Risk control method and device for multi-tenant platform, equipment and storage medium
By rating and weighting the driver-side behavior data of the multi-tenant platform, calculating credit scores and formulating risk control strategies, the problem of insufficient granularity of risk control in traditional technologies is solved, and more refined risk management is achieved.
Patent Information
- Application Number
- CN202410895127.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-04
- Publication Date
- 2024-11-26
AI Technical Summary
In a multi-tenant platform, traditional risk control methods cannot effectively reflect the differences between different tenants, resulting in the granularity of risk control and it is difficult to effectively monitor and manage fraud.
By obtaining the driver's historical and real-time behavior data, using a pre-trained scoring model to calculate the credit score, and determining the risk control strategy based on the weight and credit scores configured by the tenant, personalized risk management for different tenants is achieved.
It realizes more refined risk control under the multi-tenant platform, and can formulate targeted risk control strategies based on the credit scores and behavioral data of different tenants, improving the accuracy and effectiveness of risk management.
Smart Images

Figure CN119026899A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of computer technology, and in particular to a risk control method, apparatus, device and storage medium for a multi-tenant platform. Background Art
[0002] With the rapid development of cloud computing and big data technologies, the multi-tenant model has become a common service provision method. In a multi-tenant environment, different tenants share the same infrastructure and use virtualization technology to isolate and allocate resources. However, this model also brings problems such as high fraud risk and difficulty in risk control.
[0003] In traditional technology, the driver's behavior is generally monitored to identify whether there are behaviors such as fake trips, detours, malicious cancellations, etc. When these behaviors are monitored, corresponding measures are taken to control them. However, in a multi-tenant platform, the data of each tenant is isolated from each other, and there are certain differences between each tenant. Risk control based on traditional technology cannot reflect the differences between different tenants, resulting in insufficient granularity of risk control. Summary of the invention
[0004] Based on this, it is necessary to provide a risk control method, device, equipment and storage medium for a multi-tenant platform that can perform more sophisticated risk control on the multi-tenant platform in response to the above technical problems.
[0005] In a first aspect, an embodiment of the present disclosure provides a risk control method for a multi-tenant platform, comprising:
[0006] Obtain the historical behavior data and real-time behavior data of the current driver terminal. The current driver terminal is pre-bound to at least one tenant. The real-time behavior data is the behavior data generated by the current driver terminal under the current tenant.
[0007] Input historical behavior data into the pre-trained scoring model to obtain the scores of key factors;
[0008] Get the weight of the pre-configured key factor, which is the weight of the key factor corresponding to the current tenant;
[0009] Determine the current driver's credit score based on the weights of the key factors and the scores of the key factors;
[0010] Determine the risk control strategy based on the current driver's credit score and real-time behavior data, and perform corresponding operations on the current driver based on the risk control strategy.
[0011] In some embodiments, the risk control method for a multi-tenant platform further includes:
[0012] Get the credit score of each driver bound to the current tenant;
[0013] Determine the average credit score of the current tenants based on the credit scores of the drivers;
[0014] The traffic of the current tenant is controlled according to the average credit score of the current tenant.
[0015] In some embodiments, the flow of the current tenant is controlled according to the average credit score corresponding to the current tenant, including:
[0016] Receive a ride request from the passenger app;
[0017] Determine the car booking price for the current tenant according to the car booking request and the charging policy pre-configured by the current tenant;
[0018] When the ride-hailing price is within a preset range, and / or the average credit score is higher than a preset threshold, the ride-hailing request will be assigned to the driver bound to the current tenant to control the traffic of the current tenant.
[0019] In some embodiments, the risk control method for a multi-tenant platform further includes:
[0020] Get the traffic of each tenant;
[0021] The resource usage of each tenant is adjusted according to the traffic of each tenant, wherein the resource usage includes computing resource usage and storage resource usage.
[0022] In some embodiments, the key factors include multiple factors, and the credit score of the current driver is determined according to the weights of the key factors and the scores of the key factors, including:
[0023] According to the weight of each key factor and the score of each key factor, the weighted average of each key factor is calculated to obtain the credit score.
[0024] In some embodiments, the key factors include any one or more of the following:
[0025] Service quality, passenger evaluation, order completion rate, violation record and safety record.
[0026] In some embodiments, a risk control strategy is determined based on the current driver's credit score and real-time behavior data, including:
[0027] Use the stream processing framework to process real-time behavior data to detect whether the current driver has abnormal behavior;
[0028] When the current driver has abnormal behavior and the credit score is lower than the preset threshold, the corresponding risk control strategy is determined based on the abnormal behavior.
[0029] In a first aspect, an embodiment of the present disclosure provides a risk control device for a multi-tenant platform, comprising:
[0030] The first acquisition module is used to acquire the historical behavior data and real-time behavior data of the current driver terminal. The current driver terminal is pre-bound to at least one tenant. The real-time behavior data is the behavior data generated by the current driver terminal under the current tenant.
[0031] The scoring module is used to input historical behavior data into the pre-trained scoring model to obtain the scores of key factors;
[0032] The second acquisition module is used to acquire the weight of the pre-configured key factor, where the weight is the weight corresponding to the key factor under the current tenant;
[0033] A processing module, for determining the credit score of the current driver according to the weights of the key factors and the scores of the key factors;
[0034] The control module is used to determine the risk control strategy based on the current driver's credit score and real-time behavior data, and perform corresponding operations on the current driver according to the risk control strategy.
[0035] In a third aspect, an embodiment of the present disclosure provides a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the risk control method for a multi-tenant platform provided in any embodiment of the present disclosure in the first aspect are implemented.
[0036] In a fourth aspect, an embodiment of the present disclosure provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the risk control method for a multi-tenant platform provided in any embodiment of the present disclosure in the first aspect.
[0037] The risk control method, apparatus, device and storage medium of the multi-tenant platform configure each tenant in advance and configure the weights of key factors under each tenant, wherein the key factors are factors used to calculate the driver's credit score, and the weights of key factors corresponding to different tenants are configured differently. The credit score of the current driver is calculated according to the configured weights of the key factors under the current tenant, and a risk control strategy is obtained based on the calculated credit score combined with the real-time behavior data of the current driver, so that when performing risk control, different risk control strategies are formulated according to different tenants, thereby realizing a more fine-grained risk control mechanism under the multi-tenant platform. BRIEF DESCRIPTION OF THE DRAWINGS
[0038] Figure 1 It is an application environment diagram of the risk control method of a multi-tenant platform in some embodiments;
[0039] Figure 2 A schematic diagram of a process of risk control method for a multi-tenant platform in some embodiments;
[0040] Figure 3 A principle block diagram of a risk control method for a multi-tenant platform in some embodiments;
[0041] Figure 4 It is a structural block diagram of a risk control device for a multi-tenant platform in some embodiments;
[0042] Figure 5 1 is a diagram of the internal structure of a computer device in some embodiments. DETAILED DESCRIPTION
[0043] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.
[0044] Please refer to Figure 1 , Figure 1 A schematic diagram of an application environment of a risk control method for a multi-tenant platform provided by an exemplary embodiment of the present application. Figure 1 As shown, the application environment includes a server 100 and a current driver terminal 101. The server 100 and the current driver terminal 101 can be communicatively connected through a network 102 to implement the risk control method of the multi-tenant platform of the present application.
[0045] The server 100 is used to obtain the historical behavior data and real-time behavior data of the current driver terminal 101. The current driver terminal 101 is pre-bound to at least one tenant. The real-time behavior data is the behavior data generated by the current driver terminal 101 under the current tenant. The historical behavior data is input into a pre-trained scoring model to obtain the scores of key factors, and the pre-configured weights of the key factors are obtained. The weights are the corresponding weights of the key factors under the current tenant. The credit score of the current driver is determined based on the weights of the key factors and the scores of the key factors. The risk control strategy is determined based on the current driver's credit score and real-time behavior data, and corresponding operations are performed on the current driver terminal 101 according to the risk control strategy.
[0046] The server 100 may be implemented as an independent server or a server cluster consisting of multiple servers.
[0047] The current driver terminal 101 is used to upload historical behavior data and real-time behavior data to the server 100. The current driver terminal 101 can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, and portable wearable devices.
[0048] The network 102 is used to realize the network connection between the terminal 101 and the server 100. Specifically, the network 102 may include various types of wired or wireless networks.
[0049] In a first aspect, an embodiment of the present disclosure provides a risk control method for a multi-tenant platform, such as Figure 2 As shown, this method is applied to Figure 1 The server in the example is used to illustrate the following steps:
[0050] Step S21, obtaining the historical behavior data and real-time behavior data of the current driver terminal. The current driver terminal is pre-bound to at least one tenant, and the real-time behavior data is the behavior data generated by the current driver terminal under the current tenant.
[0051] The historical behavior data may specifically include audio data and video data. The historical behavior data may specifically include completed orders and order details, wherein the order details may include the distance, time, speed, and audio data during driving of the driver.
[0052] Real-time behavior data refers to the current driver-side behavior data obtained in real time. Specifically, when the driver receives a ride request, the ride request carries tenant information, and real-time behavior data refers to the data generated under the tenant corresponding to the tenant information.
[0053] The multi-tenant platform may be an aggregated taxi-hailing platform. The tenants may be transportation companies registered on the aggregated taxi-hailing platform. For example, the tenants may include Feidi Taxi, Qimiao Travel, and Huojianchuxing.
[0054] Specifically, this application pre-creates multiple tenants and pre-configures each tenant. Each driver can select the tenant to which he needs to bind by registering on the multi-tenant platform, and a driver can bind one or more tenants. When the driver receives a ride request, the ride request will carry tenant information, where the tenant information may include the tenant's identifier, and the tenant's identifier may include the tenant's name or ID number, etc.
[0055] Exemplarily, the current driver receives a ride request that carries tenant information such as QiMiao Travel. After receiving the order, the driver starts driving and obtains real-time behavior data of the driver during driving.
[0056] Step S22, inputting the historical behavior data into the pre-trained scoring model to obtain the scores of the key factors.
[0057] The scoring model is a model for scoring key factors, which are scoring items that affect the driver's credit score and are used to calculate the driver's credit score. For example, the key factors may be items related to the driver's service or items such as the price parity of the passenger to the driver.
[0058] The present application pre-sets a plurality of key factors, performs model training based on the plurality of key factors, and trains a model for scoring the key factors.
[0059] Specifically, the risk control method of the multi-tenant platform of the present application may also include pre-training the scoring model. The training of the scoring model may specifically include:
[0060] The historical behavior data of each driver associated with each tenant is obtained, wherein the historical behavior data may include data related to the service quality or violation of each driver, and the obtained data is trained.
[0061] Step S23, obtaining the pre-configured weight of the key factor, where the weight is the weight corresponding to the key factor under the current tenant.
[0062] Specifically, each tenant can set their own corresponding weight through the multi-tenant platform, where the weight refers to the weight of the key factor under the tenant, and the key factor is the factor used to evaluate the driver's credit score.
[0063] Specifically, a tenant can submit a configuration request to the multi-tenant platform, where the configuration request includes the key factors that need to be configured and the weights corresponding to each key factor.
[0064] Please refer to Table 1 below, which is a configuration table for each tenant in an embodiment.
[0065] Table 1
[0066]
[0067]
[0068] In Table 1, the key factors that affect the driver's credit score include key factor A, key factor B, and key factor C. For tenant 1, the corresponding weight of key factor A is A1, the weight of key factor B is B1, and the weight of key factor C is C1.
[0069] For tenant 2, the corresponding weight of key factor A is configured as A2, the weight of key factor B is configured as B2, and the weight of key factor C is configured as C2.
[0070] For tenant 3, the corresponding weight of key factor A is configured as A3, the weight of key factor B is configured as B3, and the weight of key factor C is configured as C3.
[0071] It can be seen that different weights can be configured for key factors for different tenants, and the driver's credit score can be calculated based on the configured weights, making the credit score calculation more granular and accurate.
[0072] Step S24, determining the current driver's credit score based on the weights of the key factors and the scores of the key factors.
[0073] Specifically, the present application can calculate the product between the weight of the key factor and the score of the key factor to obtain the credit score.
[0074] Step S25, determine the risk control strategy based on the current driver's credit score and real-time behavior data, and perform corresponding operations on the current driver according to the risk control strategy.
[0075] The risk control strategy refers to the strategy of controlling or taking corresponding measures against the abnormal behavior of drivers, which may include detours, mismatch between driver and vehicle, fake trips, and malicious cancellations.
[0076] For example, in response to the driver's detour behavior, the corresponding execution operation may be sending corresponding prompt information to the driver, such as voice broadcast information, to prompt the driver to follow the correct route.
[0077] In one embodiment, the risk control method of the multi-tenant platform may further include:
[0078] Get the credit score of each driver bound to the current tenant;
[0079] Determine the average credit score of the current tenants based on the credit scores of the drivers;
[0080] The traffic of the current tenant is controlled according to the average credit score of the current tenant.
[0081] Specifically, the present application can calculate the average credit score of the corresponding driver terminal under each tenant based on the driver terminal bound to each tenant, and control the traffic of each tenant based on the average credit score corresponding to each tenant.
[0082] Among them, the flow of tenants with high average credit scores can be controlled to be large, and conversely, the flow of tenants with low average credit scores can be controlled to be small.
[0083] Specifically, when a ride-hailing request is received from a passenger, the requests may be sorted according to the average credit scores of the drivers corresponding to each tenant, and the ride-hailing requests may be allocated according to the sorting to control the traffic of each tenant.
[0084] The beneficial effects of this embodiment are:
[0085] By calculating the credit score of each driver associated with each tenant, and further calculating the average credit score of the drivers under each tenant, the traffic of each tenant is controlled based on the average credit score, so as to achieve reasonable traffic distribution on the multi-tenant platform.
[0086] In some embodiments, the flow of the current tenant is controlled according to the average credit score corresponding to the current tenant, including:
[0087] Receive a ride request from the passenger app;
[0088] Determine the car booking price for the current tenant according to the car booking request and the charging policy pre-configured by the current tenant;
[0089] When the ride-hailing price is within a preset range, and / or the average credit score is higher than a preset threshold, the ride-hailing request will be assigned to the driver bound to the current tenant to control the traffic of the current tenant.
[0090] Specifically, each tenant can pre-configure a corresponding billing strategy, which is used to calculate the booking price corresponding to each order. When receiving a booking request sent by a waiting passenger, the matching candidate driver terminal is obtained according to the booking request, the tenant to which each candidate driver terminal belongs is obtained, the billing strategy of the corresponding tenant is obtained, and the booking price corresponding to each tenant is calculated according to the billing strategy.
[0091] It is further determined whether the car-hailing price is within a preset range, where the preset price range refers to a range set in advance based on actual needs or actual application scenarios. When the car-hailing price is within the preset range, the car-hailing request is assigned to the driver bound to the current tenant to control the traffic of the current tenant.
[0092] Traffic flow control can also be performed based on the average credit score of each tenant. Specifically, when the average credit score is higher than a preset threshold, the ride request is assigned to the driver bound to the current tenant to control the traffic flow of the current tenant.
[0093] Traffic control can also be performed based on the ride-hailing price and the average credit score. Specifically, when the ride-hailing price is within a preset range and the average credit score is higher than a preset threshold, the ride-hailing request will be assigned to the driver bound to the current tenant to control the current tenant's traffic.
[0094] In some of the embodiments, the risk control method for a multi-tenant platform may further include:
[0095] Determine an average price for multiple tenants based on the origin and destination in the ride request and the billing policies of the multiple tenants;
[0096] Determine the preset price range based on the average price.
[0097] Wherein, determining the average price of multiple tenants according to the origin and destination in the ride-hailing request and the billing strategies of multiple tenants may include:
[0098] Get the pre-created billing strategies for each tenant, determine the driving distance according to the origin and destination, obtain the car-hailing price for each tenant for the car-hailing request based on the driving distance and the billing strategy, and calculate the average price for each tenant.
[0099] Determine a preset price range based on the average price, which can include:
[0100] The preset price range can be obtained by adding or subtracting a preset value from the average price.
[0101] The beneficial effects of this embodiment are:
[0102] This application can control the traffic of each tenant based on their average credit score and the car-hailing price of each tenant, so as to achieve reasonable traffic distribution on the multi-tenant platform.
[0103] In some embodiments, the risk control method for a multi-tenant platform further includes:
[0104] Get the traffic of each tenant;
[0105] The resource usage of each tenant is adjusted according to the traffic of each tenant, wherein the resource usage includes computing resource usage and storage resource usage.
[0106] The resource usage may include but is not limited to computing resource usage, storage resource usage, and bandwidth usage.
[0107] Specifically, the traffic of each tenant in a preset time period is obtained, and the usage of computing resources is dynamically adjusted according to the traffic of each tenant in the preset time period.
[0108] In some embodiments, the key factors include multiple factors, and the credit score of the current driver is determined according to the weights of the key factors and the scores of the key factors, including:
[0109] According to the weight of each key factor and the score of each key factor, the weighted average of each key factor is calculated to obtain the credit score.
[0110] In one embodiment, the formula for calculating the credit score may be:
[0111] Credit score = w1*key factor 1 + w2*key factor 2 + w3*key factor 3
[0112] Among them, w1, w2 and w3 represent the weights of the three key factors, key factor 1, key factor 2 and key factor 3 respectively. The weights of the key factors are set according to the scenario requirements and importance judgment, and are set by each tenant.
[0113] In this application, each tenant can configure their own corresponding weight, and the data of each tenant is isolated from each other, so that each tenant can set the corresponding weight according to its own driver needs, reflecting the differences between each tenant and achieving better risk control for each tenant.
[0114] In some embodiments, the key factors include any one or more of the following:
[0115] Service quality, passenger evaluation, order completion rate, violation record and safety record.
[0116] In one embodiment, the formula for calculating the credit score may also be:
[0117] Credit score = w4*service quality score + w5*passenger evaluation score + w6*order completion rate score - w7*violation record score - w8*safety record score
[0118] Among them, w4, w5, w6, w7 and w8 represent the weights of the five key factors: service quality, passenger evaluation, completion rate, violation record and safety record respectively.
[0119] In some embodiments, a risk control strategy is determined based on the current driver's credit score and real-time behavior data, including:
[0120] Use the stream processing framework to process real-time behavior data to detect whether the current driver has abnormal behavior;
[0121] When the current driver has abnormal behavior and the credit score is lower than the preset threshold, the corresponding risk control strategy is determined based on the abnormal behavior.
[0122] Among them, a stream processing framework is a data processing framework that is different from a batch processing framework, which operates on the entire data set. The stream processing framework is designed to process "unbounded" data, that is, every data item that enters the system at any time. This means that the full data set currently being processed only represents the amount of data currently entering the system, and only one piece of data (in true stream processing) or a small amount of data (in micro-batch processing) can be processed at the same time. Only a minimal amount of state is maintained between different records.
[0123] Specifically, the present application may pre-train a recognition model for identifying abnormal behavior. The training of the recognition model may include:
[0124] During the journey, sensors and cameras collect sound, image, and motion data. The vehicle's location information is collected through the vehicle's onboard GPS (Global Positioning System) and uploaded to the server to monitor the journey of passengers or drivers.
[0125] By processing data streams in real time, data is cleaned, for example, audio data is denoised, and image data in video files is further processed using convolutional neural networks (CNN) to identify facial expressions, body postures and other actions. Time series data is processed using recurrent neural networks (RNN) or long short-term memory networks (LSTM) to complete the processing of audio data. Then, a scoring model is trained based on the data set to identify abnormal behaviors, such as offensive language and dangerous actions.
[0126] In one embodiment, please refer to Figure 3 , Figure 3 FIG. 1 is a block diagram of a risk control method for a multi-tenant platform in an embodiment. Figure 3 In the multi-tenant platform, the risk control method is deployed in the anti-fraud billing system, which can obtain the data uploaded by the client. The client includes the passenger side and the driver side bound to each tenant. The anti-fraud billing system can specifically perform data collection, data stream processing, algorithm model and credit score calculation, and finally output the risk control strategy, which is equivalent to Figure 3 Fraud prevention strategies in .
[0127] In a second aspect, an embodiment of the present disclosure provides a risk control device for a multi-tenant platform, such as Figure 4 As shown, the risk control device of the multi-tenant platform includes: a first acquisition module 41, a scoring module 42, a second acquisition module 43, a processing module 44 and a control module 45, wherein:
[0128] The first acquisition module 41 is used to acquire historical behavior data and real-time behavior data of the current driver terminal, where the current driver terminal is pre-bound to at least one tenant, and the real-time behavior data is the behavior data generated by the current driver terminal under the current tenant;
[0129] A scoring module 42 is used to input the historical behavior data into a pre-trained scoring model to obtain the scores of key factors;
[0130] The second acquisition module 43 is used to acquire the weight of the pre-configured key factor, where the weight is the weight corresponding to the key factor under the current tenant;
[0131] A processing module 44, for determining the credit score of the current driver according to the weights of the key factors and the scores of the key factors;
[0132] The control module 45 is used to determine the risk control strategy based on the credit score and real-time behavior data of the current driver, and perform corresponding operations on the current driver according to the risk control strategy.
[0133] In some of the embodiments, the first acquisition module 41 can also obtain the credit scores of each driver bound to the current tenant, determine the average credit score corresponding to the current tenant based on the credit scores of each driver, and control the traffic of the current tenant based on the average credit score corresponding to the current tenant.
[0134] In some of the embodiments, the first acquisition module 41 may also receive a ride request sent by a passenger, determine the ride price for the current tenant based on the ride request and a charging policy pre-configured by the current tenant, and when the ride price is within a preset range and / or the average credit score is higher than a preset threshold, assign the ride request to the driver bound to the current tenant to control the traffic of the current tenant.
[0135] In some of the embodiments, the first acquisition module 41 may also acquire the traffic of each tenant, and adjust the resource usage of each tenant according to the traffic of each tenant, wherein the resource usage includes computing resource usage and storage resource usage.
[0136] In some embodiments, the above-mentioned key factors include multiple ones, and the processing module 44 can calculate the weighted average of each key factor according to the weight of each key factor and the score of each key factor to obtain the credit score.
[0137] In some embodiments, the above-mentioned key factors include any one or more of the following:
[0138] Service quality, passenger evaluation, order completion rate, violation record and safety record.
[0139] In some of the embodiments, the control module 45 can process the real-time behavior data using a stream processing framework to detect whether the current driver has abnormal behavior. When the current driver has abnormal behavior and the credit score is lower than a preset threshold, a corresponding risk control strategy is determined based on the abnormal behavior.
[0140] In a third aspect, an embodiment of the present disclosure provides a computer device, which may be a server, and its internal structure diagram may be as follows: Figure 5As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used 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 the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a risk control method for a multi-tenant platform is implemented.
[0141] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:
[0142] Obtain the historical behavior data and real-time behavior data of the current driver terminal. The current driver terminal is pre-bound to at least one tenant. The real-time behavior data is the behavior data generated by the current driver terminal under the current tenant.
[0143] Input historical behavior data into the pre-trained scoring model to obtain the scores of key factors;
[0144] Get the weight of the pre-configured key factor, which is the weight of the key factor corresponding to the current tenant;
[0145] Determine the current driver's credit score based on the weights of the key factors and the scores of the key factors;
[0146] Determine the risk control strategy based on the current driver's credit score and real-time behavior data, and perform corresponding operations on the current driver based on the risk control strategy.
[0147] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:
[0148] Get the credit score of each driver bound to the current tenant;
[0149] Determine the average credit score of the current tenants based on the credit scores of the drivers;
[0150] The traffic of the current tenant is controlled according to the average credit score of the current tenant.
[0151] In one embodiment, when the processor executes the computer program to implement the above-mentioned step of controlling the traffic of the current tenant according to the average credit score corresponding to the current tenant, the following steps are specifically implemented:
[0152] Receive a ride request from the passenger app;
[0153] Determine the car booking price for the current tenant according to the car booking request and the charging policy pre-configured by the current tenant;
[0154] When the ride-hailing price is within a preset range, and / or the average credit score is higher than a preset threshold, the ride-hailing request will be assigned to the driver bound to the current tenant to control the traffic of the current tenant.
[0155] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:
[0156] Get the traffic of each tenant;
[0157] The resource usage of each tenant is adjusted according to the traffic of each tenant, wherein the resource usage includes computing resource usage and storage resource usage.
[0158] In one embodiment, the key factors include multiple factors, and when the processor executes the computer program to implement the above-mentioned step of determining the credit score of the current driver according to the weights of the key factors and the scores of the key factors, the following steps are specifically implemented:
[0159] According to the weight of each key factor and the score of each key factor, the weighted average of each key factor is calculated to obtain the credit score.
[0160] In one embodiment, the key factors include any one or more of the following:
[0161] Service quality, passenger evaluation, order completion rate, violation record and safety record.
[0162] In one embodiment, when the processor executes the computer program to implement the above-mentioned step of determining the risk control strategy based on the current driver's credit score and real-time behavior data, the following steps are specifically implemented:
[0163] Use the stream processing framework to process real-time behavior data to detect whether the current driver has abnormal behavior;
[0164] When the current driver has abnormal behavior and the credit score is lower than the preset threshold, the corresponding risk control strategy is determined based on the abnormal behavior.
[0165] In a fourth aspect, an embodiment of the present disclosure provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the following steps are implemented:
[0166] Obtain the historical behavior data and real-time behavior data of the current driver terminal. The current driver terminal is pre-bound to at least one tenant. The real-time behavior data is the behavior data generated by the current driver terminal under the current tenant.
[0167] Input historical behavior data into the pre-trained scoring model to obtain the scores of key factors;
[0168] Get the weight of the pre-configured key factor, which is the weight of the key factor corresponding to the current tenant;
[0169] Determine the current driver's credit score based on the weights of the key factors and the scores of the key factors;
[0170] Determine the risk control strategy based on the current driver's credit score and real-time behavior data, and perform corresponding operations on the current driver based on the risk control strategy.
[0171] In one embodiment, when the computer program is executed by a processor, the following steps are specifically implemented:
[0172] Get the credit score of each driver bound to the current tenant;
[0173] Determine the average credit score of the current tenants based on the credit scores of the drivers;
[0174] The traffic of the current tenant is controlled according to the average credit score of the current tenant.
[0175] In one embodiment, when the computer program is executed by a processor to implement the above-mentioned step of controlling the traffic of the current tenant according to the average credit score corresponding to the current tenant, the following steps are specifically implemented:
[0176] Receive a ride request from the passenger app;
[0177] Determine the car booking price for the current tenant according to the car booking request and the charging policy pre-configured by the current tenant;
[0178] When the ride-hailing price is within a preset range, and / or the average credit score is higher than a preset threshold, the ride-hailing request will be assigned to the driver bound to the current tenant to control the traffic of the current tenant.
[0179] In one embodiment, when the computer program is executed by a processor, the following steps are specifically implemented:
[0180] Get the traffic of each tenant;
[0181] The resource usage of each tenant is adjusted according to the traffic of each tenant, wherein the resource usage includes computing resource usage and storage resource usage.
[0182] In one embodiment, the key factors include multiple factors, and when the computer program is executed by the processor to implement the above-mentioned step of determining the credit score of the current driver according to the weights of the key factors and the scores of the key factors, the following steps are specifically implemented:
[0183] According to the weight of each key factor and the score of each key factor, the weighted average of each key factor is calculated to obtain the credit score.
[0184] In one embodiment, the key factors include any one or more of the following:
[0185] Service quality, passenger evaluation, order completion rate, violation record and safety record.
[0186] In one embodiment, when the computer program is executed by a processor to implement the above-mentioned step of determining the risk control strategy based on the current driver's credit score and real-time behavior data, the following steps are specifically implemented:
[0187] Use the stream processing framework to process real-time behavior data to detect whether the current driver has abnormal behavior;
[0188] When the current driver has abnormal behavior and the credit score is lower than the preset threshold, the corresponding risk control strategy is determined based on the abnormal behavior.
[0189] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing related hardware through a computer program, and the above-mentioned computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.
[0190] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0191] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.
Claims
1. A risk control method for a multi-tenant platform, characterized in that: The method comprises: Obtain historical behavior data and real-time behavior data of the current driver terminal, where the current driver terminal is pre-bound to at least one tenant, and the real-time behavior data is the behavior data generated by the current driver terminal under the current tenant; Inputting the historical behavior data into a pre-trained scoring model to obtain a score of a key factor; Obtaining a pre-configured weight of the key factor, where the weight is a weight corresponding to the key factor under the current tenant; Determining the credit score of the current driver according to the weight of the key factor and the score of the key factor; A risk control strategy is determined based on the credit score of the current driver and the real-time behavior data, and corresponding operations are performed on the current driver terminal based on the risk control strategy.
2. The method according to claim 1, characterized in that The method further comprises: Obtain the credit score of each driver bound to the current tenant; Determining an average credit score corresponding to the current tenant based on the credit scores of each of the drivers; The flow of the current tenant is controlled according to the average credit score corresponding to the current tenant.
3. The method according to claim 2, characterized in that The controlling the flow of the current tenant according to the average credit score corresponding to the current tenant includes: Receive a ride request from the passenger app; Determining a car booking price for the current tenant according to the car booking request and a charging policy pre-configured by the current tenant; When the ride-hailing price is within a preset range, and / or the average credit score is higher than a preset threshold, the ride-hailing request is assigned to the driver bound to the current tenant to control the traffic of the current tenant.
4. The method according to claim 1 or 3, characterized in that: The method further comprises: Obtaining the traffic of each of the tenants; The resource usage of each tenant is adjusted according to the traffic of each tenant, wherein the resource usage includes computing resource usage and storage resource usage.
5. The method according to claim 1, characterized in that The key factors include a plurality of key factors, and determining the credit score of the current driver according to the weights of the key factors and the scores of the key factors includes: According to the weights of the key factors and the scores of the key factors, a weighted average of the key factors is calculated to obtain the credit score.
6. The method according to claim 5, characterized in that The key factors include any one or more of the following: Service quality, passenger evaluation, order completion rate, violation record and safety record.
7. The method according to claim 1, characterized in that The step of determining a risk control strategy based on the current driver's credit score and the real-time behavior data includes: Processing the real-time behavior data using a stream processing framework to detect whether the current driver has abnormal behavior; When the current driver has abnormal behavior and the credit score is lower than a preset threshold, a corresponding risk control strategy is determined according to the abnormal behavior.
8. A risk control device for a multi-tenant platform, characterized in that: The device comprises: A first acquisition module is used to acquire historical behavior data and real-time behavior data of a current driver terminal, where the current driver terminal is pre-bound to at least one tenant, and the real-time behavior data is behavior data generated by the current driver terminal under the current tenant; A scoring module, used to input the historical behavior data into a pre-trained scoring model to obtain a score of a key factor; A second acquisition module is used to acquire a pre-configured weight of the key factor, where the weight is the weight corresponding to the key factor under the current tenant; A processing module, configured to determine the credit score of the current driver according to the weight of the key factor and the score of the key factor; The control module is used to determine a risk control strategy based on the credit score of the current driver and the real-time behavior data, and perform corresponding operations on the current driver according to the risk control strategy.
9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.
Citation Information
Cited By
Credit payment processing method and system for transaction platform
CN120612085A
