Illegal account identification method and device

Abstract

The embodiment of the invention provides an illegal account identification method and device, and relates to the technical field of data processing, and the method comprises the steps: obtaining behavior data of an account for a target transaction behavior in a transaction period; according to the behavior data, determining a similarity characterization value of transaction of two accounts included in each account group in the transaction period; based on the similarity representation value corresponding to each account group in a plurality of continuous transaction cycles, determining an accumulated similarity of transaction of two accounts included in each account group; and according to the accumulated similarity corresponding to each account group, determining an illegal account group, and determining that accounts contained in the illegal account group are illegal accounts. By applying the illegal account identification scheme provided by the embodiment of the invention, the accuracy of identifying the illegal account can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to a method and device for identifying illegal accounts. Background Art

[0002] Convergent trading belongs to an abnormal trading in the securities trading market, and the accounts conducting convergent trading are illegal accounts. In order to ensure the normal operation of the securities trading market, it is necessary to identify the illegal accounts conducting convergent trading.

[0003] Existing identification schemes use neural network models to identify illegal accounts, and this model needs to use the transaction data of illegal accounts conducting convergent trading as samples for model training. However, since the amount of real transaction data of illegal accounts conducting convergent trading in the securities trading market is small, the samples used for training the neural network model are few. Therefore, the inference ability of the trained neural network model is low, resulting in low accuracy in identifying illegal accounts using the neural network model. Summary of the Invention

[0004] The purpose of the embodiments of the present invention is to provide a method and device for identifying illegal accounts to improve the accuracy of identifying illegal accounts. The specific technical solutions are as follows:

[0005] In a first aspect, an embodiment of the present application provides a method for identifying illegal accounts, and the method includes:

[0006] Obtain the behavior data of an account for a target trading behavior during a trading period, where the target trading behavior includes: the behavior of buying and / or selling a securities product, and the behavior data represents whether the account has a target trading behavior for each securities product;

[0007] According to the behavior data, determine the similarity representation value of the transactions between two accounts included in each account group during the trading period;

[0008] Based on the similarity representation values corresponding to each account group in consecutive multiple trading periods, determine the cumulative similarity of the transactions between two accounts included in each account group;

[0009] According to the cumulative similarity corresponding to each account group, determine the illegal account group, and determine the accounts included in the illegal account group as illegal accounts.

[0010] In a second aspect, an embodiment of the present application provides a device for identifying illegal accounts, and the device includes:

[0011] A data acquisition module, configured to acquire behavioral data of an account for a target trading behavior during a trading period, where the target trading behavior includes: a behavior of buying and / or selling a securities product, and the behavioral data characterizes whether the account has a target trading behavior for each securities product;

[0012] A characterization value determination module, configured to determine a similarity characterization value of trading between two accounts included in each account group during the trading period according to the behavioral data;

[0013] A similarity determination module, configured to determine a cumulative similarity of trading between two accounts included in each account group based on the similarity characterization values corresponding to each account group in a plurality of consecutive trading periods;

[0014] An account determination module, configured to determine a group of violating accounts according to the cumulative similarity corresponding to each account group, and determine the accounts included in the group of violating accounts as violating accounts.

[0015] In a third aspect, an embodiment of the present application provides an electronic device, including a processor, a communication interface, a memory, and a communication bus, where the processor, the communication interface, and the memory complete communication with each other through the communication bus;

[0016] The memory is configured to store a computer program;

[0017] The processor is configured to implement the method described in the first aspect when executing the program stored on the memory.

[0018] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, where a computer program is stored in the computer-readable storage medium, and the computer program implements the method described in the first aspect when executed by a processor.

[0019] Advantageous effects of the embodiments of the present invention:

[0020] As can be seen from the above, when identifying a violation account by applying the solution provided in the embodiments of the present application, the behavior data can characterize whether the account has a target trading behavior for each securities product. According to the behavior data, the similarity characterization value of the transactions between the two accounts included in each account group within the trading cycle can be accurately determined. Then, based on the similarity characterization values corresponding to each account group within multiple consecutive trading cycles, the cumulative similarity of the transactions between the two accounts included in each account group is determined. This cumulative similarity can be understood as: the similarity of the transactions between the two accounts included in the account group within a relatively long period of multiple consecutive trading cycles. If the similarity of the transactions between the two accounts included in the account group is relatively high within this relatively long period, it can be accurately determined that the trading behaviors of these two accounts are relatively similar, and most of the securities products they trade are the same. Therefore, based on the cumulative similarity corresponding to each account group, the violation account group can be accurately and comprehensively determined, and further, the accounts included in the violation account group can be accurately and comprehensively determined as violation accounts. It can be seen that applying the violation account identification solution provided in the embodiments of the present application can improve the accuracy and identification effect of identifying violation accounts. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention, and those of ordinary skill in the art can also obtain other embodiments based on these drawings.

[0022] Figure 1 It is a flowchart of the first method for identifying a violation account provided in the embodiments of the present application;

[0023] Figure 2 It is a flowchart of the second method for identifying a violation account provided in the embodiments of the present application;

[0024] Figure 3 It is a flowchart of the third method for identifying a violation account provided in the embodiments of the present application;

[0025] Figure 4 It is a structural schematic diagram of a violation account identification device provided in the embodiments of the present application;

[0026] Figure 5 It is a structural schematic diagram of an electronic device provided in the embodiments of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0027] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art based on this application belong to the scope of protection of the present invention.

[0028] First, some concepts involved in the embodiments of this application will be introduced.

[0029] 1. Convergent trading

[0030] Convergent trading refers to the trading in which different accounts conduct the same trading behavior on the same security product within the same time period.

[0031] There are three types of convergent trading, namely, buy-convergent and sell-divergent, buy-divergent and sell-convergent, and buy-sell both convergent trading.

[0032] Buy-convergent means that different accounts all buy the same security product within the same time period; sell-convergent means that different accounts all sell the same security product within the same time period.

[0033] 2. Account group

[0034] An account group is a combination of two accounts. When identifying illegal accounts, the combination of every two accounts can be used as an account group, and based on various data generated by the trading of the two accounts included in the account group, it is determined whether the two accounts included in the account group are illegal accounts.

[0035] 3. Security product

[0036] The security products involved in this solution are any one of various types of products such as stocks, bonds, funds, etc.

[0037] Next, the illegal account identification method and device provided in the embodiments of this application will be described in detail.

[0038] In one embodiment of this application, referring to Figure 1 , a flow diagram of the first illegal account identification method is provided. In this embodiment, the above method includes the following steps S101-S104.

[0039] Step S101: Obtain the behavior data of the account for the target trading behavior during the trading cycle.

[0040] Among them, the behavior data represents whether the account has a target trading behavior for each security product.

[0041] The target trading behavior includes: the behavior of buying and / or selling security products.

[0042] There are three situations of target trading behaviors, namely: the first situation that the target trading behavior only includes the behavior of buying securities products, the second situation that the target trading behavior only includes the behavior of selling securities products, and the third situation that the target trading behavior includes the behavior of buying and selling securities products.

[0043] The methods for identifying illegal accounts in the first and second cases are similar. The first case is taken as an example to illustrate the method for identifying illegal accounts provided by the embodiment of the present application. In addition, the method for identifying illegal accounts in the third case can be found in the subsequent embodiments, which will not be described in detail here.

[0044] In the first case of the target transaction behavior, the above behavior data indicates whether the account has purchased each securities product. If account a purchases securities product b during the transaction cycle, the behavior data indicates that account a has generated the target transaction behavior for securities product b, that is, account a has purchased securities product b.

[0045] Specifically, accounts in the securities trading market continuously trade securities products, and the identification equipment for identifying illegal accounts can collect transaction cost data, and every other trading cycle, based on the collected transaction cost data generated in the most recent trading cycle, generate behavioral data of the account for the target trading behavior in the most recent trading cycle, that is, generate behavioral data of the account's buying securities products in the most recent trading cycle.

[0046] Based on the transaction cost exchange data, behavioral data can be generated in either of the following two implementation methods.

[0047] In the first implementation, the accounts that have concluded transactions within the transaction cycle and the securities products traded can be determined based on the transaction result exchange data. For each determined account, if the account has generated a target transaction behavior for the determined securities product, sub-behavior data representing the target transaction behavior of the account for the determined securities product is generated; if the account has not generated a target transaction behavior for the determined securities product, sub-behavior data representing the target transaction behavior of the account for the determined securities product is generated. After the above processing is performed on each determined account, behavior data containing each generated sub-behavior data can be obtained.

[0048] For example, if account a1 buys products b1 and b2 during a trading cycle, and account a2 buys products b1 and b3, it can be determined that there are two accounts that have completed transactions during the trading cycle, namely account a1 and account a2, and there are three securities products traded, namely products b1, b2, and b3. For account a1, sub-behavior data 1 representing that account a1 has a target trading behavior for security product b1, sub-behavior data 2 representing that account a1 has a target trading behavior for security product b2, sub-behavior data 3 representing that account a1 has no target trading behavior data for security product b3, sub-behavior data 4 representing that account a2 has a target trading behavior for security product b1, sub-behavior data 5 representing that account a2 has no target trading behavior data for security product b2, and sub-behavior data 6 representing that account a2 has a target trading behavior for security product b3 can be generated. Finally, behavior data containing sub-behavior data 1-6 is obtained.

[0049] In the second implementation manner, the registered accounts and the listed securities products can be determined in advance, and the combinations of each account and each securities product can be determined. Each combination includes one account and one securities product.

[0050] According to the transaction completion transaction data, the accounts that have completed transactions during the trading cycle and the securities products traded by each account are determined, so as to retrieve the target combinations including the determined accounts and the securities products traded by the accounts in each combination, generate sub-behavior data representing that the accounts in the target combinations have target trading behaviors for the securities products in the target combinations, and for other combinations, generate sub-behavior data representing that the accounts in other combinations have no target trading behaviors for the securities products in other combinations. After generating the sub-behavior data corresponding to each combination in this way, behavior data containing the generated sub-behavior data can be obtained.

[0051] Step S102: According to the behavior data, determine the similarity representation value of the transactions between the two accounts included in each account group during the trading cycle.

[0052] First, the account group and the similarity representation value are introduced.

[0053] 1. Account group

[0054] In one case, according to the behavior data, the accounts that have target trading behaviors during the trading cycle can be determined, and every two of the determined accounts are combined into an account group.

[0055] In another case, all the registered accounts can be determined, and every two of the determined accounts are combined into an account group.

[0056] 2. Similarity representation value

[0057] Similarity characterization value representation: It represents the similarity degree of transactions between two accounts. If there are more identical securities products among the securities products traded by the two accounts, it indicates a higher similarity degree of transactions between the two accounts, and the above similarity characterization value is larger; if there are fewer identical securities products among the securities products traded by the two accounts, it indicates a lower similarity degree of transactions between the two accounts, and the above similarity characterization value is smaller.

[0058] In the first case of the target trading behavior, the above similarity characterization value represents the similarity degree of buying securities products between two accounts.

[0059] In the second case of the target trading behavior, the above similarity characterization value represents the similarity degree of selling securities products between two accounts.

[0060] In the third case of the target trading behavior, the above similarity characterization value can include two sub-characterization values, respectively representing the similarity degrees of buying and selling securities products between two accounts.

[0061] Specifically, each pair of accounts in the accounts corresponding to the obtained behavior data can be used as an account group. For each account group, according to the behavior data corresponding to the two accounts included in the account group, determine the similarity characterization value of the transactions between the two accounts included in the account group.

[0062] For further introduction of the similarity characterization value and the specific determination method of the similarity characterization value, reference can be made to the subsequent embodiments, which will not be elaborated here for the time being.

[0063] Step S103: Based on the similarity characterization values corresponding to each account group within a continuous plurality of trading cycles, determine the cumulative similarity of the transactions between the two accounts included in each account group.

[0064] The similarity characterization values corresponding to each account group are: the similarity characterization values of the transactions between the two accounts included in each account group.

[0065] The above similarity characterization value can be represented in a variety of different ways (reference can be made to the further introduction of the similarity characterization value in the subsequent embodiments). If the representation method of the similarity characterization value is different, the method for determining the cumulative similarity is also different. For details, reference can be made to the subsequent embodiments, which will not be elaborated here for the time being.

[0066] Step S104: According to the cumulative similarity corresponding to each account group, determine the illegal account group, and determine the accounts included in the illegal account group as illegal accounts.

[0067] The cumulative similarity corresponding to each account group is: the cumulative similarity of the transactions between the two accounts included in each account group.

[0068] In one implementation, the determined cumulative similarity can be compared with a preset similarity threshold. If the cumulative similarity corresponding to an account group is greater than the similarity threshold, it indicates that the trading behaviors of the two accounts included in the account group are relatively similar in multiple consecutive trading cycles. At this time, the account group corresponding to the cumulative similarity greater than the similarity threshold can be determined as a violation account group, and the accounts included in the violation account group can be determined as violation accounts.

[0069] In another implementation, the cumulative similarities corresponding to each account group can be sorted in descending order, and the top preset number of account groups can be determined as violation account groups. For example, the top 10 account groups can be determined as violation account groups. Then, the accounts included in the violation account groups can be determined as violation accounts.

[0070] As can be seen from the above, when applying the solution provided by the embodiments of the present application to identify violation accounts, the behavior data can characterize whether the target trading behavior is generated for each security product by the accounts. According to the behavior data, the similarity characterization value of the trading of the two accounts included in each account group within the trading cycle can be accurately determined. Then, based on the similarity characterization values corresponding to each account group in multiple consecutive trading cycles, the cumulative similarity of the trading of the two accounts included in each account group can be determined. This cumulative similarity can be understood as: the similarity of the trading of the two accounts included in the account group within a relatively long period of multiple consecutive trading cycles. If the similarity of the trading of the two accounts included in the account group is relatively high within this relatively long period, it can be accurately determined that the trading behaviors of these two accounts are relatively similar, and most of the traded security products are the same. Therefore, based on the cumulative similarities corresponding to each account group, the violation account groups can be accurately and comprehensively determined, and further, the accounts included in the violation account groups can be accurately and comprehensively determined as violation accounts. It can be seen that applying the violation account identification solution provided by the embodiments of the present application can improve the accuracy and identification effect of identifying violation accounts.

[0071] In addition, the solution provided by the embodiments of the present application identifies violation accounts based on the similarity of trading of different accounts. In addition to being applicable to the scenario of identifying accounts conducting convergent trading, the solution provided by the embodiments of the present application can also be applied to other scenarios such as margin trading account identification and anti-money laundering identification, etc., which identify based on the similarity of trading of different accounts. It can be seen that applying the violation account identification solution provided by the embodiments of the present application can expand the application scope of identifying violation accounts.

[0072] Next, the similarity characterization value mentioned in the above step S102 and its specific determination method will be introduced.

[0073] In one embodiment of the present application, the number of first products of securities products in which both accounts included in each account group generate target transaction behavior can be determined based on behavioral data; and the similarity characterization value of transactions between the two accounts included in each account group within the transaction cycle can be determined based on the number of first products corresponding to each account group.

[0074] The number of first products corresponding to the account group is the number of securities products in which both accounts included in the account group generate target trading behaviors. The larger the number of first products, the more securities products in which both accounts included in the account group generate target trading behaviors, which means that the trading behaviors of the two accounts included in the account group are more similar. Therefore, based on the number of first products corresponding to each account group, the similarity characterization value of transactions between the two accounts included in each account group within the trading cycle can be accurately determined, thereby identifying illegal accounts based on the determined similarity characterization value, which can improve the accuracy of identifying illegal accounts.

[0075] Specifically, for each account group, the similarity representation value corresponding to the account group can be determined according to the following process:

[0076] Based on the behavioral data of the two accounts included in the account group, two product sets of securities products traded by the two accounts included in the account group are respectively determined, and the securities products existing in both product sets are determined, so as to obtain the product quantity of the determined securities products by statistics as the first product quantity corresponding to the account group.

[0077] After determining the number of first products corresponding to each account group, the similarity representation value corresponding to the account group may be determined in either of the following two ways according to the number of first products corresponding to the account group.

[0078] In the first method, since the number of first products is the number of securities products that generate target trading behaviors in both accounts included in the account group, the larger the number of first products, the more securities products that generate target trading behaviors in both accounts, that is, the more similar the trading behaviors of the two accounts are. Therefore, the number of first products can accurately represent the similarity of transactions between the two accounts included in the account group.

[0079] In view of this, the quantity of the first product corresponding to each account group can be directly determined as the similarity representation value of transactions between two accounts contained in each account group within the transaction period. In this way, subsequent processing based on the determined similarity representation value corresponding to each account group can improve the accuracy of identifying illegal accounts.

[0080] In the second method, based on the behavioral data, the number of second products of the securities products that generate target trading behavior for each account included in each account group during the trading cycle can be determined, and based on the number of first products and the number of second products corresponding to each account group, the similarity characterization value of transactions between two accounts included in each account group during the trading cycle can be determined.

[0081] In this way, for each account group, after determining the number of the first product and the two numbers of the second products corresponding to the account group, the number of the first product can be divided by the two numbers of the second products respectively to obtain two calculation results, and then the similarity characterization value corresponding to the account group is determined based on the two calculation results.

[0082] For example, both calculation results are used as the similarity characterization value corresponding to the account group, or the average of the two calculation results is calculated and used as the similarity characterization value corresponding to the account group, or the maximum or minimum value of the two calculation results is determined as the similarity characterization value corresponding to the account group, or one of the two calculation results is selected as the similarity characterization value corresponding to the account group.

[0083] Alternatively, after determining the quantity of the first product and the quantities of two second products corresponding to the account group, add the quantities of the two second products to obtain the total quantity of products, divide the quantity of the first product by the total quantity of products, and obtain the calculation result as the similarity representation value corresponding to the account group.

[0084] It can be seen that in this method, the number of first products is the number of securities products for which both accounts included in the account group generate target trading behaviors, and the number of second products is the number of securities products for which each account included in the account group generates target trading behaviors. Based on the number of first products and the number of second products, it can be known that the proportion of securities products traded by both accounts included in the account group in the securities products traded by each account. The larger the proportion, the higher the similarity of the trading behaviors of the two accounts included in the account group, and the more likely these two accounts are illegal accounts. Therefore, based on the number of first products and the number of second products corresponding to each account group, the similarity characterization value corresponding to each account group can be accurately determined, thereby identifying illegal accounts based on the determined similarity characterization value, which can improve the accuracy of identifying illegal accounts.

[0085] In another embodiment of the present application, when determining the similarity characterization value corresponding to each account group, a neural network model for calculating the similarity of transactions between two accounts can be pre-trained. In this way, after obtaining the behavior data, for each account group, the behavior data of the two accounts included in the account group can be input into the neural network model, and the neural network model calculates the similarity of transactions between the two accounts included in the account group based on the input behavior data, and outputs the calculation result as the similarity characterization value of the transactions between the two accounts included in the account group.

[0086] In the case where the first product quantity corresponding to each account group is used as the similarity characterization value of the transactions between the two accounts included in each account group, the cumulative similarity of the transactions between the two accounts included in each account group can be determined in the manner mentioned in the following embodiments.

[0087] In one embodiment of the present application, when determining the cumulative similarity of the transactions between the two accounts included in each account group, the second product quantity of the securities product for which each account included in each account group generates a target transaction behavior can be determined according to the behavior data; for each account group, the first product quantities corresponding to the account group in consecutive multiple trading periods are accumulated to obtain a first accumulated quantity, and the second product quantities corresponding to the account group in consecutive multiple trading periods are accumulated to obtain a second accumulated quantity, and the cumulative similarity of the transactions between the two accounts included in the account group is determined according to the first accumulated quantity and the second accumulated quantity.

[0088] First, the process of accumulating the first product quantity is introduced.

[0089] For each account group, each first product quantity corresponding to the account group is the quantity of the securities product for which the two accounts included in the account group both generate target transaction behaviors within a trading period. The first product quantities corresponding to the account group in consecutive multiple trading periods are accumulated to obtain a first accumulated quantity, and the first accumulated quantity is the quantity of the securities product for which the two accounts included in the account group both generate target transaction behaviors within consecutive multiple trading periods.

[0090] Secondly, the process of accumulating the second product quantity is introduced.

[0091] An account group includes two accounts, and each account corresponds to a second product quantity. When accumulating the second product quantities corresponding to the account group, the second product quantities corresponding to the same account in consecutive multiple trading periods in the account group can be accumulated to obtain a second accumulated quantity, so that there are two second accumulated quantities corresponding to one account group; or the second product quantities corresponding to the two accounts in consecutive multiple trading periods in the account group can be accumulated to obtain a second accumulated quantity, so that there is one second accumulated quantity corresponding to one account group.

[0092] For each account group, after obtaining the first cumulative quantity and the second cumulative quantity corresponding to the account group, the first cumulative quantity can be divided by the second cumulative quantity to obtain a calculation result, which is used as the cumulative similarity of transactions between the two accounts included in the account group.

[0093] In the case where one account corresponds to two second cumulative quantities, the first cumulative quantity can be divided by the two second cumulative quantities respectively to obtain two calculation results, and then based on these two calculation results, the cumulative similarity of transactions between the two accounts included in the account group can be determined.

[0094] For example, it is determined that both calculation results are used as the cumulative similarity, or the average value of the two calculation results is determined as the cumulative similarity, or the maximum or minimum value of the two calculation results is determined as the cumulative similarity, or any one of the two calculation results is determined as the cumulative similarity, etc.

[0095] In the case where it is determined that both calculation results are used as the cumulative similarity, the two obtained cumulative similarities can be understood as the transaction similarities determined from the perspectives of the two accounts included in the account group respectively. In this way, when identifying illegal accounts based on the two cumulative similarities subsequently, the information used is richer, thereby improving the accuracy of identifying illegal accounts.

[0096] As can be seen from the above, when applying the solution provided by the embodiments of the present application to identify illegal accounts, the first cumulative quantity corresponding to each account group is: the total quantity of securities products for which the two accounts included in the account group have generated target transaction behaviors in a continuous plurality of trading periods; the second cumulative quantity is: the total quantity of securities products for which each account included in the account group has generated target transaction behaviors in a continuous plurality of trading periods. According to the first cumulative quantity and the second cumulative quantity corresponding to each account group, the cumulative similarity of transactions between the two accounts included in each account group can be accurately determined in a relatively long time of a continuous plurality of trading periods, and then illegal accounts can be identified based on the determined cumulative similarity, which can improve the accuracy of identifying illegal accounts.

[0097] In an embodiment of the present application, the above behavior data can be represented in the form of a two-dimensional matrix. One dimension of the matrix is accounts, and the other dimension is securities products. The value of the matrix element is 0 or m. If the value of the matrix element is m, it means that the account corresponding to the element has generated a target transaction behavior for the securities product corresponding to the element. If the value of the matrix element is 0, it means that the account corresponding to the element has not generated a target transaction behavior for the securities product corresponding to the element, where m is greater than or equal to 1.

[0098] In the example of the above step S101, the obtained behavior data is a 2×3 two-dimensional matrix as follows:

[0099]

[0100] When executing the above step S102 to determine the similarity representation value of transactions between two accounts included in each account group within the transaction cycle, the two-dimensional matrix of the behavior data can be matrix-dot multiplied with its transposed matrix to obtain a product quantity matrix as the similarity representation value of transactions between two accounts included in each account group within the transaction cycle.

[0101] The two dimensions of the above product quantity matrix correspond to the two accounts contained in the account group. The element in the i-th row and j-th column of the product quantity matrix can be expressed as n ij Indicates that n ij The element value of represents the number of first products of securities products for which both the account corresponding to the i-th row and the account corresponding to the j-th column of the matrix generate the target transaction behavior. When i and j are the same, n ij The element value represents the number of securities products for which the account corresponding to the i-th row or j-th column generates the target transaction behavior.

[0102] When executing the above step S103 to determine the cumulative similarity of transactions between two accounts included in each account group, the values ​​of the matrix elements corresponding to the two accounts included in the same account group in the multiple product quantity matrices corresponding to multiple consecutive transaction cycles are added to obtain a cumulative quantity matrix. For each row (or each column) in the cumulative quantity matrix, each element in the row can be divided by the element belonging to the row (or the column) and the matrix diagonal to obtain the divided matrix as the cumulative similarity of transactions between the two accounts included in each account group.

[0103] When executing the above step S104 to determine the illegal account group, the account group where two accounts corresponding to the elements whose element values ​​meet the screening conditions are located can be screened out in the divided matrix as the illegal account group.

[0104] The above screening condition may be that the element value exceeds a threshold, or the element value belongs to a maximum preset number of element values ​​among the element values ​​of each element.

[0105] In this solution, the matrix dot multiplication method is used to greatly improve the efficiency of determining the quantity of the first product, thereby greatly improving the efficiency of identifying illegal accounts.

[0106] In addition, when collecting the transaction cost data and generating a two-dimensional matrix of behavior data based on the transaction cost data, after the identification device collects the transaction cost data, the collected transaction cost data can be stored using big data storage technology. When generating behavior data, the stored transaction cost data generated within the transaction cycle can be called, and a two-dimensional matrix of behavior data can be generated based on the called data, and the generated two-dimensional matrix can be stored using big data storage technology.

[0107] In one embodiment of the present application, when the number of accounts and / or securities products involved in the identification of illegal accounts is large, the data volume of the above-mentioned two-dimensional matrix is ​​relatively large. In this case, the above-mentioned identification device can use the computing resources of its own GPU and / or the data processing flow written in Python language to perform data processing, such as obtaining a transposed matrix, matrix dot multiplication, and matrix accumulation, etc. In this way, using the computing resources of the device GPU and / or the data processing flow written in Python language to perform data processing can further improve the data processing efficiency, thereby improving the recognition efficiency of identifying illegal accounts.

[0108] See also Figure 2 , a flow chart of the second method for identifying illegal accounts is provided, wherein the above-mentioned transaction cycle is 1 day. Figure 2 The process shown is as follows: the identification device collects transaction cost data, generates a two-dimensional matrix of behavior data based on the transaction cost data, that is, generates a daily transaction behavior matrix K, and determines the similarity representation value of transactions between two accounts included in each account group based on the daily transaction behavior matrix K, that is, calculates the daily transaction behavior similarity matrix M, where M = K·K , , K , is the transposed matrix of the daily transaction behavior matrix K. The recognition device can then accumulate the daily transaction behavior similarity matrix M corresponding to multiple consecutive days, for example, accumulate the daily transaction behavior similarity matrix M corresponding to one month, and obtain the cumulative similarity of transactions between the two accounts included in each account group, that is, calculate the transaction behavior similarity matrix M within the interval q , where q is the number of trading cycles. The similarity matrix M of trading behavior in the interval is obtained q After that, we can calculate the similarity matrix M of the transaction behavior in the interval q , determine the illegal account group, and determine that the accounts included in the illegal account group are illegal accounts.

[0109] In the above Figure 2 In the process, the transaction cost exchange data and the daily transaction behavior matrix K can be stored based on big data storage technology, and the daily transaction behavior similarity matrix M and the transaction behavior similarity matrix M within the calculation interval are calculated. q All of these can be achieved based on the data processing flow written in Python and the GPU computing power of the recognition device itself.

[0110] The following is an introduction to the method for identifying illegal accounts in the third case of target trading behavior (i.e., the case where the target trading behavior includes buying and selling securities products).

[0111] In one embodiment of the present application, see Figure 3, a flow chart of a third method for identifying illegal accounts is provided. In this embodiment, the target transaction behavior includes the behavior of buying and selling securities products. The above step S101 can be implemented based on the following step S101A.

[0112] Step S101A: Obtain first behavior data of the account for buying behavior and second behavior data of the account for selling behavior within the transaction cycle.

[0113] Specifically, the identification device for identifying illegal accounts can collect transaction cost exchange data. Every transaction cycle, it can generate first behavior data for the account's buying behavior in the most recent transaction cycle based on the transaction cost exchange data of buying transactions generated in the collected transaction cost exchange data; and generate second behavior data for the account's selling behavior in the most recent transaction cycle based on the transaction cost exchange data of selling transactions generated in the collected transaction cost exchange data.

[0114] After the first behavior data and the second behavior data are obtained, the above step S102 may be implemented according to the first behavior data and the second behavior data by following the steps S102A-S102B.

[0115] Step S102A: Determine, based on the first behavior data, a first similarity representation value of transactions between two accounts included in each first account group within a transaction period.

[0116] For the first account group, in one case, based on the first behavior data, each account that generates buying behavior within the trading cycle can be determined, and every two accounts among the determined accounts are formed into a first account group; in another case, every two accounts among the registered accounts can be formed into a first account group.

[0117] The method of determining the first similarity representation value corresponding to the first account group is similar to the method of determining the similarity representation value corresponding to the account group in the above step S102, and will not be repeated here.

[0118] Step S102B: determining, based on the second behavior data, a second similarity characterization value of transactions between two accounts included in each second account group within the transaction period.

[0119] For the second account group, in one case, based on the second behavior data, each account that generates selling behavior within the trading cycle can be determined, and every two accounts among the determined accounts can be formed into a second account group; in another case, every two accounts among the registered accounts can be formed into a second account group.

[0120] The method of determining the second similarity representation value corresponding to the second account group is similar to the method of determining the similarity representation value corresponding to the account group in the above step S102, and will not be repeated here.

[0121] After determining the first similarity characterization value corresponding to each first account group and the second similarity characterization value corresponding to each second account group, the above step S103 can be implemented according to the following steps S103A-S103B based on the determined first similarity characterization value and the second similarity characterization value.

[0122] Step S103A: determining a first cumulative similarity of transactions between two accounts included in each first account group based on the first similarity representation values ​​corresponding to each first account group in a plurality of consecutive transaction cycles.

[0123] The manner of determining the first cumulative similarity is similar to the manner of determining the cumulative similarity in the above step S103 , and will not be described in detail here.

[0124] Step S103B: determining a second cumulative similarity of transactions between two accounts included in each second account group based on the second similarity characterization values ​​corresponding to each second account group in a plurality of consecutive transaction cycles.

[0125] The method for determining the second cumulative similarity is similar to the method for determining the cumulative similarity in the above step S103, and will not be described in detail here.

[0126] After determining the first cumulative similarities corresponding to each first account group and the second cumulative similarities corresponding to each second account group, the above step S104 may be implemented according to the following step S104A based on the determined first cumulative similarities and second cumulative similarities.

[0127] Step S104A: Determine a violation account group according to the first cumulative similarity and the second cumulative similarity corresponding to each target account group, and determine that the accounts included in the violation account group are violation accounts.

[0128] The target account group belongs to the first account group and the second account group.

[0129] Specifically, in each first account group and each second account group, a target account group belonging to both the first account group and the second account group may be determined, and each target account group corresponds to a first cumulative similarity and a second cumulative similarity.

[0130] When determining the illegal account group, based on the first cumulative similarity and the second cumulative similarity corresponding to each target account group, it can be determined that the target account group that meets both the buying convergence condition and the selling convergence condition is the illegal account group, and then the two accounts included in the target account group are determined to be illegal accounts.

[0131] The above-mentioned buy convergence condition may be that the first cumulative similarity is greater than a preset buy similarity threshold, or the target account group is among the top first preset number of positions in the target account groups sorted in descending order of the first cumulative similarity, etc.

[0132] The above-mentioned sell convergence condition may be that the second cumulative similarity is greater than a preset sell similarity threshold, or the target account group is among the top second preset number of positions in the target account groups sorted in descending order of the second cumulative similarity, etc.

[0133] As can be seen from the above, when applying the solution provided by the embodiments of the present application to identify illegal accounts, in the case where the target trading behavior includes the behaviors of buying and selling securities products, for the behavior of buying securities products, the first behavior data is obtained, the first similarity characterization value corresponding to each first account group is determined, and the first cumulative similarity corresponding to each first account group is determined; for the behavior of selling securities products, the second behavior data is obtained, the second similarity characterization value corresponding to each second account group is determined, and the second cumulative similarity corresponding to each second account group is determined. Then, for the target account group that belongs to both the first account group and the second account group, according to the determined first and second cumulative similarities corresponding to the target account group, the illegal account group is determined. The information used includes the first cumulative similarity and the second cumulative similarity, and the information is relatively rich, so that the illegal account group can be accurately determined, and thus the illegal accounts can be accurately determined. It can be seen that applying the illegal account identification solution provided by the embodiments of the present application can improve the accuracy of identifying illegal accounts.

[0134] According to the cumulative similarity corresponding to each account group, the illegal account group can be determined in the manner mentioned in the following embodiments.

[0135] In one embodiment of the present application, when determining the illegal account group, the pending account group may be determined according to the cumulative similarity corresponding to each account group; the device information of the devices where the accounts included in the pending account group are located is obtained; according to the obtained device information, it is judged whether the devices where the accounts included in the pending account group are located are the same device; if so, the pending account group is determined to be the illegal account group.

[0136] The method for determining the pending account group may refer to the method for determining the illegal account group in the foregoing step S104, which will not be elaborated here.

[0137] The back-end server of the securities trading market can record the device information of the device where the account is located when the account logs in, such as the location of the device, the device IP, the login time, and so on. After identifying the device, after determining the pending account group, it can determine the accounts included in the pending account group, so as to obtain the device information of the devices where the determined accounts are located from the back-end server. According to the obtained device information, it can be judged whether the devices where the determined accounts are located are the same device. For example, it can be judged whether the device IPs of the devices where the determined accounts are located are the same, whether the locations of the devices are the same, whether the login times belong to the same trading cycle, and so on.

[0138] According to the obtained device information, if it is judged that the devices where the accounts included in the pending account group are not the same device, if it is judged that the devices where the accounts included in the pending account group are the same device, and since the pending account group is determined according to the cumulative similarity corresponding to each account group, it means that in a relatively long period of consecutive trading cycles, different accounts have logged in on the same device to perform highly similar trading behaviors. At this time, it can be accurately determined that the pending account group is a group of illegal accounts, and further, it can be accurately determined that the accounts included in the group of illegal accounts are illegal accounts.

[0139] Corresponding to the method for identifying illegal accounts mentioned in the foregoing embodiments, an embodiment of the present application also provides a device for identifying illegal accounts.

[0140] In an embodiment of the present application, referring to Figure 4 , a schematic structural diagram of a device for identifying illegal accounts is provided. In this embodiment, the device includes:

[0141] A data acquisition module 401, configured to acquire the behavior data of an account for a target trading behavior during a trading cycle, where the target trading behavior includes: the behavior of buying and / or selling a securities product, and the behavior data characterizes whether the account has a target trading behavior for each securities product;

[0142] A characterization value determination module 402, configured to determine a similarity characterization value for trading between two accounts included in each account group during the trading cycle according to the behavior data;

[0143] A similarity determination module 403, configured to determine the cumulative similarity for trading between two accounts included in each account group based on the similarity characterization values corresponding to each account group during a plurality of consecutive trading cycles;

[0144] An account determination module 404, configured to determine a group of illegal accounts according to the cumulative similarity corresponding to each account group, and determine that the accounts included in the group of illegal accounts are illegal accounts.

[0145] As can be seen from the above, when the scheme provided by the embodiment of the present application is used to identify illegal accounts, the behavior data can characterize whether the account generates target transaction behavior for each securities product. According to the behavior data, the similarity characterization value of the transactions between the two accounts included in each account group within the transaction cycle can be accurately determined, and then based on the similarity characterization value corresponding to each account group in multiple consecutive transaction cycles, the cumulative similarity of the transactions between the two accounts included in each account group is determined. The cumulative similarity can be understood as: the similarity of the transactions between the two accounts included in the account group within a long period of time such as multiple consecutive transaction cycles. If the similarity of the transactions between the two accounts included in the account group within this long period of time is high, it can be accurately determined that the transaction behaviors of the two accounts are relatively similar, and the securities products traded are mostly the same, so that according to the cumulative similarity corresponding to each account group, the illegal account group can be accurately determined, and then the accounts included in the illegal account group can be accurately determined as illegal accounts. It can be seen that the illegal account identification scheme provided by the embodiment of the present application can improve the accuracy of identifying illegal accounts.

[0146] In addition, the solution provided in the embodiment of the present application is to identify illegal accounts based on the similarity of transactions between different accounts. In addition to being applicable to scenarios for identifying accounts that conduct similar transactions, the solution provided in the embodiment of the present application can also be applied to other scenarios such as capital allocation account identification, anti-money laundering identification, etc. that identify accounts based on the similarity of transactions between different accounts. It can be seen that the application of the illegal account identification solution provided in the embodiment of the present application can expand the application scope of identifying illegal accounts.

[0147] In one embodiment of the present application, the characterization value determination module 402 includes:

[0148] A first product quantity determination submodule, for determining, based on the behavior data, the first product quantity of the securities product for which both of the two accounts included in each account group generate target transaction behaviors;

[0149] The characterization value determination submodule is used to determine the similarity characterization value of transactions between two accounts included in each account group within the transaction cycle according to the quantity of the first product corresponding to each account group.

[0150] In this scheme, the number of first products corresponding to the account group is the number of securities products for which both accounts included in the account group generate target trading behaviors. The larger the number of first products, the more securities products for which both accounts included in the account group generate target trading behaviors, which means that the trading behaviors of the two accounts included in the account group are more similar. Therefore, based on the number of first products corresponding to each account group, the similarity characterization value of transactions between the two accounts included in each account group within the trading cycle can be accurately determined, thereby identifying illegal accounts based on the determined similarity characterization value, which can improve the accuracy of identifying illegal accounts.

[0151] In one embodiment of the present application, the characterization value determination submodule is specifically used to:

[0152] Determining the first product quantity corresponding to each account group is a similarity characterization value of transactions between two accounts included in each account group within the transaction period.

[0153] In this solution, since the number of first products is the number of securities products that both accounts in the account group have generated target trading behaviors, the larger the number of first products, the more securities products both accounts have generated target trading behaviors, that is, the more similar the trading behaviors of the two accounts are, and therefore, the number of first products can accurately represent the similarity of transactions between the two accounts in the account group. The number of first products corresponding to each account group is determined as the similarity representation value of transactions between the two accounts in each account group within the trading cycle, and subsequent processing based on the determined similarity representation value corresponding to each account group can improve the accuracy of identifying illegal accounts.

[0154] In one embodiment of the present application, the characterization value determination submodule is specifically used to:

[0155] Based on the behavioral data, determine the number of second products of the securities products that generate target trading behavior for each account included in each account group during the trading cycle; based on the number of first products and the number of second products corresponding to each account group, determine the similarity characterization value of transactions between two accounts included in each account group during the trading cycle.

[0156] In this scheme, the number of first products is the number of securities products for which both accounts included in the account group generate target trading behaviors, and the number of second products is the number of securities products for which each account included in the account group generates target trading behaviors. Based on the number of first products and the number of second products, it can be known that the proportion of securities products traded by both accounts included in the account group in the securities products traded by each account. The larger the proportion, the higher the similarity of the trading behaviors of the two accounts included in the account group, and the more likely these two accounts are illegal accounts. Therefore, based on the number of first products and the number of second products corresponding to each account group, the similarity representation value corresponding to each account group can be accurately determined, thereby identifying illegal accounts based on the determined similarity representation value, which can improve the accuracy of identifying illegal accounts.

[0157] In one embodiment of the present application, when the quantity of the first product corresponding to each account group is used as a similarity representation value of transactions between two accounts included in each account group, the similarity determination module 403 includes:

[0158] A second product quantity determination submodule, for determining, based on the behavior data, the second product quantity of the securities product for which each account included in each account group generates a target transaction behavior;

[0159] The similarity determination submodule is used to accumulate the quantity of the first product corresponding to the account group in multiple consecutive transaction cycles for each account group to obtain a first accumulated quantity, accumulate the quantity of the second product corresponding to the account group in multiple consecutive transaction cycles to obtain a second accumulated quantity, and determine the cumulative similarity of transactions between two accounts included in the account group based on the first accumulated quantity and the second accumulated quantity.

[0160] It can be seen from the above that when the scheme provided in the embodiment of the present application is used to identify illegal accounts, the first cumulative quantity corresponding to each account group is: the total number of securities products in which the two accounts included in the account group generate target trading behaviors within multiple consecutive trading cycles, and the second cumulative quantity is: the total number of securities products in which each account included in the account group generates target trading behaviors within multiple consecutive trading cycles. Based on the first cumulative quantity and the second cumulative quantity corresponding to each account group, the cumulative similarity of transactions between the two accounts included in each account group within a longer period of time such as multiple consecutive trading cycles can be accurately determined, thereby identifying illegal accounts based on the determined cumulative similarity, which can improve the accuracy of identifying illegal accounts.

[0161] In one embodiment of the present application, the similarity determination submodule is specifically used to:

[0162] For each account group, the quantity of the first product corresponding to the account group in multiple consecutive transaction cycles is accumulated to obtain a first accumulated quantity, and the quantity of the second product of the two accounts included in the account group in multiple consecutive transaction cycles is accumulated respectively to obtain two second accumulated quantities, and the first accumulated quantity is divided by the two obtained second accumulated quantities respectively to obtain two calculation results as the cumulative similarity of transactions between the two accounts included in the account group.

[0163] When both calculation results are determined as cumulative similarities, the two cumulative similarities obtained can be understood as transaction similarities determined from the perspectives of two accounts included in the account group, respectively. In this way, when subsequently identifying illegal accounts based on the two cumulative similarities, the information used is richer, thereby improving the accuracy of identifying illegal accounts.

[0164] In one embodiment of the present application, the account determination module 404 is specifically used to:

[0165] Determine the pending account groups based on the cumulative similarities corresponding to each account group;

[0166] Obtain device information of devices where accounts included in the pending account group are located;

[0167] According to the acquired device information, determining whether the devices where the accounts included in the pending account group are located are the same device;

[0168] If yes, the pending account group is determined to be a violation account group.

[0169] In this scheme, based on the acquired device information, if it is determined that the devices where the accounts included in the pending account group are located are not the same device, if it is determined that the devices where the accounts included in the pending account group are located are the same device, and the pending account group is determined based on the cumulative similarity corresponding to each account group, it means that in a long period of time of multiple consecutive transaction cycles, different accounts have been logged in on the same device to perform highly similar transactions. At this time, the pending account group can be accurately determined as the illegal account group, and then the accounts included in the illegal account group can be accurately determined as the illegal accounts.

[0170] In one embodiment of the present application, the data acquisition module 401 is specifically used to:

[0171] Obtaining first behavior data of an account for a buying behavior and second behavior data of an account for a selling behavior within a trading cycle;

[0172] The characterization value determination module 402 is specifically used to:

[0173] Determining, according to the first behavior data, a first similarity representation value of transactions between two accounts included in each first account group within the transaction cycle;

[0174] Determining, according to the second behavior data, a second similarity representation value of transactions between two accounts included in each second account group within the transaction period;

[0175] The similarity determination module 403 is specifically used for:

[0176] Determining a first cumulative similarity of transactions between two accounts included in each first account group based on the first similarity representation value corresponding to each first account group in a plurality of consecutive transaction cycles;

[0177] Determining a second cumulative similarity of transactions between two accounts included in each second account group based on the second similarity representation value corresponding to each second account group in a plurality of consecutive transaction cycles;

[0178] The account determination module 404 is specifically used to:

[0179] The violating account group is determined according to the first cumulative similarity and the second cumulative similarity corresponding to each target account group, wherein the target account group belongs to the first account group and belongs to the second account group.

[0180] As can be seen from the above, when the scheme provided by the embodiment of the present application is used to identify illegal accounts, when the target transaction behavior includes the behavior of buying and selling securities products, for the behavior of buying securities products, the first behavior data is obtained, the first similarity characterization value corresponding to each first account group is determined, and the first cumulative similarity corresponding to each first account group is determined; for the behavior of selling securities products, the second behavior data is obtained, the second similarity characterization value corresponding to each second account group is determined, and the second cumulative similarity corresponding to each second account group is determined. After that, for the target account group belonging to the first account group and the second account group, the illegal account group is determined according to the first and second cumulative similarities corresponding to the determined target account group. The information used includes the first cumulative similarity and the second cumulative similarity, and the information is relatively rich, so that the illegal account group can be accurately determined, and thus the illegal account can be accurately determined. It can be seen that the illegal account identification scheme provided by the embodiment of the present application can improve the accuracy of identifying illegal accounts.

[0181] The embodiment of the present invention further provides an electronic device, such as Figure 5 As shown, it includes a processor 501 , a communication interface 502 , a memory 503 and a communication bus 504 , wherein the processor 501 , the communication interface 502 , and the memory 503 communicate with each other via the communication bus 504 .

[0182] Memory 503, used for storing computer programs;

[0183] The processor 501 is used to execute the program stored in the memory 503, and implements the following steps:

[0184] Obtaining behavior data of the account for target transaction behavior during a transaction cycle, wherein the target transaction behavior includes: behavior of buying and / or selling securities products, and the behavior data indicates: whether the account generates target transaction behavior for each securities product;

[0185] Determining, based on the behavior data, a similarity representation value of transactions performed by two accounts included in each account group within the transaction cycle;

[0186] Based on the similarity representation values ​​corresponding to each account group in multiple consecutive transaction cycles, the cumulative similarity of transactions between two accounts included in each account group is determined;

[0187] According to the accumulated similarities corresponding to each account group, a violation account group is determined, and the accounts included in the violation account group are determined to be violation accounts.

[0188] The communication bus mentioned in the above electronic device may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. This communication bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of convenience of representation, only a thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0189] The communication interface is used for communication between the above electronic device and other devices.

[0190] The memory may include a Random Access Memory (RAM), and may also include a Non-Volatile Memory (NVM), such as at least one disk memory. Optionally, the memory may also be at least one storage device located far from the aforementioned processor.

[0191] The above-mentioned processor may be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0192] In another embodiment provided by the present invention, a computer-readable storage medium is further provided. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of any of the above-mentioned illegal account identification methods are implemented.

[0193] In another embodiment provided by the present invention, a computer program product containing instructions is further provided. When it runs on a computer, it causes the computer to execute any of the illegal account identification methods in the above embodiments.

[0194] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present invention are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from a website, computer, server, or data center to another website, computer, server, or data center by wire (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more integrated available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid state disk (SSD)).

[0195] It should be noted that, in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variation thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising one..." does not exclude the existence of additional identical elements in the process, method, article or device comprising the element.

[0196] Each embodiment in this specification is described in a related manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the embodiments of the apparatus, electronic device, computer-readable storage medium, and computer program product, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the description of the method embodiments.

[0197] The above are only the preferred embodiments of the present invention and are not intended to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention are all included in the protection scope of the present invention.

Claims

1. A method for identifying a violating account, characterized in that, The method includes: Obtaining behavioral data of an account for a target trading behavior within a trading cycle, where the target trading behavior includes: the behavior of buying and / or selling a securities product, and the behavioral data characterizes whether the account has a target trading behavior for each securities product; Determining a similarity characterization value of trading between two accounts included in each account group within the trading cycle according to the behavioral data; Based on the similarity characterization values corresponding to each account group in consecutive multiple trading cycles, determining the cumulative similarity of trading between two accounts included in each account group; Determining a violation account group according to the cumulative similarity corresponding to each account group, and determining the accounts included in the violation account group as violation accounts.

2. The method according to claim 1, wherein The determining a similarity characterization value of trading between two accounts included in each account group within the trading cycle according to the behavioral data includes: Determining a first product quantity of securities products for which both of the two accounts included in each account group have a target trading behavior according to the behavioral data; Determining a similarity characterization value of trading between two accounts included in each account group within the trading cycle according to the first product quantity corresponding to each account group.

3. The method according to claim 2, wherein The determining a similarity characterization value of trading between two accounts included in each account group within the trading cycle according to the first product quantity corresponding to each account group includes: Determining the first product quantity corresponding to each account group as the similarity characterization value of trading between two accounts included in each account group within the trading cycle; Or Determining a second product quantity of securities products for which each account included in each account group has a target trading behavior within the trading cycle according to the behavioral data; and determining a similarity characterization value of trading between two accounts included in each account group within the trading cycle according to the first product quantity and the second product quantity corresponding to each account group.

4. The method according to claim 3, wherein In the case where the first product quantity corresponding to each account group is used as the similarity characterization value of trading between two accounts included in each account group, the determining the cumulative similarity of trading between two accounts included in each account group based on the similarity characterization values corresponding to each account group in consecutive multiple trading cycles includes: Determining a second product quantity of securities products for which each account included in each account group has a target trading behavior according to the behavioral data; For each account group, adding up the first product quantities corresponding to this account group in consecutive multiple trading cycles to obtain a first cumulative quantity, adding up the second product quantities corresponding to this account group in consecutive multiple trading cycles to obtain a second cumulative quantity, and determining the cumulative similarity of trading between two accounts included in this account group according to the first cumulative quantity and the second cumulative quantity.

5. The method according to claim 4, characterized in that The adding up the second product quantities corresponding to this account group in consecutive multiple trading cycles to obtain a second cumulative quantity includes: Adding up the second product quantities of the two accounts included in this account group in consecutive multiple trading cycles respectively to obtain two second cumulative quantities; The determining the cumulative similarity of trading between two accounts included in this account group according to the first cumulative quantity and the second cumulative quantity includes: Divide the first accumulated quantity by the two obtained second accumulated quantities respectively to obtain two calculation results, which are used as the cumulative similarity of the transactions of the two accounts included in this account group.

6. The method according to claim 1, characterized in that The determining of the illegal account group according to the cumulative similarity corresponding to each account group includes: Determine a pending account group according to the cumulative similarity corresponding to each account group; Obtain the device information of the devices where the accounts included in the pending account group are located; According to the obtained device information, determine whether the devices where the accounts included in the pending account group are located are the same device; If so, determine that the pending account group is an illegal account group.

7. The method according to claim 1, characterized in that In the case where the target trading behavior includes the behavior of buying and selling securities products, the obtaining of the behavior data of the account for the target trading behavior during the trading cycle includes: Obtain the first behavior data of the account for the buying behavior and the second behavior data of the account for the selling behavior during the trading cycle; The determining of the similarity characterization value of the transactions of the two accounts included in each account group during the trading cycle according to the behavior data includes: Determine the first similarity characterization value of the transactions of the two accounts included in each first account group during the trading cycle according to the first behavior data; Determine the second similarity characterization value of the transactions of the two accounts included in each second account group during the trading cycle according to the second behavior data; The determining of the cumulative similarity of the transactions of the two accounts included in each account group based on the similarity characterization values corresponding to each account group in consecutive trading cycles includes: Determine the first cumulative similarity of the transactions of the two accounts included in each first account group based on the first similarity characterization values corresponding to each first account group in consecutive trading cycles; Determine the second cumulative similarity of the transactions of the two accounts included in each second account group based on the second similarity characterization values corresponding to each second account group in consecutive trading cycles; The determining of the illegal account group according to the cumulative similarity corresponding to each account group includes: Determine the illegal account group according to the first cumulative similarity and the second cumulative similarity corresponding to each target account group, where the target account group belongs to the first account group and belongs to the second account group.

8. An illegal account identification device, characterized in that, The device includes: A data acquisition module, configured to acquire the behavior data of the account for the target trading behavior during the trading cycle, where the target trading behavior includes: the behavior of buying and / or selling securities products, and the behavior data characterizes whether the account has a target trading behavior for each securities product; A characterization value determination module, configured to determine the similarity characterization value of the transactions of the two accounts included in each account group during the trading cycle according to the behavior data; A similarity determination module, configured to determine the cumulative similarity of the transactions of the two accounts included in each account group based on the similarity characterization values corresponding to each account group in consecutive trading cycles; An account determination module, configured to determine the illegal account group according to the cumulative similarity corresponding to each account group, and determine the accounts included in the illegal account group as illegal accounts.

9. An electronic device, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus. Among them, the processor, the communication interface, and the memory complete mutual communication through the communication bus; The memory is used to store computer programs; The processor is used to implement the method described in any one of claims 1-7 when executing the program stored on the memory.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the method described in any one of claims 1-7 is implemented.

Images