CN115018400B

CN115018400B - Wind control data processing method and device, electronic equipment and medium

Info

Publication number: CN115018400B
Application number: CN202210946995.5A
Authority: CN
Inventors: 胡荣良
Original assignee: Jiangsu Yincheng Network Technology Co Ltd
Current assignee: Jiangsu Yincheng Network Technology Co Ltd
Priority date: 2022-08-09
Filing date: 2022-08-09
Publication date: 2022-10-25
Anticipated expiration: 2042-08-09
Also published as: CN115018400A

Abstract

The application provides a wind control data processing method, a wind control data processing device, electronic equipment and a medium, wherein the method comprises the following steps: acquiring metadata required by a metadata calculation layer of a target wind control strategy from target service data of online bill service, and sending the metadata to the metadata calculation layer; the metadata computing layer processes the metadata according to a plurality of metadata computing nodes which are configured in advance to obtain a metadata computing result and sends the metadata computing result to the sub-index computing layer; the sub-index calculation layer processes the metadata calculation result through a plurality of pre-configured sub-index calculation nodes to obtain a sub-index judgment result and sends the sub-index judgment result to the target index calculation node; and the target index calculation node judges whether the target service data hits a target wind control strategy according to the received sub-index judgment result, thereby acquiring and processing the target service data of the online bill service in real time and quickly judging whether the service data hits the wind control strategy.

Description

Wind control data processing method and device, electronic equipment and medium

Technical Field

The present application relates to the field of computer technologies, and in particular, to a method and an apparatus for processing wind control data, an electronic device, and a medium.

Background

The appearance of online bill business greatly improves the safety, transparency and transaction efficiency of the bill market. The wind control system for the online bill service plays a crucial role in guaranteeing the safety of the online bill service. The traditional wind control model belongs to a pre-calculation model, data and indexes related to the wind control model are calculated in advance through preset calculation steps through an event triggering or timing mechanism, and finally a wind control result is output through a characteristic vector or neural network model output mode. This approach has the following disadvantages: real-time calculation cannot be carried out, namely basic data relied on by the wind control model needs to be calculated in advance, and the real-time wind control requirement of a service system cannot be met; the model is difficult to adjust, and the adjustment requirement of the real-time changing wind control strategy cannot be met based on the staticized data indexes and the calculation link; the construction difficulty is high, and the rapid development and iteration scenes of the system are not facilitated.

Disclosure of Invention

In view of the above, an object of the present application is to provide a method, an apparatus, an electronic device, and a medium for processing wind control data, which can obtain and process target service data of an online ticket service in real time, and quickly determine whether the service data hits a wind control policy, and the model is easy to adjust and has low construction difficulty.

The wind control data processing method provided by the embodiment of the application comprises the following steps:

acquiring metadata required by a metadata calculation layer of a target wind control strategy from target service data of online bill service, and sending the acquired metadata to the metadata calculation layer;

the metadata computing layer processes the received metadata through a plurality of preconfigured metadata computing nodes to obtain a metadata computing result, and sends the metadata computing result to the sub-index computing layer; wherein, the metadata processing rules in different metadata computing nodes are different;

the sub-index calculation layer processes the received metadata calculation result to obtain a sub-index judgment result through a plurality of pre-configured sub-index calculation nodes, and sends the sub-index judgment result to a target index calculation node; the sub-index processing rules in different index calculation nodes are different;

and the target index calculation node judges whether the target service data hits a target wind control strategy or not according to the received sub-index judgment result.

In some embodiments, in the wind control data processing method, obtaining metadata required by a metadata computation layer of a target wind control policy from target service data of an online ticket service includes:

a metadata computing node of a first level in the metadata computing layer sends a target metadata obtaining request carrying a node identifier and request information to a service data source, so that the service data source determines target metadata from target service data of an online bill service according to the request information and sends the determined target metadata to the metadata computing node according to the node identifier;

the metadata computing node receives the target metadata.

In some embodiments, in the method for processing wind control data, the processing, by the metadata calculation layer, the received metadata through a plurality of metadata calculation nodes configured in advance to obtain a metadata calculation result, and sending the metadata calculation result to the sub-index calculation layer includes:

a metadata computing node in the metadata computing layer acquires target metadata required by the metadata computing node, and processes the target metadata according to a metadata processing rule to obtain a metadata computing result;

according to a pre-configured link connection relation, when the metadata computing node is judged to be linked with the sub-index computing node, the metadata computing result is sent to the linked sub-index computing node;

and when the metadata computing node is judged to be linked with the metadata computing node of the next level, sending the metadata computing result as target metadata to the linked metadata computing node.

In some embodiments, in the method for processing wind control data, the sub-indicator calculation layer processes the metadata calculation result to obtain a sub-indicator determination result according to the received metadata calculation result and a plurality of pre-configured sub-indicator calculation nodes, and sends the sub-indicator determination result to a target indicator calculation node, including:

the sub-index calculation nodes in the sub-index calculation layer acquire metadata calculation results or intermediate judgment results required by the sub-index calculation nodes, and judge whether the metadata calculation results or the intermediate judgment results meet the sub-index judgment conditions of the sub-index calculation nodes or not to obtain sub-index judgment results;

according to a preset link connection relation, when the sub-index computing node is judged to be linked with the target index computing node, the sub-index judgment result is sent to the target index computing node;

and when the metadata computing node is judged to be linked with the sub-index computing node of the next level, sending the sub-index judgment result as an intermediate judgment result to the linked sub-index computing node, so that the sub-index computing node of the next level obtains a new sub-index judgment result according to the received intermediate judgment result and the sub-index processing rule.

In some embodiments, the determining whether the target service data hits the target wind control policy in the wind control data processing method includes:

and if the target index calculation node judges that the received sub-index judgment results all accord with preset conditions, judging that the target business data hits a target wind control strategy.

In some embodiments, the wind-controlled data processing method further includes:

at least one computing node receives a strategy adjusting instruction and updates the configuration of the computing node according to the strategy adjusting instruction, so that a target wind control strategy is updated; the computing nodes are sub-index computing nodes or metadata computing nodes;

the strategy adjusting instruction comprises a stopping instruction and an operating instruction; the stopping instruction is used for controlling the computing node to stop sending the computing result to the computing node of the next level; the operation instruction is used for controlling the computing node to send the computing result to the computing node of the next level.

In some embodiments, in the wind control data processing method, the wind control policy model formed by the metadata computation layer, the sub-index computation layer and the target index computation node is constructed by the following method:

selecting a prestored target index computing node from a unit database according to the target wind control strategy, and constructing a target index computing node;

selecting prestored sub-index computing nodes from a unit database according to sub-indexes depended by target indexes in the target index computing nodes, and establishing a link connection relationship between the sub-index computing nodes and the target index computing nodes and/or a link connection relationship between the sub-index computing nodes so as to construct a sub-index computing layer configured with a plurality of sub-index computing nodes;

and selecting prestored metadata computing nodes from the unit database according to metadata depended by at least part of the sub-index computing nodes, and establishing a link connection relation between the metadata computing nodes and part of the sub-index computing nodes and/or a link connection relation between the metadata computing nodes so as to construct a metadata computing layer configured with a plurality of metadata computing nodes.

In some embodiments, there is also provided a wind-controlled data processing apparatus, the apparatus comprising:

the acquisition module is used for acquiring metadata required by a metadata calculation layer of a target wind control strategy from target service data of an online bill service and sending the acquired metadata to the metadata calculation layer;

the first processing module is used for enabling the metadata computing layer to process the received metadata through a plurality of metadata computing nodes which are configured in advance to obtain a metadata computing result and sending the metadata computing result to the sub-index computing layer; wherein, the metadata processing rules in different metadata computing nodes are different;

the second processing module is used for enabling the sub-index calculation layer to process the received metadata calculation result through a plurality of pre-configured sub-index calculation nodes to obtain a sub-index judgment result and sending the sub-index judgment result to a target index calculation node; the sub-index processing rules in different index calculation nodes are different;

and the judging module is used for enabling the target index computing node to judge whether the target business data hits the target wind control strategy or not according to the received sub-index judgment result.

In some embodiments, there is also provided an electronic device comprising: the wind control data processing system comprises a processor, a memory and a bus, wherein the memory stores machine readable instructions executable by the processor, the processor and the memory are communicated through the bus when the electronic device runs, and the machine readable instructions are executed by the processor to execute the steps of the wind control data processing method.

In some embodiments, a computer-readable storage medium is also provided, on which a computer program is stored, which computer program, when being executed by a processor, performs the steps of the method for processing wind-controlled data.

The embodiment of the application provides a wind control data processing method, a wind control data processing device, electronic equipment and a medium, which are used for acquiring metadata required by a metadata computing layer of a target wind control strategy in real time, enabling each metadata computing node to process the received metadata according to the metadata computing layer through a plurality of metadata computing nodes configured in advance, and sending the metadata computing result to a sub-index computing layer; the method comprises the steps that metadata processing rules in different metadata computing nodes process corresponding parts of service data differently to obtain metadata computing results quickly, then each sub-index computing node processes corresponding parts of the metadata computing results through a plurality of pre-configured sub-index computing nodes to obtain sub-index judging results, judging results of all sub-index computing nodes are fused quickly to obtain a total sub-index judging result, the fused sub-index judging results are sent to a target index computing node, and whether risks are hit or not is judged quickly; each metadata computing node, each sub-index computing node and each target index computing node respectively perform their own functions, the computing efficiency is high, each computing node is independently managed and computed, the coupling degree between the computing nodes is low, and the link relation between the nodes can be flexibly adjusted or nodes are added and deleted when a wind control strategy is adjusted, so that a wind control model is quickly adjusted; and the computing units corresponding to the nodes are pre-stored in the unit database, one computing unit can be reused by multiple wind control strategies, and the processing rule of the single computing unit is relatively simple, so that the construction difficulty is low, and an iterative scene is easy to develop quickly.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained from the drawings without inventive effort.

Fig. 1 shows a method flowchart of a method for processing wind control data according to an embodiment of the present application;

FIG. 2 is a flowchart illustrating a method for constructing a wind control strategy model according to an embodiment of the present application;

FIG. 3 illustrates a directed acyclic graph model according to an embodiment of the present application;

FIG. 4 shows a directed acyclic graph model of a "first small amount" wind control target according to an embodiment of the present application;

FIG. 5 is a flowchart illustrating a method for processing the metadata to obtain a metadata computation result according to an embodiment of the present application;

FIG. 6 illustrates that the metadata calculation result is processed to obtain a sub-index determination result according to the embodiment of the present application;

fig. 7 shows a schematic structural diagram of a wind-controlled data processing apparatus according to an embodiment of the present application;

fig. 8 shows a schematic structural diagram of an electronic device according to an embodiment of the present application.

Detailed Description

In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it should be understood that the drawings in the present application are for illustrative and descriptive purposes only and are not used to limit the scope of protection of the present application. Further, it should be understood that the schematic drawings are not drawn to scale. The flowcharts used in this application illustrate operations implemented according to some embodiments of the present application. It should be understood that the operations of the flow diagrams may be performed out of order, and steps without logical context may be performed in reverse order or simultaneously. In addition, one skilled in the art, under the guidance of the present disclosure, may add one or more other operations to the flowchart, or may remove one or more operations from the flowchart.

In addition, the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. The components of the embodiments of the present application, as generally described and illustrated in the figures herein, could be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present application, as presented in the figures, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present application without making any creative effort, shall fall within the protection scope of the present application.

It should be noted that the term "comprising" will be used in the embodiments of the present application to indicate the presence of the features stated hereinafter, but does not exclude the addition of further features.

The online bill service refers to the daily business activities of the credit institution for the establishment, transfer and reimbursement of bills according to a certain mode and requirements, and mainly comprises bill acceptance, cash sticking and bill mortgage deposit services. The appearance of online ticket business greatly improves the security, transparency and transaction efficiency of the ticket market.

However, the ticket business has various risks such as financing risk and operation risk, and in order to prevent the occurrence of the event of the risk department, it is necessary to perform risk control. The risk control means that a risk manager takes various measures and methods to eliminate or reduce various possibilities of occurrence of a risk event or reduce loss caused by occurrence of the risk event, referred to as wind control for short, wherein a common means is to establish a wind control model, predict occurrence of the risk event according to the wind control model, and evaluate risk.

In the wind control system of the online bill service, a traditional wind control model belongs to a pre-calculation model, data and indexes related to the wind control model are calculated in advance through preset calculation steps through an event triggering or timing mechanism, and finally a wind control result is output through a characteristic vector or neural network model. This approach has the following disadvantages: real-time calculation cannot be carried out, namely basic data relied on by the wind control model needs to be calculated in advance, and the real-time wind control requirement of a service system cannot be met; the model is difficult to adjust, and the adjustment requirement of the real-time changing wind control strategy cannot be met based on the staticized data index and the calculation link; the construction difficulty is large, and the rapid development and iteration scenes of the system are not facilitated.

Based on this, an embodiment of the present application provides a wind control data processing method, which obtains metadata required by a metadata calculation layer of a target wind control policy in real time, and through a plurality of metadata calculation nodes configured in advance, enables each metadata calculation node to process the received metadata according to the metadata calculation layer through the plurality of metadata calculation nodes configured in advance to obtain a metadata calculation result, and sends the metadata calculation result to a sub-index calculation layer; the method comprises the steps that metadata processing rules in different metadata computing nodes process corresponding parts of service data differently to obtain metadata computing results quickly, then each sub-index computing node processes corresponding parts of the metadata computing results through a plurality of pre-configured sub-index computing nodes to obtain sub-index judging results, judging results of all sub-index computing nodes are fused quickly to obtain a total sub-index judging result, the fused sub-index judging results are sent to a target index computing node, and whether risks are hit or not is judged quickly; each metadata computing node, each sub-index computing node and each target index computing node respectively perform their own functions, the computing efficiency is high, each computing node is independently managed and computed, the coupling degree between the computing nodes is low, and the link relation between the nodes can be flexibly adjusted or nodes are added and deleted when a wind control strategy is adjusted, so that a wind control model is quickly adjusted; and the computing units corresponding to the nodes are pre-stored in the unit database, one computing unit can be reused by multiple wind control strategies, and the processing rule of the single computing unit is relatively simple, so that the construction difficulty is low, and an iterative scene is easy to develop quickly.

Based on this, an embodiment of the present application provides a wind control data processing method, which is applied to a wind control policy model, where the wind control policy model includes a metadata computation layer, a sub-index computation layer, and a target index computation node, the metadata computation layer includes a plurality of preconfigured metadata computation nodes, and the sub-index computation layer includes a plurality of preconfigured sub-index computation nodes; different wind control strategies correspond to different wind control strategy models; the different wind control strategies can have sub-index computing nodes corresponding to the same sub-index computing unit and/or metadata computing nodes corresponding to the same metadata computing unit; and a metadata calculation unit corresponding to the metadata calculation node, a sub index calculation unit corresponding to the sub index calculation node and a target index calculation node corresponding to the target index calculation node are designed in advance and stored in a unit database.

Referring to fig. 1, fig. 1 is a flowchart illustrating a method of processing wind control data according to an embodiment of the present application; specifically, the method comprises the following steps S101-S104;

s101, acquiring metadata required by a metadata calculation layer of a target wind control strategy from target service data of online bill service, and sending the acquired metadata to the metadata calculation layer;

s102, the metadata computing layer processes the received metadata through a plurality of metadata computing nodes configured in advance to obtain a metadata computing result, and sends the metadata computing result to a sub-index computing layer; wherein, the metadata processing rules in different metadata computing nodes are different;

s103, the sub-index calculation layer processes the received metadata calculation result through a plurality of pre-configured sub-index calculation nodes to obtain a sub-index judgment result, and sends the sub-index judgment result to a target index calculation node; the sub-index processing rules in different index calculation nodes are different;

and S104, the target index calculation node judges whether the target service data hits a target wind control strategy according to the received sub-index judgment result.

In the implementation of the application, the method is applied to a wind control strategy model; referring to fig. 2, fig. 2 is a flowchart illustrating a method for constructing a wind control policy model according to an embodiment of the present application, and specifically, the wind control policy model formed by the metadata calculation layer, the sub-index calculation layer, and the target index calculation node is constructed by the following method:

s201, selecting a prestored target index calculation node from a unit database according to the target wind control strategy, and constructing a target index calculation node;

s202, according to sub-indexes depended on by target indexes in the target index computing nodes, pre-stored sub-index computing nodes are selected from a unit database, and a link connection relation between the sub-index computing nodes and the target index computing nodes and/or a link connection relation between the sub-index computing nodes are established to build a sub-index computing layer configured with a plurality of sub-index computing nodes;

s203, selecting prestored metadata computing nodes from a unit database according to metadata depended by at least part of the sub-index computing nodes, and establishing a link connection relationship between the metadata computing nodes and part of the sub-index computing nodes and/or a link connection relationship between the metadata computing nodes to construct a metadata computing layer configured with a plurality of metadata computing nodes.

Target index calculation nodes stored in the unit database in advance are configured with preset target index processing rules in advance, and the preset target index processing rules comprise preset conditions corresponding to sub-index judgment results;

the method comprises the steps that sub-index processing rules are pre-configured on pre-stored sub-index computing nodes in a unit database;

target index calculation nodes stored in the unit database in advance are configured with metadata processing rules in advance.

Different wind control strategies correspond to different wind control strategy models; the wind control system of the whole online bill service comprises a plurality of different wind control strategy models.

That is to say, for one wind control strategy, the scheme adopts a real-time reverse calculation mode to reversely split a target index in the wind control strategy into multi-level sub-targets according to the structure of a directed acyclic graph, and constructs a computing node corresponding to the sub-targets, wherein each computing node is independently calculated and managed, and reversely summarizes a final wind control result after real-time calculation through a logical relationship; meanwhile, the processing rule of each computing node is configured in advance, and the key point for constructing the wind control strategy model is to determine the needed target metadata computing node, the target index computing node, the sub-index computing node and the link relation between the nodes according to the directed acyclic graph.

Referring to fig. 3, fig. 3 shows a directed acyclic graph model according to an embodiment of the present application. Each circle node in the graph is a computing node. Each metadata calculation node corresponds to metadata, each sub-index calculation node corresponds to a sub-index, and the target index calculation node corresponds to a target index.

The calculation nodes are divided into two types, index calculation nodes and metadata calculation units. The index calculation nodes include a target index calculation node and a sub-index calculation node. And setting the lowest level of the target index, wherein the rightmost metadata is the uppermost level. Firstly, decomposing a target index into a plurality of sub indexes, wherein each sub index is obtained based on the sub index or metadata of a level on the right side of the sub index; the metadata may be derived based on the metadata of the hierarchy on its right side or directly from the service data source where the service data is stored.

That is, firstly, according to the target index of the target wind control strategy, determining a target index calculation node, and constructing the target index calculation node; determining a sub-index calculation unit of a level on the right side of the target index calculation node according to the sub-index on which the target index depends, and constructing the sub-index calculation node of the level, and if the sub-index of the level is determined depending on other sub-indexes, determining the sub-index calculation unit corresponding to other sub-indexes, and constructing the sub-index calculation node of the level on which the level depends; if the sub-index of the hierarchy is determined by depending on the metadata, determining a metadata computing unit corresponding to the metadata, and constructing a sub-index computing node of the previous hierarchy of the hierarchy; if the metadata is obtained by depending on other metadata, determining other metadata calculation units of the previous level corresponding to the metadata level, and constructing other metadata calculation nodes of the previous level corresponding to the metadata level, so as to reversely calculate the directed acyclic graph model of the wind control target index.

Here, both the index calculation node and the metadata calculation node may be multiplexed by the next hierarchy node.

Specifically, the index calculation node produces a yes or no result for a certain rule; the calculation of the metrics depends on the next level metrics or metadata. The metadata computation unit yields metadata that is used to compute metrics, or higher level abstractions. The index node with the out degree of 0 is called a target index, and the target index is the basis for whether the wind control strategy is hit or not.

That is to say, in the embodiment of the present application, each metadata computing node performs individual processing on metadata received by the node, and each sub-indicator computing node performs individual judgment on the received metadata or sub-indicator; the coupling between the nodes is low, and the quick construction and modification of the wind control model are facilitated.

When the target wind control strategy is updated, the wind control model is modified by one of the following methods:

adding, deleting or replacing at least one computing node of the wind control model, and determining a link relation between the computing node and an original computing node in the model;

and modifying the link relation between the original computing nodes in the wind control model.

Here, the computation node is a child index computation node or a metadata computation node.

Specifically, when a computing node is added, according to the sub-index or metadata in the computing node, the upper-level computing node on which the computing node depends and the lower-level computing node on which the computing node depends are determined, and the link connection relationship between the added computing node and the upper-level computing node and the link connection relationship between the added computing node and the lower-level computing node are established.

When a computing node is replaced, according to the sub-indexes or metadata in the replaced computing node, determining the upper-level computing node depended on by the replaced computing node and the lower-level computing node depended on by the replaced computing node, and establishing the link connection relation among the replaced computing node, the upper-level computing node and the lower-level computing node.

The following takes a wind control strategy of "receive order quota" as an example, and describes how to construct the directed acyclic graph wind control model according to the embodiment of the present application.

According to historical experience, people find that normal transactions can be tried by small amount when money laundering molecules perform initial transactions, and large-amount transaction money laundering activities are performed after the transactions are completed smoothly. According to the actual situation, the business department makes the following rules to identify the first order small amount.

Rule 1: the two sides of the order have three histories to complete the transaction;

rule 2: the first completed transaction amount of both parties of the order is less than 10000;

rule 3: the second and third completed transaction amounts of both sides of the order are more than or equal to 10000;

rule 4: both sides of the order have not completed the transaction with the history of other third parties;

rule 5: the buyer roles may be interchanged, such as the first pen being (a, b) and the second pen being (b, a);

rule 6: the time of the commercial registration of both orders is less than or equal to 3 months from the initial transaction initiation time (the order receiving of the buyer).

When the conditions are met simultaneously, money laundering activity is considered, and wind control is triggered in the fourth transaction.

Referring to fig. 4, fig. 4 shows a directed acyclic graph model of the "first small amount" wind control target according to the embodiment of the present application, which is specifically constructed according to the above rules 1 to 6.

The directed acyclic graph model of the wind control target of the 'first single small amount' comprises 6 metadata computing nodes which are respectively a metadata computing node 1, a metadata computing node 2, a metadata computing node 3, a metadata computing node 4, a metadata computing node 5 and a metadata computing node 6; each metadata computing node respectively acquires metadata shown in fig. 4; the metadata computing node 1 acquires the business registration time of both parties, the metadata computing node 2 acquires the first transaction time of both parties, the metadata computing node 3 acquires the historical transaction number of both parties, the metadata computing node 4 acquires four transaction details of both parties before the history, the metadata computing node 5 acquires the historical transaction number of buyers and third parties, and the metadata computing node 6 acquires the historical transaction number of sellers and third parties; the system comprises 6 sub-index computing nodes, namely a sub-index computing node 1, a sub-index computing node 2, a sub-index computing node 3, a sub-index computing node 4, a sub-index computing node 5 and a sub-index computing node 6; each sub-index computing node corresponds to a sub-index and sub-index metadata computing layer, and through a plurality of preconfigured metadata computing nodes, the received metadata are processed to obtain a metadata computing result, and the metadata computing result is sent to the sub-index computing layer; wherein, the metadata processing rules in different metadata computing nodes are different; the sub-index computing node 1 corresponds to a sub-index that a buyer does not transact with a third party, and the specific metadata computing layer processes the received metadata through a plurality of preconfigured metadata computing nodes to obtain a metadata computing result and sends the metadata computing result to the sub-index computing layer; the metadata processing rules in different metadata computing nodes are different, and whether the buyer transacts with a third party is judged according to the data of the metadata computing node 5; the sub-index computing node 2 corresponds to a sub-index that a seller has not transacted with a third party, and the specific metadata computing layer processes the received metadata through a plurality of metadata computing nodes configured in advance to obtain a metadata computing result and sends the metadata computing result to the sub-index computing layer; the metadata processing rules in different metadata computing nodes are different, namely whether the seller transacts with a third party is judged according to the data of the metadata computing node 6; the sub-index corresponding to the sub-index calculation node 3 is 'no transaction with a third party', and the specific metadata calculation layer processes the received metadata through a plurality of metadata calculation nodes configured in advance to obtain a metadata calculation result and sends the metadata calculation result to the sub-index calculation layer; the metadata processing rules in different metadata computing nodes are different, and whether the metadata are transacted with a third party is judged according to the judgment results of the sub-index computing node 1 and the sub-index computing node 2; the sub-index calculation node 4 corresponds to a sub-index of 'first small amount and then two large amounts', and the specific metadata calculation layer processes the received metadata through a plurality of metadata calculation nodes configured in advance to obtain a metadata calculation result and sends the metadata calculation result to the sub-index calculation layer; wherein, the different metadata processing rules in different metadata computing nodes are to judge whether the first small amount is followed by the second large amount; the sub-index computing nodes 5 correspond to sub-indexes of 'three historical transactions exist', and the specific metadata computing layer processes the received metadata through a plurality of preconfigured metadata computing nodes to obtain metadata computing results and sends the metadata computing results to the sub-index computing layer; the metadata processing rules in different metadata computing nodes are different, and whether three historical transactions exist is judged; the sub-index calculation node 6 corresponds to a sub-index of 'business registration transaction within three months', and the specific metadata calculation layer processes the received metadata through a plurality of preconfigured metadata calculation nodes to obtain a metadata calculation result and sends the metadata calculation result to the sub-index calculation layer; the different metadata processing rules in different metadata computing nodes are used for judging whether to trade within three months of industrial and commercial registration.

The sub-index calculation node 3, the sub-index calculation node 4, the sub-index calculation node 5 and the sub-index calculation node 6 respectively send the judgment results as sub-index judgment results to the target index calculation node; and the target index calculation node judges that the user hits the first single small amount of wind control strategy according to the received 4 sub-index judgment results.

In step S101, obtaining metadata required by the metadata calculation layer of the target wind control policy from the target service data of the online ticket service, and sending the obtained metadata to the metadata calculation layer, specifically: and when the triggering condition of the target wind control strategy is met, acquiring metadata required by a metadata calculation layer of the target wind control strategy in real time from target service data of the online ticket service, and sending the acquired metadata to the metadata calculation layer.

The triggering conditions of different target wind control strategies are different, for example, the triggering condition of the first single small wind control strategy is as follows: the user makes a fourth transaction.

When the triggering condition of the target wind control strategy is met, data are acquired in real time, and whether the target wind control strategy is hit or not is judged, so that the target wind control strategy is more timely.

The method for acquiring the metadata required by the metadata calculation layer of the target wind control strategy from the target service data of the online bill service comprises the following steps:

the metadata computation node receives the target metadata.

The node identification, the ID of the node can be calculated for the metadata.

The request information is, for example, time of business registration for both parties, time of first transaction for both parties, and the like.

The service data source may be a service database, a computational flow, a network service, or the like.

In the step S102, please refer to fig. 5, fig. 5 shows a flowchart of a method for processing the metadata to obtain a metadata calculation result according to an embodiment of the present application, and specifically, the flowchart of the method for processing the received metadata to obtain the metadata calculation result and sending the metadata calculation result to the sub-index calculation layer by using a plurality of metadata calculation nodes configured in advance by the metadata calculation layer includes the following steps S501 to S502:

s501, the metadata computing node in the metadata computing layer obtains target metadata required by the metadata computing node, and the target metadata are processed according to metadata processing rules to obtain a metadata computing result;

s502, according to a preset link connection relation, when the metadata computing node is judged to be linked with the sub-index computing node, the metadata computing result is sent to the linked sub-index computing node;

and S503, when the metadata computing node is judged to be linked with the metadata computing node of the next hierarchy, sending the metadata computing result as target metadata to the linked metadata computing node.

Wherein the metadata computation node processes the metadata according to metadata processing rules.

In connection with the directed acyclic graph model shown in FIG. 3, there may be multiple levels in the metadata computation layer; the metadata computing node in the metadata computing layer acquires target metadata required by the metadata computing node, and the metadata computing node acquires original metadata from a service data source directly under two conditions; alternatively, the processed metadata is acquired from a metadata computation node of an upper level.

If the metadata computing node is linked with a sub-index computing node, the metadata computing result is sent to the linked sub-index computing node; and if the metadata calculation node of the next hierarchy is linked, sending the metadata calculation result serving as target metadata to the linked metadata calculation node.

That is, in the metadata calculation layer, only the metadata calculation node at the top level directly obtains the original metadata from the service data source, and only the metadata calculation result output by the metadata calculation node at the top level is the metadata calculation result output by the metadata calculation layer.

In the step S103, please refer to fig. 6, where fig. 6 illustrates that the sub-indicator determination result is obtained by processing the metadata calculation result according to the embodiment of the present application; specifically, the method for the sub-index calculation layer to process the metadata calculation result to obtain a sub-index determination result according to the received metadata calculation result and a plurality of pre-configured sub-index calculation nodes and send the sub-index determination result to the target index calculation node includes the following steps S601 to S603:

s601, acquiring a metadata calculation result or an intermediate judgment result required by a sub-index calculation node by the sub-index calculation node in the sub-index calculation layer, and judging whether the metadata calculation result or the intermediate judgment result conforms to a sub-index judgment condition of the sub-index calculation node to obtain a sub-index judgment result;

s602, according to a preset link connection relation, when the sub-index calculation node is judged to be linked with the target index calculation node, the sub-index judgment result is sent to the target index calculation node;

and S603, when the metadata computing node is judged to be linked with the sub-index computing node of the next level, sending the sub-index judgment result to the linked sub-index computing node as an intermediate judgment result, so that the sub-index computing node of the next level obtains a new sub-index judgment result according to the received intermediate judgment result and the sub-index processing rule.

And the intermediate judgment result is a sub-index judgment result of a sub-index calculation node of a level above the sub-index calculation node.

And the sub-index calculation nodes process the metadata calculation results or the intermediate judgment results according to the sub-index processing rules.

In connection with the directed acyclic graph model shown in FIG. 3, there may be multiple levels in the sub-indicator computation layer; the sub-index computing nodes in the sub-index computing layer obtain target metadata required by the metadata computing nodes, and under two conditions, the sub-index computing nodes obtain metadata computing results from the metadata computing nodes; or, acquiring an intermediate judgment result obtained by judgment from the sub-index calculation node of the previous level.

The metadata calculation result or the intermediate judgment result obtained by the calculation of the sub-index calculation node is divided into two cases, if the sub-index calculation node is linked with the target index calculation node, the sub-index judgment result is sent to the linked target index calculation node; and if the sub-index computing nodes of the next level are linked, sending the sub-index computing nodes to the linked sub-index computing nodes as intermediate judgment results.

That is, in the sub-indicator calculation layer, a plurality of levels of calculation are performed, and the sub-indicator calculation node at the uppermost level receives the metadata calculation result, obtains an intermediate judgment result, and sends the intermediate judgment result to the sub-indicator calculation node at the next level until the sub-indicator calculation node at the lowermost level. Here, the sub-index calculation node of the lowermost hierarchy actually fuses the determination results of the plurality of sub-index calculation nodes, and transmits the fused sub-index determination result to the target index calculation node.

In step S104, determining whether the target service data hits a target wind control policy includes:

For example, the target index calculation node shown in fig. 4 receives the sub-index judgment results of 4 sub-index calculation nodes; if the sub-index judgment result is: sub-indicator computation node 3-no, i.e. no transaction with a third party; sub-index calculation node 4 — yes, i.e., the first small amount followed by the two large amounts; the sub-index calculation node 5-is, namely, three historical transactions exist; and the sub-index calculation node 6 is that the transaction is carried out within three months of the business registration, and the target index calculation node judges that the user hits the wind control strategy of the first single small amount.

In an embodiment of the present application, the method for processing wind control data further includes:

the strategy adjusting instruction comprises a stopping instruction and an operating instruction; the stop instruction is used for controlling the stop of the computing node to send the computing result to the computing node of the next level; the operation instruction is used for controlling the computing node to send the computing result to the computing node of the next level.

That is, if the configuration of at least one metadata computing node in the metadata computing layer changes and/or the configuration of at least one metadata computing node in the sub-index computing layer changes, the rule of the target wind control policy changes accordingly.

For example, as shown in fig. 4, if the sub-indicator calculating node 1 receives a stop instruction, the sub-indicator determining result is stopped from being sent to the sub-indicator calculating node 3; at this time, in the first single-small-amount wind control strategy, rule 4: the two sides of the order have not finished trading with the history of other third parties, and the change is that: the seller has no history of having completed the transaction with other third parties.

If the sub-index calculation node 3 receives the stop instruction, stopping sending the sub-index judgment result to the target index calculation node; at this time, in the first single-small-amount wind control strategy, rule 4: both parties of the order have no history of completing the transaction with other third parties, and the transaction has no effect, and the first small amount of wind control strategy only needs to meet the rules 1, 2 and 3 and the rules 4 and 5.

At this time, the target index computing node receives sub-index judgment results of 3 sub-index computing nodes; if the sub-index computing node 4-is, the first small amount is followed by the two large amounts; the sub-index calculation node 5-is, namely, three historical transactions exist; and the sub-index calculation node 6 is that the transaction is carried out within three months of the business registration, and the target index calculation node judges that the user hits the wind control strategy of the first single small amount.

On the basis, if the target index calculation node judges that the received sub-index judgment results all accord with preset conditions, the target business data is judged to hit a target wind control strategy; the number of the sub-index judgment results received by the target index calculation node is judged, and the number is related to the operation state of the sub-index calculation node which links the target index calculation node, and the operation state of the sub-index calculation node is determined by the configuration of the sub-index calculation node.

Furthermore, the processing procedure of the received metadata by the metadata computation layer in step S102 is determined by the operating state of each metadata computation node in the metadata computation layer, and the operating state of the metadata computation node is determined by the configuration of the metadata computation node.

In step S103, the processing procedure of the sub-index calculation layer on the received metadata is determined by the operating state of each sub-index calculation node in the sub-index calculation layer, and the operating state of the sub-index calculation node is determined by the configuration of the sub-index calculation node.

at least one computing node receives a rule modification instruction, updates a metadata computing layer of the computing node according to the rule modification instruction, processes the received metadata through a plurality of metadata computing nodes configured in advance to obtain a metadata computing result, and sends the metadata computing result to a sub-index computing layer; the metadata processing rules in different metadata computing nodes are different, so that the target wind control strategy is updated; the computing nodes are sub-index computing nodes, metadata computing nodes or target data computing nodes.

For example, as shown in the first single small amount wind control strategy in fig. 4, the business can be modified to trade within four months of the business registration time.

Referring to fig. 4, the following describes the processing procedure of the first small amount of wind control data:

a: and acquiring the configuration of a 'first single small amount' wind control model, checking to ensure that the index is in a starting state, and positioning to a specific actuator.

B: the sub-index calculation node 6 calculates a sub-index of 'business registration transaction within three months'; the sub-index "business registration within three months" of the sub-index computing node 6 relies on two metadata of "business registration time of both parties" and "first transaction time of both parties" of the upper level, and then inquire and obtain these two metadata first; and calculating the sub-index of 'trade within three months by industry and commerce registration' according to the dependence.

C: the sub-index calculation node 5 calculates a sub-index of 'three historical transactions exist', the sub-index of 'three historical transactions exist' depends on the metadata of 'four transaction details before history of two parties' at the upper stage, and the metadata of 'four transaction details before history of two parties' is inquired and acquired firstly; three historical transaction sub-indexes are calculated according to the dependence.

D: the sub-index calculation node 4 calculates a sub-index of 'first small amount followed by two large amounts'; the sub-index of the first small amount and the second large amount depends on the metadata of the historical transaction number of the two parties at the next stage, and the metadata of the historical transaction number of the two parties is inquired and obtained firstly; the sub-indicator of "first small forehead followed by two large forehead" is calculated according to the dependency.

E: the sub-index calculation node 1 calculates two metadata, namely the sub-index that the buyer does not transact with the third party and the sub-index that the seller does not transact with the third party are respectively dependent on the buyer and the third party historical transaction number, and the two metadata are inquired and obtained firstly; calculating two sub-indexes of 'buyer does not transact with the third party' and 'seller does not transact with the third party' according to dependence; the "sub-indicator is not traded with a third party according to the dependency calculation.

F: and calculating a first single small amount wind control target according to the dependence, if the four dependent sub-indexes are hit at the same time, considering that the wind control is hit, and otherwise, not hitting the wind control.

In the embodiment of the application, a wind control data processing device is further provided, which is used for acquiring metadata required by a metadata calculation layer of a target wind control strategy in real time, enabling each metadata calculation node to process the received metadata according to the metadata calculation layer through a plurality of metadata calculation nodes configured in advance, and sending the metadata calculation result to a sub-index calculation layer; the method comprises the steps that metadata processing rules in different metadata computing nodes process corresponding parts of service data differently to obtain metadata computing results quickly, then each sub-index computing node processes corresponding parts of the metadata computing results through a plurality of pre-configured sub-index computing nodes to obtain sub-index judging results, judging results of all sub-index computing nodes are fused quickly to obtain a total sub-index judging result, the fused sub-index judging results are sent to a target index computing node, and whether risks are hit or not is judged quickly; each metadata computing node, each sub-index computing node and each target index computing node respectively perform their own functions, the computing efficiency is high, each computing node is independently managed and computed, the coupling degree between the computing nodes is low, and the link relation between the nodes can be flexibly adjusted or nodes are added and deleted when a wind control strategy is adjusted, so that a wind control model is quickly adjusted; and the computing units corresponding to the nodes are pre-stored in the unit database, one computing unit can be reused by multiple wind control strategies, and the processing rule of the single computing unit is relatively simple, so that the construction difficulty is low, and the iterative scene is easy to develop quickly.

Referring to fig. 7, fig. 7 is a schematic structural diagram illustrating a wind-controlled data processing apparatus according to an embodiment of the present application; specifically, the apparatus comprises:

the obtaining module 701 is configured to obtain metadata required by a metadata calculation layer of a target wind control policy from target service data of an online ticket service, and send the obtained metadata to the metadata calculation layer;

a first processing module 702, configured to enable the metadata calculation layer to process the received metadata through a plurality of metadata calculation nodes configured in advance to obtain a metadata calculation result, and send the metadata calculation result to a sub-index calculation layer; wherein, the metadata processing rules in different metadata computing nodes are different;

a second processing module 703, configured to enable the sub-index calculation layer to process the received metadata calculation result to obtain a sub-index determination result through a plurality of pre-configured sub-index calculation nodes, and send the sub-index determination result to a target index calculation node; the sub-index processing rules in different index computing nodes are different;

a determining module 704, configured to determine, by the target index computing node, whether the target service data hits the target wind control policy according to the received sub-index determination result.

In some embodiments, when the obtaining module in the wind control data processing apparatus obtains metadata required by a metadata calculation layer of a target wind control policy from target service data of an online ticket service, the obtaining module is specifically configured to:

the metadata computation node receives the target metadata.

In some embodiments, when the metadata calculation layer is caused to process the received metadata through a plurality of metadata calculation nodes configured in advance to obtain a metadata calculation result, and send the metadata calculation result to the sub-index calculation layer, the first processing module in the wind control data processing apparatus is specifically configured to:

enabling a metadata computing node in the metadata computing layer to obtain target metadata required by the metadata computing node, and processing the target metadata according to a metadata processing rule to obtain a metadata computing result;

according to a preset link connection relation, when the metadata computing node is judged to be linked with the sub-index computing node, the metadata computing result is sent to the linked sub-index computing node;

In some embodiments, the second processing module in the wind control data processing apparatus is configured to, in a case that the sub-indicator calculation layer is enabled to process the metadata calculation result according to the received metadata calculation result and a plurality of pre-configured sub-indicator calculation nodes, obtain a sub-indicator determination result, and send the sub-indicator determination result to a target indicator calculation node, specifically, to:

enabling sub-index computing nodes in the sub-index computing layer to obtain metadata computing results or intermediate judging results required by the sub-index computing nodes, and judging whether the metadata computing results or the intermediate judging results meet sub-index judging conditions of the sub-index computing nodes or not to obtain sub-index judging results;

and when the metadata computing node is judged to be linked with the sub-index computing node of the next level, sending the sub-index judgment result serving as an intermediate judgment result to the linked sub-index computing node, so that the sub-index computing node of the next level can obtain a new sub-index judgment result according to the received intermediate judgment result and the sub-index processing rule.

In some embodiments, the determining module in the wind control data processing apparatus is specifically configured to, when determining whether the target service data hits the target wind control policy,:

In some embodiments, the wind-controlled data processing apparatus further includes:

the updating module is used for updating the configuration of the computing node according to the strategy adjusting instruction when at least one computing node receives the strategy adjusting instruction, so that the target wind control strategy is updated; the computing nodes are sub-index computing nodes or metadata computing nodes;

the system comprises a construction module, a data processing module and a data processing module, wherein the construction module is used for constructing a wind control strategy model, and the wind control strategy model comprises a metadata calculation layer, a sub-index calculation layer and a target index calculation node;

specifically, when the building module builds the wind control strategy model, the building module is specifically configured to: selecting a prestored target index computing node from a unit database according to the target wind control strategy, and constructing a target index computing node;

In an embodiment of the present application, an electronic device is further provided, please refer to fig. 8, and fig. 8 shows a schematic structural diagram of the electronic device 800 according to an embodiment of the present application, where the electronic device 800 includes: a processor 802, a memory 801 and a bus, wherein the memory 801 stores machine-readable instructions executable by the processor 802, when the electronic device 800 is running, the processor 802 communicates with the memory 801 via the bus, and the machine-readable instructions, when executed by the processor 802, perform the steps of the method for processing wind-controlled data.

In an embodiment of the present application, a computer-readable storage medium is further provided, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the computer program performs the steps of the wind-controlled data processing method.

It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the system and the apparatus described above may refer to corresponding processes in the method embodiments, and are not described in detail in this application. In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. The above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is merely a logical division, and there may be other divisions in actual implementation, and for example, a plurality of modules or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed coupling or direct coupling or communication connection between each other may be through some communication interfaces, indirect coupling or communication connection between devices or modules, and may be in an electrical, mechanical or other form.

The modules described as separate parts may or may not be physically separate, and parts displayed as modules may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.

The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a non-volatile computer-readable storage medium executable by a processor. Based on such understanding, the technical solutions of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling a computer device (which may be a personal computer, a platform server, or a network device) to execute all or part of the steps of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a U disk, a removable hard disk, a ROM, a RAM, a magnetic disk, or an optical disk.

The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A method of processing wind-controlled data, the method comprising the steps of:

when the triggering condition of the target wind control strategy is met, acquiring metadata required by a metadata calculation layer of the target wind control strategy in real time from target service data of the online bill service, and sending the acquired metadata to the metadata calculation layer;

the target index calculation node judges whether the target service data hits a target wind control strategy according to the received sub-index judgment result; the wind control strategy model formed by the metadata calculation layer, the sub index calculation layer and the target index calculation node is reversely determined according to a target index; according to the target wind control strategy, pre-stored target index computing nodes are selected from a unit database, pre-stored sub-index computing nodes are selected from the unit database according to sub-indexes, on which target indexes in the target index computing nodes depend, and pre-stored metadata computing nodes are selected from the unit database according to metadata, on which at least part of the sub-index computing nodes depend.

2. The method for processing wind control data according to claim 1, wherein obtaining metadata required by a metadata computation layer of a target wind control policy from target service data of an online ticket service includes:

the metadata computation node receives the target metadata.

3. The wind control data processing method according to claim 1, wherein the metadata computing layer processes the received metadata through a plurality of metadata computing nodes configured in advance to obtain a metadata computing result, and sends the metadata computing result to the sub-index computing layer, and the method includes:

4. The wind control data processing method according to claim 1, wherein the sub-index calculation layer processes the metadata calculation result to obtain a sub-index judgment result according to the received metadata calculation result and a plurality of pre-configured sub-index calculation nodes, and sends the sub-index judgment result to a target index calculation node, and the method includes:

according to a preset link connection relation, when the sub-index calculation node is judged to be linked with the target index calculation node, the sub-index judgment result is sent to the target index calculation node;

5. The method of claim 1, wherein determining whether the target business data hits in a target wind control policy comprises:

6. The method of processing wind-controlled data according to claim 1, further comprising:

7. The wind control data processing method according to claim 1, wherein a wind control strategy model formed by the metadata calculation layer, the sub index calculation layer and the target index calculation node is constructed by the following method:

according to metadata depended by at least part of the sub-index computing nodes, pre-stored metadata computing nodes are selected from the unit database, and link connection relations between the metadata computing nodes and part of the sub-index computing nodes and/or link connection relations between the metadata computing nodes are established, so that a metadata computing layer configured with a plurality of metadata computing nodes is constructed.

8. A wind-controlled data processing apparatus, characterized in that the apparatus comprises:

the acquisition module is used for acquiring metadata required by a metadata calculation layer of the target wind control strategy in real time from target service data of the online bill service when the trigger condition of the target wind control strategy is met, and transmitting the acquired metadata to the metadata calculation layer;

the second processing module is used for enabling the sub-index calculation layer to process the received metadata calculation result through a plurality of pre-configured sub-index calculation nodes to obtain a sub-index judgment result and send the sub-index judgment result to a target index calculation node; the sub-index processing rules in different index calculation nodes are different;

the judging module is used for enabling the target index computing node to judge whether the target service data hit a target wind control strategy or not according to the received sub-index judging result; the wind control strategy model formed by the metadata calculation layer, the sub-index calculation layer and the target index calculation node is reversely determined according to the target index; according to the target wind control strategy, pre-stored target index computing nodes are selected from a unit database, pre-stored sub-index computing nodes are selected from the unit database according to sub-indexes, on which target indexes in the target index computing nodes depend, and pre-stored metadata computing nodes are selected from the unit database according to metadata, on which at least part of the sub-index computing nodes depend.

9. An electronic device, comprising: a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, the processor and the memory communicating via the bus when the electronic device is operating, the machine-readable instructions when executed by the processor performing the steps of the method of processing wind-controlled data according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a computer program which, when being executed by a processor, carries out the steps of the method for processing wind-controlled data according to any one of claims 1 to 7.