CN112926984A

CN112926984A - Information prediction method based on block chain safety big data and block chain service system

Info

Publication number: CN112926984A
Application number: CN202110396663.XA
Authority: CN
Inventors: 郭栋; 花壮林
Original assignee: Individual
Current assignee: Shenzhen Xinhui Technology Co Ltd
Priority date: 2021-04-13
Filing date: 2021-04-13
Publication date: 2021-06-08
Anticipated expiration: 2041-04-13
Also published as: CN112926984B

Abstract

The embodiment of the disclosure provides an information prediction method based on big block chain safety data and a block chain service system, which can predict risk behavior coverage information of a target payment scene according to the payment prediction flow direction characteristic, the risk tendency value information and the target behavior probability. The risk behavior probability of the risk behavior of the paired payment behavior information is determined according to the paired payment behavior information and the payment behavior knowledge graph, namely the risk behavior probability of the risk behavior of the paired payment behavior information considers that the risk behavior probability of the risk behavior is a multi-path condition; the problem that the risk behavior coverage information of the target payment scene is inaccurate due to the fact that the risk behavior probability of the risk behaviors of different risk behavior labels is multipath can be solved, accuracy of the risk behavior coverage information can be improved, and prediction accuracy of the risk behaviors can be improved.

Description

Information prediction method based on block chain safety big data and block chain service system

Technical Field

The disclosure relates to the technical field of block chain payment security, and exemplarily relates to an information prediction method based on big block chain security data and a block chain service system.

Background

The block chain is used as an integrated innovation of a plurality of technologies such as a point-to-point network, cryptography, a sharing mechanism, an intelligent contract and the like, and provides a trusted channel for information and value transfer and exchange in an untrusted network. The block chain technology is a popular concept in the world no matter in the aspect of constructing the internet with free value circulation or in the aspect of data sharing based on 'establishing a joint multi-center' of an enterprise, and has wide market prospect.

In the related art, under a safety protection operation program based on block chain payment, risk behavior coverage information needs to be predicted to determine the range of various parameter values of risk behaviors under the current safety protection operation program, so that a reference basis can be provided for a subsequent research and development plan conveniently, but the problem that the risk behavior coverage information of a target payment scene is inaccurate in the current information prediction scheme still exists.

Disclosure of Invention

In order to overcome at least the above-mentioned deficiencies in the prior art, the present disclosure provides an information prediction method based on big data of block chain security and a block chain service system.

In a first aspect, the present disclosure provides an information prediction method based on big safety data of a block chain, which is applied to a block chain service system, where the block chain service system is in communication connection with a plurality of block chain node terminals, and the method includes:

obtaining key payment safety data of a target payment scene searched based on block chain safety big data generated by online payment of a block chain, wherein the block chain safety big data is used for representing payment safety verification events among block chain payment services, and the block chain payment services are payment services related to a target safety protection operation program;

acquiring paired payment behavior information in a paired payment scene associated with a target payment scene from the key payment safety data according to a pre-trained AI network corresponding to the target safety protection operation program, and acquiring a payment behavior knowledge graph corresponding to the paired payment behavior information;

determining payment prediction flow direction characteristics of the paired payment behavior information in the paired payment scene according to the payment behavior knowledge graph;

predicting the risk behavior probability of the risk behavior of the paired payment behavior information according to the paired payment behavior information and the payment behavior knowledge graph to serve as a prediction probability, and determining the risk tendency value information of the risk behavior of the paired payment behavior information in the paired payment scene according to the prediction probability;

and acquiring the risk behavior probability of the risk behavior in the target payment scene as a target behavior probability, and predicting the risk behavior coverage information of the target payment scene according to the payment prediction flow direction characteristic, the risk tendency value information and the target behavior probability.

In a second aspect, an embodiment of the present disclosure further provides an information prediction system based on big block chain security data, where the information prediction system based on big block chain security data includes a block chain service system and a plurality of block chain link point terminals communicatively connected to the block chain service system;

the block chain service system is configured to:

Based on any one of the above aspects, the present disclosure may predict the risk behavior coverage information of the target payment scenario according to the payment prediction flow direction feature, the risk tendency value information, and the target behavior probability. The risk behavior probability of the risk behavior of the paired payment behavior information is determined according to the paired payment behavior information and the payment behavior knowledge graph, namely the risk behavior probability of the risk behavior of the paired payment behavior information considers that the risk behavior probability of the risk behavior is a multi-path condition; the problem that the risk behavior coverage information of the target payment scene is inaccurate due to the fact that the risk behavior probability of the risk behaviors of different risk behavior labels is multipath can be solved, accuracy of the risk behavior coverage information can be improved, and prediction accuracy of the risk behaviors can be improved.

Drawings

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

Fig. 1 is a schematic view of an application scenario of an information prediction system based on big block chain security data according to an embodiment of the present disclosure;

fig. 2 is a schematic flowchart of an information prediction method based on big block chain security data according to an embodiment of the present disclosure;

fig. 3 is a functional block diagram of an information prediction apparatus based on big data for block chain security according to an embodiment of the present disclosure;

fig. 4 is a schematic block diagram illustrating structural components of a blockchain service system for implementing the above information prediction method based on blockchain security big data according to an embodiment of the present disclosure.

Detailed Description

The present disclosure is described in detail below with reference to the drawings, and the specific operation methods in the method embodiments can also be applied to the device embodiments or the system embodiments.

Fig. 1 is an interaction diagram of an information prediction system 10 based on big data block chain security according to an embodiment of the present disclosure. The system 10 for predicting information based on big data of blockchain security may include a blockchain service system 100 and a blockchain endpoint terminal 200 communicatively connected to the blockchain service system 100. The information prediction system 10 based on the blockchain security big data shown in fig. 1 is only one possible example, and in other possible embodiments, the information prediction system 10 based on the blockchain security big data may also include only at least some of the components shown in fig. 1 or may also include other components.

In an embodiment, the blockchain service system 100 and the blockchain node terminal 200 in the information prediction system 10 based on blockchain security big data may cooperatively perform the information prediction method based on blockchain security big data described in the following method embodiments, and the detailed description of the method embodiments may be referred to in the following steps for the specific blockchain service system 100 and the blockchain node terminal 200.

To solve the technical problem in the foregoing background, fig. 2 is a flowchart illustrating an information prediction method based on big data for block chain security according to an embodiment of the present disclosure, where the information prediction method based on big data for block chain security according to the embodiment may be executed by the block chain service system 100 shown in fig. 1, and the information prediction method based on big data for block chain security is described in detail below.

Step S110, obtaining key payment safety data of the target payment scene searched based on the block chain safety big data generated by block chain online payment.

In this embodiment, the big blockchain security data is used to characterize a payment security verification event between blockchain payment services, where the blockchain payment services are payment services related to the target security protection operating program. The specific implementation will be described in detail later.

Step S120, obtaining pairing payment behavior information in a pairing payment scene associated with the target payment scene from the key payment safety data according to a pre-trained AI network corresponding to the target safety protection operation program, and obtaining a payment behavior knowledge graph corresponding to the pairing payment behavior information.

In this embodiment, when the risk behavior coverage information of the target payment scenario needs to be obtained, the blockchain service system 100 may obtain relevant payment scenario data of the paired payment scenario, where the relevant payment scenario data includes at least one of dynamic payment scenario data and static payment scenario data. The block chain service system 100 may identify, according to a pre-trained AI network corresponding to the target security protection operating program, relevant payment scene data of the paired payment scene to obtain risk behavior information in the paired payment scene, where the risk behavior information of the paired payment scene includes paired payment behavior information in the paired payment scene, a payment behavior knowledge graph of the paired payment behavior information, and the like; paired payment behavior information herein may refer to any of the risk behavior tags in the paired payment scenario.

It should be noted that the pre-trained AI network may be obtained by collecting training data samples under a target security protection operation program and corresponding risk behavior preset attributes for training, so that the pre-trained AI network may obtain a plurality of paired payment behavior information based on the key payment security data, and then screen out the paired payment behavior information in the paired payment scenario associated with the target payment scenario, thereby obtaining a payment behavior knowledge graph corresponding to the paired payment behavior information.

It can be understood that the specific training process of the pre-trained AI network may be implemented in a conventional training manner in the prior art, and details of this training process are not described in this disclosure.

Step S130, determining a payment prediction flow direction feature of the paired payment behavior information in the paired payment scenario according to the payment behavior knowledge graph.

In this embodiment, the payment prediction flow direction feature of the target risky behavior tag in the target payment scene is matched with the payment prediction flow direction feature of the paired payment behavior information in the paired payment scene, and the payment behavior knowledge graph (for example, the statistical number of the risky behavior tags) corresponding to the target risky behavior tag in the target payment scene is unknown, that is, it is difficult to obtain the payment prediction flow direction feature of the target risky behavior tag in the target payment scene. Therefore, the blockchain service system 100 may determine the payment prediction flow direction feature of the paired payment behavior information in the paired payment scenario according to the payment behavior knowledge map, which is beneficial to determine the payment prediction flow direction feature of the target risk behavior tag in the target payment scenario according to the payment prediction flow direction feature of the paired payment behavior information in the paired payment scenario; namely, the payment prediction flow characteristics of the paired payment behavior information in the paired payment scene are used for reflecting: matching behavior features between a payment behavior knowledge graph of paired payment behavior information in a paired payment scenario and risk behavior distribution information in the paired payment scenario.

Step S140, predicting the risk behavior probability of the risk behavior of the paired payment behavior information according to the paired payment behavior information and the payment behavior knowledge graph, as a prediction probability, and determining the risk tendency value information of the risk behavior of the paired payment behavior information in the paired payment scenario according to the prediction probability.

In this embodiment, the behavior probability described above may be used to characterize the attention degree of the risk generating behavior group, and the higher the behavior probability is, the higher the attention degree of the risk subject behavior group with a wide surface is.

In this embodiment, the blockchain service system 100 may predict, according to the paired payment behavior information and the payment behavior knowledge graph in the paired payment scenario, a risk behavior probability of a risk behavior of the paired payment behavior information as a prediction probability; the predicted probability may include at least one of a risk behavior probability when the risk behavior of the pair of payment behavior information is not triggered and a risk behavior probability when the risk behavior of the pair of payment behavior information is triggered.

In an embodiment, the implementation manners of determining the risk tendency value information of the risk behavior of the pair payment behavior information in the pair payment scenario according to the prediction probability include the following three implementation manners.

In a first implementation manner, when the predicted probability includes a risk behavior probability when the risk behavior of the paired payment behavior information is not triggered, the block chain service system 100 may use the risk behavior probability when the risk behavior of the paired payment behavior information is not triggered as a first behavior probability; and generating a risk tendency reference value of the risk behavior of the paired payment behavior information when the risk behavior is not triggered according to the first behavior probability, wherein the risk tendency reference value is used as the first reference value. The first reference value is used for reflecting the ratio of the first behavior probability to the first total behavior probability, and the first total behavior probability is the total behavior probability of the risk behavior in the paired payment scene when the risk behavior is not triggered; and taking the first reference value as risk tendency value information of the risk behaviors of the pair payment behavior information in the pair payment scene.

If none of the risk behaviors in the paired payment scenario is triggered, the blockchain service system 100 may use the risk behavior probability when the risk behavior of the paired payment behavior information is not triggered as the first behavior probability; and accumulating and summing the first behavior probability to obtain a total behavior probability of the risk behavior when the risk behavior in the paired payment scene is not triggered, wherein the total behavior probability is used as the first total behavior probability. Further, the ratio of the first action probability to the first total action probability is used as a risk tendency reference value of the risk action of the paired payment action information when the risk action is not triggered; taking a risk tendency reference value of the risk behavior of the paired payment behavior information when the risk behavior is not triggered as a first reference value, namely that the risk tendency value information of the risk behavior of the paired payment behavior information in the paired payment scene comprises the first reference value; the risk tendency value information is used for reflecting the proportion of the risk behavior probability of the risk behavior of the paired payment behavior information when the risk behavior is not triggered.

In a second implementation manner, when the predicted probability includes the risk behavior probability when the risk behavior of the paired payment behavior information is triggered, the block chain service system 100 may use the risk behavior probability when the risk behavior of the paired payment behavior information is triggered as a second behavior probability; and generating a risk tendency reference value of the risk behavior of the paired payment behavior information when the risk behavior is triggered according to the second behavior probability, wherein the risk tendency reference value is used as a second reference value. The second reference value is used for reflecting the ratio of the first secondary behavior probability and the second total behavior probability, and the second total behavior probability is the total risk behavior probability of the risk behavior in the paired payment scene when the risk behavior is triggered; and taking the second reference value as risk tendency value information of the risk behaviors of the pair payment behavior information in the pair payment scene.

If the risk behaviors in the paired payment scenario are all triggered, the block chain service system 100 may use the risk behavior probability when the risk behaviors of the paired payment behavior information are triggered as a second behavior probability; and the second behavior probabilities are accumulated and summed to obtain the total behavior probability of the risk behaviors when the risk behaviors in the paired payment scene are triggered, and the total behavior probability is used as the second total behavior probability. Further, the ratio of the second behavior probability to the second total behavior probability is used as a risk tendency reference value of the risk behavior of the paired payment behavior information when triggered; taking a risk tendency reference value of the risk behavior of the paired payment behavior information when triggered as a second reference value, wherein the risk behavior of the paired payment behavior information in the paired payment scene comprises the risk tendency value information; namely, the risk tendency value information is used for reflecting the proportion of the risk behavior probability of the risk behavior of the paired payment behavior information when the risk behavior is not triggered.

In a third implementation manner, when the prediction probability may include the risk behavior probability when the risk behavior of the paired payment behavior information is not triggered and the risk behavior probability when the risk behavior of the paired payment behavior information is triggered; the blockchain service system 100 may use the risk behavior probability when the risk behavior of the paired payment behavior information is not triggered as the first behavior probability; and generating a risk tendency reference value of the risk behavior of the paired payment behavior information when the risk behavior is not triggered according to the first behavior probability, wherein the risk tendency reference value is used as the first reference value. Taking the risk behavior probability when the risk behavior of the paired payment behavior information is triggered as a second behavior probability; generating a risk tendency reference value of the risk behavior of the paired payment behavior information when the risk behavior is triggered according to the second behavior probability, and taking the risk tendency reference value as a second reference value; further, the first reference value and the second reference value may be used as risk tendency value information of the risk behavior of the pair payment behavior information in the pair payment scenario.

If the risk behavior in the paired payment scenario is triggered and not triggered, the blockchain service system 100 may refer to the above steps to obtain a first reference value and a second reference value, and use the first reference value and the second reference value as risk tendency value information of the risk behavior of the paired payment behavior information in the paired payment scenario; namely, the risk tendency value information of the risk behavior of the paired payment behavior information in the paired payment scenario is used for reflecting: the proportion of the risk behavior probability of the risk behavior of the paired payment behavior information when the risk behavior is not triggered is larger than the proportion of the risk behavior probability of the risk behavior of the paired payment behavior information when the risk behavior is triggered.

Step S150, acquiring the risk behavior probability of the risk behavior in the target payment scene as the target behavior probability, and predicting the risk behavior coverage information of the target payment scene according to the payment prediction flow direction characteristic, the risk tendency value information and the target behavior probability.

The blockchain service system 100 may obtain, from the traffic management device in the target payment scenario, a risk behavior probability of a risk behavior in the target payment scenario as a target behavior probability, where the target behavior probability is a total risk behavior probability of the risk behavior in the target payment scenario. In one embodiment, when none of the risky behaviors of the target payment scenario is triggered, the target behavior probability is the risky behavior probability of the risky behavior in the target payment scenario when the risky behavior is not triggered. In one embodiment, when all the risky behaviors in the target payment scenario are triggered, the target behavior probability is the risky behavior probability of the risky behaviors in the target payment scenario at the moment of triggering. In one embodiment, when there are triggered risky behaviors and non-triggered risky behaviors in the target payment scenario, the target behavior probability is the sum of the risky behavior probability of the triggered risky behaviors in the target payment scenario and the risky behavior probability of the non-triggered risky behaviors in the target payment scenario.

The payment prediction flow direction feature is used for reflecting the matching behavior feature between the payment behavior knowledge graph corresponding to the paired payment behavior information and the risk behavior distribution information in the paired payment scene, that is, the payment prediction flow direction feature can be used for reflecting the relationship between the payment behavior knowledge graph corresponding to the paired payment behavior information and the risk behavior distribution information in the paired payment scene. The risk tendency value information is used for reflecting the ratio between the risk behavior probability of the risk behavior of the paired payment behavior information and the risk behavior total behavior probability of the risk behavior in the paired payment scene, and the risk behavior probability of the risk behavior of the paired payment behavior information is determined according to the payment behavior knowledge graph corresponding to the paired payment behavior information, so that the risk tendency value information can be used for reflecting the relationship among the payment behavior knowledge graph, the risk behavior probability of the risk behavior of the paired payment behavior information and the risk behavior total behavior probability of the risk behavior in the paired payment scene. The target payment scene is associated with the paired payment scene, namely, the payment prediction flow direction characteristic and the risk tendency value information corresponding to the target payment scene are respectively matched with the payment prediction flow direction characteristic and the risk tendency value information corresponding to the paired payment scene, namely, the relationship between the payment behavior knowledge graph corresponding to the target risk behavior tag and the risk behavior distribution information in the target payment scene is matched with the relationship between the payment behavior knowledge graph corresponding to the paired payment behavior information and the risk behavior distribution information in the paired payment scene; and the relation among the payment behavior knowledge graph corresponding to the target risk behavior label, the risk behavior probability of the risk behavior of the paired payment behavior information and the risk behavior total behavior probability of the risk behavior in the paired payment scene is matched with the relation among the payment behavior knowledge graph corresponding to the paired payment behavior information, the risk behavior probability of the risk behavior of the paired payment behavior information and the risk behavior total behavior probability of the risk behavior in the paired payment scene.

The blockchain service system 100 may use the payment prediction flow direction feature corresponding to the paired payment scenario as the payment prediction flow direction feature corresponding to the target payment scenario, and use the risk tendency value information corresponding to the paired payment scenario as the risk tendency value information corresponding to the target payment scenario. Further, risk behavior coverage information of the target payment scene can be predicted according to the payment prediction flow direction characteristic, the risk tendency value information and the target behavior probability, wherein the risk behavior coverage information of the target payment scene is used for reflecting a risk behavior statistic value passing through the target payment scene in a unit time period.

Therefore, information pushing can be performed based on the risk behavior coverage information of the target payment scene, for example, information pushing corresponding to risk behaviors can be performed sequentially according to the size sequence of the risk behavior statistic.

In the embodiment of the application, the payment prediction flow direction feature is used for reflecting the matching behavior feature between the payment behavior knowledge graph corresponding to the paired payment behavior information and the risk behavior distribution information in the paired payment scene, that is, the payment prediction flow direction feature may be used for reflecting the relationship between the payment behavior knowledge graph corresponding to the paired payment behavior information and the risk behavior distribution information in the paired payment scene. The risk tendency value information is used for reflecting the ratio of the risk behavior probability of the risk behavior of the paired payment behavior information to the risk behavior total behavior probability of the risk behavior in the paired payment scene, and the risk behavior probability of the risk behavior of the paired payment behavior information is determined according to the payment behavior knowledge graph corresponding to the paired payment behavior information, so that the risk tendency value information can be used for reflecting the relationship among the payment behavior knowledge graph, the risk behavior probability of the risk behavior of the paired payment behavior information and the risk behavior total behavior probability of the risk behavior in the paired payment scene. The target payment scene is associated with the paired payment scene, namely, the payment prediction flow direction characteristic and the risk tendency value information corresponding to the target payment scene are respectively matched with the payment prediction flow direction characteristic and the risk tendency value information corresponding to the paired payment scene, namely, the relationship between the payment behavior knowledge graph corresponding to the target risk behavior tag and the risk behavior distribution information in the target payment scene is matched with the relationship between the payment behavior knowledge graph corresponding to the paired payment behavior information and the risk behavior distribution information in the paired payment scene; and the relation among the payment behavior knowledge graph corresponding to the target risk behavior label, the risk behavior probability of the risk behavior of the paired payment behavior information and the risk behavior total behavior probability of the risk behavior in the paired payment scene is matched with the relation among the payment behavior knowledge graph corresponding to the paired payment behavior information, the risk behavior probability of the risk behavior of the paired payment behavior information and the risk behavior total behavior probability of the risk behavior in the paired payment scene. Therefore, risk behavior coverage information of the target payment scene can be predicted according to the payment prediction flow direction feature, the risk tendency value information and the target behavior probability. The risk behavior probability of the risk behavior of the paired payment behavior information is determined according to the paired payment behavior information and the payment behavior knowledge graph, namely the risk behavior probability of the risk behavior of the paired payment behavior information considers that the risk behavior probability of the risk behavior is a multi-path condition; the problem that the risk behavior coverage information of the target payment scene is inaccurate due to the fact that the risk behavior probability of the risk behaviors of different risk behavior labels is multipath can be solved, accuracy of the risk behavior coverage information can be improved, and prediction accuracy of the risk behaviors can be improved.

In an embodiment, the information prediction method based on the big block chain security data may further include the following steps S101 to S103.

Step S101, obtaining historical payment prediction flow characteristics of the target risk behavior tag in candidate payment scenes of the candidate payment scene sequence in a historical time period.

Step S102, obtaining historical payment prediction flow characteristics of the target risk behavior label in the target payment scene in the historical time period.

Step S103, using the candidate payment scene in which the corresponding historical payment prediction flow direction feature in the candidate payment scene sequence matches the historical payment prediction flow direction feature in the target payment scene as a paired payment scene associated with the target payment scene.

In steps S101 to S103, the blockchain service system 100 may obtain a historical tag statistic corresponding to the target risky behavior tag in the candidate payment scenes of the candidate payment scene sequence in the historical time period, as a first historical tag statistic, and determine a historical payment prediction flow direction feature of the target risky behavior tag in the candidate payment scenes of the candidate payment scene sequence in the historical time period according to the first historical tag statistic; and the historical payment prediction flow direction characteristic of the target risk behavior tag in the candidate payment scene is used for reflecting the matching behavior characteristic between the first historical tag statistic and the risk behavior distribution information in the candidate payment scene. Further, historical tag statistics of the target risk behavior tag in the target payment scenario within the historical time period may be obtained as a second historical tag statistics; determining historical payment prediction flow direction characteristics of the target risk behavior tag in the target payment scene in the historical time period according to the second historical tag statistic; namely, the historical payment prediction flow direction characteristic of the target risky behavior tag in the target payment scene is used for reflecting the ratio between the second historical tag statistic and the risky behavior distribution information in the target payment scene. And if the difference value between the matching behavior characteristic between the first historical label statistic and the risk behavior distribution information in the candidate payment scene and the ratio between the second historical label statistic and the risk behavior distribution information in the target payment scene is larger than or equal to the difference threshold value, determining that the historical payment prediction flow direction characteristic corresponding to the candidate payment scene is not matched with the historical payment prediction flow direction characteristic in the target payment scene. If the difference between the matching behavior characteristic between the first historical tag statistic and the risk behavior distribution information in the candidate payment scene and the ratio between the second historical tag statistic and the risk behavior distribution information in the target payment scene is smaller than a difference threshold value, it is determined that the historical payment prediction flow direction characteristic corresponding to the candidate payment scene is matched with the historical payment prediction flow direction characteristic in the target payment scene, and the candidate payment scene can be used as a paired payment scene associated with the target payment scene.

In one embodiment, step S130 may include the following steps S131 and S132.

Step S131, acquiring risk behavior distribution information in the paired payment scenario.

Step S132, acquiring a matching behavior feature between the payment behavior knowledge graph and the risk behavior distribution information in the paired payment scenario, and taking the matching behavior feature as a payment prediction flow direction feature of the paired payment behavior information in the paired payment scenario.

In steps S131 and S132, since the payment prediction flow direction feature of the target risky behavior tag in the target payment scenario is matched with the payment prediction flow direction feature of the paired payment behavior information in the paired payment scenario, meanwhile, the payment prediction flow direction feature corresponding to the target risky behavior tag is difficult to directly obtain; therefore, the blockchain service system 100 may obtain the payment prediction flow direction characteristic of the target risk behavior tag in the target payment scenario by obtaining the payment prediction flow direction characteristic of the pair payment behavior information in the pair payment scenario. Specifically, the blockchain service system 100 may identify relevant payment scene data in a paired payment scene to obtain risk behavior distribution information in the paired payment scene, obtain a matching behavior feature between the payment behavior knowledge graph and the risk behavior distribution information in the paired payment scene, and use the matching behavior feature as a payment prediction flow direction feature of the paired payment behavior information in the paired payment scene, that is, use a payment prediction flow direction feature of the paired payment behavior information in the paired payment scene as a payment prediction flow direction feature of a target risk behavior tag in a target payment scene.

In one embodiment, step S140 may include the following steps S141-S145.

And a substep S141, determining the risk behavior probability of the risk behavior with the paired payment behavior information when the risk behavior is not triggered according to the paired payment behavior information and the payment behavior knowledge graph, and taking the risk behavior probability as the first behavior probability.

Substep S142, taking the ratio between the first behavior probability and the first total behavior probability as a risk tendency reference value of the risk behavior of the paired payment behavior information when the risk behavior is not triggered, and taking the ratio as a first reference value; the first total behavior probability is the total behavior probability of the risk behaviors in the paired payment scene when the risk behaviors are not triggered.

Substep S143, determining the risk behavior probability of the risk behavior with the paired payment behavior information when triggered according to the paired payment behavior information and the payment behavior knowledge graph, and using the risk behavior probability as a second behavior probability; the first behavior probability and the second behavior probability both belong to the prediction probability.

Substep S144, generating a risk tendency reference value of the risk behavior of the paired payment behavior information when triggered by the ratio between the second behavior probability and the second total behavior probability, as a second reference value; the second total behavior probability is a risk behavior total behavior probability of the risk behavior in the paired payment scene when triggered.

And a substep S145, generating risk tendency value information of the risk behavior of the paired payment behavior information in the paired payment scenario according to the first reference value and the second reference value.

In steps S141 to S145, since the risk tendency value information of the risk behavior of the target risk behavior tag in the target payment scenario matches the risk tendency value information of the risk behavior of the paired payment scenario, the block chain service system 100 may obtain the risk tendency value information of the risk behavior of the target risk behavior tag in the target payment scenario by obtaining the risk tendency value information of the risk behavior of the paired payment behavior information in the paired payment scenario. Specifically, the blockchain service system 100 may determine, according to the paired payment behavior information and the payment behavior knowledge graph, a risk behavior probability of a risk behavior having the paired payment behavior information when the risk behavior is not triggered, as a first behavior probability; and taking the ratio of the first action probability and the first total action probability as a risk tendency reference value of the risk action of the paired payment action information when the risk action is not triggered as a first reference value. The first behavior probability is the total behavior probability of the risk behaviors with the paired payment behavior information when the risk behaviors are not triggered, and the first behavior probability comprises the risk behavior probability of each service risk generation behavior configured when the risk behaviors of the reference type are not triggered; the condition that the risk behaviors have multi-path configuration risk behavior subscription data is fully considered, service risk generating behaviors in the risk behaviors can be prevented from being mistaken for the risk behaviors in the paired payment scene, and the accuracy of acquiring the risk behavior coverage information of the risk behaviors is improved. Similarly, the blockchain service system 100 may determine, according to the paired payment behavior information and the payment behavior knowledge graph, a risk behavior probability of the risk behavior having the paired payment behavior information when triggered, as a second behavior probability, and generate a risk tendency reference value of the risk behavior of the paired payment behavior information when triggered, as a second reference value, by using a ratio between the second behavior probability and a second total behavior probability; the second behavior probability is the total behavior probability of the risky behaviors with the paired payment behavior information at the time of triggering, and the first behavior probability comprises the risk behavior probability of each service risk generating behavior configured by the reference type of risky behaviors at the time of triggering; the condition that the risk behaviors have multi-path configuration risk behavior subscription data is fully considered, service risk generating behaviors in the risk behaviors can be prevented from being mistaken for the risk behaviors in the paired payment scene, and the accuracy of acquiring the risk behavior coverage information of the risk behaviors is improved.

In one embodiment, the substep S141 may be implemented by the following exemplary embodiments.

(1) And determining authority limit information configured when the risk behaviors with the pair payment behavior information are not triggered as first authority limit information.

(2) And acquiring corresponding probability of the same authority limit quantity between the first authority limit information and the payment behavior knowledge graph as the risk behavior probability of the risk behavior with the paired payment behavior information when the risk behavior is not triggered.

(3) And taking the risk behavior probability of the risk behavior with the paired payment behavior information when the risk behavior is not triggered as the first behavior probability.

Generally, configured service risk generating behaviors of risk behaviors of paired payment behavior information all carry terminals, and the terminals carried by the service risk generating behaviors can configure risk behavior subscription data; therefore, the blockchain service system 100 may determine the risk behavior probability of the risk behavior with the paired payment behavior information when the risk behavior is not triggered according to the configured authority limit information of the risk behavior with the paired payment behavior information. Specifically, the blockchain service system 100 may determine, as the first permission restriction information, permission restriction information configured when the risky behavior having the paired payment behavior information is not triggered, that is, the first permission restriction information may be the maximum permission restriction information to which the risky behavior having the paired payment behavior information can be subscribed when the risky behavior having the paired payment behavior information is not triggered; the corresponding probability of the same authority limit quantity between the first authority limit information and the payment behavior knowledge graph is obtained and used as the risk behavior probability of the risk behavior with the paired payment behavior information when the risk behavior is not triggered, and the risk behavior probability of the risk behavior with the paired payment behavior information when the risk behavior is not triggered can be used as the first behavior probability.

In one embodiment, the substep S143 may be implemented by the following exemplary embodiments.

(2) And acquiring historical trigger times of the risk behaviors with the paired payment behavior information, and generating trigger weights according to the historical trigger times.

(3) And determining the authority limit information configured when the risk behavior with the paired payment behavior information is triggered as second authority limit information according to the first authority limit information and the trigger weight.

(4) And acquiring the corresponding probability of the same authority limit quantity between the second authority limit information and the payment behavior knowledge graph as the risk behavior probability of the risk behavior with the paired payment behavior information when triggered.

(5) And taking the risk behavior probability of the risk behavior with the paired payment behavior information at the triggering time as the second behavior probability.

The block chain service system 100 may obtain a seat statistical value of a risk behavior of the paired payment behavior information, and determine the seat statistical value of the risk behavior of the paired payment behavior information as authority limit information configured when the risk behavior having the paired payment behavior information is not triggered, as the first authority limit information. Further, the historical trigger times of the risky behaviors with the paired payment behavior information can be determined according to the historical trigger records of the risky behaviors with the paired payment behavior information, and the historical trigger times are the ratio of the risky behavior trigger authority limit information and the first authority limit information of the paired payment behavior information; generating a trigger weight according to the historical trigger times, wherein the trigger weight and the historical trigger times have a direct proportional relation, namely the larger the historical trigger times, the larger the trigger weight; conversely, the smaller the historical number of triggers, the smaller the trigger weight. Further, the product of the first permission limit information and the trigger weight can be obtained, and permission limit information configured when the risk behavior with the paired payment behavior information is triggered is obtained and serves as second permission limit information; acquiring corresponding probabilities of the same authority limit quantity between the second authority limit information and the payment behavior knowledge graph, and taking the probabilities as the risk behavior probabilities of the risk behaviors with the paired payment behavior information when triggered; and taking the risk behavior probability of the risk behavior with the paired payment behavior information at the triggering time as the second behavior probability.

In one embodiment, step S150 may include steps S151-S152 as follows.

Step S151, generating a risk behavior coverage range of a target risk behavior label according to the payment prediction flow direction characteristic, the risk tendency value information and the target behavior probability; the target risk behavior tag is a risk behavior tag in the target payment scene, wherein the risk behavior tag is the same as the paired payment behavior information.

And step S152, determining the risk behavior coverage information of the target payment scene from the risk behavior coverage range of the target risk behavior label.

In steps S151-S152, the blockchain service system 100 may generate a risk behavior coverage of a target risk behavior label according to the payment prediction flow direction characteristic, the risk tendency value information, and the target behavior probability; the risk behavior statistic of the target risk behavior label in the target payment scene belongs to the risk behavior coverage range; therefore, the risk behavior coverage information of the target payment scene can be determined from the risk behavior coverage range of the target risk behavior tag, that is, the risk behavior configuration numerical value is obtained from the risk behavior coverage range of the target risk behavior tag, and the risk behavior coverage information of the target payment scene is determined according to the risk behavior coverage value. The risk behavior coverage information of the target payment scene is obtained according to the risk behavior coverage information of the different risk behavior labels, and the accuracy of obtaining the risk behavior coverage information of the risk behaviors is improved.

In an embodiment, the risk tendency value information includes a first reference value and a second reference value, and the step S151 may be implemented by the following exemplary embodiments.

(1) And acquiring the maximum ratio and the minimum ratio in the matching behavior characteristics between the payment behavior knowledge graph and the risk behavior distribution information in the paired payment scene, the ratio between the first behavior probability and the first total behavior probability, and the ratio between the second behavior probability and the second total behavior probability.

(2) And acquiring the product of the maximum ratio and the target behavior probability as a first risk behavior coverage value.

(3) And acquiring the product of the minimum ratio and the target behavior probability as a second risk behavior coverage value.

(4) And taking an interval formed by the first risk behavior coverage value and the second risk behavior coverage value as the risk behavior coverage range of the target risk behavior label.

Usually, there are triggered risk behaviors and non-triggered risk behaviors in the target payment scenario, and therefore, the risk behavior coverage of the target risk behavior tag can be predicted according to the first reference value corresponding to the risk behavior of the paired payment behavior information in the paired payment scenario when the risk behavior is not triggered and the second reference value corresponding to the risk behavior when the risk behavior is triggered. Specifically, the blockchain service system 100 may compare the matching behavior characteristics between the payment behavior knowledge graph and the risk behavior distribution information in the paired payment scenario, the ratio between the first behavior probability and the first total behavior probability, and the ratio between the second behavior probability and the second total behavior probability to obtain the maximum ratio and the minimum ratio among the matching behavior characteristics between the payment behavior knowledge graph and the risk behavior distribution information in the paired payment scenario, the ratio between the first behavior probability and the first total behavior probability, and the ratio between the second behavior probability and the second total behavior probability. Further, obtaining a product between the maximum ratio and the target behavior probability as a first risk behavior coverage value; and acquiring the product of the minimum ratio and the target behavior probability as a second risk behavior coverage value, and taking an interval formed by the first risk behavior coverage value and the second risk behavior coverage value as the risk behavior coverage range of the target risk behavior label. By considering the risk tendency value information corresponding to the risk behavior when triggered and the risk tendency value information corresponding to the risk behavior when not triggered, and meanwhile, considering the condition that service risk generating behavior in the risk behavior configures risk behavior subscription data; the accuracy of obtaining the risk behavior coverage of the target risk behavior label can be improved.

In one embodiment, the risk tendency value information includes a first reference value, and the step S151 includes: acquiring a product of a matching behavior characteristic between the payment behavior knowledge graph and risk behavior distribution information in the paired payment scene and a target behavior probability, and taking the product as a first risk behavior coverage value; obtaining the product of the ratio of the first behavior probability to the first total behavior probability and the target behavior probability as a second risk behavior coverage value; and taking an interval formed by the first risk behavior coverage value and the second risk behavior coverage value as the risk behavior coverage range of the target risk behavior label. By considering corresponding risk tendency value information of the risk behaviors when the risk behaviors are not triggered, and meanwhile, considering the condition that service risk generating behaviors exist in the risk behaviors to configure risk behavior subscription data; the accuracy of obtaining the risk behavior coverage of the target risk behavior label can be improved.

In one embodiment, the risk tendency value information includes a second reference value, and the step S151 may include: acquiring a product of a matching behavior characteristic between the payment behavior knowledge graph and risk behavior distribution information in the paired payment scene and a target behavior probability, and taking the product as a first risk behavior coverage value; obtaining the product of the ratio of the second behavior probability to the second total behavior probability and the target behavior probability as a second risk behavior coverage value; and taking an interval formed by the first risk behavior coverage value and the second risk behavior coverage value as the risk behavior coverage range of the target risk behavior label. By considering the corresponding risk tendency value information of the risk behaviors during triggering, and simultaneously, considering the condition that service risk generating behaviors exist in the risk behaviors to configure risk behavior subscription data; the accuracy of obtaining the risk behavior coverage of the target risk behavior label can be improved.

In one embodiment, the target payment scenario includes multiple risk behavior tags, the target risk behavior tag belongs to the multiple risk behavior tags, and each risk behavior tag corresponds to a risk behavior coverage value range; the above step S152 may be implemented by the following exemplary embodiments.

(1) And respectively acquiring a risk behavior coverage value from the risk behavior coverage value range corresponding to each risk behavior label as a target risk behavior coverage value.

(2) And acquiring the sum of the target risk behavior coverage values, and determining the risk behavior coverage information of the target payment scene according to the sum of the target risk behavior coverage values.

The block chain service system 100 may obtain a risk behavior coverage value from a risk behavior coverage value range corresponding to each risk behavior tag, and use the risk behavior coverage value as a target risk behavior coverage value; if so, randomly acquiring a risk behavior coverage value from a risk behavior coverage value range corresponding to each risk behavior label as a target risk behavior coverage value; or, the risk behavior coverage value may be obtained from the risk behavior coverage value range corresponding to each risk behavior tag according to the risk behavior coverage value interval, and the obtained risk behavior coverage value is used as the target risk behavior coverage value, for example, the risk behavior coverage value interval is 5, that is, the 5 th risk behavior coverage value in the risk behavior coverage value range corresponding to each risk behavior tag may be used as the target risk behavior coverage value. The target risk behavior coverage value is used for reflecting a risk behavior statistic value of a corresponding risk behavior label in a target payment scene; therefore, the sum of the target risk behavior coverage values can be obtained, and the risk behavior coverage information of the target payment scene is determined according to the sum of the target risk behavior coverage values.

In one embodiment, in (2) of the step S152, the following exemplary embodiments may be implemented.

(1) And generating a first risk behavior coverage threshold according to the target behavior probability.

(2) Generating a second risk behavior coverage threshold according to the target behavior probability and the second behavior probability; the first risk behavior coverage threshold is greater than the second risk behavior coverage threshold.

(3) And if the sum of the target risk behavior coverage values is smaller than the first risk behavior coverage threshold and is greater than or equal to the second risk behavior coverage threshold, taking the sum of the target risk behavior coverage values as risk behavior coverage information of the target payment scene.

(4) If the sum of the target risk behavior coverage values is greater than or equal to the first risk behavior coverage threshold or less than the second risk behavior coverage threshold, respectively obtaining the risk behavior coverage values from the risk behavior coverage value ranges corresponding to each risk behavior label, using the risk behavior coverage values as updated target risk behavior coverage values, obtaining the sum of the updated target risk behavior coverage values, and determining the risk behavior coverage information of the target payment scene according to the sum of the updated target risk behavior coverage values.

The block chain service system 100 may use the target behavior probability as a first risk behavior coverage threshold, and obtain a risk behavior probability of each paired payment behavior information in the paired payment scenario when the risk behavior is triggered, as a second behavior probability; obtaining the maximum behavior probability from the second behavior probability; and taking the ratio of the target behavior probability to the maximum behavior probability as a second risk behavior coverage threshold. The target behavior probability comprises the risk behavior probability of a service risk generation behavior in the risk behavior, namely the target behavior probability is larger than the actual risk behavior distribution information in the target payment scene; the ratio between the target behavior probability and the maximum behavior probability may be used to reflect: and when the risk behavior probability of the risk behavior of each risk behavior label in the target payment scene is the maximum behavior probability, obtaining the minimum risk behavior statistic value in the target payment scene. The actual risk behavior coverage information in the target payment scene is smaller than a first risk behavior coverage threshold and is larger than or equal to a second risk behavior coverage threshold; therefore, if the sum of the target risk behavior coverage values is smaller than the first risk behavior coverage threshold and greater than or equal to the second risk behavior coverage threshold, which indicates that the sum of the target risk behavior coverage values is within a reasonable range, the sum of the target risk behavior coverage values is used as the risk behavior coverage information of the target payment scenario. If the sum of the target risk behavior coverage values is greater than or equal to the first risk behavior coverage threshold or less than the second risk behavior coverage threshold, which indicates that the sum of the target risk behavior coverage values exceeds a reasonable range, respectively obtaining the risk behavior coverage values from the risk behavior coverage value ranges corresponding to each risk behavior label, as updated target risk behavior coverage values, obtaining the sum of the updated target risk behavior coverage values, and determining the risk behavior coverage information of the target payment scene according to the sum of the updated target risk behavior coverage values.

In an embodiment, the foregoing step S110 can be implemented by the following steps.

Step A110, acquiring block chain safety big data generated by paying on a block chain line; the blockchain security big data is used for representing a payment security verification event between blockchain payment services, and the blockchain payment services are payment services related to a target security protection operating program.

In one embodiment, the payment service data may be data to be analyzed, wherein the association between the payment services is represented by a knowledge graph, payment items in the payment service data correspond to the payment services, and payment security verification events between the payment services correspond to the payment security verification events between the payment services; for example, "payment service a" is one payment service in the payment service data, "payment service B" is another payment service in the payment service data, and the payment verification process authentication is performed between "payment service a" and "payment service B" in the payment service data through "payment verification process S", and "payment service a" - "payment verification process S" - "payment service B" constitutes one payment security verification event.

In one embodiment, the blockchain service system 100 may first obtain blockchain security big data generated by online payment of blockchains, where the blockchain security big data can represent a payment security verification event between blockchain payment services, where the blockchain payment services are payment services related to a target security protection operating program. Considering that payment service data constructed based on all information in a network is usually very huge, a lot of calculation amount is consumed for updating a protection association rule of a safety protection operation unit on a target safety protection operation program based on the payment service data, and a lot of information irrelevant to the target safety protection operation program exists in the network; based on this, the manner provided by the embodiment of the present disclosure starts with big blockchain security data used for characterizing a payment security verification event between blockchain payment services related to a target security protection operating program, and updates a protection association rule of a security protection operating unit on the target security protection operating program.

In one embodiment, in practical applications, the blockchain service system 100 can autonomously extract blockchain security big data from initial payment service data, where the initial payment service data is payment service data constructed based on all information in the network; the blockchain security big data can also be directly acquired from other devices, and the disclosure does not make any limitation on the implementation manner of acquiring the blockchain security big data generated by paying on the blockchain line by the blockchain service system 100.

An implementation of the blockchain service system 100 extracting big blockchain security data from the initial payment service data is described below.

The blockchain service system 100 may select a payment service satisfying a predetermined condition from the initial payment service data as a blockchain payment service, where the predetermined condition includes at least one of the following: the payment service type is a preset type, and the online rate of the payment service exceeds a preset online rate threshold value; and then, determining the big block chain security data according to the payment security verification event of the selected block chain payment service in the initial payment service data.

In one embodiment, the initial payment service data may be comprised of a number of payment security verification events having payment security verification events, each payment security verification event comprised of an initial payment service, a payment service relationship, and an ending payment service. For example, assume that the initial payment service included in the payment security verification event is "payment service a", the payment service relationship is "payment verification process S", and the terminating payment service is "payment service B". Each payment service in the initial payment service data further includes a set of associated data corresponding thereto, and in one embodiment, the associated data included in each payment service includes, but is not limited to, a payment service type, a payment service name, a payment service online rate, and the like.

When the blockchain service system 100 extracts the big blockchain security data from the initial payment service data, a payment service meeting a preset condition may be selected from the initial payment service data as a blockchain payment service. In one embodiment, the blockchain service system 100 may select a payment service with a payment service type of a preset type from the initial payment service data as a blockchain payment service; in one embodiment, the blockchain service system 100 may also select a payment service with a payment service rate exceeding a preset online rate threshold from the initial payment service data as the blockchain payment service.

In an embodiment, in practical application, the blockchain service system 100 may screen the blockchain payment service based on only the payment service type, may also screen the blockchain payment service based on only the online rate of the payment service, may also screen the blockchain payment service based on both the payment service type and the online rate of the payment service, or the blockchain service system 100 may also screen the blockchain payment service based on other payment service related data, where the disclosure does not make any limitation on a preset condition according to when the blockchain payment service is selected from the initial payment service data.

After the block chain service system 100 selects the block chain payment service related to the target security protection operating program from the initial payment service data, the payment security verification event of each selected block chain payment service in the initial payment service data can be extracted, and further, based on each selected block chain payment service and the payment security verification event of each block chain payment service in the initial payment service data, big block chain security data suitable for updating the protection association rule of the security protection operating unit for the target security protection operating program is constructed.

Step A120, generating a block chain payment service sequence based on the block chain security big data; the blockchain payment service sequence is a sequence consisting of a plurality of the blockchain payment services with payment security verification events in the blockchain security big data.

After the blockchain service system 100 obtains the blockchain security big data, several blockchain payment service sequences may be generated based on the blockchain security big data. In one embodiment, the blockchain service system 100 may compose a sequence of blockchain payment services with a series of blockchain payment services having a payment security verification event therebetween based on a payment security verification event between blockchain payment services in the blockchain security big data.

Step A130, determining payment risk behavior information corresponding to the blockchain payment service in the blockchain security big data according to the blockchain payment service sequence, and determining a first risk behavior association degree between safety protection operation units on the target safety protection operation program according to a risk behavior association degree between the payment risk behavior information corresponding to the blockchain payment service in the blockchain security big data and mapping information between the blockchain payment service and the safety protection operation units on the target safety protection operation program.

After the blockchain service system 100 generates a plurality of blockchain payment service sequences based on the blockchain security big data, the payment risk behavior information corresponding to the blockchain payment service in the blockchain payment service sequence may be correspondingly determined based on each generated blockchain payment service sequence, and thus, the payment risk behavior information corresponding to each blockchain payment service in the blockchain security big data is determined through traversal in the above manner. Since the payment risk behavior information corresponding to the above blockchain payment service is determined by the blockchain service system 100 based on the blockchain payment service sequence including the blockchain payment service, the payment risk behavior information corresponding to the blockchain payment service can reflect the relevance between the blockchain payment service and other blockchain payment services to a certain extent.

A specific implementation manner of determining the payment risk behavior information corresponding to the blockchain payment service is described below.

The blockchain service system 100 may first perform unique hot code feature extraction, such as unique hot coding, on each blockchain payment service in the blockchain security big data, so as to obtain unique hot code feature information corresponding to each blockchain payment service; then, training a payment risk behavior extraction network based on the unique hot coded feature information corresponding to the block chain payment service in the block chain payment service sequence, and continuously adjusting an embedded word vector (payment risk behavior prediction information) of the block chain payment service in the process of training the payment risk behavior extraction network; and finally, taking the payment risk behavior prediction information of the block chain payment service after the training of the payment risk behavior extraction network as the payment risk behavior information corresponding to the block chain payment service.

After the block chain service system 100 calculates and obtains the payment risk behavior information corresponding to each target vector in the block chain security big data, the block chain payment services in the block chain security big data can be combined pairwise, and the risk behavior association degree between the payment risk behavior information corresponding to each two block chain payment services is calculated. Furthermore, the blockchain service system 100 may obtain mapping information between the blockchain payment service and the security protection operating units on the target security protection operating program, convert the risk behavior association degree between the payment risk behavior information corresponding to the blockchain payment service into the risk behavior association degree between the security protection operating units on the target security protection operating program, and record the risk behavior association degree between the security protection operating units thus determined as the first risk behavior association degree between the security protection operating units.

In an embodiment, in practical application, if two blockchain payment services are mapped to the same security protection operating unit, when determining the first risk behavior association degree between the security protection operating units, the risk behavior association degree between the payment risk behavior information corresponding to each of the two blockchain payment services may not be considered. In other words, the first risk behavior association degree between the safety protection operating units is substantially determined based on the risk behavior association degree between the payment risk behavior information corresponding to the blockchain payment services respectively mapped to different safety protection operating units.

A specific implementation manner corresponding to the first risk behavior association degree between the safety protection operating units on the target safety protection operating program is determined as follows.

The blockchain service system 100 may first determine, according to the payment risk behavior information corresponding to each blockchain payment service in the blockchain security big data, first risk information (which may be distributed in a form of a matrix) corresponding to a first risk behavior association degree, where each risk node in the first risk information corresponding to the first risk behavior association degree is used to represent a risk behavior association degree between a blockchain payment service corresponding to a first risk dimension (e.g., a matrix row) where the risk node is located and a blockchain payment service corresponding to a second risk dimension (e.g., a matrix column) where the risk node is located. Furthermore, the blockchain service system 100 may convert, according to mapping information between each blockchain payment service and each safety protection operating unit on the target safety protection operating program, the first risk information corresponding to the first risk behavior association degree into second risk information corresponding to a second risk behavior association degree, where each risk node in the second risk information corresponding to the second risk behavior association degree is used to represent a risk behavior association degree between a safety protection operating unit corresponding to a first risk dimension where the risk node is located and a safety protection operating unit corresponding to a second risk dimension where the risk node is located.

In an embodiment, after determining the payment risk behavior information corresponding to each blockchain payment service in the blockchain security big data, the blockchain service system 100 may combine every two blockchain payment services in the blockchain security big data to obtain a plurality of blockchain payment service pairs; then, calculating cosine risk behavior association degree between payment risk behavior information corresponding to two block chain payment services for each block chain payment service pair to serve as the risk behavior association degree corresponding to the block chain payment service pair; furthermore, first risk information corresponding to the first risk behavior association degree is constructed on the basis of the risk behavior association degree corresponding to each block chain payment service, the first risk information corresponding to the first risk behavior association degree takes each block chain payment service as a row, and takes each block chain payment service as a second risk dimension, and risk nodes located in the x-th first risk dimension and the y-th second risk dimension in the first risk information corresponding to the first risk behavior association degree are actually cosine risk behavior association degrees between the payment risk behavior information of the block chain payment service corresponding to the x-th first risk dimension and the payment risk behavior information of the block chain payment service corresponding to the y-th second risk dimension.

In an embodiment, in practical application, the block chain service system 100 may construct the first risk information corresponding to the first risk behavior association degree by using the cosine risk behavior association degree between the payment risk behavior information, and may also construct the first risk information corresponding to the first risk behavior association degree by using the risk behavior association degree between the payment risk behavior information calculated based on other algorithms.

After the blockchain service system 100 constructs the first risk information corresponding to the first risk behavior association degree, the first risk information corresponding to the first risk behavior association degree may be converted into the second risk information corresponding to the second risk behavior association degree corresponding to the risk behavior association degree between the safety protection operation units for representing according to mapping information between each blockchain payment service and each safety protection operation unit on the target safety protection operation program. In an embodiment, when the blockchain service system 100 converts the second risk information corresponding to the second risk behavior association degree, for the risk behavior association degree between the payment risk behavior information corresponding to the two blockchain payment services mapped to the same safety protection operation unit, the risk behavior association degree may be discarded without adding the second risk information corresponding to the second risk behavior association degree.

It should be noted that, in practical applications, in order to more accurately update the protection association rule of the security protection operating unit on the target security protection operating program based on the risk behavior association degree between the security protection operating units on the target security protection operating program, the method provided in the embodiment of the present disclosure may determine the risk behavior association degree between the security protection operating units on the target security protection operating program from the dimension of the payment service data, and may also determine the risk behavior association degree between the security protection operating units on the target security protection operating program from the protection association rule of the existing security protection operating unit on the target security protection operating program.

A specific implementation manner corresponding to the second risk behavior association degree between the safety protection operating units on the target safety protection operating program is determined as follows.

The blockchain service system 100 may first construct risk information of a protection association rule of an initial security protection operation unit according to a protection association rule of a security protection operation unit on a target security protection operation program, where each risk node in the risk information of the protection association rule of the initial security protection operation unit is used to represent an influence weight of a risk subject behavior corresponding to a first risk dimension where the risk node is located on a security protection operation unit corresponding to a second risk dimension where the risk node is located. Furthermore, the blockchain service system 100 may train a target feature vector based on the risk information of the protection association rule of the starting safety protection operation unit, and use the trained target feature vector as third risk information corresponding to a third risk behavior association degree, where each risk node in the third risk information corresponding to the third risk behavior association degree is used to represent the risk behavior association degree between a label corresponding to the first risk dimension where the risk node is located and a safety protection operation unit corresponding to the second risk dimension where the risk node is located.

Step A140, based on the first risk behavior association degree, updating a protection association rule of a safety protection operation unit on the target safety protection operation program, performing key payment safety data collection based on the updated protection association rule of the safety protection operation unit, and performing safety portrait generation on the key payment safety data based on the safety protection operation rule corresponding to the safety protection operation program.

After determining the first risk behavior association degree between the safety protection operating units on the target safety protection operating program, the block chain service system 100 may update the protection association rule of the safety protection operating unit on the target safety protection operating program based on the first risk behavior association degree between the safety protection operating units. The idea is that based on the first risk behavior association degree between the safety protection operation units, the safety protection operation unit which is similar to the original safety protection operation unit is updated in the protection association rule of the safety protection operation unit.

In a case that the first risk behavior association degree between the safety protection operating units on the target safety protection operating program represents second risk information corresponding to the second risk behavior association degree, the blockchain service system 100 may update the protection association rule of the safety protection operating unit on the target safety protection operating program in the following manner:

determining risk information for updating the protection association rule of the safety protection operation unit according to second risk information corresponding to the second risk behavior association degree and risk information of the protection association rule of the starting safety protection operation unit; the risk information of the protection association rule of the initial safety protection operation unit is constructed according to the protection association rule of the existing safety protection operation unit on the target safety protection operation program, and each risk node in the risk information of the protection association rule of the initial safety protection operation unit and the risk information of the protection association rule of the updated safety protection operation unit is used for representing the influence weight of the risk subject behavior corresponding to the first risk dimension where the risk node is located on the safety protection operation unit corresponding to the second risk dimension where the risk node is located.

If the blockchain service system 100 determines the second risk behavior association degree between the safety protection operating units on the target safety protection operating program according to the protection association rule of the safety protection operating unit on the target safety protection operating program, the blockchain service system 100 may update the protection association rule of the safety protection operating unit on the target safety protection operating program based on the first risk behavior association degree and the second risk behavior association degree between the safety protection operating units at this time.

In an embodiment, the block chain service system 100 may fuse the first risk behavior association degree and the second risk behavior association degree between the safety protection operation units on the target safety protection operation program to obtain the target risk behavior association degree between the safety protection operation units, which is matched with the service mining target on the target safety protection operation program and is matched with the relationship between the payment services in the block chain safety big data, and further, the block chain service system 100 may update the protection association rule of the safety protection operation units on the target safety protection operation program according to the target risk behavior association degree between the safety protection operation units.

Under the condition that the first risk behavior association degree between the safety protection operation units on the target safety protection operation program represents second risk information corresponding to the second risk behavior association degree, and the second risk behavior association degree between the safety protection operation units represents third risk information corresponding to the third risk behavior association degree, the block chain service system 100 may update the protection association rule of the safety protection operation unit on the target safety protection operation program in the following manner:

fusing second risk information corresponding to the second risk behavior association degree and third risk information corresponding to the third risk behavior association degree to obtain target risk information; and determining the risk information for updating the protection association rule of the safety protection operation unit according to the target risk information and the risk information of the protection association rule of the starting safety protection operation unit.

In an embodiment, the blockchain service system 100 may perform weighted summation processing on the second risk information L corresponding to the second risk behavior association degree and the third risk information W corresponding to the third risk behavior association degree according to a preset weight; for example, assuming that the blockchain service system 100 sets a weight value x1 for the second risk information L corresponding to the second risk behavior association degree and sets a weight value x2 for the third risk information W corresponding to the third risk behavior association degree, the blockchain service system 100 may calculate the target risk information Q by the following formula:

Q＝P*x1+W*x2

in one embodiment, the weighted values x1 and x2 are determined by the blockchain service system 100 according to the attention degrees of the second risk information corresponding to the second risk behavior association degree and the third risk information corresponding to the third risk behavior association degree, if the protection association rule of the safety protection operation unit is updated, the risk behavior association degree between the safety protection operation units determined based on the payment safety verification events among the payment services in the blockchain safety big data is concerned more, x1 may be set to be greater than x2, if the risk behavior association degree between the safety protection operation units determined based on the protection association rule of the safety protection operation unit on the target safety protection operation program is more concerned when updating the protection association rule of the safety protection operation unit, x2 may be set to be greater than x1, and the weight values x1 and x2 set are not specifically limited by this disclosure.

Further, the blockchain service system 100 may multiply the target risk information Q by the risk information V of the protection association rule of the initial security protection operation unit to obtain the risk information V' of the protection association rule of the updated security protection operation unit.

It should be noted that, in practical application, after determining, by the blockchain service system 100, the risk information of the protection association rule of the updated safety protection operation unit in any one of the manners described above, the block chain service system 100 needs to update the protection association rule of the existing safety protection operation unit on the target safety protection operation program based on the risk information of the protection association rule of the updated safety protection operation unit. In order to enable the blockchain service system 100 to update the protection association rule of the existing safety protection operation unit more conveniently based on the risk information of the protection association rule of the updated safety protection operation unit, after the blockchain service system 100 generates the risk information of the protection association rule of the updated safety protection operation unit, the generated risk information of the protection association rule of the updated safety protection operation unit may be updated in the following implementation manner.

In a possible implementation manner, the blockchain service system 100 may determine, for a risk node in the same dimension in the risk information of the protection association rule updating the security protection operating unit and the risk information of the protection association rule of the initial security protection operating unit, whether a risk node in the risk information of the protection association rule of the initial security protection operating unit, which is in the node, is greater than a first target value, and if so, may set the risk node in the risk information of the protection association rule updating the security protection operating unit, which is in the node, to 0.

In one embodiment, if it is determined that the risk generating action a has an operational tendency for the security protection operating unit 1 according to the risk information of the protection association rule of the starting security protection operating unit, it is also determined that the risk generating action a has an operational tendency for the security defending operating unit 1 according to the risk information for updating the defending association rule of the security defending operating unit, the blockchain service system 100 needs to set a risk node in the risk information for updating the protection association rule of the safeguard operating unit for characterizing the influence weight of the risk generating action a on the safeguard operating unit 1 to 0, thereby avoiding that when the risk information based on the protection association rule of the updated safety protection operation unit is updated to the risk generation behavior safety protection operation unit, and adding the original safety protection operation unit in the risk information of the protection association rule of the safety protection operation unit.

In general, in the risk information of the protection association rule of the initial security protection operation unit, if a certain risk node is not 0, the security protection operation unit corresponding to the second risk dimension where the risk node is located should be in the protection association rule of the security protection operation unit of the risk subject behavior corresponding to the first risk dimension where the risk node is located, in this case, the blockchain service system 100 needs to set the first target value to 0. Of course, in some cases, in the risk information of the protection association rule of the initial safety protection operation unit, only if a certain risk node is greater than the preset value a, the safety protection operation unit corresponding to the second risk dimension where the risk node is located will be in the protection association rule of the safety protection operation unit of the risk subject behavior corresponding to the first risk dimension where the risk node is located, and in this case, the blockchain service system 100 needs to set the first target value as a. The first target value is not specifically limited by the present disclosure.

In another possible implementation manner, the blockchain service system 100 may determine, for each risk node in the risk information of the protection association rule updating the safety protection operating unit, whether the risk node is less than or equal to the second target value, and if so, set the risk node in the risk information of the protection association rule updating the safety protection operating unit to 0.

In one embodiment, in order to ensure sparsity of the risk information of updating the protection association rule of the safety protection operation unit, the blockchain service system 100 needs to set a second target value between 0 and 1, and for the risk nodes in the risk information of updating the protection association rule of the safety protection operation unit, which are less than or equal to the second target value, the blockchain service system 100 needs to set 0 accordingly.

After the risk information of the protection association rule of the updated safety protection operation unit is processed through the two implementation manners, the safety protection operation unit corresponding to the second risk dimension of the non-zero item in each first risk dimension in the risk information of the protection association rule of the updated safety protection operation unit is actually the safety protection operation unit updated according to the risk subject behavior corresponding to the first risk dimension.

In an embodiment, in practical application, in addition to performing update processing on the risk information of the protection association rule of the updated security protection operating unit through the two implementation manners, other manners may also be used to perform update processing on the risk information of the protection association rule of the updated security protection operating unit.

The protection association rule updating method of the safety protection operation unit starts with payment service data covering a large number of payment services and relations among the payment services, determines the risk behavior association degree among the payment services in the payment service data, converts the risk behavior association degree among the payment services into the risk behavior association degree among the safety protection operation units according to mapping information between the payment services and the safety protection operation unit, and updates the protection association rule of the safety protection operation unit based on the risk behavior association degree among the safety protection operation units. Therefore, the protection association rule of the safety protection operation unit is quickly and accurately updated, and further, the method is favorable for generating a more accurate safety portrait for the key payment safety data collected according to the updated protection association rule of the safety protection operation unit based on the safety protection operation program.

Fig. 3 is a schematic functional block diagram of an information prediction apparatus 300 based on big data for blockchain security according to an embodiment of the present disclosure, and the functions of the functional blocks of the information prediction apparatus 300 based on big data for blockchain security are described in detail below.

The first obtaining module 310 is configured to obtain key payment security data of a target payment scene found based on blockchain security big data generated by online payment of blockchains, where the blockchain security big data is used to represent a payment security verification event between blockchain payment services, and the blockchain payment services are payment services related to a target security protection operating program.

The second obtaining module 320 is configured to obtain, according to the pre-trained AI network corresponding to the target security protection operating program, paired payment behavior information in a paired payment scenario associated with the target payment scenario from the key payment security data, and obtain a payment behavior knowledge graph corresponding to the paired payment behavior information.

The determining module 330 is configured to determine, according to the payment behavior knowledge graph, a payment prediction flow direction feature of the paired payment behavior information in the paired payment scenario.

The first prediction module 340 is configured to predict, according to the paired payment behavior information and the payment behavior knowledge graph, a risk behavior probability of a risk behavior of the paired payment behavior information, as a prediction probability, and determine, according to the prediction probability, risk tendency value information of the risk behavior of the paired payment behavior information in a paired payment scenario.

The second prediction module 350 is configured to obtain a risk behavior probability of a risk behavior in the target payment scene as a target behavior probability, and predict risk behavior coverage information of the target payment scene according to the payment prediction flow direction characteristic, the risk tendency value information, and the target behavior probability.

Fig. 4 is a schematic diagram illustrating a hardware structure of a blockchain service system 100 for implementing the information prediction method based on blockchain security big data according to an embodiment of the present disclosure, and as shown in fig. 4, the blockchain service system 100 may include a processor 110, a machine-readable storage medium 120, a bus 130, and a transceiver 140.

In a specific implementation process, at least one processor 110 executes computer-executable instructions stored in the machine-readable storage medium 120, so that the processor 110 may execute the information prediction method based on the blockchain security big data according to the above method embodiment, the processor 110, the machine-readable storage medium 120, and the transceiver 140 are connected through the bus 130, and the processor 110 may be configured to control the transceiving action of the transceiver 140, so as to perform data transceiving with the aforementioned blockchain link point terminal 200.

For a specific implementation process of the processor 110, reference may be made to the various method embodiments executed by the block chain service system 100, which have similar implementation principles and technical effects, and further description of the embodiments is omitted here.

In addition, the embodiment of the disclosure also provides a readable storage medium, where the readable storage medium is preset with a computer execution instruction, and when a processor executes the computer execution instruction, the information prediction method based on the big block chain security data is implemented as above.

Finally, it should be understood that the examples in this specification are only intended to illustrate the principles of the examples in this specification. Other variations are also possible within the scope of this description. Thus, by way of example, and not limitation, alternative configurations of the embodiments of the specification can be considered consistent with the teachings of the specification. Accordingly, the embodiments of the present description are not limited to only those embodiments explicitly described and depicted herein.

Claims

1. An information prediction method based on block chain security big data is applied to a block chain service system, wherein the block chain service system is in communication connection with a plurality of block chain node terminals, and the method comprises the following steps:

2. The method for predicting information based on big block chain security data according to claim 1, wherein the determining payment prediction flow direction characteristics of the paired payment behavior information in the paired payment scenario according to the payment behavior knowledge graph comprises:

acquiring risk behavior distribution information in the paired payment scene;

and acquiring matching behavior characteristics between the payment behavior knowledge graph and risk behavior distribution information in the paired payment scenes as payment prediction flow direction characteristics of the paired payment behavior information in the paired payment scenes.

3. The information prediction method based on big block chain security data according to claim 2, wherein the step of predicting risk behavior probability of risk behavior of the paired payment behavior information according to the paired payment behavior information and the payment behavior knowledge graph, and determining risk tendency value information of risk behavior of the paired payment behavior information in the paired payment scenario according to the predicted probability comprises:

predicting and determining the risk behavior probability of the risk behavior with the paired payment behavior information when the risk behavior is not triggered according to the paired payment behavior information and the payment behavior knowledge graph, wherein the risk behavior probability is used as a first behavior probability;

taking the ratio of the first action probability to the first total action probability as a risk tendency reference value of the risk action of the paired payment action information when the risk action is not triggered as a first reference value; the first total behavior probability is the total behavior probability of the risk behaviors in the paired payment scene when the risk behaviors are not triggered;

predicting and determining the risk behavior probability of the risk behavior with the paired payment behavior information when the risk behavior is triggered according to the paired payment behavior information and the payment behavior knowledge graph, wherein the risk behavior probability is used as a second behavior probability; the first behavior probability and the second behavior probability both belong to the prediction probability;

generating a risk tendency reference value of the risk behavior of the paired payment behavior information when triggered by a ratio between the second behavior probability and a second total behavior probability, wherein the second total behavior probability is the total risk behavior probability of the risk behavior of the paired payment scene when triggered;

and generating risk tendency value information of the risk behaviors of the paired payment behavior information in the paired payment scene according to the first reference value and the second reference value.

4. The information prediction method based on block chain security big data as claimed in claim 3, wherein the step of predicting the risk behavior probability of the risk behavior determined to have the paired payment behavior information when the risk behavior is not triggered according to the paired payment behavior information and the payment behavior knowledge graph as the first behavior probability comprises:

determining authority limit information configured when the risk behaviors with the paired payment behavior information are not triggered as first authority limit information;

acquiring corresponding probabilities of the same authority limit quantity between the first authority limit information and the payment behavior knowledge graph, wherein the probabilities are used as the risk behavior probabilities of the risk behaviors with the paired payment behavior information when the risk behaviors are not triggered;

and taking the risk behavior probability of the risk behavior with the paired payment behavior information when the risk behavior is not triggered as the first behavior probability.

5. The method for predicting information based on big block chain security data according to claim 3, wherein the determining the risk behavior probability of the risk behavior with the paired payment behavior information at the triggering time according to the paired payment behavior information and the payment behavior knowledge graph as the second behavior probability comprises:

acquiring historical trigger times of the risk behaviors with the paired payment behavior information, and generating trigger weights according to the historical trigger times;

determining authority limit information configured when the risk behavior with the paired payment behavior information is triggered according to the first authority limit information and the trigger weight, wherein the authority limit information is used as second authority limit information;

acquiring corresponding probabilities of the same authority limit quantity between the second authority limit information and the payment behavior knowledge graph as risk behavior probabilities of risk behaviors with the paired payment behavior information when triggered;

and taking the risk behavior probability of the risk behavior with the paired payment behavior information at the triggering time as the second behavior probability.

6. The information prediction method based on big block chain safety data according to claim 3, wherein predicting risk behavior coverage information of the target payment scenario according to the payment prediction flow direction feature, the risk tendency value information and the target behavior probability comprises:

generating a risk behavior coverage range of a target risk behavior label according to the payment prediction flow direction characteristic, the risk tendency value information and the target behavior probability; the target risk behavior tag is a risk behavior tag which is the same as the paired payment behavior information in the target payment scene;

determining risk behavior coverage information of the target payment scene from the risk behavior coverage range of the target risk behavior tag;

wherein the generating of the risk behavior coverage of the target risk behavior label according to the payment prediction flow direction characteristics, the risk tendency value information and the target behavior probability includes:

acquiring a matching behavior characteristic between the payment behavior knowledge graph and risk behavior distribution information in the paired payment scene, a maximum ratio and a minimum ratio among a ratio between the first behavior probability and the first total behavior probability, and a ratio between the second behavior probability and the second total behavior probability;

obtaining the product of the maximum ratio and the target behavior probability as a first risk behavior coverage value;

obtaining the product of the minimum ratio and the target behavior probability as a second risk behavior coverage value;

and taking an interval formed by the first risk behavior coverage value and the second risk behavior coverage value as a risk behavior coverage range of the target risk behavior label.

7. The information prediction method based on big block chain security data according to claim 6, wherein the target payment scenario includes multiple risk behavior tags, the target risk behavior tag belongs to the multiple risk behavior tags, and each risk behavior tag corresponds to a risk behavior coverage value range;

the determining risk behavior coverage information of the target payment scene from the risk behavior coverage range of the target risk behavior tag includes:

respectively acquiring risk behavior coverage values from the risk behavior coverage value ranges corresponding to the risk behavior labels, and taking the risk behavior coverage values as target risk behavior coverage values;

acquiring the sum of the target risk behavior coverage values, and determining the risk behavior coverage information of the target payment scene according to the sum of the target risk behavior coverage values;

wherein the determining risk behavior coverage information of the target payment scenario according to the sum of the target risk behavior coverage values includes:

generating a first risk behavior coverage threshold according to the target behavior probability;

generating a second risk behavior coverage threshold according to the target behavior probability and the second behavior probability; the first risk behavior coverage threshold is greater than the second risk behavior coverage threshold;

if the sum of the target risk behavior coverage values is smaller than the first risk behavior coverage threshold and is larger than or equal to the second risk behavior coverage threshold, taking the sum of the target risk behavior coverage values as risk behavior coverage information of the target payment scene;

if the sum of the target risk behavior coverage values is greater than or equal to the first risk behavior coverage threshold or less than the second risk behavior coverage threshold, respectively obtaining the risk behavior coverage values again from the risk behavior coverage value range corresponding to each risk behavior label as updated target risk behavior coverage values, obtaining the sum of the updated target risk behavior coverage values, and determining the risk behavior coverage information of the target payment scene according to the sum of the updated target risk behavior coverage values.

8. The method for predicting information based on big block chain security data according to any one of claims 1 to 7, wherein the method further comprises:

acquiring historical payment prediction flow characteristics of a target risk behavior tag in a candidate payment scene of a candidate payment scene sequence in a historical time period;

obtaining historical payment prediction flow characteristics of the target risk behavior tag in the target payment scene in the historical time period;

and taking the candidate payment scene with the matched historical payment prediction flow direction characteristic in the target payment scene as a paired payment scene associated with the target payment scene, wherein the corresponding historical payment prediction flow direction characteristic in the candidate payment scene sequence is matched with the historical payment prediction flow direction characteristic in the target payment scene.

9. The information prediction method based on blockchain security big data according to any one of claims 1 to 8, wherein the step of obtaining the key payment security data of the target payment scene searched based on the blockchain security big data generated by block chain online payment comprises:

acquiring block chain security big data generated by online payment of a block chain, wherein the block chain security big data is used for representing a payment security verification event between block chain payment services, and the block chain payment services are payment services related to a target security protection operation program;

generating a blockchain payment service sequence based on the blockchain security big data, wherein the blockchain payment service sequence is a sequence formed by a plurality of blockchain payment services with payment security verification events in the blockchain security big data;

determining payment risk behavior information corresponding to the blockchain payment service in the blockchain security big data according to the blockchain payment service sequence, and determining a first risk behavior association degree between safety protection operation units on the target safety protection operation program according to risk behavior association degree between the payment risk behavior information corresponding to the blockchain payment service in the blockchain security big data and mapping information between the blockchain payment service and the safety protection operation units on the target safety protection operation program;

and updating the protection association rule of the safety protection operation unit on the target safety protection operation program based on the first risk behavior association degree, and collecting key payment safety data based on the updated protection association rule of the safety protection operation unit so as to generate a safety portrait of the key payment safety data based on a pre-trained AI network corresponding to the safety protection operation program.

10. A blockchain service system, comprising a processor, a machine-readable storage medium, and a network interface, wherein the machine-readable storage medium, the network interface, and the processor are connected through a bus system, the network interface is configured to be communicatively connected to at least one blockchain link point terminal, the machine-readable storage medium is configured to store a program, an instruction, or a code, and the processor is configured to execute the program, the instruction, or the code in the machine-readable storage medium to perform the information prediction method based on blockchain security big data according to any one of claims 1 to 9.