Method and device for placing products at bank branches

Abstract

The present invention provides a method and device for placing products in bank branches, which are applied to the field of financial technology. The method includes: classifying bank branches according to the business data of the bank branches to obtain multiple branch categories; determining the safety coefficient vector and the upper bound of error convergence of each product in each branch category according to the product transaction data of each branch category; determining the extreme value products of each branch category according to the safety coefficient vector and the upper bound of error convergence of each product in each branch category; determining the safety coefficient vector of the placed product according to the product transaction data of the placed product; and determining the bank branch corresponding to the placed product according to the safety coefficient vector of the placed product and the extreme value products of each branch category. The present invention can determine the products to be placed in bank branches so as to reduce the banking business risks.

Description

Technical Field

[0001] The present invention relates to the field of financial technologies, and in particular, to a method and device for placing products at bank branches. Background Art

[0002] When placing products in banks, not only the business needs of customers but also the possible risks of the products need to be considered. Currently, at bank branches, the placement of products lacks a data basis and digital risk control cannot be achieved. Summary of the Invention

[0003] An embodiment of the present invention provides a method for placing products at bank branches to determine the products to be placed at bank branches and reduce banking business risks. The method includes:

[0004] Classifying bank branches according to the business data of bank branches to obtain multiple branch categories;

[0005] Determining the safety coefficient vector and the upper bound of error convergence of each product in each branch category according to the product transaction data of each branch category;

[0006] Determining the extreme value products of each branch category according to the safety coefficient vector and the upper bound of error convergence of each product in each branch category;

[0007] Determining the safety coefficient vector of the product to be placed according to the product transaction data of the product to be placed;

[0008] Determining the bank branches corresponding to the product to be placed according to the safety coefficient vector of the product to be placed and the extreme value products of each branch category.

[0009] An embodiment of the present invention provides a device for placing products at bank branches to determine the products to be placed at targeted bank branches and reduce banking business risks. The device includes:

[0010] A branch category obtaining module, configured to classify bank branches according to the business data of bank branches to obtain multiple branch categories;

[0011] A product transaction data analysis module, configured to determine the safety coefficient vector and the upper bound of error convergence of each product in each branch category according to the product transaction data of each branch category;

[0012] An extreme value product determination module for branch categories, configured to determine the extreme value products of each branch category according to the safety coefficient vector and the upper bound of error convergence of each product in each branch category;

[0013] A safety coefficient vector determination module, configured to determine the safety coefficient vector of the product to be placed according to the product transaction data of the product to be placed;

[0014] A product placement determination module, configured to determine the bank branches corresponding to the products to be placed according to the safety factor vectors of the products to be placed and the extreme value products of each branch category.

[0015] An embodiment of the present invention further provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the method for placing products at bank branches as described above is implemented.

[0016] An embodiment of the present invention further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for placing products at bank branches as described above is implemented.

[0017] An embodiment of the present invention further provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the method for placing products at bank branches as described above is implemented.

[0018] In the embodiment of the present invention, according to the business data of bank branches, the bank branches are classified to obtain multiple branch categories; according to the product transaction data of each branch category, the safety factor vectors and error convergence upper bounds of each product in the branch category are determined; according to the safety factor vectors and error convergence upper bounds of each product in each branch category, the extreme value products of the branch category are determined; according to the product transaction data of the product to be placed, the safety factor vector of the product to be placed is determined; according to the safety factor vector of the product to be placed and the extreme value products of each branch category, the bank branches corresponding to the product to be placed are determined. In the above process, according to the product transaction data of each branch category, the safety factor vectors and error convergence upper bounds of each product in the branch category are determined, and then the extreme value products of the branch category are determined, and the safety factor vector of the product to be placed is determined. Finally, the bank branches corresponding to the product to be placed are determined, so that the product to be placed is controllable and the product placement risk of bank branches is reduced. Description of the Drawings

[0019] 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. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts. In the drawings:

[0020] Figure 1 It is a flowchart of the method for placing products at bank branches in the embodiment of the present invention;

[0021] Figure 2 It is a flowchart of obtaining multiple branch categories in the embodiment of the present invention;

[0022] Figure 3 This is a flowchart for analyzing product transaction data in an embodiment of the present invention;

[0023] Figure 4 This is a flowchart for determining extreme value products of network categories in an embodiment of the present invention;

[0024] Figure 5 This is a flowchart for determining the safety factor vector of products to be placed in an embodiment of the present invention;

[0025] Figure 6 This is a flowchart for determining the bank branches corresponding to the products to be placed in an embodiment of the present invention;

[0026] Figure 7 This is a schematic diagram of a device for placing products at bank branches in an embodiment of the present invention. Detailed implementation manners

[0027] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer and more understandable, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings. Herein, the illustrative embodiments and descriptions thereof of the present invention are used to explain the present invention, but not to limit the present invention.

[0028] In the description of this specification, the terms "including", "comprising", "having", "containing", etc. are all open-ended terms, that is, they are intended to include but not be limited to. The descriptions referring to terms such as "an embodiment", "a specific embodiment", "some embodiments", "for example", etc. mean that the specific features, structures, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. The step sequences involved in each embodiment are used to schematically illustrate the implementation of the present application, and the step sequences are not limited and can be adjusted appropriately as needed.

[0029] Figure 1 This is a flowchart of the method for placing products at bank branches in an embodiment of the present invention. As Figure 1 shown, it includes:

[0030] Step 101: Classify bank branches according to the business data of bank branches to obtain multiple network categories;

[0031] Step 102: Determine the safety factor vector and the upper bound of error convergence of each product of the network category according to the product transaction data of each network category;

[0032] Step 103: Determine the extreme value products of each network point category based on the safety factor vectors and error convergence upper bounds of the products in each network point category.

[0033] Step 104: Determine the safety factor vector of the product to be launched based on the product transaction data of the product to be launched.

[0034] Step 105: Determine the bank branches corresponding to the product to be launched based on the safety factor vector of the product to be launched and the extreme value products of each network point category.

[0035] In the embodiment of the present invention, based on the product transaction data of each network point category, determine the safety factor vectors and error convergence upper bounds of the products in each network point category, and then determine the extreme value products of the network point category, and determine the safety factor vector of the product to be launched, and finally determine the bank branches corresponding to the product to be launched, so that the product to be launched is controllable and the product launch risk of bank branches is reduced.

[0036] Figure 2 This is a flowchart for obtaining multiple network point categories in the embodiment of the present invention. In one embodiment, classify bank branches based on the business data of bank branches to obtain multiple network point categories, including:

[0037] Step 201: For each bank branch, determine the business volume corresponding to each business in the business data of the bank branch.

[0038] Step 202: Determine the distance function corresponding to each business, where the distance function determines the distance between any two bank branches as the difference in the business volume corresponding to the business in the business data of the two bank branches.

[0039] Step 203: Determine the distance function corresponding to the bank branch based on the distance functions corresponding to each business.

[0040] Step 204: Classify bank branches based on the distance function corresponding to the bank branch to obtain multiple network point categories.

[0041] In one embodiment, classify bank branches based on the distance function corresponding to the bank branch to obtain multiple network point categories, including:

[0042] Select a clustering algorithm to perform clustering analysis on bank branches based on the distance function corresponding to the bank branch to obtain multiple network point subsets.

[0043] For each obtained network point subset, determine the main customer types of the bank branches in the network point subset, and the quantity ratio of the bank branches corresponding to each main customer type in the network point subset.

[0044] Loop and execute the following steps until there is a major customer type in each subset of bank branches such that the proportion of the number of bank branches corresponding to this major customer type is greater than or equal to a set threshold:

[0045] Select a subset of bank branches, where the proportion of the number of bank branches corresponding to each major customer type in this subset of bank branches is less than the set threshold;

[0046] Based on the business distance function of the bank branches, perform clustering analysis on all bank branches in this subset of bank branches, and replace this subset of bank branches with multiple new subsets of bank branches obtained;

[0047] For each new subset of bank branches, determine the major customer type of each bank branch in this new subset of bank branches, and the proportion of the number of bank branches corresponding to each major customer type in this new subset of bank branches;

[0048] When there is a major customer type in each subset of bank branches such that the proportion of the number of bank branches corresponding to this major customer type is greater than or equal to the set threshold, regard each subset of bank branches as a bank branch category.

[0049] Figure 3 This is a flowchart for product transaction data analysis in an embodiment of the present invention. In one embodiment, based on the product transaction data of each bank branch category, determine the safety factor vector and the upper bound of error convergence of each product in this bank branch category, including:

[0050] Step 301, extract the transaction data corresponding to each data dimension of each product in this bank branch category from the product transaction data of this bank branch category;

[0051] Step 302, divide the transaction data corresponding to each data dimension of each product in this bank branch category into multiple subsets of transaction data corresponding to each data dimension of this product in chronological order;

[0052] Step 303, regard the proportion of the number of transactions not involving risks in the transactions included in each subset of transaction data corresponding to each data dimension of each product in this bank branch category as the safety factor sample of this product corresponding to this data dimension;

[0053] Step 304, determine the safety factor vector of each product in this bank branch category, where the components of this safety factor vector correspond one-to-one with the data dimensions, and the value of each component of this safety factor vector is equal to the mean value of the safety factor samples of this product corresponding to the data dimension corresponding to this component;

[0054] Step 305, set the safety factor error threshold;

[0055] Step 306, take the product of the square of the safety factor error threshold and the number of subsets of transaction data for each data dimension corresponding to each product of this network point category as the product of this product corresponding to this data dimension;

[0056] Step 307, based on the safety factor samples of each product of this network point category corresponding to each data dimension, determine the variance of this product corresponding to this data dimension, and take the ratio of the square of this variance to the product of this product corresponding to this data dimension as the upper bound of error convergence of this product corresponding to this data dimension;

[0057] Step 308, for each product of this network point category, determine the upper bound of error convergence of this product as the maximum value of the upper bounds of error convergence of this product corresponding to each data dimension.

[0058] Figure 4 This is a flowchart for determining the extreme value products of the network point category in the embodiment of the present invention. In one embodiment, based on the safety factor vectors and the upper bounds of error convergence of each product of each network point category, determine the extreme value products of this network point category, including:

[0059] Step 401, set the maximum probability that the acceptable safety factor error is greater than the safety factor error threshold;

[0060] Step 402, select multiple products from the products of this network point category whose corresponding upper bounds of error convergence are less than this maximum probability;

[0061] Step 403, determine the partial order of the products according to the safety factor vectors, where this partial order is used to determine whether the first product is greater than the second product among any two selected products;

[0062] Step 404, determine the extreme value products of this network point category according to the partial order of the products, where for each extreme value product, there is no other product among the selected multiple products that satisfies that this other product is greater than this extreme value product.

[0063] In one embodiment, determining the partial order of the products according to the safety factor vectors includes:

[0064] For any two selected products, if for each component, the value of the safety factor vector of the first product among the two products is less than or equal to the value of the safety factor vector of the second product among the two products in this component, then determine that the first product is greater than the second product.

[0065] Figure 5 This is a flowchart for determining the safety factor vector of the product to be put in the embodiment of the present invention. In one embodiment, based on the product transaction data of the product to be put, determine the safety factor vector of the product to be put, including:

[0066] Step 501: Extract the transaction data corresponding to each data dimension of the launched product from the product transaction data of the launched product;

[0067] Step 502: Divide the transaction data corresponding to each data dimension of the launched product into multiple transaction data subsets corresponding to each data dimension of the launched product in chronological order;

[0068] Step 503: Take the proportion of the number of transactions without risks in each transaction data subset corresponding to each data dimension of the launched product as the safety factor sample of the launched product corresponding to this data dimension;

[0069] Step 504: Determine the safety factor vector of the launched product, where the components of the safety factor vector correspond one by one to the data dimensions, and the value of each component of the safety factor vector is equal to the mean value of the safety factor samples of the launched product corresponding to the data dimension corresponding to this component.

[0070] Figure 6 This is the flowchart for determining the bank outlets corresponding to the launched product in the embodiments of the present invention. In one embodiment, based on the safety factor vector of the launched product and the extreme value products of each outlet category, determine the bank outlets corresponding to the launched product, including:

[0071] Step 601: For each extreme value product of each outlet category, calculate the vector difference between the safety factor vector of the extreme value product and the safety factor vector of the launched product. If each component of the vector difference is less than or equal to 0, then determine this extreme value product as the product of the launched product corresponding to this outlet category;

[0072] Step 602: For each outlet category, if the product of the launched product corresponding to this outlet category is not empty, then take this outlet category as the bank outlet corresponding to the launched product.

[0073] In one embodiment, the method includes:

[0074] For each outlet category corresponding to the launched product, determine the priority probability of each product of the launched product corresponding to this outlet category;

[0075] At this bank outlet, at regular intervals, execute the following steps:

[0076] Based on the priority probability, select a product from the products of the launched product corresponding to this outlet category, and perform safety control on the launched product using the safety control method of this product.

[0077] In one embodiment, for each outlet category corresponding to the launched product, determining the priority probability of each product of the launched product corresponding to this outlet category includes:

[0078] For each product of the product corresponding to the network point category of the product to be placed, set the safety distance of the product to the modulus of the safety factor vector of the product;

[0079] According to the safety distance of each product of the product corresponding to the network point category of the product to be placed, determine the priority probability of each product of the product corresponding to the network point category of the product to be placed according to the following formula:

[0080]

[0081] where, t i and s i are respectively the priority probability and the safety distance of the i-th product of the product corresponding to the network point category of the product to be placed, s j is the safety distance of the j-th product of the product corresponding to the network point category of the product to be placed, f is a monotonically increasing function, and f > 0.

[0082] In one embodiment, based on the priority probability, select a product from the products of the product corresponding to the network point category of the product to be placed, and perform safety control on the product to be placed using the safety control method of the product, including:

[0083] Select a function g with a value range greater than 0, and determine according to the domain D of the function g and d = Min(D);

[0084] Sort the multiple products of the product corresponding to the network point category of the product to be placed to form a product list;

[0085] For each product in the product list, use the sum of the priority probabilities of all products before this product in the product list and the priority probability of this product as the probability sum of this product, and then determine the right endpoint value of this product according to the following formula:

[0086]

[0087] where, t k is the probability sum of the k-th product in the product list, u k is the right endpoint value of the k-th product in the product list;

[0088] According to generate a random value r;

[0089] Select the smallest right endpoint value greater than r from the right endpoint values corresponding to all products in the product list;

[0090] Based on the safety control method of the product corresponding to the selected smallest right endpoint value, perform safety control on the product to be placed.

[0091] In summary, in the method proposed in the embodiment of the present invention, according to the business data of bank branches, the bank branches are classified to obtain multiple branch categories; according to the product transaction data of each branch category, the safety factor vector and the upper bound of error convergence of each product in the branch category are determined; according to the safety factor vector and the upper bound of error convergence of each product in each branch category, the extreme value products in the branch category are determined; according to the product transaction data of the launched product, the safety factor vector of the launched product is determined; according to the safety factor vector of the launched product and the extreme value products of each branch category, the bank branch corresponding to the launched product is determined. In the above process, according to the product transaction data of each branch category, the safety factor vector and the upper bound of error convergence of each product in the branch category are determined, and then the extreme value products in the branch category are determined, and the safety factor vector of the launched product is determined. Finally, the bank branch corresponding to the launched product is determined, making the launched product controllable and reducing the product launch risk of bank branches.

[0092] The embodiment of the present invention also proposes a device for launching products at bank branches. The principle is similar to the method for launching products at bank branches and will not be elaborated here.

[0093] Figure 7 The following is a schematic diagram of the device for launching products at bank branches in the embodiment of the present invention, including:

[0094] A branch category obtaining module 701, configured to classify bank branches according to the business data of bank branches to obtain multiple branch categories;

[0095] A product transaction data analysis module 702, configured to determine the safety factor vector and the upper bound of error convergence of each product in the branch category according to the product transaction data of each branch category;

[0096] An extreme value product determination module 703 for the branch category, configured to determine the extreme value products in the branch category according to the safety factor vector and the upper bound of error convergence of each product in each branch category;

[0097] A safety factor vector determination module 704, configured to determine the safety factor vector of the launched product according to the product transaction data of the launched product;

[0098] A launched product determination module 705, configured to determine the bank branch corresponding to the launched product according to the safety factor vector of the launched product and the extreme value products of each branch category.

[0099] In one embodiment, the branch category obtaining module is specifically configured to:

[0100] For each bank branch, determine the business volume corresponding to each business in the business data of the bank branch;

[0101] Determine the distance function corresponding to each service, where the distance function determines the distance between any two bank branches as the difference in the service volume corresponding to the service in the service data of the two bank branches;

[0102] Determine the distance function corresponding to the bank branch according to the distance functions corresponding to each service;

[0103] Classify the bank branches according to the distance function corresponding to the bank branch to obtain multiple branch categories.

[0104] In one embodiment, the product transaction data analysis module is specifically configured to:

[0105] Extract the transaction data corresponding to each data dimension of each product of the branch category from the product transaction data of the branch category;

[0106] Divide the transaction data corresponding to each data dimension of each product of the branch category into multiple transaction data subsets corresponding to each data dimension of the product in chronological order;

[0107] Take the proportion of the number of transactions not involving risks in the transactions included in each transaction data subset corresponding to each data dimension of each product of the branch category as the safety factor sample of the product corresponding to the data dimension;

[0108] Determine the safety factor vector of each product of the branch category, where the components of the safety factor vector correspond one-to-one with the data dimensions, and the value of each component of the safety factor vector is equal to the mean value of the safety factor samples of the product corresponding to the data dimension corresponding to the component;

[0109] Set the safety factor error threshold;

[0110] Take the product of the square of the safety factor error threshold and the number of transaction data subsets corresponding to each data dimension of each product of the branch category as the product of the product corresponding to the data dimension;

[0111] Based on the safety factor samples corresponding to each data dimension of each product of the branch category, determine the variance of the product corresponding to the data dimension, and take the ratio of the square of the variance to the product of the product corresponding to the data dimension as the error convergence upper bound of the product corresponding to the data dimension;

[0112] For each product of the branch category, determine the error convergence upper bound of the product as the maximum value of the error convergence upper bounds corresponding to each data dimension of the product.

[0113] In one embodiment, the extreme value product determination module of the branch category is specifically configured to:

[0114] Set the maximum probability that the acceptable safety factor error is greater than the safety factor error threshold;

[0115] From the products of this network point category, select multiple products whose corresponding upper bound of error convergence is less than this maximum probability;

[0116] Determine the partial order of the products according to the safety factor vector, where this partial order is used to determine whether the first product is greater than the second product among any two selected products;

[0117] Determine the extreme value products of this network point category according to the partial order of the products. For each extreme value product, among the multiple selected products, there is no other product other than this extreme value product that satisfies that this other product is greater than this extreme value product.

[0118] In one embodiment, the extreme value product determination module of the network point category is specifically used for:

[0119] For any two selected products, if for each component, the value of the safety factor vector of the first product among the two products is less than or equal to the value of the safety factor vector of the second product among the two products in this component, then determine that the first product is greater than the second product.

[0120] In one embodiment, the safety factor vector determination module is specifically used for:

[0121] Extract the transaction data of the placed product corresponding to each data dimension from the product transaction data of the placed product;

[0122] Divide the transaction data of the placed product corresponding to each data dimension into multiple transaction data subsets of the placed product corresponding to each data dimension in chronological order;

[0123] Take the proportion of the number of transactions not involving risks in the transactions included in each transaction data subset corresponding to each data dimension of the placed product as the safety factor sample of the placed product corresponding to this data dimension;

[0124] Determine the safety factor vector of the placed product, where the components of this safety factor vector correspond one-to-one with the data dimensions, and the value of each component of this safety factor vector is equal to the mean value of the safety factor samples of the placed product corresponding to the data dimension corresponding to this component.

[0125] In one embodiment, the placed product determination module is specifically used for:

[0126] For each extreme value product of each network point category, calculate the vector difference between the safety factor vector of this extreme value product and the safety factor vector of the placed product. If each component of this vector difference is less than or equal to 0, then determine this extreme value product as the product of the placed product corresponding to this network point category;

[0127] For each network point category, if the products corresponding to the network point category of the placed product are not empty, then use this network point category as the bank network point corresponding to the placed product.

[0128] In summary, in the device proposed in the embodiment of the present invention, according to the business data of bank network points, classify the bank network points to obtain multiple network point categories; according to the product transaction data of each network point category, determine the safety factor vector and the upper bound of error convergence of each product in this network point category; according to the safety factor vector and the upper bound of error convergence of each product in each network point category, determine the extreme value products of this network point category; according to the product transaction data of the placed product, determine the safety factor vector of the placed product; according to the safety factor vector of the placed product and the extreme value products of each network point category, determine the bank network point corresponding to the placed product. In the above process, according to the product transaction data of each network point category, determine the safety factor vector and the upper bound of error convergence of each product in this network point category, and then determine the extreme value products of the network point category, and determine the safety factor vector of the placed product, and finally determine the bank network point corresponding to the placed product, making the placed product controllable and reducing the product placement risk of bank network points.

[0129] The embodiment of the invention also provides a computer device, which includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the method for placing products at bank network points as described above.

[0130] The embodiment of the invention also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the method for placing products at bank network points as described above.

[0131] The embodiment of the invention also provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the method for placing products at bank network points as described above.

[0132] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program business system. Therefore, the present invention can be implemented in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can be implemented in 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 code.

[0133] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program business systems according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows 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 the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or in multiple blocks.

[0134] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that implement the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or in multiple blocks.

[0135] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or in multiple blocks.

[0136] The specific embodiments described above further elaborate on the objectives, technical solutions, and beneficial effects of the present invention. It should be understood that the above are only specific embodiments of the present invention and are not used to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for placing products in bank branches, characterized in that, it includes: Classifying bank branches according to the business data of bank branches to obtain multiple branch categories; Determining the safety coefficient vector and the upper bound of error convergence of each product in the branch category according to the product transaction data of each branch category; Determining the extreme value products of the branch category according to the safety coefficient vector and the upper bound of error convergence of each product in the branch category; Determining the safety coefficient vector of the product to be placed according to the product transaction data of the product to be placed; Determining the corresponding bank branch of the product to be placed according to the safety coefficient vector of the product to be placed and the extreme value products of each branch category; Determining the safety coefficient vector and the upper bound of error convergence of each product in the branch category according to the product transaction data of each branch category, including: extracting the transaction data corresponding to each data dimension of each product in the branch category from the product transaction data of the branch category; dividing the transaction data corresponding to each data dimension of each product in the branch category into multiple transaction data subsets corresponding to each data dimension of the product in chronological order; taking the proportion of the number of transactions not involving risks in the transactions included in each transaction data subset corresponding to each data dimension of each product in the branch category as the safety coefficient sample of the product corresponding to the data dimension; determining the safety coefficient vector of each product in the branch category, where the components of the safety coefficient vector correspond one by one to the data dimensions, and the value of each component of the safety coefficient vector is equal to the mean value of the safety coefficient samples of the product corresponding to the data dimension corresponding to the component; setting a safety coefficient error threshold; taking the product of the square of the safety coefficient error threshold and the number of transaction data subsets corresponding to each data dimension of each product in the branch category as the product corresponding to the data dimension of the product; determining the variance of the product corresponding to the data dimension based on the safety coefficient samples corresponding to each data dimension of each product in the branch category, and taking the ratio of the square of the variance to the product corresponding to the data dimension of the product as the upper bound of error convergence of the product corresponding to the data dimension; for each product in the branch category, determining the upper bound of error convergence of the product as the maximum value of the upper bounds of error convergence corresponding to each data dimension of the product; Determining the extreme value products of the branch category according to the safety coefficient vector and the upper bound of error convergence of each product in the branch category, including: setting the maximum probability that the acceptable safety coefficient error is greater than the safety coefficient error threshold; selecting multiple products from the products in the branch category whose corresponding upper bounds of error convergence are less than the maximum probability; determining the partial order of the products according to the safety coefficient vector, where the partial order is used to determine whether the first product is greater than the second product among any two selected products; determining the extreme value products of the branch category according to the partial order of the products, where for each extreme value product, there is no other product among the selected multiple products that satisfies that the other product is greater than the extreme value product; Determine the bank outlets corresponding to the products to be launched based on the safety coefficient vector of the products to be launched and the extreme value products of each outlet category, including: for each extreme value product of each outlet category, calculate the vector difference between the safety coefficient vector of the extreme value product and the safety coefficient vector of the product to be launched. If each component of the vector difference is less than or equal to 0, then determine the extreme value product as the product of the product to be launched corresponding to this outlet category; for each outlet category, if the product of the product to be launched corresponding to this outlet category is not empty, then use this outlet category as the bank outlet corresponding to the product to be launched.

2. The method according to claim 1, characterized in that, classify bank outlets based on the business data of bank outlets to obtain multiple outlet categories, including: for each bank outlet, determine the business volume corresponding to each business in the business data of this bank outlet; determine the distance function corresponding to each business, where the distance function determines the distance between any two bank outlets as the difference in the business volume corresponding to this business in the business data of the two bank outlets; determine the distance function corresponding to the bank outlet based on the distance functions corresponding to each business; classify bank outlets based on the distance function corresponding to the bank outlet to obtain multiple outlet categories.

3. The method according to claim 1, characterized in that, determine the partial order of products based on the safety coefficient vector, including: for any two selected products, if for each component, the value of the safety coefficient vector of the first product among the two products in this component is less than or equal to the value of the safety coefficient vector of the second product among the two products in this component, then determine that the first product is greater than the second product.

4. The method according to claim 1, characterized in that, determine the safety coefficient vector of the product to be launched based on the product transaction data of the product to be launched, including: extract the transaction data corresponding to each data dimension of the product to be launched from the product transaction data of the product to be launched; divide the transaction data corresponding to each data dimension of the product to be launched into multiple transaction data subsets corresponding to each data dimension of the product to be launched in chronological order; use the proportion of the number of transactions not involving risks in the transactions included in each transaction data subset corresponding to each data dimension of the product to be launched as the safety coefficient sample of the product to be launched corresponding to this data dimension; determine the safety coefficient vector of the product to be launched, where the components of the safety coefficient vector correspond one-to-one with the data dimensions, and the value of each component of the safety coefficient vector is equal to the mean value of the safety coefficient samples of the product to be launched corresponding to the data dimension corresponding to this component.

5. An apparatus for launching products at bank outlets, characterized in that, comprising: an outlet category obtaining module, configured to classify bank outlets based on the business data of bank outlets to obtain multiple outlet categories; a product transaction data analysis module, configured to determine the safety coefficient vector and the upper bound of error convergence of each product of this outlet category based on the product transaction data of each outlet category; an extreme value product determination module for the outlet category, configured to determine the extreme value products of this outlet category based on the safety coefficient vectors and the upper bound of error convergence of each product of this outlet category; A safety factor vector determination module, configured to determine a safety factor vector of the product to be placed according to the product transaction data of the product to be placed; A product to be placed determination module, configured to determine the corresponding bank branch of the product to be placed according to the safety factor vector of the product to be placed and the extreme value products of each branch category; The product transaction data analysis module is specifically configured to: extract the transaction data of each product of the branch category corresponding to each data dimension from the product transaction data of the branch category; divide the transaction data of each product of the branch category corresponding to each data dimension into multiple transaction data subsets corresponding to each data dimension of the product in chronological order; use the proportion of the number of transactions not involving risks in the transactions included in each transaction data subset of each product of the branch category corresponding to each data dimension as the safety factor sample of the product corresponding to the data dimension; Determine the safety factor vector of each product of the branch category, where the components of the safety factor vector correspond one by one to the data dimensions, and the value of each component of the safety factor vector is equal to the mean value of the safety factor samples of the product corresponding to the data dimension corresponding to the component; set a safety factor error threshold; use the product of the square of the safety factor error threshold and the number of transaction data subsets of each product of the branch category corresponding to each data dimension as the product of the product corresponding to the data dimension; based on the safety factor samples of each product of the branch category corresponding to each data dimension, determine the variance of the product corresponding to the data dimension, and use the ratio of the square of the variance to the product of the product corresponding to the data dimension as the error convergence upper bound of the product corresponding to the data dimension; for each product of the branch category, determine the error convergence upper bound of the product as the maximum value of the error convergence upper bounds of the product corresponding to each data dimension; The extreme value product determination module of the branch category is specifically configured to: set the maximum probability that the acceptable safety factor error is greater than the safety factor error threshold; select multiple products from the products of the branch category whose corresponding error convergence upper bounds are less than the maximum probability; determine the partial order of the products according to the safety factor vector, where the partial order is used to determine whether the first product is greater than the second product among any two selected products; determine the extreme value products of the branch category according to the partial order of the products, where for each extreme value product, there is no other product other than the extreme value product among the selected products that satisfies that the other product is greater than the extreme value product; the product to be placed determination module is specifically configured to: for each extreme value product of each branch category, calculate the vector difference between the safety factor vector of the extreme value product and the safety factor vector of the product to be placed, and if each component of the vector difference is less than or equal to 0, then determine the extreme value product as the product of the product to be placed corresponding to the branch category; for each branch category, if the product of the product to be placed corresponding to the branch category is not empty, then use the branch category as the bank branch corresponding to the product to be placed.

6. The apparatus according to claim 5, wherein, The branch category acquisition module is specifically configured to: For each bank branch, determine the business volume corresponding to each business in the business data of the bank branch; Determine the distance function corresponding to each service, where the distance function determines the distance between any two bank branches as the difference in the service volume corresponding to the service in the service data of the two bank branches; Determine the distance function corresponding to the bank branch according to the distance functions corresponding to each service; Classify the bank branches according to the distance function corresponding to the bank branch to obtain multiple branch categories.

7. The device according to claim 5, wherein, The extreme value product determination module of the branch category is specifically used for: For any two selected products, if for each component, the value of the safety factor vector of the first product among the two products in this component is less than or equal to the value of the safety factor vector of the second product among the two products in this component, then determine that the first product is greater than the second product.

8. The device according to claim 5, wherein, The safety factor vector determination module is specifically used for: Extract the transaction data corresponding to each data dimension of the launched product from the product transaction data of the launched product; Divide the transaction data corresponding to each data dimension of the launched product into multiple transaction data subsets corresponding to each data dimension of the launched product in chronological order; Take the proportion of the number of transactions not involving risks in the transactions included in each transaction data subset corresponding to each data dimension of the launched product as the safety factor sample corresponding to the launched product for this data dimension; Determine the safety factor vector of the launched product, where the components of the safety factor vector correspond one-to-one with the data dimensions, and the value of each component of the safety factor vector is equal to the mean value of the safety factor samples corresponding to the data dimension corresponding to this component of the launched product.

9. A computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein, When the processor executes the computer program, it implements the method according to any one of claims 1 to 7.

10. 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, it implements the method according to any one of claims 1 to 7.

11. A computer program product, wherein, The computer program product includes a computer program, and when the computer program is executed by a processor, it implements the method according to any one of claims 1 to 7.

Images