-module further comprises:

12. The apparatus of claim 7, wherein the apparatus further comprises:

13. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 6 when the computer program is executed.

14. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 6.

15. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 6.