Account data processing method, device, computer equipment and storage medium

Abstract

The present application relates to an account data processing method, apparatus, computer device, storage medium, and computer program product. The method comprises: determining the abnormal account association between a target account and each passive account; obtaining the initial active association confidence between the target account and each passive account, and determining the current active credibility corresponding to the target account; determining the current passive credibility of each passive account; determining the current active association confidence between the target account and each passive account; and adjusting the corresponding initial active association confidence based on the difference between the initial active association confidence and the current active association confidence between the target account and the same passive account until a convergence condition is met, thereby obtaining each target active association confidence and thus obtaining the target active credibility of the target account. The account status of the target account is determined based on the target active credibility of the target account. The use of this method can improve the accuracy of account data processing.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to an account data processing method, apparatus, computer equipment, storage medium, and computer program product. Background Art

[0002] With the development of computer technology, more and more companies are providing various online services to people. People can register accounts to access these services. However, these companies need to analyze and process massive amounts of account data and determine account status. A common method for determining account status is to process account data and determine account status based on machine learning technology.

[0003] However, in traditional methods, when using machine learning technology to process account data, it is easily affected by sample imbalance, and thus it is impossible to obtain accurate account data processing results. Summary of the Invention

[0004] Based on this, it is necessary to provide an account data processing method, apparatus, computer equipment, computer-readable storage medium and computer program product that can improve the accuracy of account data processing in response to the above technical problems.

[0005] This application provides an account data processing method. The method includes:

[0006] Obtain the number of passive accounts corresponding to the target account, obtain the number of active accounts corresponding to each passive account of the target account, determine the number of abnormal passive accounts from the number of passive accounts, and determine the number of abnormal active accounts from the number of active accounts;

[0007] Based on the number of passive accounts, the number of active accounts, the number of abnormal passive accounts, and the number of abnormal active accounts, the abnormal account correlation between the target account and the passive accounts is calculated to obtain the abnormal account correlation between the target account and each passive account;

[0008] Obtain the initial active association confidence between the target account and the corresponding passive account. Based on the number of passive accounts, the initial active association confidence, and the association degree of each abnormal account, obtain the current active credibility of the target account.

[0009] Obtaining the initial passive association confidence between the passive account and the corresponding active account, and obtaining the current passive credibility of each passive account based on the number of active accounts corresponding to the same passive account and the initial passive association confidence;

[0010] Based on the abnormal account association degree between the target account and the passive account, the current active credibility of the target account, and the current passive credibility of the passive account, the current active association confidence between the target account and the passive account is calculated to obtain the current active association confidence between the target account and each passive account;

[0011] Based on the difference between the initial active association confidence and the current active association confidence between the target account and the same passive account, the corresponding initial active association confidence is adjusted until the convergence condition is met, thereby obtaining the target active association confidence between the target account and each passive account.

[0012] Based on the active association confidence of each target corresponding to the target account, the association degree of each abnormal account and the number of passive accounts, the active credibility of the target corresponding to the target account is obtained;

[0013] Based on the target active credibility, determine the account status corresponding to the target account.

[0014] The present application also provides an account data processing device. The device includes:

[0015] An account quantity information acquisition module is used to obtain the number of passive accounts corresponding to the target account, obtain the number of active accounts corresponding to each passive account of the target account, determine the number of abnormal passive accounts from the number of passive accounts, and determine the number of abnormal active accounts from the number of active accounts;

[0016] An abnormal account association calculation module is used to calculate the abnormal account association between the target account and the passive account based on the number of passive accounts, the number of active accounts, the number of abnormal passive accounts, and the number of abnormal active accounts, and obtain the abnormal account association between the target account and each passive account;

[0017] The current active credibility calculation module is used to obtain the initial active association confidence between the target account and the corresponding passive account. Based on the number of passive accounts, the initial active association confidence and the association degree of each abnormal account, the current active credibility corresponding to the target account is obtained;

[0018] The current passive credibility calculation module is used to obtain the initial passive association confidence between the passive account and the corresponding active account, and obtain the current passive credibility corresponding to each passive account based on the number of active accounts corresponding to the same passive account and the initial passive association confidence.

[0019] A current active association confidence calculation module is used to calculate the current active association confidence between the target account and the passive account based on the abnormal account association between the target account and the passive account, the current active credibility corresponding to the target account, and the current passive credibility corresponding to the passive account, and obtain the current active association confidence between the target account and each passive account;

[0020] a target active association confidence determination module, configured to adjust the corresponding initial active association confidence based on the difference between the initial active association confidence and the current active association confidence between the target account and the same passive account until a convergence condition is met, thereby obtaining the target active association confidence between the target account and each passive account;

[0021] A target active credibility calculation module is used to obtain the target active credibility corresponding to the target account based on the target active association confidence level, the association level of each abnormal account, and the number of passive accounts corresponding to the target account;

[0022] The target account status determination module is used to determine the account status corresponding to the target account based on the target active credibility.

[0023] A computer device includes a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above-mentioned account data processing method when executing the computer program.

[0024] A computer-readable storage medium stores a computer program, which implements the steps of the above-mentioned account data processing method when executed by a processor.

[0025] A computer program product includes a computer program, which implements the steps of the above-mentioned account data processing method when executed by a processor.

[0026] The above-described account data processing method, apparatus, computer device, storage medium, and computer program product calculate the abnormal account association degree between the target account and each passive account based on the number of passive accounts and the number of abnormal passive accounts corresponding to the target account, as well as the number of active accounts and the number of abnormal active accounts corresponding to each passive account. This method fully utilizes the account status information of each account directly or indirectly associated with the target account, making the calculated abnormal account association degree more reliable, thereby improving the accuracy of account data processing. The current active credibility of the target account is calculated based on the abnormal account association degree and initial active association confidence between the target account and each passive account, as well as the number of passive accounts corresponding to the target account. This method fully utilizes information such as the abnormal account association degree between the target account and each passive account, the initial active association confidence, and the number of passive accounts associated with the target account, and takes into account the interrelationships between different data information. This makes the calculated current active credibility more accurate and effectively reflects the target account's trustworthiness with respect to each passive account. The current passive credibility of a passive account is calculated based on the initial passive association confidence between the passive account and each active account, as well as the number of active accounts corresponding to the passive account. The passive association confidence reflects the confidence of the corresponding passive account's association. The calculation of the current passive credibility takes into account the passive association confidence information between the passive account and each active account. By fully integrating the passive association confidence information corresponding to each passive account, the current passive credibility obtained is more accurate and reliable. The current active association confidence between the target account and each passive account is calculated based on the abnormal account association between the target account and the passive account, the current active credibility corresponding to the target account, and the current passive credibility corresponding to the passive account. This calculation of the active association confidence fully considers data reflecting the credibility of the target and passive accounts, data reflecting the status of the association between the target and passive accounts, and the interplay between these data. This effectively improves the reliability of the active association confidence, thereby enhancing the accuracy of account data processing. Based on the difference between the initial active association confidence and the current active association confidence between the target account and the same passive account, the initial active association confidence is adjusted until convergence conditions are met, resulting in target active association confidences between the target account and each passive account. When the active association confidences corresponding to the target account converge, the active association confidences are now stable and can more effectively reflect the confidence of the abnormal account associations between the target account and each passive account. Consequently, the target active credibility of the target account, calculated based on the target active association confidences, the abnormal account associations, and the number of passive accounts, can also more effectively reflect the credibility of the target account. Determining the account status of the target account based on the target active credibility can improve the reliability and accuracy of account data processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 This is a diagram of an application environment of an account data processing method in one embodiment;

[0028] Figure 2 1 is a flowchart of a method for processing account data in one embodiment;

[0029] Figure 3 A schematic diagram of a process for determining target active association confidence in one embodiment;

[0030] Figure 4 A schematic diagram of account feature information in one embodiment;

[0031] Figure 5 Schematic diagram of the association between a target account and other accounts in one embodiment;

[0032] Figure 6 A schematic diagram of the process of processing account data in one embodiment;

[0033] Figure 7 is a structural block diagram of an account data processing device in one embodiment;

[0034] Figure 8 is a diagram of the internal structure of a computer device in one embodiment;

[0035] Figure 9 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0036] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0037] The account data processing method provided in the embodiment of the present application can be applied to Figure 1In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. The terminal 102 can be, but is not limited to, various personal computers, laptops, smart phones, tablets, Internet of Things devices and portable wearable devices. The Internet of Things devices can be smart TVs, smart car-mounted devices, etc. Portable wearable devices can be smart watches, smart head-mounted devices, etc. The server 104 can be implemented as an independent server or a server cluster or cloud server composed of multiple servers. The terminal 102 and the server 104 can be directly or indirectly connected via wired or wireless communication, and this application is not limited here.

[0038] Both the terminal and the server can be used independently to execute the account data processing method provided in the embodiments of the present application.

[0039] For example, the terminal obtains the number of passive accounts corresponding to the target account, obtains the number of active accounts corresponding to each passive account of the target account, determines the number of abnormal passive accounts from the number of passive accounts, and determines the number of abnormal active accounts from the number of active accounts. Based on the number of passive accounts, the number of active accounts, the number of abnormal passive accounts, and the number of abnormal active accounts, the terminal calculates the abnormal account association degree between the target account and the passive accounts, and obtains the abnormal account association degree between the target account and each passive account. The terminal obtains the initial active association confidence between the target account and the corresponding passive account, and obtains the current active confidence degree corresponding to the target account based on the number of passive accounts, the initial active association confidence degrees, and the abnormal account association degrees. The terminal obtains the initial passive association confidence between the passive account and the corresponding active account, and obtains the current passive confidence degree corresponding to each passive account based on the number of active accounts corresponding to the same passive account and the initial passive association confidence degrees. Based on the abnormal account association degree between the target account and the passive account, the current active confidence degree corresponding to the target account, and the current passive confidence degree corresponding to the passive account, the terminal calculates the current active association confidence degree between the target account and the passive account, and obtains the current active association confidence degree between the target account and each passive account. Based on the difference between the initial active association confidence and the current active association confidence between the target account and the same passive account, the terminal adjusts the corresponding initial active association confidence until convergence conditions are met, thereby obtaining the target active association confidence between the target account and each passive account. Based on the target active association confidences corresponding to each target account, the associations of each abnormal account, and the number of passive accounts, the terminal obtains the target active credibility corresponding to the target account. Based on the target active credibility, the terminal determines the account status corresponding to the target account.

[0040] The terminal and the server can also be used in conjunction to execute the account data processing method provided in the embodiments of the present application.

[0041] For example, a terminal sends an account data processing request to a server, the account data processing request carrying the account identifier corresponding to the target account. Based on the account identifier, the server obtains the number of passive accounts corresponding to the target account, obtains the number of active accounts corresponding to each passive account of the target account, determines the number of abnormal passive accounts from the number of passive accounts, and determines the number of abnormal active accounts from the number of active accounts. Based on the number of passive accounts, the number of active accounts, the number of abnormal passive accounts, and the number of abnormal active accounts, the server calculates the abnormal account association degree between the target account and the passive accounts, and obtains the abnormal account association degree between the target account and each passive account. The server obtains the initial active association confidence between the target account and the corresponding passive account, and obtains the current active credibility of the target account based on the number of passive accounts, the initial active association confidences, and the abnormal account associations. The server obtains the initial passive association confidence between the passive account and the corresponding active account, and obtains the current passive credibility of each passive account based on the number of active accounts corresponding to the same passive account and the initial passive association confidences. The server calculates the current active association confidence between the target account and the passive account based on the abnormal account association between the target account and the passive account, the current active credibility of the target account, and the current passive credibility of the passive account, thereby obtaining the current active association confidence between the target account and each passive account. Based on the difference between the initial active association confidence and the current active association confidence between the target account and the same passive account, the server adjusts the corresponding initial active association confidence until convergence conditions are met, thereby obtaining the target active association confidence between the target account and each passive account. Based on the target active association confidences of each target account, the abnormal account associations, and the number of passive accounts, the server obtains the target active credibility of the target account. Based on the target active credibility, the server determines the account status of the target account. The server transmits the account status of the target account to the terminal. The terminal may display the account status of the target account or perform corresponding processing on the target account based on the account status of the target account.

[0042] In one embodiment, Figure 2 As shown, a method for processing account data is provided, which is described by taking the method applied to a computer device as an example. The computer device can be a terminal or a server, and includes the following steps:

[0043] Step S202: Obtain the number of passive accounts corresponding to the target account, obtain the number of active accounts corresponding to each passive account of the target account, determine the number of abnormal passive accounts from the number of passive accounts, and determine the number of abnormal active accounts from the number of active accounts.

[0044] The target account refers to the account that requires account data processing to determine the account status. The number of passive accounts refers to the number of all passive accounts associated with the target account. When two accounts have an association relationship, the role of one of the accounts in the association relationship can be used to determine whether the account is an active account or a passive account in the association relationship. In one embodiment, if there is interaction information between the accounts, then the accounts have an association relationship. The active account and the passive account can be distinguished based on the interaction information between the accounts. The interaction initiator corresponding to the interaction information is regarded as the active account, and the interaction recipient corresponding to the interaction information is regarded as the passive account. For example, if the interaction information between account A and account B is that account A invites account B to register an account, then there is an association relationship between account A and account B. Account A is the active account in this association relationship, that is, account A is an active account associated with account B, and account B is the passive account in this association relationship, that is, account B is a passive account associated with account A.

[0045] The number of active accounts refers to the number of active accounts associated with each passive account associated with the target account. The number of abnormal passive accounts refers to the number of passive accounts with known abnormal account status among the passive accounts associated with the target account. The number of abnormal active accounts refers to the number of active accounts with known abnormal account status among the active accounts corresponding to each passive account associated with the target account. For example, if target account A is associated with accounts B and C as an active account, account B is associated with accounts B1 and B2 as a passive account, and account C is associated with accounts C1 and C2 as a passive account, and the account statuses of accounts B, B1, B2, and C1 are abnormal, then the number of passive accounts corresponding to target account A is 2, the number of abnormal passive accounts corresponding to target account A is 1, the number of active accounts corresponding to accounts B and C are both 2, the number of abnormal active accounts corresponding to account B is 2, and the number of abnormal active accounts corresponding to account C is 1.

[0046] Specifically, the computer device searches for passive accounts associated with the target account, counts the number of passive accounts corresponding to the target account, searches for active accounts associated with each passive account, counts the number of active accounts corresponding to each passive account associated with the target account, searches for accounts with abnormal account status among the passive accounts associated with the target account, counts the number of abnormal passive accounts corresponding to the target account, searches for accounts with abnormal account status among the active accounts corresponding to each passive account associated with the target account, and counts the number of abnormal active accounts corresponding to each passive account associated with the target account.

[0047] Step S204 : Based on the number of passive accounts, the number of active accounts, the number of abnormal passive accounts, and the number of abnormal active accounts, the abnormal account association degree between the target account and the passive accounts is calculated to obtain the abnormal account association degree between the target account and each passive account.

[0048] The abnormal account association degree indicates the likelihood that the relationship between two accounts is abnormal. The greater the number of abnormal accounts associated with a relationship, the greater the likelihood that the relationship is abnormal, and the higher the abnormal account association degree.

[0049] Specifically, the computer device may calculate the abnormal account association degree between the target account and the passive accounts based on the number of passive accounts corresponding to the target account, the number of abnormal passive accounts, and the number of active accounts and the number of abnormal active accounts corresponding to a single passive account of the target account. If the target account has multiple passive accounts, the abnormal account association degree between the target account and each passive account may be calculated.

[0050] In one embodiment, the computer device may obtain the abnormal account association degree between the target account and the passive account based on the ratio of the number of abnormal passive accounts to the number of passive accounts and the ratio of the number of abnormal active accounts to the number of active accounts.

[0051] Step S206: Obtain the initial active association confidence between the target account and the corresponding passive account, and obtain the current active credibility corresponding to the target account based on the number of passive accounts, the initial active association confidences, and the associations of the abnormal accounts.

[0052] The active association confidence level refers to the confidence level of the association between the target account and its passive account, and is used to measure the reliability of the association between the target account and the passive account. A corresponding active association confidence level exists between the target account and each passive account. For example, if account a, as the active account, is associated with accounts b and c, then the abnormal account association level between accounts a and b is S(a, b), the active association confidence level between accounts a and b is C(a, b), the abnormal account association level between accounts a and c is S(a, c), and the active association confidence level between accounts a and c is C(a, c). A higher active association confidence level for the target account indicates, to a certain extent, a more accurate assessment of the association relationship using the target account's abnormal account association level. The initial active association confidence level refers to the initial active association confidence level and is included in the calculation of the current active credibility level. It is understood that the final active credibility requires an iterative process to obtain. The initial active association confidence is the initial active association confidence during each iteration and is the data used to calculate the current active credibility corresponding to the current round during each iteration. During the first iteration, the value of the initial active association confidence can be set according to actual needs. In one embodiment, when the current active credibility corresponding to the target account is calculated for the first time, that is, during the first iteration, each initial active association confidence corresponding to the target account can be assigned the same initial value, for example, 0.5.

[0053] Active credibility is used to characterize the trustworthiness and reliability of an account as an active account. The higher the active credibility of an account, the less likely it is to be an anomalous account. Active credibility can be used to calculate the corresponding active association confidence. The current active credibility refers to the currently calculated active credibility of the target account, and is the active credibility of the target account calculated in the current round.

[0054] Specifically, the computer device obtains the initial active association confidence between the target account and each passive account, fuses the abnormal account association between the target account and the same passive account with the initial active association confidence to obtain a fusion result, and calculates the current active credibility of the target account based on the fusion result between the target account and each passive account and the number of passive accounts corresponding to the target account. This method of calculating the current active credibility of the target account comprehensively considers multiple different indicators corresponding to the target account, which can improve the accuracy of calculating the current active credibility.

[0055] In one embodiment, the current active credibility is negatively correlated with the number of passive accounts, the current active credibility is positively correlated with the initial active association confidence, and the current active credibility is negatively correlated with the abnormal account association. It is understandable that if the target account has a larger number of passive accounts, it means that the target account is associated with a large number of accounts, and the target account is prone to abnormalities. Therefore, the current active credibility is negatively correlated with the number of passive accounts. The smaller the abnormal account association, the less likely the association between the target account and the passive account is abnormal. The larger the initial active association confidence, the higher the confidence of the abnormal account association between the target account and the passive account, that is, the more accurate the abnormal account association is in assessing the association relationship. Therefore, the current active credibility is negatively correlated with the abnormal account association, and positively correlated with the initial active association confidence.

[0056] Step S208 , obtaining the initial passive association confidence between the passive account and the corresponding active account, and obtaining the current passive credibility corresponding to each passive account based on the number of active accounts corresponding to the same passive account and each initial passive association confidence.

[0057] The passive association confidence level refers to the confidence level of the association between a passive account and its active account, and is used to measure the reliability of the association between the passive and active accounts. For example, if account a is a passive account and is associated with accounts b and c, then the abnormal account association level between accounts a and b is S(b,a), the passive association confidence level between accounts a and b is C(b,a), the abnormal account association level between accounts a and c is S(c,a), and the passive association confidence level between accounts a and c is C(c,a). The initial passive association confidence level refers to the initial passive association confidence level and is used in the calculation of the current passive credibility level. It is understood that the final passive credibility level is obtained through an iterative process. The initial passive association confidence level is the initial passive association confidence level in each iteration and is used to calculate the current passive credibility level for each iteration. During the first iteration, the value of the initial passive association confidence level can be set according to actual needs. In one embodiment, when the current passive credibility corresponding to the passive account is calculated for the first time, the same initial value may be assigned to each initial passive association confidence corresponding to the passive account. For example, the initial value may be 0.5.

[0058] Passive credibility is used to indicate the degree of trust an account holds from active accounts when acting as a passive account. The more connections with high passive confidence an account has, the higher its passive credibility. Current passive credibility refers to the currently calculated passive credibility, i.e., the passive credibility of a passive account calculated in the current round.

[0059] Specifically, the computer device obtains the initial passive association confidence between each passive account and its corresponding active account. Based on the number of active accounts corresponding to the same passive account and the initial passive association confidence, the computer device obtains the current passive credibility corresponding to the single passive account. Since the target account has at least one passive account, the current passive credibility corresponding to each passive account can be ultimately obtained.

[0060] In one embodiment, the current passive credibility is positively correlated with the initial passive association confidence, and negatively correlated with the number of active accounts. The greater the number of active accounts associated with a passive account, the more accounts are associated with the passive account, making the passive account more susceptible to anomalies. Therefore, the current passive credibility is negatively correlated with the number of active accounts. The higher the passive association confidence between a passive account and each active account, the closer the association between the passive account and the active account, and the higher the current passive credibility corresponding to the passive account. Therefore, the current passive credibility is positively correlated with the initial passive association confidence.

[0061] In one embodiment, the initial passive association confidences between a single passive account and each of its own active accounts are counted to obtain statistical results. Each passive account has its own corresponding statistical result. Based on the number of active accounts corresponding to the same passive account and the statistical results, the current passive credibility of each passive account corresponding to the target account is obtained.

[0062] Step S210, based on the abnormal account association between the target account and the passive account, the current active credibility corresponding to the target account, and the current passive credibility corresponding to the passive account, calculate the current active association confidence between the target account and the passive account, and obtain the current active association confidence between the target account and each passive account.

[0063] The current active association confidence refers to the active association confidence calculated based on the current active credibility and the current passive credibility.

[0064] In one embodiment, the active association confidence, passive association confidence, active credibility, and passive credibility are updated through a recursive iteration process. During each iteration, the current active credibility of the target account is calculated based on the initial active association confidences of the target account, the associations of each abnormal account, and the number of passive accounts. The current passive credibility of each passive account is calculated based on the initial passive association confidences of the same passive account and the number of active accounts. The current active association confidence between the target account and each passive account is calculated based on the current active credibility of the target account, the current passive credibility of each passive account, and the number of passive accounts. Similarly, the current passive association confidence between the passive account and each active account is calculated based on the current passive credibility of the passive account, the current active credibility of each active account, and the number of active accounts. In the next iteration, the current active association confidence calculated in the previous iteration is used as the initial active association confidence for the current iteration, and the current passive association confidence calculated in the previous iteration is used as the initial passive association confidence for the current iteration. That is, when calculating the current active credibility of the target account and the current passive credibility of each passive account next time, the current active association confidence between the target account and each passive account will be used as the initial active association confidence, and the current passive association confidence between the passive account and each active account will be used as the initial passive association confidence.

[0065] Specifically, the computer device calculates the current active association confidence between the target account and the single passive account based on the abnormal account association between the target account and the single passive account, the current active credibility corresponding to the target account, and the current passive credibility corresponding to the single passive account. There is at least one passive account in the target account, so the current active association confidence between the target account and each passive account can be finally obtained.

[0066] In one embodiment, the current active association confidence is negatively correlated with the abnormal account association, the current passive credibility is negatively correlated with the current active credibility, and the current active association confidence is positively correlated with the current passive credibility. A passive account with a higher current passive credibility is more likely to have an association with a higher current passive association confidence. If the current passive credibility of a single passive account corresponding to a target account is higher, it indicates that the current active association confidence between the target account and the single passive account is high. Therefore, the current active association confidence and the current passive credibility are positively correlated. If the current active credibility of a target account is higher, it indicates that the target account is more trusted by the passive account, and the probability of the target account's account status being abnormal is lower, then the probability of the association between the target account and the passive account being abnormal is lower. In other words, the abnormal account association between the target account and the passive account is lower, and the current active association confidence is higher. Therefore, the current active association confidence is positively correlated with the current active credibility and negatively correlated with the abnormal account association.

[0067] Step S212: Based on the difference between the initial active association confidence and the current active association confidence between the target account and the same passive account, the corresponding initial active association confidence is adjusted until the convergence condition is met, thereby obtaining the target active association confidence between the target account and each passive account.

[0068] The convergence condition includes that the difference between each current active association confidence and the corresponding initial active association confidence is less than a first threshold. In one embodiment, the convergence condition also includes that the difference between the current active credibility and the initial active credibility is less than a second threshold, and the difference between each current passive credibility and the corresponding initial passive credibility is less than a third threshold. The first threshold, the second threshold, and the third threshold can be set according to actual needs. The first threshold, the second threshold, and the third threshold can be the same or different. For example, the first threshold, the second threshold, and the third threshold are set to 0.001.

[0069] The target active association confidence refers to the current active association confidence that meets the convergence conditions. For example, the target active association confidence refers to the current active association confidence between the target account and the passive account when the difference between the current active credibility and the initial active credibility, the difference between each current passive credibility and the corresponding initial passive credibility, and the difference between each current active association confidence and the corresponding initial active association confidence are all less than the corresponding convergence threshold.

[0070] Specifically, the computer device determines the difference between the initial active association confidence and the current active association confidence between the target account and the same passive account, and determines whether the convergence condition is met. If the convergence condition is not met, the initial active association confidence corresponding to the target account is adjusted, and a new current active association confidence is calculated based on the adjusted initial active association confidence. It is again determined whether the convergence condition is met, and so on, until the convergence condition is met, and the current active association confidence between the target account and each passive account that meets the convergence condition is used as the corresponding target active association confidence.

[0071] Step S214: Based on the target active association confidence levels, abnormal account association levels, and the number of passive accounts corresponding to the target account, the target active credibility level corresponding to the target account is obtained.

[0072] The target active credibility refers to the active credibility of the target account calculated by the target active association confidence, the association degree of each abnormal account, and the number of passive accounts.

[0073] Specifically, similar to the process of calculating the current active credibility, the computer device can obtain the target active credibility corresponding to the target account based on the target active association confidence levels, abnormal account association levels, and the number of passive accounts corresponding to the target account. The computer device fuses the abnormal account association levels and target active association confidence levels between the target account and the same passive account to obtain a fusion result, and obtains the target active credibility corresponding to the target account based on the fusion result between the target account and each passive account and the number of passive accounts corresponding to the target account.

[0074] Step S216: Determine the account status corresponding to the target account based on the target active credibility.

[0075] Among them, the account status refers to the status corresponding to the account, which is divided into abnormal status and normal status.

[0076] Specifically, the computer device determines the account status corresponding to the target account based on the target active credibility corresponding to the target account. For example, if the target active credibility is greater than a preset threshold, the account status is normal, otherwise, the account status is abnormal. If the account status of the target account is abnormal, the computer device can generate a prompt message and send it to the relevant terminal to prompt the relevant personnel. For example, if the target account is a bank account and the account status is abnormal, a prompt message is sent to the terminal corresponding to the bank staff to prompt the bank staff that there is an abnormality in the bank account. If the account status is abnormal, the computer device can restrict the account permissions of the target account. For example, in a virtual resource transfer application, if the target account is in an abnormal state, the virtual resource borrowing share of the target account is limited, such as limiting the virtual resource borrowing share of the target account to be less than a preset share.

[0077] In one embodiment, the characteristic data of an account can be obtained, and an account knowledge graph can be established based on the characteristic data of the account. The relationship between accounts can be reflected through the account knowledge graph. If only account A is included in the account knowledge graph, the active credibility and passive credibility of account A need to be initialized. For example, the active credibility and passive credibility of account A are initialized to 0.5. When account B is added to the account knowledge graph where account A resides as the active account of account A, the association confidence between account A and account B and the passive credibility of account B need to be initialized. For example, the association confidence C(B, A) is initialized to 0.5, and the passive credibility of account B is initialized to 0.5. The current active credibility of account B can be calculated based on the association confidence C(B, A), the abnormal account association S(B, A), and the number of passive accounts corresponding to account B. Next, the active credibility corresponding to account B and the passive credibility corresponding to account A, as well as the association confidence C(B, A) between accounts A and B, are iteratively updated until the data information corresponding to accounts A and B converges. This results in the target active credibility corresponding to accounts A and B. Based on the target active credibility, the account status of accounts A and B is determined. When a new account is added to the account knowledge graph, initial values ​​are assigned to the association confidences corresponding to the new account. If the new account is not associated with any account, it has no corresponding association confidence. For the active credibility and passive credibility of a new account, if the new account is not associated with any passive account as an active account after being added to the account knowledge graph, it means that the active credibility of the new account cannot be calculated through the existing information at this time, and the active credibility of the new account needs to be initialized. If the new account is not associated with any active account as a passive account after being added to the account knowledge graph, it means that the passive credibility of the new account cannot be calculated through the existing information at this time, and the passive credibility of the new account needs to be initialized. If the new account is not associated with any account after being added to the account knowledge graph, the active credibility and passive credibility of the new account need to be initialized. For example, when account C is added to the account knowledge graph as a passive account of account A, since account C is not associated with any passive account at this time, the active credibility corresponding to account C cannot be calculated. It is necessary to initialize the association confidence between account A and account C and the active credibility of account C. For example, the association confidence C(A, C) is initialized to 0.5, and the active credibility of account C is initialized to 0.5. The current passive credibility corresponding to account C can be calculated based on the association confidence C(A, C) and the number of active accounts corresponding to account C. When account C is added to the account knowledge graph as an isolated account, the active credibility and passive credibility of account C need to be initialized. For example, the active credibility and passive credibility corresponding to account C are initialized to 0.5.Then, the data information corresponding to all accounts in the knowledge graph and all associated confidences are iteratively updated until the data information corresponding to all accounts converges, and the target active credibility corresponding to each account is obtained. The account status corresponding to each account is determined based on the target active credibility.

[0078] It is understandable that as the account knowledge graph expands, new accounts will affect old accounts, causing the target active credibility of each account to change accordingly, and the corresponding account status of each account will also change accordingly. As the account knowledge graph expands, the final account status will become increasingly accurate.

[0079] In the above-described account data processing method, by obtaining the number of passive accounts corresponding to the target account, the abnormal account association degree between the target account and each passive account is calculated based on the number of passive accounts corresponding to the target account and the number of abnormal passive accounts, as well as the number of active accounts and the number of abnormal active accounts corresponding to each passive account. This method fully utilizes the account status information of each account directly or indirectly associated with the target account, making the calculated abnormal account association degree more reliable, thereby improving the accuracy of account data processing. The current active credibility of the target account is calculated based on the abnormal account association degree and initial active association confidence between the target account and each passive account, as well as the number of passive accounts corresponding to the target account. This method fully utilizes information such as the abnormal account association degree between the target account and each passive account, the initial active association confidence, and the number of passive accounts associated with the target account, and takes into account the interrelationships between different data information. This makes the calculated current active credibility more accurate and effectively reflects the target account's trustworthiness in each passive account. The current passive credibility of a passive account is calculated based on the initial passive association confidence between the passive account and each active account, as well as the number of active accounts corresponding to the passive account. The passive association confidence reflects the confidence of the corresponding association relationship for the passive account. The calculation of the current passive credibility takes into account the passive association confidence information between the passive account and each active account. By fully integrating the passive association confidence information corresponding to each passive account, the current passive credibility obtained is more accurate and reliable. The current active association confidence between the target account and each passive account is calculated based on the abnormal account association between the target account and the passive account, the current active credibility corresponding to the target account, and the current passive credibility corresponding to the passive account. This calculation of the active association confidence fully considers data reflecting the credibility of the active and passive accounts, data reflecting the status of the association between the target account and the passive account, and the interplay between these data. This effectively improves the reliability of the active association confidence, thereby enhancing the accuracy of account data processing. Based on the difference between the initial active association confidence and the current active association confidence between the target account and the same passive account, the initial active association confidence is adjusted until convergence conditions are met, resulting in the target active association confidence between the target account and each passive account. When the active association confidences corresponding to the target account converge, the active association confidence for this event is stable and can more effectively reflect the confidence of the abnormal account associations between the target account and each passive account. Therefore, the target active credibility of the target account, calculated based on each target active association confidence, each abnormal account association, and the number of passive accounts, can also more effectively reflect the credibility of the target account. Determining the account status of the target account based on the target active credibility can improve the reliability and accuracy of account data processing.

[0080] In one embodiment, step S202 includes:

[0081] Obtain the account knowledge graph corresponding to the target account; determine the number of passive accounts, the number of active accounts, the number of abnormal passive accounts, and the number of abnormal active accounts from the account knowledge graph.

[0082] Among them, the account knowledge graph refers to the knowledge graph that reflects the relationship between various accounts.

[0083] Specifically, the computer device obtains an account knowledge graph corresponding to the target account from a local device or another device, obtains passive accounts associated with the target account from the account knowledge graph, counts the number of passive accounts associated with the target account, obtains active accounts associated with each passive account, and counts the number of active accounts associated with each passive account. The computer device obtains abnormal accounts with known abnormal account status from the passive accounts corresponding to the target account, counts the number of abnormal passive accounts, obtains abnormal accounts with known abnormal account status from the active accounts corresponding to each passive account, and counts the number of abnormal active accounts.

[0084] In the above embodiment, the computer device fully utilizes the account information in the account knowledge graph corresponding to the target account, and obtains information such as the number of abnormal accounts corresponding to the target account and each passive account, the number of passive accounts corresponding to the target account, and the number of active accounts corresponding to each passive account, thereby improving the reliability of account data processing.

[0085] In one embodiment, step S204 includes:

[0086] A first ratio is obtained based on the ratio of the number of abnormal passive accounts to the number of passive accounts; a target passive account is determined from each passive account, and a second ratio is obtained based on the ratio of the number of abnormal active accounts corresponding to the target passive account to the number of active accounts; and a degree of abnormal account association between the target account and the target passive account is obtained based on the first ratio and the second ratio.

[0087] The target passive account refers to an account selected from the passive accounts corresponding to the target account. When calculating the abnormal account correlation, each passive account needs to be used as the target passive account to calculate the abnormal account correlation between the target account and each passive account.

[0088] Specifically, the computer device calculates the ratio of the number of abnormal passive accounts to the number of passive accounts to obtain a first ratio. A passive account is selected from each passive account corresponding to the target account as the target passive account, and the ratio of the number of abnormal active accounts corresponding to the target passive account to the number of active accounts is calculated to obtain a second ratio. Based on the first ratio and the second ratio, the abnormal account association degree between the target account and the target passive account is calculated. The computer device selects each passive account corresponding to the target account as the target passive account, calculates the abnormal account association degree between the target account and the target passive account, and thereby obtains the abnormal account association degree between the target account and each passive account.

[0089] In the above embodiment, the ratio of the number of abnormal passive accounts corresponding to the target account to the number of passive accounts is calculated to obtain a first ratio, and the ratio of the number of abnormal active accounts corresponding to each passive account to the number of active accounts is calculated to obtain a second ratio. The first ratio and each second ratio make full use of the account information associated with the target account and the passive account, and can reflect the degree of association between the target account and the abnormal account and the degree of association between each passive account and the abnormal account. The abnormal account association degree between the target account and each passive account is obtained based on the first ratio and each second ratio, which can ensure the reliability of the abnormal account association degree, thereby improving the accuracy of account data processing.

[0090] In one embodiment, the abnormal account association degree between the target account and each passive account can be calculated using the following formula:

[0091]

[0092] Among them, u is the target account, v is the passive account, S(u,v) is the abnormal account correlation between the target account and the passive account, num(u) is the number of abnormal passive accounts corresponding to the target account, num(v) is the number of abnormal active accounts corresponding to the passive account, Out(u) is the number of passive accounts corresponding to the target account, In(v) is the number of active accounts corresponding to the passive account, and the value range of S(u,v) is [0,1].

[0093] In one embodiment, step S206 includes:

[0094] The abnormal account association degree and the initial active association confidence between the target account and the same passive account are integrated to obtain the normal account association value between the target account and each passive account respectively; the normal account association value is counted to obtain a first statistical result; based on the ratio of the first statistical result and the number of passive accounts, the current active credibility corresponding to the target account is obtained.

[0095] The normal account association value represents the probability that the relationship between two accounts is normal. The fewer abnormal accounts associated with a relationship, the lower the abnormal account association. Furthermore, the greater the active association confidence for this relationship, the greater the probability that the relationship is normal. The greater the corresponding normal account association value, the higher the active credibility of the target account.

[0096] Specifically, the computer device obtains the abnormal account association degree and initial active association confidence between the target account and each passive account, and integrates the abnormal account association degree and initial active association confidence between the target account and the same passive account to obtain the normal account association value between the target account and each passive account. For example, the computer device calculates the difference between the abnormal account association degree of the target account and the passive account and a preset value, and multiplies the difference by the corresponding initial active association confidence as the normal account association value; calculates the difference between the abnormal account association degree of the target account and the passive account and a preset value, and multiplies the difference by the corresponding initial active association confidence by a constant value as the normal account association value; and so on. The computer device calculates the normal account association values ​​between the target account and each passive account to obtain a first statistical result. For example, the computer device calculates the sum of the normal account association values ​​as the first statistical result; the computer device calculates the weighted sum of the normal account association values ​​as the first statistical result; and so on. The computer device calculates the ratio of the first statistical result to the passive account to obtain the current active credibility of the target account.

[0097] In the above embodiment, the abnormal account association degree and initial active association confidence between the target account and each passive account are obtained to calculate the normal account association value between the target account and each passive account. The larger the normal account association value between the target account and each passive account, the larger the first statistical result, and the higher the current active credibility of the corresponding target account. The calculation of the current active credibility fully utilizes data information such as the abnormal account association degree and the initial active association confidence that can reflect the status of the association relationship between the target account and each passive account, as well as the mutual influence between various data information, to ensure the reliability of the current active credibility of the target account, thereby improving the accuracy of account data processing.

[0098] In one embodiment, the active credibility of the target account can be calculated using the following formula:

[0099]

[0100] Among them, T(u) is the active credibility of the target account u, (u,v) is the association relationship corresponding to the target account u, Out(u,v) is the association relationship between the target account and each passive account, (1-S(u,v))×C(u,v) is the normal account association degree, C(u,v) is the active association confidence between the target account and the passive account, and the value range of T(u) is [0,1].

[0101] For example, account A corresponds to three passive accounts, namely B, C and D. In this case, Out(u,v) contains the association relationships (A,B), (A,C) and (A,D).

[0102] In one embodiment, step S208 includes:

[0103] A target passive account is determined from each passive account; each initial passive association confidence level corresponding to the target passive account is counted to obtain a second statistical result; and a current passive credibility level corresponding to the target passive account is obtained based on a ratio of the second statistical result to the number of active accounts corresponding to the target passive account.

[0104] The target passive account refers to an account selected from the passive accounts corresponding to the target account. When calculating the current passive credibility, each passive account needs to be used as the target passive account to calculate the current passive credibility corresponding to each passive account.

[0105] Specifically, the computer device selects a target passive account from among the passive accounts of the target account, obtains the initial passive association confidences corresponding to the target passive account, and calculates the initial passive association confidences to obtain a second statistical result. For example, the sum of the initial passive association confidences corresponding to the target passive account may be used as the second statistical result; the weighted sum of the initial passive association confidences corresponding to the target passive account may be used as the second statistical result; and so on. The computer device calculates the ratio of the second statistical result to the number of active accounts corresponding to the target passive account to obtain the current passive confidence of the target passive account.

[0106] In the above embodiment, since an account with a higher passive credibility is more likely to have some associations with a high passive association confidence, the initial passive association confidence corresponding to each passive account is obtained to calculate the passive credibility of the passive account, which can improve the reliability of the passive credibility and thus improve the accuracy of account data processing.

[0107] In one embodiment, the passive credibility of a passive account can be calculated using the following formula:

[0108]

[0109] Among them, R(v) is the passive credibility of passive account v, (u,v) is the account association information corresponding to passive account v, In(u,v) is the association relationship between passive account v and each active account associated with the passive account, C(u,v) is the passive association confidence between passive account v and active accounts, In(v) is the number of active accounts corresponding to the passive account, and the value range of R(v) is [0,1].

[0110] For example, account A corresponds to three active accounts: B, C, and D. In this case, the associations contained in In(u,v) are (B,A), (C,A), and (D,A). C(B,A), C(C,A), and C(D,A) are the passive association confidences between account A and each active account, respectively, and are also the active association confidences corresponding to accounts B, C, and D, respectively.

[0111] In one embodiment, step S210 includes:

[0112] A target passive account is determined from each passive account; based on the abnormal account association between the target account and the target passive account and the current active credibility corresponding to the target account, an intermediate association confidence value is obtained; based on the difference between the current passive credibility corresponding to the target passive account and the intermediate association confidence value, a current active association confidence value between the target account and the target passive account is obtained.

[0113] The target passive account refers to an account selected from the passive accounts corresponding to the target account. When calculating the current active association confidence, each passive account is used as the target passive account to calculate the current active association confidence between the target account and each passive account. The median association confidence value is the fusion of the current active credibility of the target account and the abnormal account association degree, resulting in data reflecting the active association confidence between the target account and the target passive account. The smaller the median association confidence value, the greater the active association confidence.

[0114] Specifically, the higher the active association confidence, the closer the corresponding non-abnormal account association will be to the active account's current active credibility. The non-abnormal account association refers to the difference between the preset value and the abnormal account association, and is used to indicate the likelihood that the association between two accounts is normal. The non-abnormal account association and the abnormal account association are negatively correlated. The higher the active association confidence corresponding to the association, the higher the confidence in the abnormal account association. When the abnormal account association corresponding to the target account is larger, the likelihood that this association is abnormal is greater, that is, the non-abnormal account association is smaller, and the current active credibility of the target account is smaller. When the abnormal account association corresponding to the target account is smaller, the likelihood that this association is abnormal is smaller, that is, the non-abnormal account association is larger, and the current active credibility of the target account is larger. Therefore, the higher the active association confidence, the closer the corresponding non-abnormal account association is to the active account's current active credibility. The computer device determines the target passive account from among the passive accounts corresponding to the target account, obtains the abnormal account association degree between the target account and the target passive account and the current active credibility corresponding to the target account, and calculates the non-abnormal account association degree. An intermediate association confidence value is obtained based on the non-abnormal account association degree and the current active credibility. For example, the difference between the non-abnormal account association degree and the current active credibility is used as the intermediate association confidence value; the difference between the non-abnormal account association degree and the current active credibility is calculated, and the difference is multiplied by a preset weight to obtain the intermediate association confidence value. Since a passive account with a higher passive credibility is more likely to have an association relationship with a higher passive association confidence, the active association confidence corresponding to the target account is obtained based on the difference between the passive credibility and the intermediate association confidence value.

[0115] In the above embodiment, a target passive account is identified from among the passive accounts corresponding to the target account, the abnormal account association degree between the target account and the target passive account and the active credibility of the target account are obtained, an intermediate association confidence value is obtained based on the abnormal account association degree and the active credibility of the target account, and the active association confidence value corresponding to the target account is obtained based on the intermediate association confidence value and the passive credibility value. This calculation of the active association confidence value fully considers data information reflecting the credibility of the target account and the passive account, data information reflecting the status of the association relationship between the target account and the passive account, and the mutual influence between various data information, effectively improving the reliability of the active association confidence value and thereby improving the accuracy of account data processing.

[0116] In one embodiment, the active association confidence can be calculated using the following formula:

[0117]

[0118] Among them, C(u,v) is the active association confidence between the target account u and the passive account v, and the value range of C(u,v) is [0,1].

[0119] In one embodiment, Figure 3 As shown, step S212 includes:

[0120] Step S302, based on the difference between the initial active association confidence and the current active association confidence between the target account and the same passive account, the difference between the initial active credibility and the current active credibility corresponding to the target account, and the difference between the initial passive credibility and the current passive credibility corresponding to the same passive account, adjust the corresponding initial active association confidence, initial active credibility, and initial passive credibility to obtain the intermediate active association confidence, intermediate active credibility, and intermediate passive credibility.

[0121] Step S304 : Using the intermediate active association confidence, the intermediate active credibility, and the intermediate passive credibility as the initial active association confidence, the initial active credibility, and the initial passive credibility, respectively.

[0122] Step S306 returns to the step of obtaining the current active credibility corresponding to the target account based on the number of passive accounts, each initial active association confidence, and each abnormal account association, and executes until the convergence condition is met to obtain the target active association confidence between the target account and each passive account.

[0123] Among them, the convergence conditions include that the differences between each current active association confidence and the corresponding initial active association confidence are less than a first threshold, the difference between the current active credibility and the initial active credibility is less than a second threshold, and the difference between each current passive credibility and the corresponding initial passive credibility is less than a third threshold.

[0124] Specifically, each indicator (initial active association confidence, initial active credibility and initial passive credibility) can be updated through the idea of ​​recursive iteration. When all indicators converge, the target active association confidence, target active credibility and target passive credibility are obtained. Subsequently, the account status is determined based on the target active association confidence that integrates relevant information of active credibility, passive credibility and abnormal account correlation.

[0125] During recursive iterations, the computer device calculates the difference between the initial active association confidence and the current active association confidence between the target account and the same passive account, the difference between the initial active credibility and the current active credibility corresponding to the target account, and the difference between the initial passive credibility and the current passive credibility corresponding to the same passive account, and determines whether convergence conditions are met. If convergence conditions are met, the initial active credibility or the current active credibility is used as the target active credibility. If the convergence condition is not met, the current active association confidence, the current active credibility, and the current passive credibility are used as the intermediate active association confidence, the intermediate active credibility, and the intermediate passive credibility, and the intermediate active association confidence, the intermediate active credibility, and the intermediate passive credibility are used as the initial active association confidence, the initial active credibility, and the initial passive credibility, respectively. The step of obtaining the current active credibility corresponding to the target account based on the number of passive accounts, each initial active association confidence, and each abnormal account association is returned and executed for iterative processing until the difference between the initial active association confidence and the current active association confidence between the target account and the same passive account, the difference between the initial active credibility and the current active credibility corresponding to the target account, and the difference between the initial passive credibility and the current passive credibility corresponding to the same passive account are all less than the convergence threshold in the convergence condition, that is, until the convergence condition is met, and the current active association confidence that meets the convergence condition is used as the target active association confidence.

[0126] In the above embodiment, the computer device determines whether convergence conditions are met based on the difference between the initial active association confidence and the current active association confidence between the target account and the same passive account, the difference between the initial active credibility and the current active credibility corresponding to the target account, and the difference between the initial passive credibility and the current passive credibility corresponding to the same passive account. If the convergence conditions are not met, the iterative calculation continues until convergence. When the various data information converges, it indicates that the active association confidence is in a stable state, which can more effectively reflect the confidence level of abnormal account associations between the target account and each passive account. Determining the account status based on the target active association confidence obtained after data convergence helps improve the accuracy of account data processing.

[0127] In one embodiment, step S216 includes:

[0128] An account risk value is obtained based on the target active credibility; the account risk value and the target active credibility are negatively correlated; when the account risk value is greater than a preset threshold, the account status is determined to be abnormal; when the account risk value is less than or equal to the preset threshold, the account status is determined to be normal.

[0129] The account risk value is calculated from the target active credibility and can be used to reflect the probability that the target account is an abnormal account. The preset threshold can be set according to actual needs.

[0130] Specifically, the computer device obtains an account risk value corresponding to the target account based on the difference between the target active credibility and a preset value. It is understood that the higher the target active credibility of the target account, the smaller the difference between the target active credibility and the preset value, and the smaller the corresponding account risk value, that is, the account risk value and the target active credibility are negatively correlated. For example, the difference between the preset value and the target active credibility can be used as the account risk value; the difference between the preset value and the target active credibility can be calculated, and the product of the difference and a preset multiple can be used as the account risk value; and so on. The account risk value is compared with a preset threshold. When the account risk value is greater than the preset threshold, the account status corresponding to the target account is determined to be abnormal; when the account risk value is less than or equal to the preset threshold, the account status corresponding to the target account is determined to be normal.

[0131] In one embodiment, the account risk value can be calculated using the following formula:

[0132] O(u)=10×(1-T(u))

[0133] Where O(u) is the account risk value of account u, T(u) is the target active credibility of account u, and the value range of O(u) is [0,10].

[0134] In the above embodiment, since the target active credibility and the account risk value corresponding to the target account are negatively correlated, the account risk value calculated based on the target active credibility corresponding to the target account can effectively reflect the status information of the target account and obtain a more accurate account data processing result.

[0135] In a specific embodiment, the account data processing method of the present application can be applied to account data processing of loan accounts. An account knowledge graph is established based on the loan application information of the account, and a risk propagation model is built based on the data in the account knowledge graph. The risk propagation model includes three interdependent indicators (active association confidence, active credibility and passive credibility) and an indicator for measuring the correlation between the account and known abnormal loan accounts (abnormal account correlation). Each indicator is updated through the idea of ​​recursive iteration. When all indicators converge, the overdue risk value of each account can be calculated, and based on the relationship between the overdue risk value and the set threshold, it is determined whether the account is an abnormal loan account.

[0136] The account data processing method includes the following steps:

[0137] 1. Obtain account quantity information based on the target account’s account knowledge graph

[0138] like Figure 4 As shown in the figure, each account has five aspects of characteristic data, namely account characteristics, device characteristics, bank card characteristics, ID card characteristics, and unit characteristics. At the same time, the account status of known abnormal loan accounts is determined to be abnormal. Based on the association relationship between the target account and each account, the computer device connects the target account with other accounts to obtain the account knowledge graph corresponding to the target account. For example, Figure 5 As shown, when account 2 registers account 1 as its contact, there is an association relationship between account 2 and account 1, and in this association, account 2 is the active account and account 1 is the passive account. When account 1 registers account 4 as its contact, there is an association relationship between account 1 and account 4, and in this association, account 1 is the active account and account 4 is the passive account. When account 1 invites account 3 to register as a user, there is an association relationship between account 1 and account 3, and in this association, account 1 is the active account and account 3 is the passive account. The computer device obtains the number of passive accounts corresponding to the target account and the number of active accounts corresponding to each passive account of the target account, identifies abnormal accounts among the passive accounts corresponding to the target account, counts the number of abnormal passive accounts corresponding to the target account, identifies abnormal accounts among the active accounts corresponding to each passive account, and counts the number of abnormal active accounts corresponding to each passive account.

[0139] 2. Determine the correlation between abnormal accounts

[0140] The computer device obtains the abnormal account association degree between the target account and each passive account based on the number of abnormal passive accounts and the number of passive accounts and the number of abnormal active accounts and the number of active accounts corresponding to each passive account.

[0141] The computer device calculates the abnormal account association degree using the following formula:

[0142]

[0143] 3. Determine the current active credibility of the target account

[0144] The computer device obtains the initial active association confidence between the target account and the corresponding passive account. When first calculating the current active credibility, it assigns the same initial value of 0.5 to each initial active association confidence. Based on the number of passive accounts, each initial active association confidence, and each abnormal account association, the current active credibility corresponding to the target account is calculated.

[0145] The computer device calculates the current active credibility using the following formula:

[0146]

[0147] 4. Determine the current passive credibility of each passive account

[0148] The computer device obtains an initial passive association confidence between the passive account and the corresponding active account, and obtains a current passive credibility corresponding to each passive account based on the number of active accounts corresponding to the same passive account and each initial passive association confidence.

[0149] The computer device calculates the current passive credibility using the following formula:

[0150]

[0151] 5. Determine the confidence level of each active association

[0152] Based on the abnormal account association between the target account and the passive account, the current active credibility corresponding to the target account, and the current passive credibility corresponding to the passive account, the current active association confidence between the target account and the passive account is calculated to obtain the current active association confidence between the target account and each passive account.

[0153] The computer device calculates the current active association confidence using the following formula:

[0154]

[0155] 6. Determine the confidence level of target active association

[0156] The computer device determines whether a convergence condition is satisfied based on the difference between the initial active association confidence and the current active association confidence between the target account and the same passive account, the difference between the initial active credibility and the current active credibility corresponding to the target account, and the difference between the initial passive credibility and the current passive credibility corresponding to the same passive account. If the convergence condition is not satisfied, the current active association confidence, the current active credibility, and the current passive credibility are used as the intermediate active association confidence, the intermediate active credibility, and the intermediate passive credibility. The intermediate active association confidence, the intermediate active credibility, and the intermediate passive credibility are used as the initial active association confidence, the initial active credibility, and the initial passive credibility, respectively. The step of obtaining the current active credibility corresponding to the target account based on the number of passive accounts, each initial active association confidence, and each abnormal account association is returned and executed until the convergence condition is satisfied. When all data information has converged, the current active association confidence is used as the target active association confidence.

[0157] 7. Determine account status

[0158] The account risk value is obtained based on the target active credibility.

[0159] The computer device calculates the account risk value using the following formula:

[0160] O(u)=10×(1-T(u))

[0161] When the account risk value is greater than a preset threshold (for example, set to 7), the account status corresponding to the target account is determined to be abnormal, and the target account is an abnormal lending account. When the account risk value is less than or equal to the preset threshold, the account status corresponding to the target account is determined to be normal, and the target account is a normal lending account.

[0162] In the above embodiment, when processing the account data of the target account, the overall process is as follows: Figure 6 As shown, basic account information is collected and an account knowledge graph is established. Indicator values ​​and corresponding formulas for abnormal account association, active credibility, passive credibility, and active association confidence are designed. Each indicator value is then initialized and iteratively updated. When all indicator values ​​converge, the account risk value is calculated. Finally, the account risk value of each account is compared with the preset threshold to determine whether the account status is abnormal. This solution is not negatively affected by sample imbalance. It can utilize and disseminate effective information about each account, thereby improving the effectiveness and accuracy of detecting abnormal loan accounts in data with unbalanced samples. Compared to manual identification of abnormal loan accounts, this solution reduces labor costs and increases the speed and efficiency of detecting abnormal loan accounts in large-scale data.

[0163] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0164] Based on the same inventive concept, embodiments of the present application further provide an account data processing device for implementing the aforementioned account data processing method. The solution provided by this device is similar to the solution described in the aforementioned method. Therefore, the specific limitations of one or more of the following embodiments of the account data processing device can be found in the above-described limitations of the account data processing method and will not be further elaborated here.

[0165] In one embodiment, Figure 7As shown, an account data processing device is provided, comprising: an account quantity information acquisition module 702, an abnormal account association calculation module 704, a current active credibility calculation module 706, a current passive credibility calculation module 708, a current active association confidence calculation module 710, a target active association confidence determination module 712, a target active credibility calculation module 714, and a target account status determination module 716, wherein:

[0166] The account quantity information acquisition module 702 is used to obtain the number of passive accounts corresponding to the target account, obtain the number of active accounts corresponding to each passive account of the target account, determine the number of abnormal passive accounts from the number of passive accounts, and determine the number of abnormal active accounts from the number of active accounts.

[0167] The abnormal account association calculation module 704 is used to calculate the abnormal account association between the target account and the passive account based on the number of passive accounts, the number of active accounts, the number of abnormal passive accounts, and the number of abnormal active accounts, and obtain the abnormal account association between the target account and each passive account.

[0168] The current active credibility calculation module 706 is used to obtain the initial active association confidence between the target account and the corresponding passive account, and obtain the current active credibility corresponding to the target account based on the number of passive accounts, the initial active association confidences and the associations of the abnormal accounts.

[0169] The current passive credibility calculation module 708 is used to obtain the initial passive association confidence between the passive account and the corresponding active account, and obtain the current passive credibility corresponding to each passive account based on the number of active accounts corresponding to the same passive account and the initial passive association confidence.

[0170] The current active association confidence calculation module 710 is used to calculate the current active association confidence between the target account and the passive account based on the abnormal account association between the target account and the passive account, the current active credibility corresponding to the target account, and the current passive credibility corresponding to the passive account, and obtain the current active association confidence between the target account and each passive account.

[0171] The target active association confidence determination module 712 is used to adjust the corresponding initial active association confidence based on the difference between the initial active association confidence and the current active association confidence between the target account and the same passive account until the convergence condition is met, thereby obtaining the target active association confidence between the target account and each passive account.

[0172] The target active credibility calculation module 714 is used to obtain the target active credibility corresponding to the target account based on the target active association confidence levels, abnormal account association levels, and the number of passive accounts corresponding to the target account.

[0173] The target account status determination module 716 is configured to determine the account status corresponding to the target account based on the target active credibility.

[0174] The above-mentioned account data processing device calculates the abnormal account association degree between the target account and each passive account based on the number of passive accounts and the number of abnormal passive accounts corresponding to the target account, as well as the number of active accounts and the number of abnormal active accounts corresponding to each passive account. This device fully utilizes the account status information of each account directly or indirectly associated with the target account, making the calculated abnormal account association degree more reliable, thereby improving the accuracy of account data processing. The current active credibility of the target account is calculated based on the abnormal account association degree and initial active association confidence between the target account and each passive account, as well as the number of passive accounts corresponding to the target account. This device fully utilizes information such as the abnormal account association degree between the target account and each passive account, the initial active association confidence, and the number of passive accounts associated with the target account, and takes into account the interrelationships between different data information. This makes the calculated current active credibility more accurate and effectively reflects the trustworthiness of the target account with respect to each passive account. The current passive credibility of a passive account is calculated based on the initial passive association confidence between the passive account and each active account, as well as the number of active accounts corresponding to the passive account. The passive association confidence reflects the confidence of the corresponding association relationship for the passive account. The calculation of the current passive credibility takes into account the passive association confidence information between the passive account and each active account. By fully integrating the passive association confidence information corresponding to each passive account, the current passive credibility obtained is more accurate and reliable. The current active association confidence between the target account and each passive account is calculated based on the abnormal account association between the target account and the passive account, the current active credibility corresponding to the target account, and the current passive credibility corresponding to the passive account. This calculation of the active association confidence fully considers data reflecting the credibility of the active and passive accounts, data reflecting the status of the association between the target account and the passive account, and the interplay between these data. This effectively improves the reliability of the active association confidence, thereby enhancing the accuracy of account data processing. Based on the difference between the initial active association confidence and the current active association confidence between the target account and the same passive account, the initial active association confidence is adjusted until convergence conditions are met, resulting in target active association confidences for the target account and each passive account. When the active association confidences corresponding to the target account converge, the active association confidences are now stable and can more effectively reflect the confidence of the abnormal account associations between the target account and each passive account. Therefore, the target active credibility of the target account, calculated based on each target active association confidence, each abnormal account association, and the number of passive accounts, can also more effectively reflect the credibility of the target account. Determining the account status of the target account based on the target active credibility can improve the reliability and accuracy of account data processing.

[0175] In one embodiment, the account quantity information acquisition module 702 is further configured to:

[0176] Obtain the account knowledge graph corresponding to the target account; determine the number of passive accounts, the number of active accounts, the number of abnormal passive accounts, and the number of abnormal active accounts from the account knowledge graph.

[0177] In one embodiment, the abnormal account relevance calculation module 704 is further configured to:

[0178] A first ratio is obtained based on the ratio of the number of abnormal passive accounts to the number of passive accounts; a target passive account is determined from each passive account, and a second ratio is obtained based on the ratio of the number of abnormal active accounts corresponding to the target passive account to the number of active accounts; and a degree of abnormal account association between the target account and the target passive account is obtained based on the first ratio and the second ratio.

[0179] In one embodiment, the current active credibility calculation module 706 is further configured to:

[0180] The abnormal account association degree and the initial active association confidence between the target account and the same passive account are integrated to obtain the normal account association value between the target account and each passive account respectively; the normal account association value is counted to obtain a first statistical result; based on the ratio of the first statistical result and the number of passive accounts, the current active credibility corresponding to the target account is obtained.

[0181] In one embodiment, the current passive credibility calculation module 708 is further configured to:

[0182] A target passive account is determined from each passive account; each initial passive association confidence level corresponding to the target passive account is counted to obtain a second statistical result; and a current passive credibility level corresponding to the target passive account is obtained based on a ratio of the second statistical result to the number of active accounts corresponding to the target passive account.

[0183] In one embodiment, the current active association confidence calculation module 710 is further configured to:

[0184] A target passive account is determined from each passive account; based on the abnormal account association between the target account and the target passive account and the current active credibility corresponding to the target account, an intermediate association confidence value is obtained; based on the difference between the current passive credibility corresponding to the target passive account and the intermediate association confidence value, a current active association confidence value between the target account and the target passive account is obtained.

[0185] In one embodiment, the target active association confidence determination module 712 is further configured to:

[0186] Based on the difference between the initial active association confidence and the current active association confidence between the target account and the same passive account, the corresponding initial active association confidence is adjusted to obtain an intermediate active association confidence, and the intermediate active association confidence is used as the initial active association confidence; based on the difference between the initial active credibility and the current active credibility between the target account and the same passive account, the corresponding initial active credibility is adjusted to obtain an intermediate active credibility, and the intermediate active credibility is used as the initial active credibility; based on the difference between the initial passive credibility and the current passive credibility corresponding to the same passive account, the corresponding initial passive credibility is adjusted to obtain an intermediate passive credibility, and the intermediate passive credibility is used as the initial passive credibility; the step of obtaining the current active credibility corresponding to the target account based on the number of passive accounts, each initial active association confidence and each abnormal account association is returned and executed until the convergence condition is met, and the target active association confidence between the target account and each passive account is obtained.

[0187] In one embodiment, the target account status determination module 716 is further configured to:

[0188] An account risk value is obtained based on the target active credibility; the account risk value and the target active credibility are negatively correlated; when the account risk value is greater than a preset threshold, the account status is determined to be abnormal; when the account risk value is less than or equal to the preset threshold, the account status is determined to be normal.

[0189] Each module in the aforementioned account data processing device may be implemented in whole or in part through software, hardware, or a combination thereof. Each module may be embedded in or independent of a processor in a computer device in the form of hardware, or may be stored in a memory in the computer device in the form of software, so that the processor can call and execute the corresponding operations of each module.

[0190] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 8As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O) and a communication interface. The processor, memory and input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store data such as the number of passive accounts, the number of active accounts, the number of abnormal passive accounts, and the number of abnormal active accounts. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, an account data processing method is implemented.

[0191] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 9 As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected via a system bus, and the communication interface, the display unit and the input device are connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, an account data processing method is implemented. The display unit of the computer device is used to form a visually visible image, and can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device casing, or an external keyboard, touchpad or mouse, etc.

[0192] Those skilled in the art will understand that Figure 8 、 9The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0193] In one embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps in the above method embodiments when executing the computer program.

[0194] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0195] In one embodiment, a computer program product or computer program is provided, comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the steps of each of the above-described method embodiments.

[0196] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions.

[0197] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processor involved in the various embodiments provided herein may be, but are not limited to, a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, and the like.

[0198] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0199] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A method for processing account data, characterized in that: The method comprises: Obtaining the number of passive accounts corresponding to a target account, obtaining the number of active accounts corresponding to each passive account of the target account, determining the number of abnormal passive accounts from the number of passive accounts, and determining the number of abnormal active accounts from the number of active accounts, including: obtaining an account knowledge graph corresponding to the target account; determining the number of passive accounts, the number of active accounts, the number of abnormal passive accounts, and the number of abnormal active accounts from the account knowledge graph; Based on the number of passive accounts, the number of active accounts, the number of abnormal passive accounts, and the number of abnormal active accounts, calculating the abnormal account association degree between the target account and the passive accounts, and obtaining the abnormal account association degree between the target account and each passive account; Obtaining an initial active association confidence between the target account and the corresponding passive account, and obtaining a current active credibility corresponding to the target account based on the number of passive accounts, the initial active association confidences, and the associations of the abnormal accounts; Obtaining an initial passive association confidence between the passive account and the corresponding active account, and obtaining a current passive credibility corresponding to each passive account based on the number of active accounts corresponding to the same passive account and each initial passive association confidence; Identify target passive accounts from among various passive accounts; Obtaining an association confidence intermediate value based on the abnormal account association between the target account and the target passive account and the current active credibility corresponding to the target account; Obtaining a current active association confidence between the target account and the target passive account based on a difference between a current passive credibility corresponding to the target passive account and the intermediate value of the association confidence; Based on the difference between the initial active association confidence and the current active association confidence between the target account and the same passive account, adjusting the corresponding initial active association confidence until a convergence condition is satisfied, thereby obtaining target active association confidences between the target account and each passive account; Obtaining a target active credibility corresponding to the target account based on each target active association confidence level corresponding to the target account, each abnormal account association level, and the number of passive accounts; Based on the target active credibility, an account status corresponding to the target account is determined.

2. The method according to claim 1, characterized in that The calculating, based on the number of passive accounts, the number of active accounts, the number of abnormal passive accounts, and the number of abnormal active accounts, the abnormal account association degree between the target account and the passive accounts, and obtaining the abnormal account association degree between the target account and each passive account, includes: Obtaining a first ratio based on a ratio of the number of abnormal passive accounts to the number of passive accounts; Determining a target passive account from the passive accounts, and obtaining a second ratio based on a ratio of the number of abnormal active accounts to the number of active accounts corresponding to the target passive account; Based on the first ratio and the second ratio, an abnormal account association degree between the target account and the target passive account is obtained.

3. The method according to claim 1, characterized in that The obtaining of the current active credibility corresponding to the target account based on the number of passive accounts, each initial active association confidence level, and each abnormal account association level includes: The abnormal account association degree between the target account and the same passive account and the initial active association confidence are integrated to obtain the normal account association value between the target account and each passive account; Counting the association values ​​of each normal account to obtain a first statistical result; Based on the ratio of the first statistical result and the number of the passive accounts, the current active credibility corresponding to the target account is obtained.

4. The method according to claim 1, wherein The method of obtaining the current passive credibility corresponding to each passive account based on the number of active accounts corresponding to the same passive account and each initial passive association confidence level includes: Identify target passive accounts from among various passive accounts; Counting the initial passive association confidences corresponding to the target passive account to obtain a second statistical result; Based on the ratio of the second statistical result to the number of active accounts corresponding to the target passive account, the current passive credibility corresponding to the target passive account is obtained.

5. The method according to claim 1, wherein The step of adjusting the corresponding initial active association confidence based on the difference between the initial active association confidence and the current active association confidence between the target account and the same passive account until a convergence condition is satisfied, thereby obtaining the target active association confidence between the target account and each passive account, including: Based on the difference between the initial active association confidence and the current active association confidence between the target account and the same passive account, the difference between the initial active credibility and the current active credibility corresponding to the target account, and the difference between the initial passive credibility and the current passive credibility corresponding to the same passive account, adjusting the corresponding initial active association confidence, initial active credibility, and initial passive credibility to obtain an intermediate active association confidence, an intermediate active credibility, and an intermediate passive credibility; using the intermediate active association confidence, the intermediate active credibility, and the intermediate passive credibility as the initial active association confidence, the initial active credibility, and the initial passive credibility, respectively; The step of obtaining the current active credibility corresponding to the target account based on the number of passive accounts, the initial active association confidences, and the abnormal account associations is returned and executed until a convergence condition is met, thereby obtaining the target active association confidences between the target account and each passive account.

6. The method according to claim 1, characterized in that The determining, based on the target active credibility, an account status corresponding to the target account includes: Obtaining an account risk value based on the target active credibility; wherein the account risk value and the target active credibility are negatively correlated; When the account risk value is greater than a preset threshold, determining that the account status is an abnormal state; When the account risk value is less than or equal to a preset threshold, the account status is determined to be normal.

7. An account data processing device, characterized in that: The device comprises: An account quantity information acquisition module is configured to acquire the number of passive accounts corresponding to a target account, acquire the number of active accounts corresponding to each passive account of the target account, determine the number of abnormal passive accounts from the number of passive accounts, and determine the number of abnormal active accounts from the number of active accounts, including: acquiring an account knowledge graph corresponding to the target account; determining the number of passive accounts, the number of active accounts, the number of abnormal passive accounts, and the number of abnormal active accounts from the account knowledge graph; an abnormal account association calculation module, configured to calculate the abnormal account association between the target account and the passive account based on the number of passive accounts, the number of active accounts, the number of abnormal passive accounts, and the number of abnormal active accounts, and obtain the abnormal account association between the target account and each passive account; a current active credibility calculation module, configured to obtain the initial active association confidence between the target account and the corresponding passive account, and obtain the current active credibility corresponding to the target account based on the number of passive accounts, the initial active association confidences, and the associations of the abnormal accounts; a current passive credibility calculation module, configured to obtain an initial passive association confidence between the passive account and the corresponding active account, and obtain a current passive credibility corresponding to each passive account based on the number of active accounts corresponding to the same passive account and the initial passive association confidences; a current active association confidence calculation module configured to determine a target passive account from each passive account; obtain an association confidence median based on the abnormal account association between the target account and the target passive account and the current active credibility corresponding to the target account; and obtain a current active association confidence between the target account and the target passive account based on the difference between the current passive credibility corresponding to the target passive account and the association confidence median; a target active association confidence determination module, configured to adjust the corresponding initial active association confidence based on the difference between the initial active association confidence and the current active association confidence between the target account and the same passive account until a convergence condition is satisfied, thereby obtaining the target active association confidence between the target account and each passive account; a target active credibility calculation module, configured to obtain the target active credibility corresponding to the target account based on the target active association confidences of each target account, the association degrees of each abnormal account, and the number of passive accounts; The target account status determination module is used to determine the account status corresponding to the target account based on the target active credibility.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images