CN107423982B - Account-based service implementation method and device - Google Patents

Account-based service implementation method and device Download PDF

Info

Publication number
CN107423982B
CN107423982B CN201610348880.0A CN201610348880A CN107423982B CN 107423982 B CN107423982 B CN 107423982B CN 201610348880 A CN201610348880 A CN 201610348880A CN 107423982 B CN107423982 B CN 107423982B
Authority
CN
China
Prior art keywords
account
value
whiteness
seed
blackness
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201610348880.0A
Other languages
Chinese (zh)
Other versions
CN107423982A (en
Inventor
何慧梅
林瑞华
何帝君
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Advanced New Technologies Co Ltd
Advantageous New Technologies Co Ltd
Original Assignee
Advanced New Technologies Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Advanced New Technologies Co Ltd filed Critical Advanced New Technologies Co Ltd
Priority to CN201610348880.0A priority Critical patent/CN107423982B/en
Publication of CN107423982A publication Critical patent/CN107423982A/en
Application granted granted Critical
Publication of CN107423982B publication Critical patent/CN107423982B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q20/00—Payment architectures, schemes or protocols
    • G06Q20/38—Payment protocols; Details thereof
    • G06Q20/40—Authorisation, e.g. identification of payer or payee, verification of customer or shop credentials; Review and approval of payers, e.g. check credit lines or negative lists
    • G06Q20/401—Transaction verification
    • G06Q20/4016—Transaction verification involving fraud or risk level assessment in transaction processing

Abstract

The application provides a service implementation method based on an account, which comprises the following steps: acquiring all associated accounts in an account set and association coefficients among the associated accounts; any account in the account set is at least correlated with one other account; the account set includes at least one white seed account and at least one black seed account; determining an account whiteness value and an account blackness value of a non-seed account according to the account whiteness value and the account blackness value of at least one associated account of the non-seed accounts and the association coefficient of the associated account; generating an account gray value of a non-seed account by adopting an evidence synthesis method based on the account whiteness value and the account blackness value of the non-seed account; and determining the business strategy of the non-seed account according to the account gray value. When the technical scheme of the application is applied to a business system related to a user fund account, the verification operation of the user with high credit degree and good behavior can be reduced on the basis of not reducing the system security.

Description

Account-based service implementation method and device
Technical Field
The present application relates to the field of network communication technologies, and in particular, to a method and an apparatus for implementing a service based on an account.
Background
With the development of communication technology, people are becoming more and more accustomed to processing various work and life items using a network, and the processing of the items is generally performed by a user accessing a server of a service system providing a corresponding service through a terminal. In conducting these services, the user typically represents an account registered with a server of the service system as a representative of his identity, and the server runs the service logic associated with the user against the user's account.
In the prior art, most business systems adopt the same business logic for all user accounts and execute the same business process. For example, before executing a user instruction to change funds, to ensure that the user operates or is authorized by the user, and reduce the occurrence of malicious behaviors such as card stealing, account stealing, and marketing resource retaking, security verification is usually performed by using a mobile phone verification code, KYC (customer identity verification) problems, and the like, and the user account is also restricted or funds are frozen when an obvious abnormality occurs. These additional security checks are necessary for a small number of poorly behaving users, but add unnecessary operations and burden to the user for the majority of more confident, well behaving users.
Disclosure of Invention
In view of this, the present application provides an account-based service implementation method, including:
acquiring all associated accounts in an account set and association coefficients among the associated accounts; any account in the account set is at least correlated with one other account; the account set comprises at least one white seed account and at least one black seed account, the account whiteness value of the white seed account is a preset whiteness extreme value, and the account blackness value of the black seed account is a preset blackness extreme value;
determining an account whiteness value of a non-seed account according to an account whiteness value of at least one associated account of the non-seed accounts and an association coefficient of the associated account and the non-seed account based on a preset whiteness extreme value;
determining an account blackness value of a non-seed account according to an account blackness value of at least one associated account of the non-seed accounts and an association coefficient of the associated account and the non-seed account based on a preset blackness extreme value;
generating an account gray value of a non-seed account by adopting an evidence synthesis method based on the account whiteness value and the account blackness value of the non-seed account;
and determining the business strategy of the non-seed account according to the account gray value.
The application also provides a service implementation device based on the account, which includes:
the association relation obtaining unit is used for obtaining all association accounts in the account set and association coefficients among the association accounts; any account in the account set is at least correlated with one other account; the account set comprises at least one white seed account and at least one black seed account, the account whiteness value of the white seed account is a preset whiteness extreme value, and the account blackness value of the black seed account is a preset blackness extreme value;
the whiteness value determining unit is used for determining an account whiteness value of a non-seed account according to an account whiteness value of at least one associated account of the non-seed accounts and an association coefficient of the associated account and the non-seed account based on a preset whiteness extreme value;
the blackness value determining unit is used for determining an account blackness value of a non-seed account according to an account blackness value of at least one associated account of the non-seed accounts and an association coefficient of the associated account and the non-seed account based on a preset blackness extreme value;
the evidence synthesis unit is used for generating an account gray value of the non-seed account by adopting an evidence synthesis rule based on the account white value and the account black value of the non-seed account;
and the business strategy determining unit is used for determining the business strategy of the non-seed account according to the account gray value.
According to the technical scheme, in the embodiment of the application, the white seed account and the black seed account in the account set are respectively used as bases, the account whiteness value and the account blackness value of a non-seed account are determined according to at least one associated account of a certain non-seed account and the associated coefficient of the associated account or the associated accounts, and the account grey value used for determining the non-seed account business strategy is generated as two independent evidences by applying an evidence synthesis rule, so that the account quality degree is judged according to the user account relation, the business strategy matched with the account quality degree is applicable, and when the method is applied to a business system related to a user fund account, the verification operation of a user with high credit degree and good behavior can be reduced on the basis of not reducing the system safety.
Drawings
FIG. 1 is a flowchart of a method for implementing an account-based service according to an embodiment of the present application;
FIG. 2 is a schematic diagram of accounts, associations, and association coefficients for an example of a set of accounts of the present application;
FIG. 3 is a hardware block diagram of an apparatus in which embodiments of the present application operate;
fig. 4 is a logic structure diagram of an account-based service implementation apparatus in an embodiment of the present application.
Detailed Description
With the popularity of various instant messaging tools and social media in networks, almost everyone forms their own social circle on the network. Two people who are closely related in life tend to have more contact with the user account on the network. In real life, if most or all of a person's close relationships are well-behaved persons, then that person is typically also a well-behaved person; therefore, the network can also judge the quality degree of one account according to the quality degree of other accounts closely related to the account. Different business strategies can be applied to accounts with different degrees of quality so as to provide better service for user accounts with good behaviors.
Based on the above thought, embodiments of the present application provide a new account-based service implementation method, where a plurality of best accounts are used as white seed accounts in an account set, a plurality of worst accounts are used as black seed accounts, an account whiteness extremum is used as an account whiteness value of the white seed accounts, an account blackness extremum is used as an account blackness value of the black seed accounts, good and bad influences among the accounts are respectively reflected in the account whiteness values and the account blackness values of the non-seed accounts according to an association relationship, and then the good and bad degrees of the non-seed accounts are determined according to the account whiteness values and the account blackness values, so as to determine an applicable service policy, thereby implementing different service logics to the accounts with different good and bad degrees, and solving the problems existing in the prior art.
The embodiment of the application can be applied to any device with computing and storage capabilities, for example, a physical device or a logical device such as a mobile phone, a tablet Computer, a PC (Personal Computer), a notebook, a server, a virtual machine, and the like; the functions of the embodiments of the present application may be implemented by two or more physical or logical devices sharing different responsibilities and cooperating with each other.
In the embodiment of the present application, the accounts having an association relationship form an account set, in other words, in the account set of the embodiment of the present application, any account and at least one other account are associated accounts having an association relationship with each other, and the affinity between the two associated accounts is measured by an association coefficient. In practical application, all accounts meeting the above requirements in the business system may be used as elements of the account set, or accounts meeting the above requirements and meeting other business conditions may be used as elements of the account set according to the needs of the application scenario.
In the account set, at least one white seed account and at least one black seed account are pre-designated, wherein for specific services to which different service policies belong in the embodiment of the present application, the white seed accounts are the best user accounts, and the black seed accounts are the worst user accounts. The white seed account or the black seed account may be selected from the account set according to criteria for an account in a particular service. The other accounts in the account set that are not white seed accounts or black seed accounts are non-seed accounts.
For example, for fund-related businesses that wish to use different security verification policies, a user account with stronger fund strength and higher illicit cost may be selected as a white seed account, such as an account of a high-level manager, a famous enterprise employee, a high-credit group, a high-end bank customer, etc. of a listed company; and selecting a user account which is verified to have bad behaviors as a black seed sub-account, such as a user account which has bad records in the business system, a user account corresponding to a law court losing executant list or a public security bureau involved case list and the like.
In the embodiment of the application, the account whiteness value is adopted to reflect the performance of one account in the account set in the good aspect, the account blackness value is adopted to reflect the performance of one account in the bad aspect, and the account gray value is adopted to reflect the good or bad degree of one account after the performances of the good or bad aspects are comprehensively considered. In the account set, the account whiteness value of a white seed account can be preset to be a preset whiteness extreme value, and the account blackness value of a black seed account can be preset to be a preset blackness extreme value; in addition, the account gray value of the white seed account can be set to be the highest value in advance, and the account gray value of the black seed account can be set to be the lowest value in advance. In order to apply the evidence synthesis rule when calculating the account gray value of the non-seed account, the predetermined whiteness extreme value and the predetermined blackness extreme value may be respectively set to 1, if the account gray value represents that the account is better by a larger value, the account gray value of the white seed account may be set to 1, and the account gray value of the black seed account may be set to 0.
In the embodiment of the present application, a flow of an account-based service implementation method is shown in fig. 1.
Step 110, acquiring all the associated accounts in the account set and the association coefficients among the associated accounts.
According to the requirements of the actual application scene and the characteristics of the specific service, the association relationship can be established between the accounts of the service system according to certain conditions, and the association coefficient is given to reflect the strength of the association relationship. Wherein, the value range of the correlation coefficient is more than 0 and less than or equal to 1.
For example, according to the social networking behavior data of one user account, according to the number of interactions between the account and other accounts, the account with the number of interactions and the frequency of interactions, which both exceed the threshold value, may be used as its associated account, and the association coefficient may be assigned according to the number of interactions and the frequency. For another example, the social networking behavior may be divided into several categories, each category has a respective weight, the social networking behavior between one account and another account is divided according to the categories, and whether the two accounts are used as the associated accounts or not and the association coefficients of the two accounts are determined according to the product of the number of interactions or the frequency of interactions of each divided category and the weight of the category. For specific implementation, please refer to the prior art, which is not described in detail.
In some implementations, if two or more user accounts have the same key information, for example, the same bank card is bound, the user accounts are often used on the same terminal device, and the like, the same key information is virtualized into one device account, and an association relationship between the device account and the two or more user accounts is established. The association coefficient between the device account and the user account is usually set to a fixed value.
For a business system with a large number of accounts, the number of associated accounts owned by each account is limited to be not more than a preset maximum number n of associated accounts, namely only n accounts with the strongest association coefficient with a certain account are used as the associated accounts, so that the excessive calculation amount when calculating the account whiteness value and the account darkness value of non-seed accounts is caused by the excessive number of the associated accounts and the excessively complicated direct and indirect relationship between the accounts.
And 120, determining the account whiteness value of the non-seed account according to the account whiteness value of at least one associated account of the non-seed accounts and the association coefficient of the associated account and the non-seed account based on the white seed account and the preset whiteness extreme value.
And step 130, determining the account blackness value of the non-seed account according to the account blackness value of at least one associated account of the non-seed accounts and the association coefficient of the associated account and the non-seed account based on the black seed account and the preset blackness extreme value.
In the embodiment of the application, two associated accounts can mutually influence each other in a good aspect and a bad aspect, and if the account A and the account B are two associated accounts and the account B is a non-seed account, the influence of the account A on the account B in the good aspect is embodied by increasing the account whiteness value of the account B, and the influence degree is related to the account whiteness value of the account A and the association coefficient between the account A and the account B; similarly, the bad influence of account a on account B is reflected by increasing the account blackness value of account B, the degree of influence being related to the account blackness value of account a and the correlation coefficient between account a and account B; the effect of account B on non-seed account a is also true. Since the account whiteness value of the white seed account and the account blackness value of the black seed account have been set to extreme values, the influence of the associated account on them is no longer considered.
Each account in the account set has an association relationship with at least one other account, and the association relationship is regarded as a path, so that each non-seed account can be connected with each white seed account and each black seed account through at least one path from 0 to a plurality of accounts. Thus, the impact of a white seed account in good terms is reflected in the account whiteness values of all its associated accounts, centered on each white seed account; then spreading the influence of the associated accounts on good aspects to all the associated accounts; by repeating the steps, the non-seed account and the black seed account can know the influence degree of the associated accounts in good aspects and determine the account whiteness value according to the influence degree. A similar diffusion process, centered on each black seed account, enables the non-seed and white seed accounts to determine their account blackness values based on the degree of influence of the associated account in bad respects. That is, the account whiteness value of the non-seed account is determined according to the account whiteness value of at least one associated account of the non-seed account, the association coefficient of the associated account(s) with the non-seed account, based on the white seed account and the predetermined whiteness limit value; the account blackness value of the non-seed account is determined according to the account blackness value of at least one associated account of the non-seed accounts and the association coefficient of the associated account(s) and the non-seed account on the basis of the black seed account and the preset blackness limit value.
The specific calculation process for obtaining the account whiteness value and the account blackness value of the non-seed account can be selected according to factors such as the number of accounts in the account set in the actual application scene, the requirement on the operation speed and the like, and the embodiment of the application is not limited.
In one implementation, the account whiteness value and the account blackness value of the non-seed account are obtained by a whiteness iteration process and a blackness iteration process, respectively. Specifically, in the whiteness iteration process affected by the good diffusion aspect, iteration is started after each white seed sub-account is used as a first round of diffusion starting point, in each iteration round, all associated accounts of each diffusion starting point are used as diffusion end points subordinate to the diffusion starting point, and if the diffusion end points are the white seed sub-accounts, no processing is performed; if the diffusion end point is a non-seed account or a black-seed account, determining the account whiteness value of the account according to the influence degree of each diffusion start point to which the diffusion end point belongs on the good aspect of the account (which can be determined by the account whiteness value of the diffusion start point and the correlation coefficient between the diffusion start point and the account); if the iteration result of the current round meets the preset whiteness iteration stopping condition, stopping iteration, and determining account whiteness values of all non-seed accounts and black seed accounts; if the iteration result of the current round does not meet the preset whiteness iteration stop condition, taking all the diffusion starting points and the diffusion end points of the current round as the diffusion starting points of the next round, and entering the next round of whiteness iteration;
in the blackness iteration process for influencing the bad aspects of diffusion, each black seed sub-account is used as a diffusion starting point of the first round, then iteration is started, in each round of iteration, all the associated accounts of each diffusion starting point are used as diffusion end points subordinate to the diffusion starting point, and if the diffusion end points are the black seed sub-accounts, no processing is performed; if the diffusion end point is a non-seed account or a white-seed account, determining the account blackness value of the account according to the influence degree of each diffusion start point to which the diffusion end point belongs on the bad aspect of the account (which can be determined by the account blackness value of the diffusion start point and the correlation coefficient between the diffusion start point and the account); if the iteration result of the current round meets the preset blackness iteration stop condition, stopping iteration, and determining account blackness values of all non-seed accounts and white seed accounts; and if the iteration result of the current round does not meet the preset blackness iteration stop condition, taking all the diffusion starting points and the diffusion end points of the current round as the diffusion starting points of the next round, and entering the next round of blackness iteration.
When the whiteness iteration is carried out for enough times, the account whiteness values of the non-seed accounts and the black seed accounts tend to be more and more stable, and the same is true for the blackness iteration. If the account whiteness value and the account blackness value which are as accurate as possible are required to be obtained, the preset whiteness iteration stop condition can be that the change proportion of the sum of the account whiteness values of all the black seed sub-accounts in two adjacent iterations is smaller than a certain threshold value, and similarly, the preset blackness iteration stop condition can be that the change proportion of the sum of the account blackness values of all the white seed sub-accounts in two adjacent iterations is smaller than the threshold value.
However, in some cases, the above-mentioned iteration stop condition may cause the black seed account to have a higher account whiteness value and the white seed account to have a higher account blackness value, which usually means that the non-seed accounts also have higher or even higher account whiteness values and account blackness values, so that the account whiteness values and the account blackness values of all the accounts are concentrated in a small interval, which is not favorable for distinguishing the quality of the accounts. In order to reduce the possibility of occurrence of the situation, the number of black seed sub-accounts with the account brightness value larger than zero and the ratio of the number of the black seed sub-accounts to the total number of the black seed sub-accounts to reach the brightness iteration stop proportion are used as preset brightness iteration stop conditions, the number of white seed sub-accounts with the account brightness value larger than zero and the ratio of the number of the white seed sub-accounts to reach the black iteration stop proportion are used as preset black iteration stop conditions, so that the brightness iteration is stopped after the brightness iteration is diffused to a certain number of black seed sub-accounts, and the black iteration is stopped after the black iteration is diffused to another certain number of white seed sub-accounts, and the resolution of the quality of the non-seed accounts can be guaranteed in most cases.
In one example of such an implementation, each account may be modeled as a network node, and the association relationship between accounts may be modeled as a connection relationship between network nodes. In each round of whiteness iteration, each diffusion starting point sends a whiteness influence value to the associated account, the whiteness influence value is determined according to the account whiteness value of the diffusion starting point and the association coefficient between the diffusion starting point and the associated account receiving the whiteness influence value, the non-seed account or the black seed account receiving the whiteness influence value determines the account whiteness value of the account according to all the received whiteness influence values, and if the account whiteness value of the account exceeds a preset whiteness limit value, the preset whiteness limit value is used as the account whiteness value of the account. In each round of blackness iteration, each diffusion starting point sends a blackness influence value to the associated account, the blackness influence value is determined according to the account blackness value of the diffusion starting point and the association coefficient between the diffusion starting point and the associated account receiving the blackness influence value, the non-seed account or the white seed account receiving the blackness influence value determines the account blackness value of the account according to all the received blackness influence values, and if the account blackness value of the account exceeds a preset blackness extreme value, the preset blackness extreme value is used as the account blackness value of the account. In addition, in the whiteness iteration, the self-account whiteness value can be sent to the related account by the diffusion starting point, the non-seed account or the black seed account of the account whiteness value is received, the whiteness influence value of the diffusion starting point on the account is determined according to the received account whiteness value and the correlation coefficient between the received account whiteness value and the diffusion starting point, and the account whiteness value of the account is determined according to all the whiteness influence values in the round; similar in the blackness iteration, and will not be described in detail.
In order to facilitate the application of the evidence synthesis rule based on the account blackness value and the account whiteness value in the subsequent steps, the value ranges of the account blackness value and the account whiteness value can be limited between 0 and 1. And setting the preset whiteness extreme value and the preset blackness extreme value as 1, wherein the value range of the association coefficient is greater than 0 and less than or equal to 1, and the whiteness influence value or the blackness influence value of the associated account caused by one account is greater than 0 and less than or equal to 1. In the whiteness iteration, the non-seed account or the black seed account as the diffusion end point can normalize the sum of the whiteness influence values of the account from each diffusion start point to obtain the whiteness value of the account after the iteration. Similarly, in the blackness iteration, the non-seed account or the black seed account serving as the diffusion endpoint may be normalized with respect to the sum of the blackness influence values of the account from each diffusion endpoint to obtain the blackness value of the account after the iteration. The normalization processing can be performed based on the number of the associated accounts of the account; the normalization process may also be performed based on a predetermined maximum associated-account number for the set of accounts having the predetermined maximum associated-account number.
If the account set comprises the equipment account, the equipment account is used as a non-seed account to participate in whiteness iteration and blackness iteration, and the processing process of the equipment account in the iteration process can be simplified because the equipment account does not need to calculate the account gray value (the account gray value has no meaning to the equipment account). After the equipment account serving as the diffusion endpoint in the whiteness iteration receives the whiteness influence value of the diffusion starting point, the sum of all the received whiteness influence values can be used as the account whiteness value of the account, and if the account whiteness value of the account exceeds a preset whiteness extreme value, the preset whiteness extreme value is used as the account whiteness value of the account; after the device account serving as the diffusion endpoint in the blackness iteration receives the blackness influence value of the diffusion origin, the sum of all the received blackness influence values can be used as the account blackness value of the account, and if the account blackness value of the account exceeds the preset blackness extreme value, the preset blackness extreme value is used as the account blackness value of the account.
It should be noted that there is no timing relationship between step 120 and step 130.
And 140, generating an account gray value of the non-seed account by adopting an evidence synthesis method based on the account whiteness value and the account blackness value of the non-seed account.
The evidence theory is a method of uncertainty reasoning, and the core of the evidence theory is to use an evidence synthesis rule (i.e. Dempster synthesis rule, D-S synthesis rule) to synthesize two or more evidences which can be used for considering the same proposition, and then give the possibility of establishing the proposition.
And D is a universe set of all possible values of U, and all elements in D are mutually incompatible pairwise, so that D is the identification frame of U. Power set 2 of DDIs the set of all subsets of D. Identifying the Basic Probability Assignment (BPA) m on frame D to power set 2DEach element in (1) maps to [0,1 ]]A probability value of the interval, and satisfies formula 1:
Figure BDA0000998501140000101
Figure BDA0000998501140000102
is an empty set. For subset A of the recognition framework, such that m (A)>A of 0 is called a focal element (focuselement).
Each evidence gives a basic probability distribution on the recognition frame D, and two different and independent basic probability distributions m given by the two evidences1And m2Reliability function m of focal element A after combination of two evidences1⊕m2(A) This can be derived from equation 2:
Figure BDA0000998501140000103
power set 2 of B and CDOf (1). K is a normalization constant that can be used to measure the degree of conflict between different evidences, and is given by equation 3:
Figure BDA0000998501140000104
when K is 0, it indicates that the two evidences are completely consistent (completely compatible); 0 < K < 1 indicates that the two evidence portions are compatible; when K is 1, two evidences contradict each other. The closer K is to 1, the more serious the conflict between two evidences, and after the severity reaches a certain level, the evidence combination rule no longer applies.
Many evidence conflict handling methods are proposed in the prior art, such as: collision redistribution method, weighted average method, add-multiply synthesis method, evidence removal method, unknown perturbation method, synthesis rule under development knowledge framework, DSmT (Dezert-smart rapid theory) synthesis rule, etc. The evidence conflict processing methods adopt respective applicable evidence conflict conditions to judge whether evidence conflict occurs, and adopt respective processing algorithms to calculate evidence fusion results when the evidence conflict occurs.
In one example, when the normalization constant K in the evidence synthesis algorithm is greater than a predetermined conflict threshold KcAnd carrying out collision processing by adopting a weighted average method.
In the weighted average method, when evidence synthesis is performed, the degree of support of one evidence by other evidence is used as the weight of the evidence. That is, if an evidence is supported by other evidence, the evidence should be more credible, and a larger weight can be adopted to make the evidence have a larger influence on the evidence fusion result; if the evidence has a large conflict with other evidences and the confidence of the evidence is low, a low weight can be adopted to reduce the influence of the evidence on the evidence fusion result.
In weighted averaging methods usingThe distance between the basic probability assignments of two evidences represents the degree of collision between the two evidences, and the further the two basic probability assignments are, the more severe their collision. Distance d (m) between two basic probability assignments1,m2) Derived from equation 4:
Figure BDA0000998501140000111
in the formula 4, | | mi||2＝<mi,mi>，<m1,m2>Is the inner product of two vectors, as given by equation 5:
Figure BDA0000998501140000112
let the total number of evidences be M, the support degree of the evidence i (used to measure the degree that the evidence i is supported by other evidences) sup (M)i) Derived from equation 6:
Figure BDA0000998501140000113
the smaller the distance between the basic probability distributions of two evidences, the higher the similarity thereof; if one evidence is similar to the other evidence, the evidence is supported by the other evidence more.
Support degree sup (m) of evidencei) The evidence is normalized to obtain a confidence trm (m)i) As shown in formula 7:
Figure BDA0000998501140000121
trm(mi) Representing the confidence level of the evidence i and satisfying
Figure BDA0000998501140000122
Trm (m)i) As the weight of the evidence i, the weight is based on the formula 8The evidence conflict processing algorithm obtains an evidence fusion result of a weighted average method:
Figure BDA0000998501140000123
when other evidence conflict processing rules are adopted, please refer to the prior art for judging the condition of evidence conflict and calculating the processing algorithm of the evidence fusion result, and details are not repeated.
In an embodiment of the present application, the recognition framework D contains 2 elements, which can be expressed as { x, y }, where x is a good account, y is a bad account, and the power set is 2DIs composed of
Figure BDA0000998501140000124
The account whiteness value and the account blackness value of a non-seed account can be used as evidence respectively. Assuming that the account whiteness value is P after being normalized by a preset whiteness extreme valueg(normalization is not necessary if the predetermined whiteness threshold is 1), and the account blackness value is P after normalization by the predetermined blackness thresholdb(normalization is not necessary if the predetermined blackness limit is 1), X ═ X }, Y ═ Y }, then the base probability assignment m for evidence 1 is determined from the account whiteness value1Satisfies formula 9:
m1(X)＝Pg，m1(Y)＝1-Pgformula 9
Basic probability distribution m of evidence 2 determined according to account blackness value2Satisfies formula 10:
m2(X)＝1-Pb，m2(Y)＝Pbformula 10
In the embodiment of the application, the account gray value of the non-seed account is determined according to the evidence fusion result of the evidence 1 and the evidence 2. When the evidence 1 and the evidence 2 are synthesized by adopting an evidence synthesis rule, if the basic probability distribution of the evidence 1 and the evidence 2 meets a preset evidence conflict condition, calculating an evidence fusion result by adopting a preset evidence conflict processing method, and otherwise, calculating the evidence fusion result by adopting the evidence synthesis rule. The predetermined evidence conflict processing method may be any evidence conflict processing method used in an evidence synthesis rule, and the predetermined evidence conflict condition is a condition suitable for judging that an evidence conflict occurs in the adopted predetermined evidence conflict processing method.
Following is represented by K>KcAs the predetermined evidence conflict condition, a weighted average method is taken as an example of a predetermined evidence conflict processing method.
In the embodiment of the application, m can be1⊕m2(X) and/or m1⊕m2And (Y) as evidence fusion results, respectively representing the credibility of the non-seed account as a good account and the credibility of the non-seed account as a bad account after the account whiteness value and the account blackness value are comprehensively considered.
Equation 3 applies to the power set of the embodiments of the present application
Figure BDA0000998501140000131
Due to the fact that
Figure BDA0000998501140000132
The formula 11 can be obtained:
Figure BDA0000998501140000133
when K is>KcThen, a weighted average method is adopted to calculate an evidence fusion result m1⊕m2(X) or m1⊕m2(Y)。
Specifically, in the embodiments of the present application
Figure BDA0000998501140000134
Figure BDA0000998501140000135
<m1,m2>＝m1(X)*m2(X)+m1(Y)*m2(Y), equation 4 is applied to the basic probability distribution m of the embodiment of the present application1And m2From equation 12, two basic probability distributions m are calculated1And m2Distance d (m) therebetween1,m2)：
Figure BDA0000998501140000136
Assigning m according to two basic probabilities1And m2Distance d (m) therebetween1,m2) Determining support sup (m) for two evidences1) And sup (m)2). Applying equation 6 to the examples of the present application, equations 13 and 14 can be obtained:
Figure BDA0000998501140000137
Figure BDA0000998501140000138
normalizing the support degrees of the two evidences to obtain the credibility trm (m) of the two evidences1) And trm (m)2). Applying equation 7 to the examples of the present application, equations 15 and 16 can be obtained:
Figure BDA0000998501140000141
Figure BDA0000998501140000142
and taking the credibility of each evidence as the weight of the evidence, and obtaining an evidence fusion result by adopting an evidence conflict processing algorithm of a weighted average method. Applying equation 8 to the embodiments of the present application, the evidence conflict handling algorithm of the weighted average method shown in equations 17 and 18 respectively calculates m1⊕m2(X) and m1⊕m2(Y)：
Figure BDA0000998501140000143
Figure BDA0000998501140000144
When K is less than or equal to KcAnd calculating an evidence fusion result by adopting an evidence synthesis rule. Applying equation 2 to the examples of the present application, m is calculated from equations 19 and 20, respectively1⊕m2(X) and m1⊕m2(Y)：
Figure BDA0000998501140000145
Figure BDA0000998501140000146
Obtaining evidence fusion result m based on account whiteness value and account gray value1⊕m2(X) or m1⊕m2After (Y), the evidence fusion result may be directly used as the account gray value of the non-seed account, or a certain function using the evidence fusion result as an argument may be used as the account gray value of the non-seed account, which is not limited in the embodiment of the present application.
And 150, determining the business strategy of the non-seed account according to the account gray value.
As described above, in the embodiment of the present application, the white seed sub-account and the black seed sub-account are selected according to a certain service-to-account evaluation criterion. Therefore, the account quality degree reflected by the account gray value of the non-seed account is also obtained according to the judgment standard of the service. Therefore, the business strategy of a certain account can be determined according to the gray value of the account, so that different business strategies can be applied to accounts with different degrees of quality for the business. For example, for fund-related services, different user account security verification policies may be applied to the account gray scale value.
It can be seen that in the embodiment of the present application, the best accounts in the account set are used as white seed accounts, the worst accounts are used as black seed accounts, the white seed accounts and the black seed accounts are respectively used as a basis, the account whiteness value and the account blackness value of a non-seed account are determined according to one to all associated accounts of the non-seed account and the association coefficients of the associated accounts, and then the goodness of the non-seed account is determined by applying an evidence synthesis rule according to the account whiteness value and the account blackness value, so as to determine an applicable service policy, thereby implementing application of different service logics to accounts of different goodness.
In an application example of the application, a server of a third-party payment platform discovers and evaluates association relations among user accounts registered on the platform, specifically, establishes the association relations among the accounts according to interaction behaviors among the user accounts, and assigns values to association coefficients describing the association relations in a range of more than 0 and less than or equal to 1 according to the degree of closeness of relation among the user accounts; in addition, if two or more user accounts are bound to the same key information such as the mobile phone, the bank card and/or the identification card number, an equipment account is established to represent the same binding information, an association relationship is established between the equipment account and the user account bound with the information, and the association coefficient between the equipment account and the user account is a fixed value of 0.8.
After the association relationship and the association coefficient between all the accounts (including the device account) are found, 10 accounts (which may be device accounts) with the strongest association coefficient with each user account are selected as the association accounts of the user account (that is, the predetermined maximum association account number is 10), and the association relationship between other unselected accounts and the user account is deleted. The account set is formed with all user accounts and the associated account for each user account.
According to the factors of the asset condition of the account, the historical credit record and the like, the staff marks a plurality of user accounts in the account set which are best for fund-related services as white sub-accounts, and marks a plurality of user accounts which are worst for the fund-related services as black sub-accounts.
The server of the third party payment platform runs the following process:
1) and reading the association relation and the association coefficient among all the accounts in the generated account set.
2) According to the marks of the workers, the account whiteness value of the white seed sub-account is assigned to be 1, and the account blackness value of the black seed sub-account is assigned to be 1; assume that the account set includes 22 user accounts and 1 device account from account A, B, C through account V, where accounts A, B, C and D are white seed sub-accounts and account E is a black seed sub-account; the accounts are simulated as nodes in the network, the association relationship is simulated as the connection between the associated accounts, the association coefficient of the association relationship is marked on the connecting line, and the account set is shown in fig. 2.
3) And (3) performing whiteness iteration, wherein the preset whiteness iteration stopping conditions are as follows: the ratio of the number of the black seed sub-accounts with the account whiteness value larger than zero to the total number of the black seed sub-accounts reaches 0.1 percent.
Specifically, in the first round of the whiteness iteration, each white seed account (accounts A, B, C and D) is taken as a diffusion starting point, and the associated account (account M, N and device account) of each white seed account is taken as a diffusion ending point; each diffusion starting point sends whiteness influence values to all the associated accounts of the diffusion starting point, and the whiteness influence value sent to a certain associated account is the product of the account whiteness value of the diffusion starting point and the association coefficient between the diffusion starting point and the associated account receiving the whiteness influence value, namely: white seed account a sends 1 to its associated account M (account whiteness value 1 for white seed account a multiplied by the association coefficient between accounts a and M of 1.0), and white seed accounts B, C and D send 1, 0.8 to their respective associated account M, N and device account, respectively. Normalizing the sum of the received whiteness influence values by using a preset maximum associated account number 10 at a diffusion end point (account M, N) of each user account, taking the normalized value as an account whiteness value of the user account, and updating the account whiteness values of the accounts M and N to be 0.2 and 0.08 respectively; the equipment account takes the sum of the received whiteness influence values as the own account whiteness value, specifically 0.8 (the preset whiteness limit value is not exceeded 1). Since the account whiteness value for black seed account E is 0, the whiteness iteration continues, entering the second whiteness iteration round with all diffusion start and diffusion end points, i.e., account A, B, C, D, M, N and device accounts, as diffusion start points.
In the second round of whiteness iteration, similarly, all the associated accounts of each diffusion starting point are used as diffusion end points, and according to the first round of iteration mode, each diffusion starting point respectively sends the whiteness influence value of the diffusion starting point on the associated account to all the associated accounts, and the non-seed account or the black seed account which receives the whiteness influence value calculates the whiteness value of the account. It should be noted that the white seed account a receives 0.2 sent by the diffusion starting point M, and since the account whiteness value of the white seed account is already an extreme value, the received whiteness influence value is not processed; account M and account N become the diffusion start point and the diffusion end point at the same time in this round, account M sends 0.12 to account N, and account N sends 0.048 to account M. Because the account whiteness value of the black seed account E is still 0, the whiteness iteration continues until the fifth round, the account whiteness value of the black seed account E will be greater than 0, the whiteness iteration is terminated, and at this time, the account whiteness values of all accounts are determined. After the whiteness iteration is terminated, the account whiteness value of account M is 0.2215, and the account whiteness value of account F is 0.0016.
4) Carrying out blackness iteration, wherein the preset blackness iteration stop condition is as follows: the ratio of the number of the white seed sub-accounts with the account whiteness value larger than zero to the total number of the white seed sub-accounts reaches 0.1 percent.
The specific process of the blackness iteration is similar to the whiteness iteration, please refer to the description of the whiteness iteration, and is not repeated. And when the blackness iteration is terminated, determining account blackness values of all accounts. After the blackness iteration is terminated, the account blackness value of account M is 0.0001 and the account blackness value of account F is 0.0118.
5) According to the account whiteness value P of a certain non-seed accountgDetermining a basic probability distribution m for evidence 11(X)＝Pg，m1(Y)＝1-PgWherein m is1(X) degree of trustworthiness that the non-seed account is a good account in evidence 1, m1(Y) represents the degree of trustworthiness of the non-seed account in evidence 1 as a bad account. According to the account blackness value P of the non-seed accountbDetermining a basic probability distribution m for evidence 22(X)＝1-Pb，m2(Y)＝PbWherein m is2(X) degree of trustworthiness that the non-seed account is a good account in evidence 2, m2(Y) represents the degree of trustworthiness of the non-seed account in evidence 2 as a bad account.
Get KcThe normalization constant K of the evidence synthesis law was calculated using equation 11, 0.5.
When K is>At 0.5, the evidence fusion result m is calculated by adopting the formula 171⊕m2(X); when K is less than or equal to 0.5, m is calculated by adopting the formula 191⊕m2(X). M is to be1⊕m2(X) as the account gray value for the non-seed account.
For account M, Pg＝0.2215，Pb＝0.0001，K＝Pg*Pb+(1-Pg)×(1-Pb) 0.7784. Due to K>0.5，m1⊕m2(X)＝0.5*(Pg+1-Pb) 0.6107, account M has an account gray value of 0.6107.
To account F, Pg＝0.0016，Pb＝0.0118，K＝Pg*Pb+(1-Pg)×(1-Pb) 0.9866. Due to K>0.5，m1⊕m2(X)＝0.5*(Pg+1-Pb) 0.4999, account M has an account gray value of 0.4999.
6) After the server receives a fund operation request of a certain account, if the account gray value of the account is greater than or equal to 0.6 (such as an account M), the account is required to provide a password for security verification; if the account gray value of the account is greater than or equal to 0.5 and less than 0.6, the account is required to provide a password and a mobile phone verification code sent by the server for security verification; and if the account gray value of the account is less than 0.5 (such as account F), the account is required to provide a password, a mobile phone verification code sent by the server and an answer to the KYC question for security verification.
Corresponding to the above flow implementation, an embodiment of the present application further provides an account-based service implementation apparatus, which may be implemented by software, or may be implemented by hardware, or by a combination of hardware and software. Taking a software implementation as an example, the logical device is formed by reading a corresponding computer program instruction into a memory for running through a Central Processing Unit (CPU) of the device. In terms of hardware, in addition to the CPU, the memory, and the nonvolatile memory shown in fig. 3, the device in which the account-based service implementation apparatus is located generally includes other hardware such as a chip for performing wireless signal transmission and reception and/or other hardware such as a board for implementing a network communication function.
Fig. 4 shows an account-based service implementation apparatus provided in an embodiment of the present application, which includes an association obtaining unit, a whiteness value determining unit, a blackness value determining unit, an evidence synthesizing unit, and a service policy determining unit, where:
the association relation obtaining unit is used for obtaining all association accounts in the account set and association coefficients among the association accounts; any account in the account set is at least correlated with one other account; the account set comprises at least one white seed account and at least one black seed account, the account whiteness value of the white seed account is a preset whiteness extreme value, and the account blackness value of the black seed account is a preset blackness extreme value;
the whiteness value determination unit is used for determining an account whiteness value of a non-seed account according to an account whiteness value of at least one associated account of the non-seed accounts and an association coefficient of the associated account and the non-seed account based on a white seed account and a preset whiteness extreme value;
the blackness value determining unit is used for determining an account blackness value of a non-seed account according to an account blackness value of at least one associated account of the non-seed accounts and an association coefficient of the associated account and the non-seed account based on a black seed account and a preset blackness extreme value;
the evidence synthesis unit is used for generating an account gray value of the non-seed account by adopting an evidence synthesis rule based on the account white value and the account black value of the non-seed account;
and the business strategy determining unit is used for determining the business strategy of the non-seed account according to the account gray value.
In one implementation, the whiteness value determination unit is specifically configured to: taking each white seed sub-account as a diffusion starting point of a first round of whiteness iteration, taking an associated account of each diffusion starting point as a diffusion end point subordinate to the diffusion starting point in each round of whiteness iteration, and if the diffusion end point is not a white seed sub-account, determining an account whiteness value of the account according to an account whiteness value of each diffusion starting point and an associated coefficient of each diffusion starting point and the account; when the preset whiteness iteration stopping condition is not met, taking all diffusion starting points and diffusion end points of the current round as the diffusion starting points of the next round; the blackness value determination unit is specifically configured to: taking each black seed sub-account as a diffusion starting point of the first round of blackness iteration, taking the associated account of each diffusion starting point as a diffusion end point subordinate to the diffusion starting point in each round of blackness iteration, and if the diffusion end point is not a black seed sub-account, determining the account blackness value of the account according to the account blackness value of each diffusion starting point and the association coefficient of each diffusion starting point and the account; and when the preset blackness iteration stop condition is not met, taking all the diffusion starting points and the diffusion end points of the current round as the diffusion starting points of the next round.
In one example of the foregoing implementation, the determining unit, with the account associated with each diffusion start point as a diffusion end point subordinate to the diffusion start point, determines the account whiteness value of the account if the diffusion end point is not a white seed sub-account according to the account whiteness value of each diffusion start point and a correlation coefficient between each diffusion start point and the account, and includes: each diffusion starting point sends a whiteness influence value to the associated account, the whiteness influence value is determined according to the account whiteness value of the diffusion starting point and the association coefficient between the diffusion starting point and the associated account receiving the whiteness influence value, the non-seed account or the black seed account receiving the whiteness influence value determines the account whiteness value of the account according to all the received whiteness influence values, and if the account whiteness value of the account exceeds a preset whiteness extreme value, the preset whiteness extreme value is used as the account whiteness value of the account; the blackness value determination unit takes the account associated with each diffusion starting point as a diffusion end point subordinate to the diffusion starting point, and if the diffusion end point is not a black seed sub-account, the blackness value determination unit determines the account blackness value of the account according to the account blackness value of each diffusion starting point and the association coefficient of each diffusion starting point and the account, and the method comprises the following steps: each diffusion starting point sends a blackness influence value to the associated account, the blackness influence value is determined according to the account blackness value of the diffusion starting point and the association coefficient between the diffusion starting point and the associated account receiving the blackness influence value, the non-seed account or the white seed account receiving the blackness influence value determines the account blackness value of the account according to all the received blackness influence values, and if the account blackness value of the account exceeds a preset blackness extreme value, the preset blackness extreme value is used as the account blackness value of the account.
In the above example, the value ranges of the account whiteness value and the account blackness value are greater than or equal to 0 and less than or equal to 1, and the value range of the correlation coefficient is greater than 0 and less than or equal to 1; the determination unit of the whiteness value receives the non-seed account or black seed account of the whiteness influence value and determines the account whiteness value of the account according to all the received whiteness influence values, and the determination unit of the whiteness value comprises the following steps: receiving a non-seed account or a black seed account of the whiteness influence value, and carrying out normalization processing on the sum of all the received whiteness influence values according to the number of the associated accounts of the account or the preset maximum number of the associated accounts to obtain the account whiteness value of the account; the non-seed account or black seed account receiving the blackness influence value and determining the account blackness value of the account according to all the received blackness influence values by the blackness value determining unit comprises the following steps: receiving a non-seed account or a black seed account of the blackness influence value, and carrying out normalization processing on the sum of all the received blackness influence values according to the number of the associated accounts of the account or the preset maximum number of the associated accounts to obtain the account blackness value of the account; the predetermined maximum associated account number is the maximum number of associated accounts allowed to be owned by each account in the account set.
In the above example, the account set further includes: the device account is a non-seed account and is an account which is associated with at least two user accounts; the determination unit of the whiteness value receives the non-seed account or black seed account of the whiteness influence value and determines the account whiteness value of the account according to all the received whiteness influence values, and the determination unit of the whiteness value comprises the following steps: the equipment account receiving the whiteness influence value takes the sum of all the received whiteness influence values as the account whiteness value of the account, and if the account whiteness value of the account exceeds a preset whiteness extreme value, the preset whiteness extreme value is taken as the account whiteness value of the account; the non-seed account or the white seed account receiving the blackness influence value and determining the account blackness value of the account according to all the received blackness influence values by the blackness value determining unit comprises the following steps: and if the account blackness value of the account exceeds a preset blackness extreme value, the preset blackness extreme value is taken as the account blackness value of the account.
Optionally, the predetermined whiteness iteration stop condition includes: the ratio of the number of the black seed sub-accounts with the account whiteness value larger than zero to the total number of the black seed sub-accounts reaches the whiteness iteration stop ratio; the predetermined blackness iteration stop condition includes: and the ratio of the number of the white seed sub-accounts with the account blackness value larger than zero to the total number of the white seed sub-accounts reaches a blackness iteration stop ratio.
In one implementation, the evidence synthesis unit is specifically configured to: and determining basic probability distribution of two evidences according to the account whiteness value and the account blackness value of the non-seed account, if the two basic probability distribution meet a preset evidence conflict condition, calculating an evidence fusion result by adopting a preset evidence conflict processing method, otherwise, calculating an evidence fusion result by adopting an evidence synthesis rule, and generating the account grey value of the non-seed account according to the evidence fusion result.
In the foregoing implementation manner, the predetermined evidence conflict condition includes: the normalization constant K used for measuring the conflict degree between the evidences in the evidence synthesis rule is larger than a preset conflict threshold value; the evidence synthesizing unit calculates an evidence fusion result by adopting a preset evidence conflict processing rule, and comprises the following steps: and calculating the distance between the two basic probability distributions, determining the support degree of each evidence according to the distance between the two basic probability distributions, taking the credibility obtained after the support degree is normalized as the weight of the evidence, and obtaining an evidence fusion result by adopting an evidence conflict processing algorithm of a weighted average method.
Optionally, the service policy includes: security verification policy for user accounts.
The above description is only exemplary of the present application and should not be taken as limiting the present application, as any modification, equivalent replacement, or improvement made within the spirit and principle of the present application should be included in the scope of protection of the present application.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

Claims (14)

1. An account-based service implementation method is applied to an electronic device, and includes:
acquiring all associated accounts in an account set and association coefficients among the associated accounts; any account in the account set is at least correlated with one other account; the account set comprises at least one white seed account and at least one black seed account, the account whiteness value of the white seed account is a preset whiteness extreme value, and the account blackness value of the black seed account is a preset blackness extreme value;
determining an account whiteness value of a non-seed account according to an account whiteness value of at least one associated account of the non-seed accounts and an association coefficient of the associated account and the non-seed account based on a white seed account and a preset whiteness limit value;
determining an account blackness value of a non-seed account according to an account blackness value of at least one associated account of the non-seed accounts and an association coefficient of the associated account and the non-seed account based on a black seed account and a predetermined blackness limit value;
generating an account gray value of a non-seed account by adopting an evidence synthesis method based on the account whiteness value and the account blackness value of the non-seed account;
determining the service strategy of the non-seed account according to the account gray value, wherein the service strategy comprises the following steps: a security verification policy for a user account, the security verification policy comprising: account password verification, mobile phone verification code verification and KYC problem verification;
the determining the business strategy of the non-seed account according to the account gray value comprises the following steps:
determining different quantities of security verification strategies aiming at non-seed accounts with different account gray values;
the determining the account whiteness value of the non-seed account according to the account whiteness value of at least one associated account of the non-seed accounts and the association coefficient of the associated account and the non-seed account based on the white seed account and the preset whiteness extreme value comprises the following steps: taking each white seed sub-account as a diffusion starting point of a first round of whiteness iteration, taking an associated account of each diffusion starting point as a diffusion end point subordinate to the diffusion starting point in each round of whiteness iteration, and if the diffusion end point is not a white seed sub-account, determining an account whiteness value of the account according to an account whiteness value of each diffusion starting point and an associated coefficient of each diffusion starting point and the account; when the preset whiteness iteration stopping condition is not met, taking all diffusion starting points and diffusion end points of the current round as the diffusion starting points of the next round;
the determining the account blackness value of the non-seed account according to the account blackness value of at least one associated account of the non-seed accounts and the association coefficient of the associated account and the non-seed account based on the black seed account and the predetermined blackness limit value comprises: taking each black seed sub-account as a diffusion starting point of the first round of blackness iteration, taking the associated account of each diffusion starting point as a diffusion end point subordinate to the diffusion starting point in each round of blackness iteration, and if the diffusion end point is not a black seed sub-account, determining the account blackness value of the account according to the account blackness value of each diffusion starting point and the association coefficient of each diffusion starting point and the account; and when the preset blackness iteration stop condition is not met, taking all the diffusion starting points and the diffusion end points of the current round as the diffusion starting points of the next round.
2. The method according to claim 1, wherein the regarding the account associated with each diffusion starting point as the diffusion ending point subordinate to the diffusion starting point, and if the diffusion ending point is not a white seed sub-account, determining the account whiteness value of the account according to the account whiteness value of each diffusion starting point and the association coefficient of each diffusion starting point and the account comprises: each diffusion starting point sends a whiteness influence value to the associated account, the whiteness influence value is determined according to the account whiteness value of the diffusion starting point and the association coefficient between the diffusion starting point and the associated account receiving the whiteness influence value, the non-seed account or the black seed account receiving the whiteness influence value determines the account whiteness value of the account according to all the received whiteness influence values, and if the account whiteness value of the account exceeds a preset whiteness extreme value, the preset whiteness extreme value is used as the account whiteness value of the account;
the step of taking the account associated with each diffusion starting point as the diffusion ending point subordinate to the diffusion starting point, and if the diffusion ending point is not a black seed sub-account, determining the account blackness value of the account according to the account blackness value of each diffusion starting point and the association coefficient of each diffusion starting point and the account, includes: each diffusion starting point sends a blackness influence value to the associated account, the blackness influence value is determined according to the account blackness value of the diffusion starting point and the association coefficient between the diffusion starting point and the associated account receiving the blackness influence value, the non-seed account or the white seed account receiving the blackness influence value determines the account blackness value of the account according to all the received blackness influence values, and if the account blackness value of the account exceeds a preset blackness extreme value, the preset blackness extreme value is used as the account blackness value of the account.
3. The method according to claim 2, wherein the value ranges of the account whiteness value and the account blackness value are greater than or equal to 0 and less than or equal to 1, and the value ranges of the correlation coefficient are greater than 0 and less than or equal to 1;
the non-seed account or the black seed account receiving the whiteness influence value determines the account whiteness value of the account according to all the received whiteness influence values, and the method comprises the following steps: receiving a non-seed account or a black seed account of the whiteness influence value, and carrying out normalization processing on the sum of all the received whiteness influence values according to the number of the associated accounts of the account or the preset maximum number of the associated accounts to obtain the account whiteness value of the account;
the non-seed account or the black seed account receiving the blackness influence value determines the account blackness value of the account according to all the received blackness influence values, and the method comprises the following steps: receiving a non-seed account or a black seed account of the blackness influence value, and carrying out normalization processing on the sum of all the received blackness influence values according to the number of the associated accounts of the account or the preset maximum number of the associated accounts to obtain the account blackness value of the account;
the predetermined maximum associated account number is the maximum number of associated accounts allowed to be owned by each account in the account set.
4. The method of claim 2, wherein the set of accounts further comprises: the device account is a non-seed account and is an account which is associated with at least two user accounts;
the non-seed account or the black seed account receiving the whiteness influence value determines the account whiteness value of the account according to all the received whiteness influence values, and the method comprises the following steps: the equipment account receiving the whiteness influence value takes the sum of all the received whiteness influence values as the account whiteness value of the account, and if the account whiteness value of the account exceeds a preset whiteness extreme value, the preset whiteness extreme value is taken as the account whiteness value of the account;
the non-seed account or the white seed account receiving the blackness influence value determines the account blackness value of the account according to all the received blackness influence values, and the method comprises the following steps: and if the account blackness value of the account exceeds a preset blackness extreme value, the preset blackness extreme value is used as the account blackness value of the account.
5. The method of claim 1, wherein the predetermined whiteness iteration stop condition comprises: the ratio of the number of the black seed sub-accounts with the account whiteness value larger than zero to the total number of the black seed sub-accounts reaches the whiteness iteration stop ratio;
the predetermined blackness iteration stop condition includes: and the ratio of the number of the white seed sub-accounts with the account blackness value larger than zero to the total number of the white seed sub-accounts reaches a blackness iteration stop ratio.
6. The method of any one of claims 1 to 5, wherein generating the account gray scale value of the non-seed account using an evidence synthesis rule based on the account white value and the account black value of the non-seed account comprises: and determining basic probability distribution of two evidences according to the account whiteness value and the account blackness value of the non-seed account, if the two basic probability distribution meet a preset evidence conflict condition, calculating an evidence fusion result by adopting a preset evidence conflict processing method, otherwise, calculating an evidence fusion result by adopting an evidence synthesis rule, and generating the account grey value of the non-seed account according to the evidence fusion result.
7. The method according to claim 6, wherein the predetermined evidence conflict condition comprises: the normalization constant K used for measuring the conflict degree between the evidences in the evidence synthesis rule is larger than a preset conflict threshold value;
the calculating of the evidence fusion result by adopting the preset evidence conflict processing rule comprises the following steps: and calculating the distance between the two basic probability distributions, determining the support degree of each evidence according to the distance between the two basic probability distributions, taking the credibility obtained after the support degree is normalized as the weight of the evidence, and obtaining an evidence fusion result by adopting an evidence conflict processing algorithm of a weighted average method.
8. An account-based service implementation device, applied to an electronic device, includes:
the association relation obtaining unit is used for obtaining all association accounts in the account set and association coefficients among the association accounts; any account in the account set is at least correlated with one other account; the account set comprises at least one white seed account and at least one black seed account, the account whiteness value of the white seed account is a preset whiteness extreme value, and the account blackness value of the black seed account is a preset blackness extreme value;
the whiteness value determining unit is used for determining the account whiteness value of the non-seed account according to the account whiteness value of at least one associated account of the non-seed accounts and the association coefficient of the associated account and the non-seed account based on the white seed account and a preset whiteness extreme value;
the blackness value determining unit is used for determining an account blackness value of a non-seed account according to an account blackness value of at least one associated account of the non-seed accounts and an association coefficient of the associated account and the non-seed account based on a black seed account and a preset blackness extreme value;
the evidence synthesis unit is used for generating an account gray value of the non-seed account by adopting an evidence synthesis rule based on the account white value and the account black value of the non-seed account;
a service policy determining unit, configured to determine a service policy of the non-seed account according to an account gray value, where the service policy includes: a security verification policy for a user account, the security verification policy comprising: account password verification, mobile phone verification code verification and KYC problem verification;
the determining the business strategy of the non-seed account according to the account gray value comprises the following steps:
determining different quantities of security verification strategies aiming at non-seed accounts with different account gray values;
the whiteness value determination unit is specifically configured to: taking each white seed sub-account as a diffusion starting point of a first round of whiteness iteration, taking an associated account of each diffusion starting point as a diffusion end point subordinate to the diffusion starting point in each round of whiteness iteration, and if the diffusion end point is not a white seed sub-account, determining an account whiteness value of the account according to an account whiteness value of each diffusion starting point and an associated coefficient of each diffusion starting point and the account; when the preset whiteness iteration stopping condition is not met, taking all diffusion starting points and diffusion end points of the current round as the diffusion starting points of the next round;
the blackness value determination unit is specifically configured to: taking each black seed sub-account as a diffusion starting point of the first round of blackness iteration, taking the associated account of each diffusion starting point as a diffusion end point subordinate to the diffusion starting point in each round of blackness iteration, and if the diffusion end point is not a black seed sub-account, determining the account blackness value of the account according to the account blackness value of each diffusion starting point and the association coefficient of each diffusion starting point and the account; and when the preset blackness iteration stop condition is not met, taking all the diffusion starting points and the diffusion end points of the current round as the diffusion starting points of the next round.
9. The apparatus according to claim 8, wherein the whiteness value determination unit takes the account associated with each diffusion start point as a diffusion end point belonging to the diffusion start point, and if the diffusion end point is not a white seed sub-account, determines the account whiteness value of the own account according to the account whiteness value belonging to each diffusion start point and the association coefficient of each diffusion start point and the own account, and includes: each diffusion starting point sends a whiteness influence value to the associated account, the whiteness influence value is determined according to the account whiteness value of the diffusion starting point and the association coefficient between the diffusion starting point and the associated account receiving the whiteness influence value, the non-seed account or the black seed account receiving the whiteness influence value determines the account whiteness value of the account according to all the received whiteness influence values, and if the account whiteness value of the account exceeds a preset whiteness extreme value, the preset whiteness extreme value is used as the account whiteness value of the account;
the blackness value determination unit takes the account associated with each diffusion starting point as a diffusion end point subordinate to the diffusion starting point, and if the diffusion end point is not a black seed sub-account, the blackness value determination unit determines the account blackness value of the account according to the account blackness value of each diffusion starting point and the association coefficient of each diffusion starting point and the account, and the method comprises the following steps: each diffusion starting point sends a blackness influence value to the associated account, the blackness influence value is determined according to the account blackness value of the diffusion starting point and the association coefficient between the diffusion starting point and the associated account receiving the blackness influence value, the non-seed account or the white seed account receiving the blackness influence value determines the account blackness value of the account according to all the received blackness influence values, and if the account blackness value of the account exceeds a preset blackness extreme value, the preset blackness extreme value is used as the account blackness value of the account.
10. The apparatus according to claim 9, wherein the value ranges of the account whiteness value and the account blackness value are greater than or equal to 0 and less than or equal to 1, and the value ranges of the correlation coefficient are greater than 0 and less than or equal to 1;
the determination unit of the whiteness value receives the non-seed account or black seed account of the whiteness influence value and determines the account whiteness value of the account according to all the received whiteness influence values, and the determination unit of the whiteness value comprises the following steps: receiving a non-seed account or a black seed account of the whiteness influence value, and carrying out normalization processing on the sum of all the received whiteness influence values according to the number of the associated accounts of the account or the preset maximum number of the associated accounts to obtain the account whiteness value of the account;
the non-seed account or black seed account receiving the blackness influence value and determining the account blackness value of the account according to all the received blackness influence values by the blackness value determining unit comprises the following steps: receiving a non-seed account or a black seed account of the blackness influence value, and carrying out normalization processing on the sum of all the received blackness influence values according to the number of the associated accounts of the account or the preset maximum number of the associated accounts to obtain the account blackness value of the account;
the predetermined maximum associated account number is the maximum number of associated accounts allowed to be owned by each account in the account set.
11. The apparatus of claim 9, wherein the set of accounts further comprises: the device account is a non-seed account and is an account which is associated with at least two user accounts;
the determination unit of the whiteness value receives the non-seed account or black seed account of the whiteness influence value and determines the account whiteness value of the account according to all the received whiteness influence values, and the determination unit of the whiteness value comprises the following steps: the equipment account receiving the whiteness influence value takes the sum of all the received whiteness influence values as the account whiteness value of the account, and if the account whiteness value of the account exceeds a preset whiteness extreme value, the preset whiteness extreme value is taken as the account whiteness value of the account;
the non-seed account or the white seed account receiving the blackness influence value and determining the account blackness value of the account according to all the received blackness influence values by the blackness value determining unit comprises the following steps: and if the account blackness value of the account exceeds a preset blackness extreme value, the preset blackness extreme value is taken as the account blackness value of the account.
12. The apparatus of claim 8, wherein the predetermined whiteness iteration stop condition comprises: the ratio of the number of the black seed sub-accounts with the account whiteness value larger than zero to the total number of the black seed sub-accounts reaches the whiteness iteration stop ratio;
the predetermined blackness iteration stop condition includes: and the ratio of the number of the white seed sub-accounts with the account blackness value larger than zero to the total number of the white seed sub-accounts reaches a blackness iteration stop ratio.
13. The apparatus according to any one of claims 8 to 12, wherein the evidence synthesis unit is specifically configured to: and determining basic probability distribution of two evidences according to the account whiteness value and the account blackness value of the non-seed account, if the two basic probability distribution meet a preset evidence conflict condition, calculating an evidence fusion result by adopting a preset evidence conflict processing method, otherwise, calculating an evidence fusion result by adopting an evidence synthesis rule, and generating the account grey value of the non-seed account according to the evidence fusion result.
14. The apparatus according to claim 13, wherein the predetermined evidence conflict condition comprises: the normalization constant K used for measuring the conflict degree between the evidences in the evidence synthesis rule is larger than a preset conflict threshold value;
the evidence synthesizing unit calculates an evidence fusion result by adopting a preset evidence conflict processing rule, and comprises the following steps: and calculating the distance between the two basic probability distributions, determining the support degree of each evidence according to the distance between the two basic probability distributions, taking the credibility obtained after the support degree is normalized as the weight of the evidence, and obtaining an evidence fusion result by adopting an evidence conflict processing algorithm of a weighted average method.
CN201610348880.0A 2016-05-24 2016-05-24 Account-based service implementation method and device Active CN107423982B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610348880.0A CN107423982B (en) 2016-05-24 2016-05-24 Account-based service implementation method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610348880.0A CN107423982B (en) 2016-05-24 2016-05-24 Account-based service implementation method and device

Publications (2)

Publication Number Publication Date
CN107423982A CN107423982A (en) 2017-12-01
CN107423982B true CN107423982B (en) 2020-12-29

Family

ID=60422615

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610348880.0A Active CN107423982B (en) 2016-05-24 2016-05-24 Account-based service implementation method and device

Country Status (1)

Country Link
CN (1) CN107423982B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111598713B (en) * 2020-07-24 2021-12-14 北京淇瑀信息科技有限公司 Cluster recognition method and device based on similarity weight updating and electronic equipment
CN115526616B (en) * 2022-09-19 2023-08-22 青岛畅联科技有限公司 MaaS payment settlement system based on artificial intelligence

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2011163525A1 (en) * 2010-06-23 2011-12-29 Obopay, Inc. Mobile networked payment system
CN104917739A (en) * 2014-03-14 2015-09-16 腾讯科技（北京）有限公司 False account identification method and device
CN105404947A (en) * 2014-09-02 2016-03-16 阿里巴巴集团控股有限公司 User quality detection method and device

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8744880B2 (en) * 2008-10-16 2014-06-03 James G. C. T. Gerber System and method to provide for and communicate about safer and better returning asset-liability investment programs
US20120022610A1 (en) * 2010-04-22 2012-01-26 Kenneth Blum Devices for Delivering Neuro Electro Adaptive Therapy NEAT
US9202249B1 (en) * 2014-07-03 2015-12-01 Palantir Technologies Inc. Data item clustering and analysis

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2011163525A1 (en) * 2010-06-23 2011-12-29 Obopay, Inc. Mobile networked payment system
CN104917739A (en) * 2014-03-14 2015-09-16 腾讯科技（北京）有限公司 False account identification method and device
CN105404947A (en) * 2014-09-02 2016-03-16 阿里巴巴集团控股有限公司 User quality detection method and device

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
《基于跳_扩散过程的信用风险模型研究》;彭玉梅;《商业研究》;20040525;期刊第30-32页，本文第1-3页 *

Also Published As

Publication number Publication date
CN107423982A (en) 2017-12-01

Similar Documents

Publication Publication Date Title
KR102197371B1 (en) Identity verification method and device
CN109598281B (en) Service risk prevention and control method, device and equipment
US20180046939A1 (en) Automated Machine Learning Feature Processing
US10609087B2 (en) Systems and methods for generation and selection of access rules
AU2008323784B2 (en) Network rating
US10587596B1 (en) Framework for authenticating new users
US20190295085A1 (en) Identifying fraudulent transactions
CN104965844A (en) Information processing method and apparatus
US20210264337A1 (en) Risk decision-making method and apparatus
CN110738473B (en) Wind control method, system, device and equipment
US11310236B2 (en) Deriving confidence scores based on device sharing
WO2021098274A1 (en) Method and apparatus for evaluating risk of leakage of private data
CN112927061B (en) User operation detection method and program product
CN110348978A (en) The recognition methods of risk clique, device, equipment and the storage medium calculated based on figure
US11379591B2 (en) Methods and devices for user authorization
CN110930218B (en) Method and device for identifying fraudulent clients and electronic equipment
CN110032846B (en) Identity data anti-misuse method and device and electronic equipment
CN110796269A (en) Method and device for generating model, and method and device for processing information
CN110851482B (en) Method and device for providing data model for multiple data parties
CN110011960A (en) Data sharing method, system, server and readable storage medium storing program for executing based on block chain
CN107423982B (en) Account-based service implementation method and device
CN110008986B (en) Batch risk case identification method and device and electronic equipment
US20190295090A1 (en) Methods and apparatuses for fraud handling
CN113919664A (en) API asset risk control method and device, computer equipment and storage medium
CN111476668B (en) Identification method and device of credible relationship, storage medium and computer equipment

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
TA01 Transfer of patent application right
TA01 Transfer of patent application right

Effective date of registration: 20200921

Address after: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman Islands

Applicant after: Innovative advanced technology Co.,Ltd.

Address before: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman Islands

Applicant before: Advanced innovation technology Co.,Ltd.

Effective date of registration: 20200921

Address after: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman Islands

Applicant after: Advanced innovation technology Co.,Ltd.

Address before: A four-storey 847 mailbox in Grand Cayman Capital Building, British Cayman Islands

Applicant before: Alibaba Group Holding Ltd.

GR01 Patent grant
GR01 Patent grant