Bank branch mobile phone number risk control method and device

Abstract

The present invention provides a mobile phone number risk control method and device for bank branches, which is applicable to the field of financial technology. The method includes: constructing a customer mobile phone number relationship diagram of a bank branch; for each mobile phone number, determining a potential associated mobile phone number set of the mobile phone number; when a customer modifies a mobile phone number at a bank branch, if the new mobile phone number is in the potential associated mobile phone number set of the old mobile phone number, performing face recognition on the customer based on the potential associated mobile phone number set and the customer's face recognition threshold, and judging whether the mobile phone number modification is supported according to the face recognition result; if the new mobile phone number is not in the potential associated mobile phone number set of the old mobile phone number, determining the association distance between the old mobile phone number and the new mobile phone number, correcting the customer's face recognition threshold according to the association distance, performing face recognition on the customer based on the corrected face recognition threshold, and judging whether the mobile phone number modification is supported according to the face recognition result. The present invention can realize mobile phone number risk control at bank branches.

Description

Technical Field

[0001] The present invention relates to the field of financial technology, and in particular to a method and device for controlling mobile phone number risks at bank outlets. Background Art

[0002] Mobile phone numbers are important information for customers. To some extent, mobile phone numbers are closely related to customer identities. Especially when customers make payments or modify information on the bank system, they need to verify their identity information through mobile phone verification codes and other methods. At the same time, it also brings risks to customers' funds, especially potential risks to customers from within the bank. At present, there is a lack of a risk control method for mobile phone numbers in bank branches. Summary of the invention

[0003] The embodiment of the present invention provides a bank branch mobile phone number risk control method for implementing bank branch mobile phone number risk control, the method comprising:

[0004] Build a customer mobile phone number relationship map for bank branches;

[0005] Based on the customer mobile phone number relationship graph and historical modification data, for each mobile phone number, determine the set of potential associated mobile phone numbers for the mobile phone number;

[0006] When a customer changes his / her mobile phone number at a bank branch, if the new mobile phone number is in the set of potential associated mobile phone numbers of the old mobile phone number, the customer's face recognition is performed based on the set of potential associated mobile phone numbers and the customer's face recognition threshold, and whether the mobile phone number change is supported is determined based on the face recognition result;

[0007] If the new mobile phone number is not in the set of potentially associated mobile phone numbers of the old mobile phone number, determine the association distance between the old mobile phone number and the new mobile phone number, correct the customer's face recognition threshold based on the association distance, perform face recognition on the customer based on the corrected face recognition threshold, and determine whether the mobile phone number modification is supported based on the face recognition result.

[0008] The embodiment of the present invention provides a bank branch mobile phone number risk control device, which is used to implement bank branch mobile phone number risk control. The device includes:

[0009] Customer mobile phone number relationship graph construction module, used to construct the customer mobile phone number relationship graph in the bank;

[0010] Potentially associated personnel determination module, used to determine the set of potentially associated mobile phone numbers for each mobile phone number based on the customer mobile phone number relationship diagram and historical modification data;

[0011] The mobile phone number modification judgment module is used when a customer modifies his mobile phone number at a bank branch. If the new mobile phone number is in the potential associated mobile phone number set of the old mobile phone number, the customer's face recognition is performed based on the potential associated mobile phone number set and the customer's face recognition threshold, and whether the mobile phone number modification is supported is determined based on the face recognition result;

[0012] A face recognition threshold correction module is used to determine the correlation distance between the old mobile phone number and the new mobile phone number if the new mobile phone number is not in the set of potential associated mobile phone numbers of the old mobile phone number, correct the customer's face recognition threshold according to the correlation distance, perform face recognition on the customer based on the corrected face recognition threshold, and determine whether the mobile phone number modification is supported based on the face recognition result.

[0013] An embodiment of the present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned bank branch mobile phone number risk control method when executing the computer program.

[0014] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned bank branch mobile phone number risk control method is implemented.

[0015] An embodiment of the present invention further provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the above-mentioned bank branch mobile phone number risk control method is implemented.

[0016] In an embodiment of the present invention, a customer mobile phone number relationship graph of a bank branch is constructed; based on the customer mobile phone number relationship graph and historical modification data, for each mobile phone number, a set of potential associated mobile phone numbers of the mobile phone number is determined; when a customer modifies a mobile phone number at a bank branch, if the new mobile phone number is in the set of potential associated mobile phone numbers of the old mobile phone number, facial recognition is performed on the customer based on the set of potential associated mobile phone numbers and the customer's face recognition threshold, and whether the mobile phone number modification is supported is determined based on the face recognition result; if the new mobile phone number is not in the set of potential associated mobile phone numbers of the old mobile phone number, the association distance between the old mobile phone number and the new mobile phone number is determined, the customer's face recognition threshold is corrected based on the association distance, facial recognition is performed on the customer based on the corrected face recognition threshold, and whether the mobile phone number modification is supported is determined based on the face recognition result. In the above process, whether the mobile phone number modification is supported is determined by using the customer mobile phone number relationship graph, historical modification data and face recognition threshold of the bank branch, and the above face recognition threshold can be corrected based on the association distance between the old mobile phone number and the new mobile phone number, so that the risk control of mobile phone number modification can be performed in real time and accurately. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0018] Figure 1 Flow chart of a method for controlling mobile phone number risks at bank outlets in an embodiment of the present invention;

[0019] Figure 2 A flow chart of constructing a customer mobile phone number relationship diagram of a bank branch in an embodiment of the present invention;

[0020] Figure 3 A flowchart for calculating the degree of association between each mobile phone number and the mobile phone number associated with the mobile phone number in an embodiment of the present invention;

[0021] Figure 4 A flowchart of constructing a customer mobile phone number relationship diagram according to the association distance in an embodiment of the present invention;

[0022] Figure 5 A flowchart of determining a set of potentially associated mobile phone numbers for each mobile phone number in an embodiment of the present invention;

[0023] Figure 6 A flowchart of determining multiple boundary interval distances corresponding to a customer mobile phone number relationship graph and a boundary association distance corresponding to each boundary interval distance in an embodiment of the present invention;

[0024] Figure 7 The process of determining a potential associated mobile phone number of a mobile phone number in an embodiment of the present invention Figure 1 ;

[0025] Figure 8 The process of determining a potential associated mobile phone number of a mobile phone number in an embodiment of the present invention Figure 2 ;

[0026] Fig. 9 The process of determining a potential associated mobile phone number of a mobile phone number in an embodiment of the present invention Figure 3 ;

[0027] Fig.10 This is a flow chart of determining whether to support mobile phone number modification through face recognition in an embodiment of the present invention;

[0028] Fig.11 A flowchart of correcting a customer's face recognition threshold in an embodiment of the present invention;

[0029] Fig.12Schematic diagram of a bank branch mobile phone number risk control device in an embodiment of the present invention;

[0030] Fig.13 Schematic diagram of a computer device in an embodiment of the present invention. DETAILED DESCRIPTION

[0031] To make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings. Here, the exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0032] In the description of this specification, the terms "include", "including", "have", "contain", etc. are all open terms, which mean including but not limited to. The descriptions with reference to the terms "one embodiment", "a specific embodiment", "some embodiments", "for example", etc. mean that the specific features, structures or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures or characteristics described may be combined in a suitable manner in any one or more embodiments or examples. The order of steps involved in each embodiment is used to schematically illustrate the implementation of the present application, and the order of steps is not limited and can be appropriately adjusted as needed.

[0033] Figure 1 The flowchart of the bank branch mobile phone number risk control method in the embodiment of the present invention includes:

[0034] Step 101, constructing a customer mobile phone number relationship diagram of bank branches;

[0035] Step 102, for each mobile phone number, determine a set of potential associated mobile phone numbers of the mobile phone number according to the customer mobile phone number relationship graph and historical modification data;

[0036] Step 103, when a customer changes his / her mobile phone number at a bank branch, if the new mobile phone number is in the set of potential associated mobile phone numbers of the old mobile phone number, the customer's face recognition is performed based on the set of potential associated mobile phone numbers and the customer's face recognition threshold, and whether the mobile phone number modification is supported is determined based on the face recognition result;

[0037] Step 104: If the new mobile phone number is not in the set of potential associated mobile phone numbers of the old mobile phone number, determine the association distance between the old mobile phone number and the new mobile phone number, correct the customer's face recognition threshold based on the association distance, perform face recognition on the customer based on the corrected face recognition threshold, and determine whether the mobile phone number modification is supported based on the face recognition result.

[0038] Figure 2The flowchart of constructing a customer mobile phone number relationship diagram of a bank branch in an embodiment of the present invention is as follows. In one embodiment, constructing a customer mobile phone number relationship diagram of a bank branch includes:

[0039] Step 201, determining the associated mobile phone number of each mobile phone number;

[0040] Step 202, calculating the correlation between each mobile phone number and the mobile phone number associated with the mobile phone number;

[0041] Step 203, based on the degree of association between each mobile phone number and the mobile phone number associated with the mobile phone number, calculate the association distance between each mobile phone number and the mobile phone number associated with the mobile phone number;

[0042] Step 204: construct a customer mobile phone number relationship graph based on the association distance.

[0043] In one embodiment, determining the associated mobile phone number of each mobile phone number includes:

[0044] The two accounts involved in each transfer data are obtained, and the two mobile phone numbers corresponding to the two accounts are determined as the associated mobile phone numbers of the other party.

[0045] Figure 3 The flowchart of calculating the association degree between each mobile phone number and the mobile phone number associated with the mobile phone number in an embodiment of the present invention, in one embodiment, calculating the association degree between each mobile phone number and the mobile phone number associated with the mobile phone number includes:

[0046] Step 301, for each mobile phone number, determine the amount of transfer data involved between the mobile phone number and each associated mobile phone number of the mobile phone number as the amount of associated data corresponding to each associated mobile phone number of the mobile phone number;

[0047] Step 302, determining the maximum value among all the associated data amounts as the associated data maximum value;

[0048] Step 303: for each mobile phone number, the degree of association between the mobile phone number and each associated mobile phone number of the mobile phone number is determined as the ratio of the amount of associated data corresponding to each associated mobile phone number of the mobile phone number to the maximum value of the associated data.

[0049] It should be noted that the above transfer data includes transfer transactions and payment transactions.

[0050] In one embodiment, calculating the degree of association between each mobile phone number and the mobile phone number associated with the mobile phone number includes:

[0051] Obtain the customer and teller involved in each business handling data, and determine the customer's mobile phone number as the associated mobile phone number of the teller's mobile phone number;

[0052] For each customer's mobile phone number, the amount of business handling data involving the customer and the teller is determined as the amount of associated data corresponding to the customer's mobile phone number and the teller's mobile phone number;

[0053] The maximum value among all the associated data amounts is determined as the associated data maximum value;

[0054] For each customer's mobile phone number, the correlation degree between the customer's mobile phone number and the teller's mobile phone number is set to the ratio of the amount of correlation data between the customer's mobile phone number and the teller's mobile phone number to the maximum value of the correlation data.

[0055] In one embodiment, based on the degree of association between each mobile phone number and the mobile phone number associated with the mobile phone number, calculating the association distance between each mobile phone number and the mobile phone number associated with the mobile phone number includes:

[0056] For each mobile phone number, the association distance between the mobile phone number and each associated mobile phone number of the mobile phone number is determined to be -lg(r), where r is the degree of association between the mobile phone number and each associated mobile phone number of the mobile phone number, and lg is a logarithmic function with base 10.

[0057] Figure 4 The flowchart of constructing a customer mobile phone number relationship graph according to the association distance in an embodiment of the present invention is as follows. In one embodiment, constructing a customer mobile phone number relationship graph according to the association distance includes:

[0058] Step 401, taking each mobile phone number as a node, if and only if the two mobile phone numbers are each other's associated mobile phone numbers, the nodes corresponding to the two mobile phone numbers have an edge;

[0059] Step 402, determining the association distance of the edge as the association distance between two mobile phone numbers corresponding to the edge, and determining the interval distance of the edge as 1.

[0060] Among them, if there is no edge between two nodes, the shortest distance algorithm is needed to calculate.

[0061] Figure 5 The flowchart of determining a set of potentially associated mobile phone numbers for each mobile phone number in an embodiment of the present invention is as follows. In one embodiment, based on a customer mobile phone number relationship graph and historical modification data, for each mobile phone number, a set of potentially associated mobile phone numbers for the mobile phone number is determined, including:

[0062] Step 501, determining a plurality of boundary interval distances corresponding to a customer mobile phone number relationship graph and a boundary association distance corresponding to each boundary interval distance based on historical modification data;

[0063] Step 502: for each mobile phone number, determine a set of potential associated mobile phone numbers for the mobile phone number based on a plurality of boundary interval distances and a boundary association distance corresponding to each boundary interval distance.

[0064] Among them, the historical modification data is the historical modification data of the mobile phone number.

[0065] Figure 6 The flowchart of determining multiple boundary interval distances corresponding to the customer mobile phone number relationship graph and the boundary association distance corresponding to each boundary interval distance in an embodiment of the present invention, in one embodiment, determining multiple boundary interval distances corresponding to the customer mobile phone number relationship graph and the boundary association distance corresponding to each boundary interval distance based on historical modification data includes:

[0066] Step 601, for each piece of historical modification data, determine the mobile phone number involved in the historical modification data;

[0067] Step 602, determining the association distance and interval distance between the mobile phone numbers involved in the historical modification data according to the customer mobile phone number relationship diagram;

[0068] Step 603, determining a partial order of the historical modification data according to the association distance and the interval distance, wherein, for any two historical modification data, the partial order is used to determine whether the first historical modification data of the two historical modification data is greater than the second historical modification data of the two historical modification data; wherein, of the two historical modification data, one is the first historical modification data and the other is the second historical modification data;

[0069] Step 604, determining a plurality of maximum historical modification data among all the historical modification data according to the partial order of the historical modification data, wherein the maximum historical modification data is a maximum element of the partial order;

[0070] Step 605, for each maximum historical modification data, the interval distance between the mobile phone numbers involved in the maximum historical modification data is determined as the boundary interval distance, and the association distance between the mobile phone numbers involved in the maximum historical modification data is determined as the boundary association distance corresponding to the boundary interval distance.

[0071] It should be noted that, in the present application, determining the association distance between the mobile phone numbers involved in the historical modification data refers to determining the shortest distance between the mobile phone numbers involved based on the customer mobile phone number relationship graph, the association distance corresponding to each edge on the graph, and the shortest path algorithm, and using the shortest distance as the association distance between the mobile phone numbers involved; similarly, determining the interval distance between the mobile phone numbers involved in the historical modification data refers to determining the shortest distance between the mobile phone numbers involved based on the customer mobile phone number relationship graph, the interval distance corresponding to each edge on the graph, and the shortest path algorithm, and using the shortest distance as the interval distance between the mobile phone numbers involved.

[0072] It should be noted that the maximum element of the partial order is that in the set corresponding to the partial order, there is no other element that is better than the maximum element. In this application, the maximum historical modification data among all historical modification data means that there is no other historical modification data among all other historical modification data except the maximum historical modification data, so that the other historical modification data is greater than the maximum historical modification data.

[0073] In one embodiment, a method of determining a plurality of maximum historical modification data among all historical modification data according to the partial order of the historical modification data in step 604 is as follows:

[0074] 1. Initialize the maximum flag corresponding to each historical modification data in all historical modification data to possible, and initialize the corresponding comparison flag to yes;

[0075] 2. For each historical modification data in all the historical modification data, if the maximum mark corresponding to the historical modification data is possible, select multiple other historical modification data with corresponding comparison marks as yes from all other historical modification data except the historical modification data in all the historical modification data, and then set the historical modification data to be compared corresponding to the historical modification data to the selected multiple other historical modification data; if the maximum mark corresponding to the historical modification data is not possible, continue to execute step 2 for the next historical modification data;

[0076] 3. Select each historical modification data to be compared corresponding to the historical modification data in turn; if the historical modification data to be compared is better than the historical modification data, set the maximum flag corresponding to the historical modification data to no, and then continue to perform the above step 2 on the next historical modification data; if the historical modification data is better than the historical modification data to be compared, set the maximum flag corresponding to the historical modification data to be compared to no, and determine the historical modification data to be compared as the secondary historical modification data of the historical modification data; otherwise, the maximum flag corresponding to the historical modification data and the maximum flag corresponding to the historical modification data to be compared remain unchanged;

[0077] 4. If it is confirmed that all the historical modification data to be compared with the historical modification data are not better than the historical modification data (that is, after sequentially comparing the historical modification data with each corresponding historical modification data to be compared, the maximum mark corresponding to the historical modification data is still possible), the historical modification data is determined as the maximum historical modification data among all the historical modification data, and the comparison mark of each secondary historical modification data of the maximum historical modification data is updated to no;

[0078] 5. Then continue to execute step 2 for the next historical modification data until the above steps are completed for all historical modification data.

[0079] In one embodiment, determining the partial order of the historical modification data according to the association distance and the interval distance includes:

[0080] Determine the partial order of historical modification data, wherein, for any two historical modification data, if the association distance between the mobile phone numbers involved in the first historical modification data of the two historical modification data is greater than or equal to the association distance between the mobile phone numbers involved in the second historical modification data of the two historical modification data, and the interval distance between the mobile phone numbers involved in the first historical modification data is greater than or equal to the interval distance between the mobile phone numbers involved in the second historical modification data, then determine that the first historical modification data is greater than the second historical modification data.

[0081] Figure 7 The process of determining a potential associated mobile phone number of a mobile phone number in an embodiment of the present invention Figure 1 In one embodiment, for each mobile phone number, based on a plurality of boundary interval distances and a boundary association distance corresponding to each boundary interval distance, determining the potential associated mobile phone number of the mobile phone number includes:

[0082] Step 701, in the customer mobile phone number relationship graph, for each mobile phone number, calculate the association distance between the mobile phone number and all other mobile phone numbers and the interval distance between the mobile phone number and all mobile phone numbers;

[0083] Step 702: For each other mobile phone number, if there is a boundary separation distance and a boundary association distance corresponding to the boundary separation distance, so that the separation distance between the mobile phone number and the other mobile phone number is less than or equal to the boundary separation distance, and the association distance between the mobile phone number and the other mobile phone number is less than or equal to the boundary association distance corresponding to the boundary separation distance, then the other mobile phone number is determined as a potential associated mobile phone number of the mobile phone number.

[0084] It should be noted that Figure 7 The corresponding method is applicable when the structural complexity of the customer mobile phone number relationship graph is less than the first threshold. Otherwise, the complexity of the corresponding method will be very high, that is, the calculation of the method is redundant. When the structural complexity of the customer mobile phone number relationship graph is not less than the first threshold, the following process can be used to determine the potential associated mobile phone numbers of a mobile phone number: Figure 2 and Process Figure 3 The corresponding method.

[0085] Figure 8 The process of determining a potential associated mobile phone number of a mobile phone number in an embodiment of the present invention Figure 2Here is another method for determining potential associated mobile phone numbers of a mobile phone number. In one embodiment, for each mobile phone number, based on a plurality of boundary interval distances and a boundary association distance corresponding to each boundary interval distance, determining the potential associated mobile phone numbers of the mobile phone number includes:

[0086] Step 801: When the structural complexity of the customer mobile phone number relationship graph is not less than a first threshold, the corresponding edge with a correlation distance greater than a specified threshold is deleted from the customer mobile phone number relationship graph, and in the customer mobile phone number relationship graph, the following steps are used to determine the potential associated mobile phone number of the mobile phone number:

[0087] Step 802: for each mobile phone number, the mobile phone numbers associated with the mobile phone number are added to the queue in order of the distance associated with the mobile phone number from small to large, and the first distance between the mobile phone number and each associated mobile phone number of the mobile phone number is determined to be the interval distance between the edges corresponding to the mobile phone number and the associated mobile phone number in the customer mobile phone number relationship graph, and the second distance between the mobile phone number and each associated mobile phone number of the mobile phone number is the association distance between the edges corresponding to the mobile phone number and the associated mobile phone number in the customer mobile phone number relationship graph. The following steps are repeated until the queue corresponding to the mobile phone number is empty, and a set of potential associated mobile phone numbers of the mobile phone number is obtained:

[0088] Step 8021, popping the first mobile phone number from the queue, and adding the first mobile phone number to the set of potential associated mobile phone numbers of the mobile phone number;

[0089] Step 8022, obtain the associated mobile phone numbers of the first mobile phone number in the team, and for each associated mobile phone number of the first mobile phone number in the team, determine whether the associated mobile phone number is added to the set of potential associated mobile phone numbers of the mobile phone number; if not, determine the first distance and the second distance between the mobile phone number and the associated mobile phone number; if there is a boundary interval distance and a boundary association distance corresponding to the boundary interval distance, so that the first distance between the mobile phone number and the associated mobile phone number is less than or equal to the boundary interval distance, and the second distance between the mobile phone number and the associated mobile phone number is less than or equal to the boundary association distance corresponding to the boundary interval distance, then the associated mobile phone number is used as the mobile phone number to be added to the first mobile phone number in the team; the mobile phone numbers to be added to the first mobile phone number in the team are added to the queue in ascending order of the second distance to the mobile phone number.

[0090] Wherein, the calculation method for determining the first distance between the mobile phone number and the associated mobile phone number (the second distance is similar and will not be repeated here): iIN(d1(i)+d2(i)), where d1(i) is the first distance between the mobile phone number and the i-th potential associated mobile phone number in the set of potential associated mobile phone numbers that have been added to the mobile phone number and have direct edges with the associated mobile phone number in the customer mobile phone number relationship graph, and d2(i) is the interval distance between the associated mobile phone number and the corresponding edge of the i-th potential associated mobile phone number in the customer mobile phone number relationship graph among all the potential associated mobile phone numbers that have been added to the set of potential associated mobile phone numbers that have been directly connected to the associated mobile phone number in the customer mobile phone number relationship graph.

[0091] Fig. 9 The process of determining a potential associated mobile phone number of a mobile phone number in an embodiment of the present invention Figure 3 Here is another method for determining potential associated mobile phone numbers of a mobile phone number. In one embodiment, for each mobile phone number, based on a plurality of boundary interval distances and a boundary association distance corresponding to each boundary interval distance, determining the potential associated mobile phone numbers of the mobile phone number includes:

[0092] Step 901, for each mobile phone number A, add all the associated mobile phone numbers of the mobile phone number A to the to-be-selected set corresponding to the mobile phone number A; and initialize the third distance corresponding to each associated mobile phone number of the mobile phone number A to the interval distance between the mobile phone number A and the corresponding edge of the associated mobile phone number in the customer mobile phone number relationship graph, and initialize the corresponding fourth distance to the association distance between the mobile phone number A and the corresponding edge of the associated mobile phone number in the customer mobile phone number relationship graph; for other mobile phone numbers except the mobile phone number A and the associated mobile phone numbers of the mobile phone number A, initialize the corresponding third distance to a value greater than all boundary interval distances, and initialize the corresponding fourth distance to a value greater than all boundary association distances;

[0093] Step 902, select the mobile phone number B with the smallest corresponding association distance from the to-be-selected set corresponding to the mobile phone number A; if there is a boundary interval distance and a boundary association distance corresponding to the boundary interval distance, so that the third distance between the mobile phone number A and the selected mobile phone number B is less than or equal to the boundary interval distance, and the fourth distance between the mobile phone number A and the selected mobile phone number B is less than or equal to the boundary association distance corresponding to the boundary interval distance, then add the selected mobile phone number B to the set of potential associated mobile phone numbers of the mobile phone number A, and delete the selected mobile phone number B from the to-be-selected set corresponding to the mobile phone number A; otherwise, stop determining the potential associated mobile phone numbers of the mobile phone number A;

[0094] Step 903, for each associated mobile phone number C of the selected mobile phone number B, determine whether the associated mobile phone number C has been added to the potential associated mobile phone number set of the mobile phone number A, and if not, add the associated mobile phone number C to the to-be-selected set corresponding to the mobile phone number A;

[0095] Step 904: for each associated mobile phone number C of the selected mobile phone number B, update the third distance of the associated mobile phone number C to min(d1 Co ,d1 B +d1(B,C)), where d1 Co is the third distance of the associated mobile phone number C before the update, d1 B is the third distance of the selected mobile phone number B, d1(B,C) is the distance between the corresponding edges of mobile phone number B and mobile phone number C in the customer mobile phone number relationship graph, and the fourth distance of the associated mobile phone number C is updated to min(d2 Co ,d2 B +d2(B,C)), where d2 Co is the fourth distance of the associated mobile phone number C before the update, d2 B is the fourth distance of the selected mobile phone number B, and d2(B,C) is the association distance of the corresponding edges of mobile phone number B and mobile phone number C in the customer mobile phone number relationship graph;

[0096] Step 905, after steps 903 and 904 are executed for all the associated mobile phone numbers of the selected mobile phone number B, execute step 902.

[0097] Fig.10 The flowchart of determining whether to support the modification of the mobile phone number through face recognition in an embodiment of the present invention, in one embodiment, if the new mobile phone number is in the set of potential associated mobile phone numbers of the old mobile phone number, the customer is subjected to face recognition according to the set of potential associated mobile phone numbers and the face recognition threshold of the customer, and whether the mobile phone number modification is supported is determined according to the face recognition result, including:

[0098] Step 1001, compare the collected customer face with the old mobile phone number and the customer faces in the set of potential associated mobile phone numbers of the old mobile phone number one by one to obtain a matching value;

[0099] Step 1002, determining the face corresponding to the mobile phone number with the largest matching value as the face of the customer who is applying to change the mobile phone number this time;

[0100] Step 1003: If the customer corresponding to the determined face and the customer corresponding to the old mobile phone number are the same customer, it is determined that the mobile phone number modification is supported; otherwise, the mobile phone number modification is not supported.

[0101] Fig.11 This is a flow chart of correcting a customer's face recognition threshold in an embodiment of the present invention. In one embodiment, determining the association distance between the old mobile phone number and the new mobile phone number, and correcting the customer's face recognition threshold according to the association distance includes:

[0102] Step 1101, determining the association distance between two nodes corresponding to the old mobile phone number and the new mobile phone number in the customer mobile phone number relationship graph according to the shortest path algorithm;

[0103] Step 1102, determining the correlation between the old mobile phone number and the new mobile phone number according to the correlation distance between two corresponding nodes in the customer mobile phone number relationship graph;

[0104] Step 1103, selecting, from the historical modification data, the historical modification data of the two corresponding mobile phone numbers whose correlation degree is equal to the correlation degree according to the correlation degree;

[0105] Step 1104, determining a face recognition threshold corresponding to the correlation degree based on the selected historical modification data, so that when the face recognition threshold of the selected historical modification data is set to the face recognition threshold corresponding to the correlation degree, the corresponding risk coefficient is less than or equal to a set value;

[0106] Step 1105, the face recognition threshold corresponding to the modified mobile phone number is corrected to the maximum value of the face recognition threshold corresponding to the association degree of the two mobile phone numbers and the previous face recognition threshold of the customer.

[0107] Among them, according to the association distance between two nodes corresponding to the old mobile phone number and the new mobile phone number in the customer mobile phone number relationship graph, the association degree between the old mobile phone number and the new mobile phone number is determined, which can be: the association degree between the old mobile phone number and the new mobile phone number is determined as: Where s1 is the association distance between the two nodes corresponding to the old mobile phone number and the new mobile phone number in the customer mobile phone number relationship graph.

[0108] For each historical modification data, the correlation between the two mobile phone numbers corresponding to the historical modification data is determined as follows: based on the customer mobile phone number relationship graph, the correlation distance corresponding to each edge on the graph and the shortest path algorithm, the shortest distance between the two mobile phone numbers is determined, and the shortest distance is used as the correlation distance s between the two mobile phone numbers; the correlation between the two mobile phone numbers is determined to be 10 -s .

[0109] Among them, the corresponding risk coefficient is determined as the proportion of risk data in the selected historical modification data when the face recognition threshold of the selected historical modification data is set to the face recognition threshold corresponding to the associated distance. Risk data is data with risks. For example, when the face recognition threshold is set relatively low, others can tamper with the customer's mobile phone number and cause risks to the customer. If the set value is set relatively high, this risk data (before modification) is no longer risk data (after modification). The higher the face recognition threshold, the smaller the risk coefficient.). The set value can be the risk coefficient corresponding to all mobile phone number modification business data (that is, the proportion of risk data in all mobile phone number modification business data), and the risk coefficient is stored in the bank server.

[0110] In summary, in the method proposed in the embodiment of the present invention, a customer mobile phone number relationship graph of a bank branch is constructed; based on the customer mobile phone number relationship graph and historical modification data, for each mobile phone number, a set of potential associated mobile phone numbers of the mobile phone number is determined; when a customer modifies a mobile phone number at a bank branch, if the new mobile phone number is in the set of potential associated mobile phone numbers of the old mobile phone number, the customer is subjected to face recognition based on the set of potential associated mobile phone numbers and the customer's face recognition threshold, and whether the mobile phone number modification is supported is determined based on the face recognition result; if the new mobile phone number is not in the set of potential associated mobile phone numbers of the old mobile phone number, the association distance between the old mobile phone number and the new mobile phone number is determined, the customer's face recognition threshold is corrected based on the association distance, the customer is subjected to face recognition based on the corrected face recognition threshold, and whether the mobile phone number modification is supported is determined based on the face recognition result. In the above process, whether the mobile phone number modification is supported is determined by using the customer mobile phone number relationship graph, historical modification data and face recognition threshold of the bank branch, and the above face recognition threshold can be corrected based on the association distance between the old mobile phone number and the new mobile phone number, so that the risk control of mobile phone number modification can be performed in real time and accurately.

[0111] The embodiment of the present invention further proposes a bank branch mobile phone number risk control device, the principle of which is similar to the bank branch mobile phone number risk control method, which will not be repeated here.

[0112] Fig.12 This is a schematic diagram of a bank branch mobile phone number risk control device in an embodiment of the present invention, comprising:

[0113] A customer mobile phone number relationship graph construction module 1201 is used to construct a customer mobile phone number relationship graph in a bank;

[0114] Potentially associated personnel determination module 1202, for determining a set of potentially associated mobile phone numbers for each mobile phone number based on the customer mobile phone number relationship graph and historical modification data;

[0115] The mobile phone number modification judgment module 1203 is used for, when a customer modifies his mobile phone number at a bank branch, if the new mobile phone number is in the potential associated mobile phone number set of the old mobile phone number, performing face recognition on the customer based on the potential associated mobile phone number set and the customer's face recognition threshold, and judging whether the mobile phone number modification is supported based on the face recognition result;

[0116] The face recognition threshold correction module 1204 is used to determine the correlation distance between the old mobile phone number and the new mobile phone number if the new mobile phone number is not in the set of potential associated mobile phone numbers of the old mobile phone number, correct the customer's face recognition threshold according to the correlation distance, perform face recognition on the customer based on the corrected face recognition threshold, and determine whether the mobile phone number modification is supported based on the face recognition result.

[0117] In one embodiment, the customer mobile phone number relationship graph construction module is specifically used to:

[0118] Determine the associated mobile phone number for each mobile phone number;

[0119] Calculate the correlation between each mobile phone number and its associated mobile phone number;

[0120] Based on the degree of association between each mobile phone number and the mobile phone number associated with the mobile phone number, calculate the association distance between each mobile phone number and the mobile phone number associated with the mobile phone number;

[0121] A customer mobile phone number relationship graph is constructed based on the association distance.

[0122] In one embodiment, the customer mobile phone number relationship graph construction module is specifically used to:

[0123] The two accounts involved in each transfer data are obtained, and the two mobile phone numbers corresponding to the two accounts are determined as the associated mobile phone numbers of the other party.

[0124] In one embodiment, the customer mobile phone number relationship graph construction module is specifically used to:

[0125] For each mobile phone number, the amount of transfer data involved between the mobile phone number and each associated mobile phone number of the mobile phone number is determined as the amount of associated data corresponding to each associated mobile phone number of the mobile phone number;

[0126] The maximum value among all the associated data amounts is determined as the associated data maximum value;

[0127] For each mobile phone number, the degree of association between the mobile phone number and each associated mobile phone number of the mobile phone number is determined as the ratio of the amount of associated data corresponding to each associated mobile phone number of the mobile phone number to the maximum value of the associated data.

[0128] In one embodiment, the customer mobile phone number relationship graph construction module is specifically used to:

[0129] For each mobile phone number, the association distance between the mobile phone number and each associated mobile phone number of the mobile phone number is determined to be -lg(r), where r is the degree of association between the mobile phone number and each associated mobile phone number of the mobile phone number, and lg is a logarithmic function with base 10.

[0130] In one embodiment, the customer mobile phone number relationship graph construction module is specifically used to:

[0131] Take each mobile phone number as a node;

[0132] If and only if the two mobile phone numbers are each other's associated mobile phone numbers, the nodes corresponding to the two mobile phone numbers have an edge;

[0133] The association distance of an edge is determined to be the association distance between the two mobile phone numbers corresponding to the edge, and the interval distance of the edge is determined to be 1.

[0134] In one embodiment, the potential associated person determination module is specifically used to:

[0135] Determine multiple boundary interval distances corresponding to the customer mobile phone number relationship graph and boundary association distances corresponding to each boundary interval distance based on historical modification data;

[0136] For each mobile phone number, a set of potential associated mobile phone numbers of the mobile phone number is determined based on a plurality of boundary interval distances and a boundary association distance corresponding to each boundary interval distance.

[0137] In one embodiment, the potential associated person determination module is specifically used to:

[0138] For each piece of historical modification data, determine the mobile phone number involved in the historical modification data;

[0139] According to the customer mobile phone number relationship diagram, determine the association distance and interval distance between the mobile phone numbers involved in the historical modification data;

[0140] Determine a partial order of the historical modification data according to the association distance and the interval distance, wherein, for any two historical modification data, the partial order is used to determine whether a first historical modification data of the two historical modification data is greater than a second historical modification data of the two historical modification data;

[0141] Determine, according to the partial order of the historical modification data, a plurality of maximum historical modification data among all the historical modification data, wherein the maximum historical modification data is a maximum element of the partial order;

[0142] For each maximum historical modification data, the interval distance between the mobile phone numbers involved in the maximum historical modification data is determined as the boundary interval distance, and the association distance between the mobile phone numbers involved in the maximum historical modification data is determined as the boundary association distance corresponding to the boundary interval distance.

[0143] In one embodiment, the potential associated person determination module is specifically used to:

[0144] Determine the partial order of historical modification data, wherein, for any two historical modification data, if the association distance between the mobile phone numbers involved in the first historical modification data of the two historical modification data is greater than or equal to the association distance between the mobile phone numbers involved in the second historical modification data of the two historical modification data, and the interval distance between the mobile phone numbers involved in the first historical modification data is greater than or equal to the interval distance between the mobile phone numbers involved in the second historical modification data, then determine that the first historical modification data is greater than the second historical modification data.

[0145] In one embodiment, the potential associated person determination module is specifically used to:

[0146] When the structural complexity of the customer mobile phone number relationship graph is less than the first threshold, in the customer mobile phone number relationship graph, for each mobile phone number, the association distance between the mobile phone number and all other mobile phone numbers and the interval distance between the mobile phone number and all mobile phone numbers are calculated;

[0147] For each other mobile phone number, if there is a boundary separation distance and a boundary association distance corresponding to the boundary separation distance, so that the separation distance between the mobile phone number and the other mobile phone number is less than or equal to the boundary separation distance, and the association distance between the mobile phone number and the other mobile phone number is less than or equal to the boundary association distance corresponding to the boundary separation distance, the other mobile phone number is determined as a potential associated mobile phone number of the mobile phone number.

[0148] In one embodiment, the mobile phone number modification determination module is specifically used to:

[0149] The collected customer face is compared with the old mobile phone number and the customer faces in the set of potential associated mobile phone numbers of the old mobile phone number one by one to obtain a matching value;

[0150] The face corresponding to the mobile phone number with the largest matching value is determined as the face of the customer who is applying to change the mobile phone number this time;

[0151] If the customer corresponding to the confirmed face and the customer corresponding to the old mobile phone number are the same customer, the mobile phone number modification is supported, otherwise the mobile phone number modification is not supported.

[0152] In one embodiment, the face recognition threshold correction module is specifically used to:

[0153] Determine the association distance between two nodes corresponding to the old mobile phone number and the new mobile phone number in the customer mobile phone number relationship graph based on the shortest path algorithm;

[0154] According to the association distance, selecting from the historical modification data, the historical modification data of the two corresponding mobile phone numbers whose association distance is equal to the association distance;

[0155] Determine, based on the selected historical modification data, a face recognition threshold corresponding to the associated distance, so that when the face recognition threshold of the selected historical modification data is set to the face recognition threshold corresponding to the associated distance, the corresponding risk coefficient is less than or equal to a set value;

[0156] The face recognition threshold corresponding to the modified mobile phone number is corrected to the face recognition threshold corresponding to the associated distance between the two mobile phone numbers and the maximum value of the previous face recognition threshold of the customer.

[0157] To summarize, in the device proposed in the embodiment of the present invention, the customer mobile phone number relationship graph construction module is used to construct a customer mobile phone number relationship graph in the bank; the potential associated personnel determination module is used to determine, for each mobile phone number, a set of potential associated mobile phone numbers of the mobile phone number based on the customer mobile phone number relationship graph and historical modification data; the mobile phone number modification judgment module is used when a customer modifies his mobile phone number at a bank branch, if the new mobile phone number is in the set of potential associated mobile phone numbers of the old mobile phone number, perform face recognition on the customer based on the set of potential associated mobile phone numbers and the customer's face recognition threshold, and judge whether the mobile phone number modification is supported according to the face recognition result; the face recognition threshold correction module is used to determine the association distance between the old mobile phone number and the new mobile phone number if the new mobile phone number is not in the set of potential associated mobile phone numbers of the old mobile phone number, correct the customer's face recognition threshold according to the association distance, correct the customer's face recognition threshold according to the association distance, perform face recognition on the customer based on the corrected face recognition threshold, and judge whether the mobile phone number modification is supported according to the face recognition result. In the above process, whether the mobile phone number modification is supported is determined through the customer mobile phone number relationship diagram, historical modification data and face recognition threshold of the bank branch, and the above face recognition threshold can be corrected according to the correlation distance between the old mobile phone number and the new mobile phone number, so that the risk control of mobile phone number modification can be carried out in real time and accurately.

[0158] An embodiment of the present invention further provides a computer device, Fig.13 It is a schematic diagram of a computer device in an embodiment of the present invention. The computer device 1300 includes a memory 1310, a processor 1320, and a computer program 1330 stored in the memory 1310 and executable on the processor 1320. When the processor 1320 executes the computer program 1330, the above-mentioned bank branch mobile phone number risk control method is implemented.

[0159] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned bank branch mobile phone number risk control method is implemented.

[0160] An embodiment of the present invention further provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the above-mentioned bank branch mobile phone number risk control method is implemented.

[0161] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program business systems. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program business system implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0162] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program business systems according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0163] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0164] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0165] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A bank branch mobile phone number risk control method, characterized in that: include: Build a customer mobile phone number relationship map for bank branches; Based on the customer mobile phone number relationship graph and historical modification data, for each mobile phone number, determine the set of potential associated mobile phone numbers for the mobile phone number; When a customer changes his / her mobile phone number at a bank branch, if the new mobile phone number is in the set of potential associated mobile phone numbers of the old mobile phone number, the customer's face recognition is performed based on the set of potential associated mobile phone numbers and the customer's face recognition threshold, and whether the mobile phone number change is supported is determined based on the face recognition result; If the new mobile phone number is not in the set of potential associated mobile phone numbers of the old mobile phone number, determine the association distance between the old mobile phone number and the new mobile phone number, modify the customer's face recognition threshold according to the association distance, perform face recognition on the customer based on the modified face recognition threshold, and determine whether the mobile phone number modification is supported based on the face recognition result; Constructing a customer mobile phone number relationship graph of a bank branch, including: determining the associated mobile phone number of each mobile phone number; calculating the degree of association between each mobile phone number and the associated mobile phone number of the mobile phone number; calculating the association distance between each mobile phone number and the associated mobile phone number of the mobile phone number based on the degree of association between each mobile phone number and the associated mobile phone number of the mobile phone number; constructing a customer mobile phone number relationship graph according to the association distance; Determining the associated mobile phone number of each mobile phone number includes: obtaining two accounts involved in each transfer data, and determining two mobile phone numbers corresponding to the two accounts as the associated mobile phone numbers of the other party; Calculating the degree of association between each mobile phone number and the mobile phone numbers associated with the mobile phone number, including: for each mobile phone number, determining the amount of transfer data involved between the mobile phone number and each mobile phone number associated with the mobile phone number as the amount of associated data corresponding to each mobile phone number associated with the mobile phone number; determining the maximum value of all the associated data amounts as the maximum value of associated data; for each mobile phone number, determining the degree of association between the mobile phone number and each mobile phone number associated with the mobile phone number as the ratio of the amount of associated data corresponding to each mobile phone number associated with the mobile phone number to the maximum value of associated data; According to the association distance, construct a customer mobile phone number relationship graph, including: taking each mobile phone number as a node; if and only if two mobile phone numbers are each other's associated mobile phone numbers, the nodes corresponding to the two mobile phone numbers have an edge; determining the association distance of the edge as the association distance between the two mobile phone numbers corresponding to the edge, and determining the interval distance of the edge as 1; Determine the association distance between the old mobile phone number and the new mobile phone number, and correct the customer's face recognition threshold according to the association distance, including: determining the association distance between two nodes corresponding to the old mobile phone number and the new mobile phone number in the customer's mobile phone number relationship graph based on the shortest path algorithm; determining the association degree between the old mobile phone number and the new mobile phone number based on the association distance between two nodes corresponding to the old mobile phone number and the new mobile phone number in the customer's mobile phone number relationship graph; based on the association degree, select the historical modification data whose association degree of the corresponding two mobile phone numbers is equal to the association degree from the historical modification data; based on the selected historical modification data, determine the face recognition threshold corresponding to the association degree, so that when the face recognition threshold of the selected historical modification data is set to the face recognition threshold corresponding to the association degree, the corresponding risk coefficient is less than or equal to the set value; correct the face recognition threshold corresponding to the currently modified mobile phone number to the face recognition threshold corresponding to the association degree of the two mobile phone numbers and the maximum value of the customer's previous face recognition threshold.

2. The method according to claim 1, characterized in that Based on the degree of association between each mobile phone number and the mobile phone number associated with the mobile phone number, the association distance between each mobile phone number and the mobile phone number associated with the mobile phone number is calculated, including: For each mobile phone number, the association distance between the mobile phone number and each associated mobile phone number of the mobile phone number is determined to be -lg(r), where r is the degree of association between the mobile phone number and each associated mobile phone number of the mobile phone number, and lg is a logarithmic function with base 10.

3. The method according to claim 1, characterized in that Based on the customer mobile phone number relationship graph and historical modification data, for each mobile phone number, determine the set of potential associated mobile phone numbers for the mobile phone number, including: Determine multiple boundary interval distances corresponding to the customer mobile phone number relationship graph and boundary association distances corresponding to each boundary interval distance based on historical modification data; For each mobile phone number, determine a set of potential associated mobile phone numbers for the mobile phone number based on a plurality of boundary interval distances and a boundary association distance corresponding to each boundary interval distance; Based on the historical modification data, multiple boundary interval distances corresponding to the customer mobile phone number relationship graph and the boundary association distance corresponding to each boundary interval distance are determined, including: for each historical modification data, the mobile phone number involved in the historical modification data is determined; based on the customer mobile phone number relationship graph, the association distance and the interval distance between the mobile phone numbers involved in the historical modification data are determined; based on the association distance and the interval distance, the partial order of the historical modification data is determined, wherein, for any two historical modification data, the partial order is used to determine whether the first historical modification data in the two historical modification data is greater than the second historical modification data in the two historical modification data; based on the partial order of the historical modification data, multiple maximum historical modification data in all the historical modification data are determined, wherein the maximum historical modification data is the maximum element of the partial order; for each maximum historical modification data, the interval distance between the mobile phone numbers involved in the maximum historical modification data is determined as the boundary interval distance, and the association distance between the mobile phone numbers involved in the maximum historical modification data is determined as the boundary association distance corresponding to the boundary interval distance; Among them, determining the association distance between the mobile phone numbers involved in the historical modification data refers to determining the shortest distance between the mobile phone numbers involved based on the customer mobile phone number relationship graph, the association distance corresponding to each edge on the customer mobile phone number relationship graph, and the shortest path algorithm, and using the shortest distance as the association distance between the mobile phone numbers involved; determining the interval distance between the mobile phone numbers involved in the historical modification data refers to determining the shortest distance between the mobile phone numbers involved based on the customer mobile phone number relationship graph, the interval distance corresponding to each edge on the customer mobile phone number relationship graph, and the shortest path algorithm, and using the shortest distance as the interval distance between the mobile phone numbers involved; The maximum element of the partial order means that in the set corresponding to the partial order, no other element is superior to the maximum element; the maximum historical modification data among all historical modification data means that among all historical modification data except the maximum historical modification data, no other historical modification data exists, such that the other historical modification data is greater than the maximum historical modification data; Determining the partial order of the historical modification data according to the association distance and the interval distance includes: determining the partial order of the historical modification data, wherein, for any two historical modification data, if the association distance between the mobile phone numbers involved in the first historical modification data of the two historical modification data is greater than or equal to the association distance between the mobile phone numbers involved in the second historical modification data of the two historical modification data, and the interval distance between the mobile phone numbers involved in the first historical modification data is greater than or equal to the interval distance between the mobile phone numbers involved in the second historical modification data, then determining that the first historical modification data is greater than the second historical modification data; For each mobile phone number, based on multiple boundary interval distances and boundary association distances corresponding to each boundary interval distance, determine the potential associated mobile phone numbers of the mobile phone number, including: in the customer mobile phone number relationship graph, for each mobile phone number, calculate the association distance between the mobile phone number and all other mobile phone numbers and the interval distance between the mobile phone number and all mobile phone numbers; for each other mobile phone number, if there is a boundary interval distance and a boundary association distance corresponding to the boundary interval distance, so that the interval distance between the mobile phone number and the other mobile phone number is less than or equal to the boundary interval distance, and the association distance between the mobile phone number and the other mobile phone number is less than or equal to the boundary association distance corresponding to the boundary interval distance, determine the other mobile phone number as the potential associated mobile phone number of the mobile phone number; According to the partial order of the historical modification data, multiple extremely large historical modification data among all the historical modification data are determined, including: Step 1: Initialize the maximum flag corresponding to each historical modification data in all historical modification data to possible, and initialize the corresponding comparison flag to yes; Step 2: for each historical modification data in all the historical modification data, if the maximum mark corresponding to the historical modification data is possible, select multiple other historical modification data with corresponding comparison marks as yes from all other historical modification data except the historical modification data in all the historical modification data, and then set the historical modification data to be compared corresponding to the historical modification data to the selected multiple other historical modification data; if the maximum mark corresponding to the historical modification data is not possible, continue to execute step 2 for the next historical modification data; Step 3: select each historical modification data to be compared corresponding to the historical modification data in turn; if the historical modification data to be compared is better than the historical modification data, set the maximum flag corresponding to the historical modification data to be no, and then continue to execute the above step 2 for the next historical modification data; if the historical modification data is better than the historical modification data to be compared, set the maximum flag corresponding to the historical modification data to be compared to no, and determine the historical modification data to be compared as the secondary historical modification data of the historical modification data; otherwise, the maximum flag corresponding to the historical modification data and the maximum flag corresponding to the historical modification data to be compared remain unchanged; Step 4: If it is confirmed that all the historical modification data to be compared with the historical modification data are not superior to the historical modification data, the historical modification data is determined as the maximum historical modification data among all the historical modification data, and the comparison flag of each secondary historical modification data of the maximum historical modification data is updated to no; Step 5: Continue to execute step 2 for the next historical modification data until the above steps are completed for all historical modification data.

4. The method according to claim 1, characterized in that If the new mobile phone number is in the set of potential associated mobile phone numbers of the old mobile phone number, the customer's face recognition is performed based on the set of potential associated mobile phone numbers and the customer's face recognition threshold, and whether the mobile phone number modification is supported is determined based on the face recognition result, including: The collected customer face is compared with the old mobile phone number and the customer faces in the set of potential associated mobile phone numbers of the old mobile phone number one by one to obtain a matching value; The face corresponding to the mobile phone number with the largest matching value is determined as the face of the customer who is applying to change the mobile phone number this time; If the customer corresponding to the confirmed face and the customer corresponding to the old mobile phone number are the same customer, the mobile phone number modification is supported, otherwise the mobile phone number modification is not supported.

5. A bank branch mobile phone number risk control device, characterized in that: include: Customer mobile phone number relationship graph construction module, used to construct the customer mobile phone number relationship graph in the bank; Potentially associated personnel determination module, used to determine the set of potentially associated mobile phone numbers for each mobile phone number based on the customer mobile phone number relationship diagram and historical modification data; The mobile phone number modification judgment module is used when a customer modifies his mobile phone number at a bank branch. If the new mobile phone number is in the potential associated mobile phone number set of the old mobile phone number, the customer's face recognition is performed based on the potential associated mobile phone number set and the customer's face recognition threshold, and whether the mobile phone number modification is supported is determined based on the face recognition result; A face recognition threshold correction module is used to determine the correlation distance between the old and new mobile phone numbers if the new mobile phone number is not in the set of potential associated mobile phone numbers of the old mobile phone number, correct the customer's face recognition threshold according to the correlation distance, perform face recognition on the customer according to the corrected face recognition threshold, and determine whether the mobile phone number modification is supported according to the face recognition result; The customer mobile phone number relationship graph construction module is specifically used to: determine the associated mobile phone number of each mobile phone number; calculate the degree of association between each mobile phone number and the associated mobile phone number of the mobile phone number; based on the degree of association between each mobile phone number and the associated mobile phone number of the mobile phone number, calculate the association distance between each mobile phone number and the associated mobile phone number of the mobile phone number; and construct the customer mobile phone number relationship graph according to the association distance; The customer mobile phone number relationship graph construction module is specifically used to: obtain the two accounts involved in each transfer data, and determine the two mobile phone numbers corresponding to the two accounts as the other party's associated mobile phone numbers; The customer mobile phone number relationship graph construction module is specifically used to: for each mobile phone number, determine the amount of transfer data involved between the mobile phone number and each associated mobile phone number of the mobile phone number as the amount of associated data corresponding to each associated mobile phone number of the mobile phone number; determine the maximum value among all the associated data amounts as the maximum associated data value; for each mobile phone number, determine the degree of association between the mobile phone number and each associated mobile phone number of the mobile phone number as the ratio of the amount of associated data corresponding to each associated mobile phone number of the mobile phone number to the maximum associated data value; The customer mobile phone number relationship graph construction module is specifically used to: take each mobile phone number as a node; if and only if two mobile phone numbers are each other's associated mobile phone numbers, the nodes corresponding to the two mobile phone numbers have an edge; determine the associated distance of the edge as the associated distance between the two mobile phone numbers corresponding to the edge, and determine the interval distance of the edge as 1; The face recognition threshold correction module is specifically used for: determining the association distance between the two nodes corresponding to the old mobile phone number and the new mobile phone number in the customer mobile phone number relationship diagram based on the shortest path algorithm; based on the association distance, selecting the historical modification data whose association distance between the corresponding two mobile phone numbers is equal to the association distance from the historical modification data; based on the selected historical modification data, determining the face recognition threshold corresponding to the association distance, so that when the face recognition threshold of the selected historical modification data is set to the face recognition threshold corresponding to the association distance, the corresponding risk coefficient is less than or equal to the set value; correcting the face recognition threshold corresponding to the currently modified mobile phone number to the face recognition threshold corresponding to the association distance between the two mobile phone numbers and the maximum value of the customer's previous face recognition threshold.

6. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 4 is implemented.

7. 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 a processor, the method according to any one of claims 1 to 4 is implemented.

8. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

Images