Banking outlet recommendation method and device, equipment and medium

Abstract

The embodiment of the invention provides a bank outlet recommendation method and device, equipment and a medium, and is applied to the technical field of intelligent resource allocation and optimization. The method comprises the steps of obtaining a user set and a bank outlet set of a current time-space period; the current space-time period is a space-time unit formed by combining the time interval of the business time of a plurality of bank outlets in the bank outlet set and the positions of the bank outlets; according to the user set and the bank outlet set, calculating to obtain a first preference list corresponding to each user and a second preference list corresponding to each bank outlet; based on the first preference list, the second preference list, the current space-time period and the next space-time period of the current space-time period, pairing the user and the bank outlet to obtain a stable pairing set; and generating bank outlet recommendation information based on the stable pairing set. The technical effect of improving the recommendation accuracy of the bank outlets is achieved.

Description

Technical Field

[0001] The present application relates to the technical field of intelligent resource allocation and optimization, and in particular to a bank branch recommendation method, device, equipment and medium. Background Art

[0002] As the primary channel for users to access financial products and services, and the primary touchpoint for commercial banks to provide services to users, the compatibility between bank branches and users often affects both user satisfaction and branch operational efficiency. User demands for banking services are diverse, including but not limited to product diversity, service quality, convenience, and speed. Furthermore, the types of services offered by bank branches and the number of users they can accommodate directly impact the accessibility and convenience of services, which in turn influences user satisfaction and preferences. Therefore, by combining information such as user needs and preferences with the operational status of bank branches to find the optimal match between branches and users, we can optimize the user experience and improve user satisfaction, while also enhancing service efficiency and competitiveness.

[0003] In the prior art, the method for seeking the best match between network points and users is as follows: receiving recommendation instructions sent by each user at the current moment, and obtaining the user location of each user in response to the recommendation instructions; determining the shortest distance from each user location to each network point based on each user location and a pre-established network recommendation model; determining the driving time and driving cost from each user location to each network point based on pre-acquired traffic data, each user's travel preference information, and the shortest distance from each user location to each network point; obtaining network point information of each network point at the current moment, and predicting the queuing time of each network point based on the network point information; determining the network point recommendation information corresponding to each user based on the driving time, driving cost, queuing time, and a pre-determined objective function, and sending the network point recommendation information to each user.

[0004] Since the existing technology only focuses on the arrival cost between users and bank branches, there is a technical problem in the existing technology of low accuracy of bank branch recommendations. Summary of the Invention

[0005] The embodiments of the present application provide a bank branch recommendation method, apparatus, equipment, and medium to achieve the technical effect of improving the accuracy of bank branch recommendations.

[0006] In a first aspect, an embodiment of the present application provides a bank branch recommendation method, comprising:

[0007] Obtain the user set and bank branch set of the current time-space period; the current time-space period is the time interval of the business hours of multiple bank branches in the bank branch set and the time-space unit formed by the combination of the locations of the bank branches;

[0008] Based on the user set and the bank branch set, a first preference list corresponding to each user and a second preference list corresponding to each bank branch are calculated;

[0009] Pairing users and bank branches based on the first preference list, the second preference list, the current time-space period, and the time-space period next to the current time-space period to obtain a stable pairing set;

[0010] Generate bank branch recommendation information based on the stable pairing set.

[0011] In a possible implementation, based on the user set and the bank branch set, a first preference list corresponding to each user and a second preference list corresponding to each bank branch are calculated, including:

[0012] Calculate the target user's expected value for each bank branch in the bank branch set to obtain the target user's first preference list; the target user is any user in the user set;

[0013] The congestion degree of the target bank branch is calculated based on the initial preference list corresponding to the target bank branch, the number of users whose initial locations are within the target bank branch, and the maximum customer flow corresponding to the target bank branch; the target bank branch is any bank branch in the bank branch set;

[0014] Based on the target user's initial satisfaction list and the target bank branch's congestion level, the target bank branch's expected value for the target user is calculated;

[0015] The expected value of each target user is sorted in descending order based on the target bank branches to obtain a second preference list of the target bank branches.

[0016] In one possible implementation, users and bank branches are paired based on the first preference list, the second preference list, the current spatiotemporal period, and the spatiotemporal period next to the current spatiotemporal period to obtain a stable pairing set, including:

[0017] Performing feasible pairing on the first preference list and the second preference list to obtain a first feasible pairing set;

[0018] Based on the user set, the bank branch set and the first feasible pairing set, a set of under-capacity bank branches is obtained;

[0019] For each under-capacity bank branch in the set of under-capacity bank branches, calculate a first probability that each user in the user set forms a first stable pairing with each under-capacity bank branch in the current spatiotemporal period, and calculate a second probability that each user in the user set forms a second stable pairing with each under-capacity bank branch in the next spatiotemporal period;

[0020] Calculate the first user revenue and the first network revenue of the first stable pairing in the current space-time period based on the first probability, and calculate the second user revenue and the second network revenue of the second stable pairing in the next space-time period based on the second probability;

[0021] A first quotient is obtained based on a ratio of the first user's income to the second user's income, and a second quotient is obtained based on a ratio of the first network point's income to the second network point's income;

[0022] When both the first quotient value and the second quotient value are greater than a preset threshold, adding the first stable pairing to the stable pairing set;

[0023] When the first quotient value or the second quotient value is less than the preset threshold value, the stable pairing calculation of the next spatiotemporal period is entered.

[0024] In one possible implementation, performing feasible pairing on the first preference list and the second preference list to obtain a first feasible pairing set includes:

[0025] Match users and bank branches in the current time and space period to obtain multiple feasible pairs;

[0026] Calculate the priority corresponding to each feasible pairing;

[0027] The feasible pairings are sorted in descending order of priority to obtain a first feasible pairing set.

[0028] In one possible implementation, obtaining a set of under-capacity bank branches based on the user set, the bank branch set, and the first feasible pairing set includes:

[0029] Based on the user set and the bank branch set, a stable pairing set is determined from the first feasible pairing set;

[0030] Based on the stable pairing set, the bank branches that are not full and still in business hours are screened out from the bank branch set to obtain the set of not full bank branches.

[0031] In one possible implementation, based on the user set and the bank branch set, determining a stable pairing set from the first feasible pairing set includes:

[0032] Traversing the first feasible pairing set, and adding the feasible pairing with the highest priority in the first feasible pairing set as a stable pairing to the stable pairing set;

[0033] Deleting other feasible pairings associated with the users in the stable pairing from the first set of feasible pairings;

[0034] When the number of stable pairings of bank branches reaches the expected number of bank branches, deleting other feasible pairings related to the bank branches from the first feasible pairing set;

[0035] After the traversal is completed, a stable pairing set and an updated first feasible pairing set are obtained.

[0036] In one possible implementation, based on the stable pairing set, bank branches that are not fully staffed and are still in business hours are screened from the bank branch set to obtain a set of not fully staffed bank branches, including:

[0037] Based on the stable pairing set, determine the number of stable pairings corresponding to each bank branch in the bank branch set;

[0038] When the number of stable pairings reaches the maximum user limit of the corresponding bank branch, or when the bank branch in the stable pairing is not in business hours, the bank branch is deleted from the bank branch set to obtain a set of under-capacity bank branches.

[0039] In a second aspect, an embodiment of the present application provides a bank branch recommendation device, comprising:

[0040] An acquisition module is used to acquire a user set and a bank branch set in a current time-space period; the current time-space period is a time interval of the business hours of multiple bank branches in the bank branch set and a time-space unit formed by the combination of the locations of the bank branches;

[0041] A first processing module is configured to calculate, based on the user set and the bank branch set, a first preference list corresponding to each user and a second preference list corresponding to each bank branch;

[0042] a second processing module, configured to pair users and bank branches based on the first preference list, the second preference list, the current spatiotemporal period, and the spatiotemporal period next to the current spatiotemporal period to obtain a stable pairing set;

[0043] The third processing module is used to generate bank branch recommendation information based on the stable pairing set.

[0044] In a possible implementation, the first processing module is further configured to:

[0045] Calculate the target user's expected value for each bank branch in the bank branch set to obtain the target user's first preference list; the target user is any user in the user set;

[0046] The congestion degree of the target bank branch is calculated based on the initial preference list corresponding to the target bank branch, the number of users whose initial locations are within the target bank branch, and the maximum customer flow corresponding to the target bank branch; the target bank branch is any bank branch in the bank branch set;

[0047] Based on the target user's initial satisfaction list and the target bank branch's congestion level, the target bank branch's expected value for the target user is calculated;

[0048] The expected value of each target user is sorted in descending order based on the target bank branches to obtain a second preference list of the target bank branches.

[0049] In a possible implementation, the second processing module is further configured to:

[0050] Performing feasible pairing on the first preference list and the second preference list to obtain a first feasible pairing set;

[0051] Based on the user set, the bank branch set and the first feasible pairing set, a set of under-capacity bank branches is obtained;

[0052] For each under-capacity bank branch in the set of under-capacity bank branches, calculate a first probability that each user in the user set forms a first stable pairing with each under-capacity bank branch in the current spatiotemporal period, and calculate a second probability that each user in the user set forms a second stable pairing with each under-capacity bank branch in the next spatiotemporal period;

[0053] Calculate the first user revenue and the first network revenue of the first stable pairing in the current space-time period based on the first probability, and calculate the second user revenue and the second network revenue of the second stable pairing in the next space-time period based on the second probability;

[0054] A first quotient is obtained based on a ratio of the first user's income to the second user's income, and a second quotient is obtained based on a ratio of the first network point's income to the second network point's income;

[0055] When both the first quotient value and the second quotient value are greater than a preset threshold, adding the first stable pairing to the stable pairing set;

[0056] When the first quotient value or the second quotient value is less than the preset threshold value, the stable pairing calculation of the next spatiotemporal period is entered.

[0057] In a possible implementation, the second processing module is further configured to:

[0058] Match users and bank branches in the current time and space period to obtain multiple feasible pairs;

[0059] Calculate the priority corresponding to each feasible pairing;

[0060] The feasible pairings are sorted in descending order of priority to obtain a first feasible pairing set.

[0061] In a possible implementation, the second processing module is further configured to:

[0062] Based on the user set and the bank branch set, a stable pairing set is determined from the first feasible pairing set;

[0063] Based on the stable pairing set, the bank branches that are not full and still in business hours are screened out from the bank branch set to obtain the set of not full bank branches.

[0064] In a possible implementation, the second processing module is further configured to:

[0065] Traversing the first feasible pairing set, and adding the feasible pairing with the highest priority in the first feasible pairing set as a stable pairing to the stable pairing set;

[0066] Deleting other feasible pairings associated with the users in the stable pairing from the first set of feasible pairings;

[0067] When the number of stable pairings of bank branches reaches the expected number of bank branches, deleting other feasible pairings related to the bank branches from the first feasible pairing set;

[0068] After the traversal is completed, a stable pairing set and an updated first feasible pairing set are obtained.

[0069] In a possible implementation, the second processing module is further configured to:

[0070] Based on the stable pairing set, determine the number of stable pairings corresponding to each bank branch in the bank branch set;

[0071] When the number of stable pairings reaches the maximum user limit of the corresponding bank branch, or when the bank branch in the stable pairing is not in business hours, the bank branch is deleted from the bank branch set to obtain a set of under-capacity bank branches.

[0072] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a memory, a processor;

[0073] Memory stores computer-executable instructions;

[0074] The processor executes the computer-executable instructions stored in the memory, so that the processor executes the above first aspect and various possible implementations of the first aspect.

[0075] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the above-mentioned first aspect and various possible implementation methods of the first aspect.

[0076] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the above first aspect and various possible implementation methods of the first aspect.

[0077] The bank branch recommendation method, device, equipment and medium provided by the embodiments of the present application. The method obtains the user set and bank branch set within the current spatiotemporal period, and calculates the first preference list corresponding to each user and the second preference list corresponding to each bank branch based on the user set and the bank branch set; uses the first preference list and the second preference list to match the user and the bank branch in the current spatiotemporal period and the next spatiotemporal period, obtains a stable pairing set between the user and the bank branch, and uses the stable pairing set to generate bank branch recommendation information. Compared with the existing technology, the present application uses the current spatiotemporal period to restrict the space and time between the bank branch and the user, and uses the user's first preference list and the bank branch's second preference list to represent the degree of preference between the user and the bank branch; thereby, based on the first preference list and the second preference list, matching within the same spatiotemporal period and matching within the next different spatiotemporal period is achieved, and the stable pairing set obtained by matching is used to generate bank branch recommendation information, thereby achieving the technical effect of improving the accuracy of bank branch recommendations. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0079] Figure 1 Schematic diagram of the application scenario of the bank branch recommendation method provided in this application;

[0080] Figure 2 Schematic diagram of the bank branch recommendation method provided for this application Figure 1 ;

[0081] Figure 3 Schematic diagram of the bank branch recommendation method provided for this application Figure 2 ;

[0082] Figure 4 Schematic diagram of the bank branch recommendation method provided for this application Figure 3 ;

[0083] Figure 5 A schematic diagram of a feasible pairing priority provided in an embodiment of the present application;

[0084] Figure 6 A schematic diagram of the structure of the bank branch recommendation device provided for this application;

[0085] Figure 7This is a schematic diagram of the structure of the electronic device provided in this application.

[0086] The above drawings illustrate specific embodiments of the present application, which will be described in more detail below. These drawings and the textual description are not intended to limit the scope of the present application in any way, but rather to illustrate the concepts of the present application to those skilled in the art by reference to specific embodiments. DETAILED DESCRIPTION

[0087] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.

[0088] In the prior art, the method for seeking the best match between network points and users is as follows: receiving recommendation instructions sent by each user at the current moment, and obtaining the user location of each user in response to the recommendation instructions; determining the shortest distance from each user location to each network point based on each user location and a pre-established network recommendation model; determining the driving time and driving cost from each user location to each network point based on pre-acquired traffic data, each user's travel preference information, and the shortest distance from each user location to each network point; obtaining network point information of each network point at the current moment, and predicting the queuing time of each network point based on the network point information; determining the network point recommendation information corresponding to each user based on the driving time, driving cost, queuing time, and a pre-determined objective function, and sending the network point recommendation information to each user.

[0089] However, in the existing technology, the driving time, driving cost, waiting time and a predetermined objective function are mainly used to determine the bank branch recommendation information corresponding to each user, without considering the matching degree between the bank branch business and the user's business. As a result, the branch finally recommended to the user may not meet the user's business needs. Therefore, there is a technical problem of low accuracy of bank branch recommendation in the existing technology.

[0090] In response to the above technical problems, the present application proposes the following technical concept: using the preference information between users and bank branches to achieve accurate bank branch recommendations. Specifically: obtaining the user set and bank branch set of the current space-time period, using the user set and bank branch set to calculate the first preference list of each user and the second preference list of each bank branch; using the two preference lists to match the users and bank branches of the current space-time period and the next space-time period, to obtain the corresponding stable pairing set, thereby generating bank branch recommendation information. Compared with the existing technology, the present application uses the space-time period to consider the temporal and spatial matching degree between bank branches and users, and uses the preference list to represent the degree of preference between users and bank branches, thereby achieving the generation of stable pairings between users and bank branches based on the preference list and the space-time period, and using the stable pairing set to achieve bank branch recommendations, thereby achieving the technical effect of improving the accuracy of bank branch recommendations.

[0091] The following specific embodiments describe in detail the technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0092] Figure 1 The application scenario diagram of the bank branch recommendation method provided in this application is as follows: Figure 1 As shown, this scenario includes: a user set and a bank branch set. In the bank branch set, there is one user currently queuing at branch 1, and four users currently queuing at branch 2. A user is on their way to a bank branch for business. A bank branch push notification is required based on two bank branches in the bank branch set to help the user determine which branch is most suitable.

[0093] Figure 2 Schematic diagram of the bank branch recommendation method provided for this application Figure 1 ,like Figure 2 As shown, the method includes:

[0094] S201. Obtain the user set and bank branch set of the current time and space period.

[0095] In this step, the current spatiotemporal period is a time interval of the business hours of multiple bank outlets in the bank outlet set and a spatiotemporal unit formed by the combination of the locations of the bank outlets.

[0096] Optionally, a possible implementation method for obtaining the user set and bank branch set of the current time-space period is:

[0097] S2011. Divide the business hours of a bank branch into multiple time intervals according to fixed time intervals; obtain the service scope of each bank branch based on the service radius of the bank branch and the location of the bank branch.

[0098] S2012: Obtain the user's real-time location and the time when the user initiates the network query request.

[0099] S2013. Match the user's real-time location and time with the bank branches operating in the current time interval, group the bank branches with users in their service areas into a forgotten bank branch set, and group the users who initiate forgotten branch query requests in the current time interval into a user set.

[0100] S202: Calculate a first preference list corresponding to each user and a second preference list corresponding to each bank branch based on the user set and the bank branch set.

[0101] In this step, the first preference list is calculated as follows: for each user, the distance between the user's current location and the location of each bank branch is calculated, the correlation between the services required by the user and the services supported by each bank branch is calculated, and the user's expected value for each bank branch is obtained through weighted calculation based on the distance and the correlation. Each bank branch in the bank branch set is sorted in descending order based on the expected value to obtain the first preference list of each user.

[0102] The second preference list is calculated as follows: for each bank branch, the expected value of each user for the bank branch and the congestion degree of the bank branch are combined to calculate the expected value of the bank branch for each user, and each user in the user set is sorted in descending order based on the expected value to obtain the second preference list of each bank branch.

[0103] It should be noted that the calculation of the first preference list and the second preference list in this step will be further explained in the following embodiments and will not be described in detail here.

[0104] S203. Pair users and bank branches based on the first preference list, the second preference list, the current spatiotemporal period, and the next spatiotemporal period of the current spatiotemporal period to obtain a stable pairing set.

[0105] In this step, the first preference list includes each bank branch in the bank branch set and the ranking corresponding to each bank branch; the second preference list includes each user in the user set and the ranking corresponding to each user pair.

[0106] Among them, the pairing for the current space-time period refers to instant pairing, which pairs users and bank branches in the same space-time period to obtain multiple feasible pairings. According to the ranking of each user in the second preference list and the ranking of each bank branch in the first preference list, the priority of each feasible pairing consisting of a bank branch and a user is calculated; according to the level of priority, the feasible pairing with the highest priority is selected as the stable pairing; when it is determined that there are bank branches that are still in business hours and are not full, it is necessary to calculate the probability of the corresponding feasible pairings forming stable pairings in the current space-time period and the next space-time period for the remaining bank branches, and calculate the corresponding benefits of the pairing in different space-time periods according to the probabilities of different space-time periods, and judge whether the pairing needs to form a stable pairing in the current space-time period by comparing the benefits; thus, by pairing in the same space-time period and pairing in different space-time periods, a stable pairing set corresponding to the user set and the bank branch set is obtained.

[0107] It should be noted that the process of obtaining stable pairing in this step is further explained in the following embodiments and will not be described in detail here.

[0108] S204: Generate bank branch recommendation information based on the stable pairing set.

[0109] Optionally, a possible implementation method of generating bank branch recommendation information is:

[0110] S2041. Extract the matching bank branch corresponding to each user and the key attributes of the stable pairing from the stable pairing set.

[0111] In this step, key attributes may be: user ID, bank branch ID, priority of stable pairing, distance between the user's current location and the bank branch, matching degree between the user's business needs and the services supported by the bank branch, and estimated waiting time for the user to handle the service.

[0112] S2042. Sort the stable pairings corresponding to each user from high to low priority, and select the bank branches corresponding to the top N stable pairs as the recommended branches for the user.

[0113] Wherein, N is a positive integer greater than 0.

[0114] In this step, if there are multiple bank branches with the same priority, the bank branch with a shorter distance to the user can be selected as the recommended branch; or the bank branch with a higher matching degree between services can be selected as the recommended branch; or the bank branch with a shorter expected waiting time can be selected as the recommended branch.

[0115] S2043: Generate a recommendation reason for each recommended outlet based on the extracted attribute information of each stable pairing, and push the recommendation reason and the recommended outlet to the user.

[0116] The bank branch recommendation method provided by the embodiment of the present application obtains a user set and a bank branch set within the current spatiotemporal period, calculates a first preference list corresponding to each user and a second preference list corresponding to each bank branch based on the user set and the bank branch set; uses the first preference list and the second preference list to match users and bank branches in the current spatiotemporal period and the next spatiotemporal period, obtains a stable pairing set between users and bank branches, and uses the stable pairing set to generate bank branch recommendation information. Compared with the prior art, the present application uses the current spatiotemporal period to restrict the space and time between bank branches and users, uses the user's first preference list and the bank branch's second preference list to represent the degree of preference between users and bank branches; thereby, matching within the current same spatiotemporal period and matching within the next different spatiotemporal period is achieved based on the first preference list and the second preference list, and uses the stable pairing set obtained by matching to generate bank branch recommendation information, thereby achieving the technical effect of improving the accuracy of bank branch recommendations.

[0117] Figure 3 Schematic diagram of the bank branch recommendation method provided for this application Figure 2 , this embodiment is for Figure 2 The calculation of the first preference list and the second preference list in step S202 of the embodiment shown is further explained as follows: Figure 3 As shown, the method includes:

[0118] S301. Calculate the target user's expected value for each bank branch in the bank branch set to obtain the target user's first preference list.

[0119] In this step, the target user is any user in the user set. For the target user, their preference mainly depends on the distance and the adaptability of the bank branch services. The expected value calculation formula is shown in Formula 1:

[0120] A i,j =α·q(p j )·χ(t i ,p j )Formula 1

[0121] Among them, A i,j Refers to user p j Relative to bank branches i The expected value of α represents the expected coefficient of converting the distance and the suitability of bank branch services into a measurable form; χ(t i ,pj ) indicates user p j Businesses and outlets handled i The matching degree between the supported services is shown in Formula 2 and is expressed as:

[0122]

[0123] Where D represents a set of business types that can be handled by bank branches. j Always prefer bank branches that can provide higher expected values, so by calculating the expected value A of each branch i,j Sorting to finally generate user p j Preference list PL(p i ).

[0124] S302: Calculate the congestion degree of the target bank branch based on the initial preference list corresponding to the target bank branch, the number of users whose initial locations are within the target bank branch, and the maximum customer flow corresponding to the target bank branch.

[0125] In this step, the target bank branch is any bank branch in the bank branch set. j For example, when user p i When conducting business at a branch, the target bank branch t j Will also get a certain degree of satisfaction F i,j , and each user's satisfaction with each bank branch is different. When users choose the same bank branch too often, the branch may become congested. In this case, if users force their way to the branch to handle business, on the one hand, the branch's passenger capacity will be exceeded, and on the other hand, users will not be able to handle business in a timely manner, resulting in a decrease in user satisfaction. Therefore, a method is used here to measure branch congestion, and the bank branch congestion degree is expressed using Formula 3, specifically:

[0126]

[0127] Among them, TL(t j ) represents the target bank branch t j The initial preference list, count(t r ) indicates that the initial position is at the target point t j The number of users within the range, K represents the target bank branch t j The maximum passenger flow, represents the target bank branch t j congestion.

[0128] S303: Calculate the target bank branch's expected value for the target user based on the target user's initial satisfaction list and the target bank branch's congestion level.

[0129] In this step, the target bank branch calculates the expected value for each target user as shown in Formula 4:

[0130]

[0131] Among them, the expected value R i,j and satisfaction F i,j Proportional to the congestion of the target bank branch Inversely proportional. max((F i,j ) i∈ξ ) indicates user p i In its corresponding first preference list PL(p i ) to select a bank branch to obtain the maximum satisfaction, min((F i,j ) i∈ξ ) indicates user p i In its corresponding first preference list PL(p i ) is the minimum satisfaction that can be obtained by choosing a bank branch.

[0132] S304 : Sort the expected value of each target user in descending order based on the target bank outlets to obtain a second preference list of the target bank outlets.

[0133] In this step, the second preference list contains each user in the user set, along with the user's desired value and ranking relative to the target bank branch. The higher the desired value, the lower the user's ranking index in the second preference list; the lower the desired value, the higher the user's ranking index in the second preference list.

[0134] In this embodiment, the distance relationship between users and bank branches, business adaptability, customer flow of bank branches, and user satisfaction with bank branches are used to calculate the mutual expectation value between each user and each bank branch, thereby obtaining a first preference list for each user to sort the set of bank branches, and a second preference list for each bank branch to sort the set of users.

[0135] Figure 4 Schematic diagram of the bank branch recommendation method provided for this application Figure 3 , this embodiment is for Figure 2 The acquisition of the stable pairing set in step S203 in the embodiment shown is further explained as follows: Figure 4 As shown, the method includes:

[0136] S401: Perform feasible pairing on the first preference list and the second preference list to obtain a first feasible pairing set.

[0137] Optionally, a possible implementation method of obtaining the first feasible pairing set is:

[0138] S4011. Match users and bank branches in the current time-space period to obtain multiple feasible pairs.

[0139] In this step, feasible pairing refers to the pairing of users and bank branches in the same time and space cycle. At the same time, there is an overlap between the services required by the users in each pairing and the services supported by the corresponding bank branches, and the time when the users arrive at the bank branches must also be within the business hours of the bank branches.

[0140] S4012. Calculate the priority corresponding to each feasible pairing.

[0141] In this step, the priority of the feasible pairing is calculated as shown in Formula 5:

[0142]

[0143] Among them, θ(t i ,p j ) represents a feasible pairing (t i ,p j ) priority, the smaller its value is, the greater the priority of the feasible pairing is, m is the number of users, ξ is the number of bank branches. i ) indicates bank branch t i In user p j The first preference list PL(p j ) in the index value, I(p j ) indicates user p j At bank branches i The second preference list TL(t i ). The priority of a feasible pairing can be specified by specifying the satisfaction level of the feasible pairing consisting of users and bank branches in a pairwise manner. The higher the satisfaction of both users and bank branches with each other, the higher the stability of the matching pair.

[0144] Figure 5 A feasible pairing priority diagram is provided for an embodiment of the present application, for example, Figure 5 As shown, there are users p1, p2, p3, and bank branches t1 and t2. A feasible pairing M(t1, p1) is composed of bank branch t1 and user p1, where user p1 has a second preference list TL(t i) in the index value is I(p1) = 1, and bank branch t1 is in the first preference list of user p1.

[0145] The index value in PL(p1) is I(t1) = 2. The feasible pairing M(t1, p2) is the one between the bank branch t1 and the user

[0146] p2, where the second preference list TL(t i ) is the index value

[0147] I(p2)=2, the index value of bank branch t1 in the first preference list PL(p2) of user p2 is

[0148] I(t1)=1. It can be seen that bank branches and users occupy the same positions ((1,2), (2,1)) in their respective preference lists; at this time, the more users available in the bank branch's second preference list, the more selectivity the bank branch has. This means that the bank branch has a higher tolerance for stability. On the contrary, if the number of users is small, the bank branch's tolerance for stability is lower. Similarly, this also applies to users' tolerance for bank branches. Therefore, in the definition of feasible pairing priority, when there are more possibilities for either bank branches or users to choose from, the bank branch's or user's demand for stability will decrease. Therefore, the final feasible pairing priority θ(t i ,p j ) explains: When user p j Compared to bank branches i More like t k hour, When the bank branch i Compared to user p j More like p k hour, When the bank branch i Compared to user p j More like p k And when user p j Compared to bank branches i More like t k When t k θ(t i ,p j )<θ(t k ,p k ).

[0149] S4013 : Sort the feasible pairings in descending order of priority to obtain a first feasible pairing set.

[0150] In this step, the smaller the priority data, the higher the priority. The first feasible pairing set obtained by sorting based on the priority includes: feasible pairs, users and bank branches corresponding to each feasible pairing, and the priority corresponding to each feasible pairing.

[0151] S402: Obtain a set of under-capacity bank branches based on the user set, the bank branch set, and the first feasible pairing set.

[0152] Optionally, a possible implementation method for obtaining the set of under-staffed network points is:

[0153] S4021. Based on the user set and the bank branch set, determine a stable pairing set from the first feasible pairing set.

[0154] Alternatively, a possible implementation of determining the stable pairing set from the first feasible pairing set is:

[0155] a1. Traverse the first feasible pairing set and add the feasible pairing with the highest priority in the first feasible pairing set as a stable pairing to the stable pairing set.

[0156] Exemplarily, there is a first feasible pairing set:

[0157] Pairing 1:

[0158] Pairing 2:

[0159] Pairing 3:

[0160] Select the pairing 1 with the highest priority and add it to the stable set.

[0161] a2. Delete other feasible pairings related to the users in the stable pairing from the first feasible pairing set.

[0162] In this step, deleting other feasible pairs related to the user means that when it is determined that the user has been assigned to a certain bank branch, all other pairs related to the user are removed from the first feasible pairing set to avoid duplicate matching.

[0163] For example, after pairing 1 in a1 is added to the stable pairing set, other feasible pairs related to user A need to be deleted.

[0164] a3. When the number of stable pairings of bank branches reaches the expected number of bank branches, other feasible pairings related to the bank branches are deleted from the first feasible pairing set.

[0165] In this step, if the number of stable matches for a bank branch has reached its maximum customer capacity, all remaining matches related to the bank branch are deleted, where the expected number refers to the maximum customer capacity of the bank branch in the current time and space period.

[0166] For example, assume that the expected capacity of bank branch X is 1 and user A is currently assigned. The remaining feasible pairing set includes pairing 4: Then, other feasible pairings including bank branch X and pairing 4 are deleted, and the first feasible pairing set obtained after the update is empty.

[0167] a4. After the traversal is completed, a stable pairing set and an updated first feasible pairing set are obtained.

[0168] In this step, by traversing the first feasible pairing set, a stable pairing set is obtained, which includes multiple bank branches and the stable pairing corresponding to each bank branch. The updated first feasible pairing set may be empty, or there may be feasible pairs that have not yet formed a stable pairing.

[0169] For example, when each bank branch in the bank branch set has reached the expected number and is no longer in business hours, there are still unassigned customers, and then there are still remaining feasible pairing sets in the first feasible pairing set.

[0170] S4022. Based on the stable pairing set, filter out bank branches that are not fully staffed and are still in business hours from the bank branch set to obtain a set of not fully staffed bank branches.

[0171] Alternatively, a possible implementation method for obtaining the set of under-staffed bank branches is:

[0172] b1. Based on the stable pairing set, determine the number of stable pairings corresponding to each bank branch in the bank branch set.

[0173] For example, the number of stable pairs that each bank has forgotten to bring can be calculated as follows: the stable pair set includes: Statistical results: The number of stable pairings for bank branch X is 2, and for bank branch Y is 1.

[0174] b2. When the number of stable matches reaches the maximum user limit of the corresponding bank branch, or when the bank branch in the stable match is not in business hours, the bank branch is deleted from the bank branch set to obtain a set of under-capacity bank branches.

[0175] In this step, a bank branch that is not fully occupied refers to a bank branch that is in business hours and has not reached the maximum user limit and still has user reception needs.

[0176] For example, a set of network points:

[0177] Bank branch X, maximum user limit 3, business hours 9:00-12:00, current hours 10:00-11:00 (open for business).

[0178] Bank branch Y, maximum user limit 3, business hours 14:00-17:00, current hours 10:00-11:00 (closed).

[0179] Bank branch Z, maximum user limit 1, business hours 9:00-17:00, current hours 10:00-11:00 (open for business).

[0180] The stable pairing numbers for each bank branch are: Bank Branch X, stable pairing number 2; Bank Branch Y, stable pairing number 1; Bank Branch Z, stable pairing number 1.

[0181] It is determined that bank branch X has not reached the maximum user limit and is within business hours; bank branch Y has not reached the maximum user limit but is not within business hours; bank branch Z is within business hours but has reached the maximum user limit. Therefore, the final set of unoccupied branches = {bank branch X}.

[0182] S403. For each under-capacity bank branch in the set of under-capacity bank branches, calculate a first probability that each user in the user set forms a first stable pairing with each under-capacity bank branch in the current spatiotemporal period, and calculate a second probability that each user in the user set forms a second stable pairing with each under-capacity bank branch in the next spatiotemporal period.

[0183] In this step, the first probability and the second probability are calculated as shown in Formula 6:

[0184]

[0185] in, Refers to bank branches i and user p j At time k i The probability of forming a stable pairing, where time k i Indicates whether the pair is in the current space-time cycle or in the next space-time cycle of the current space-time cycle; t e Refers to bank branches i Closing hours of business; Refers to the time k i Bank branches i The probability of not matching with other high-priority users; Refers to user p jIn the remaining time t e -k i Arrive at a bank branch within i probability. Indicates bank branch t i At the current time k i The probability of selecting rl stable matching users is still needed, where r is the bank branch t i At the current time k i The maximum passenger flow that can be carried, l is the bank branch t i At the current time k i of existing passenger traffic, The calculation formula is as follows:

[0186]

[0187] in, Indicates bank branch t i In the time interval [k i ,t e ] does not match user p j The probability of a higher priority user. If the bank branch t i At the closing time t e Any time step k before i+l , l=0,1,....has carried the maximum passenger flow, then at time k i+l at The calculation formula is shown in Formula 8:

[0188]

[0189] in, Indicates bank branch t i In the time interval [k i ,t e ] does not match user p j the probability of a higher priority user; Refers to user p j In the remaining time t e -k i Arrive at a bank branch within i The probability of t e Refers to bank branches i Closing hours of business; Refers to the time k i Bank branches i The probability of not matching with other high-priority users.

[0190] S404: Calculate the first user revenue and the first network revenue of the first stable pairing in the current space-time period based on the first probability, and calculate the second user revenue and the second network revenue of the second stable pairing in the next space-time period based on the second probability.

[0191] In this step, the calculation formula for the first user's revenue is shown in Formula 9:

[0192]

[0193] Among them, A i (k i ) refers to bank branches i In time slice k i The cumulative income of all matched users within the period; l(m) refers to the current matching to bank branch t i The number of users; a(p l ) refers to user p l For bank branches i revenue contribution.

[0194] The calculation formula for the first outlet's revenue is shown in Formula 10:

[0195]

[0196] Among them, R j (k i ) refers to user p j In time slice k i The cumulative revenue of all matching nodes within the network; l(ξ) refers to the user p j The number of possible matching bank branches; r(t l ) refers to bank branches l For user p j revenue contribution.

[0197] The calculation formula for the second user's revenue is shown in Formula 11:

[0198]

[0199] Among them, A i,j (k i ) refers to user p j Skip the current time slice k i , the predicted benefits that may be obtained in the next space-time cycle; A(p j ) refers to user p j Benchmark return; A(p l ) refers to user p j In the next time and space cycle and other bank branches t l Expected benefits of pairing.

[0200] The calculation formula for the second branch’s revenue is shown in Formula 12:

[0201]

[0202] Among them, R i,j (k i ) refers to bank branches i Skip the current time slice k i , the predicted benefits that may be obtained in the next space-time cycle; R(t i ) refers to bank branches i Benchmark return; R(t l ) refers to bank branches i In the next space-time cycle and other users p l Expected benefits of pairing.

[0203] S405: Obtain a first quotient based on a ratio of the first user's income to the second user's income, and obtain a second quotient based on a ratio of the first network's income to the second network's income.

[0204] In this step, the calculation formula of the first quotient is shown in Formula 13:

[0205]

[0206] Among them, A i (k i ) refers to bank branches i In time slice k i The cumulative income of all matched users, that is, the income of the first user; A i,j (k i ) refers to user p j Skip the current time slice k i , the predicted benefit that may be obtained in the next space-time cycle, that is, the second user benefit.

[0207] The calculation formula for the second quotient is shown in Formula 14:

[0208]

[0209] Among them, R j (k i ) refers to user p j In time slice k i The cumulative revenue of all matching outlets within the network, that is, the revenue of the first outlet; R i,j (k i ) refers to bank branches i Skip the current time slice k i , the predicted profit that may be obtained in the next time and space cycle, that is, the second network point profit.

[0210] S406 : When both the first quotient value and the second quotient value are greater than a preset threshold, add the first stable pairing to the stable pairing set.

[0211] In this step, the preset threshold is set to the reciprocal of ω, wherein when the first quotient value and the second quotient value are both greater than 1 / ω, it indicates that a stable pairing is formed in the current space-time period.

[0212] S407: When the first quotient or the second quotient is less than a preset threshold, enter into the stable pairing calculation of the next spatiotemporal period.

[0213] In this step, the reciprocal of the preset threshold ω refers to the profit threshold coefficient used to control the strictness of the matching, and ω is greater than 1. When ω=1, it means that the first profit of the current time-space period must be equal to the second profit in the future for the matching to occur. When ω=2, it means that the first profit of the current time-space period must be twice the second profit in the future for the matching to occur.

[0214] In this embodiment, by first matching users and bank branches within the same space-time period, multiple stable pairings are obtained, and then under-crowded bank branches are screened out. By calculating the probability of forming stable pairings within the same space-time period and the probability of forming stable pairings within different space-time periods, it is determined whether a stable pairing can be formed in the current space-time period, and a stable pairing set corresponding to the user set and the bank branch set is obtained, thereby taking into account the recommendation results at different times and improving the accuracy of the recommendation.

[0215] Figure 6 The schematic diagram of the bank branch recommendation device provided in this application is as follows: Figure 6 As shown, the bank branch recommendation device provided in this embodiment includes:

[0216] The acquisition module 601 is used to obtain the user set and bank branch set of the current time-space period; the current time-space period is the time interval of the business hours of multiple bank branches in the bank branch set and the time-space unit formed by the location combination of the bank branches.

[0217] The first processing module 602 is configured to calculate, based on the user set and the bank branch set, a first preference list corresponding to each user and a second preference list corresponding to each bank branch.

[0218] The second processing module 603 is configured to pair users and bank branches based on the first preference list, the second preference list, the current spatiotemporal period, and the spatiotemporal period next to the current spatiotemporal period to obtain a stable pairing set.

[0219] The third processing module 604 is configured to generate bank branch recommendation information based on the stable pairing set.

[0220] In a possible implementation, the first processing module 602 is further configured to:

[0221] Calculate the target user's expected value for each bank branch in the bank branch set to obtain the target user's first preference list; the target user is any user in the user set.

[0222] The congestion degree of the target bank branch is calculated based on the initial preference list corresponding to the target bank branch, the number of users whose initial location is within the range of the target bank branch, and the maximum customer flow corresponding to the target bank branch; the target bank branch is any bank branch in the bank branch set.

[0223] Based on the initial satisfaction list of target users and the congestion degree of target bank outlets, the expected value of the target bank outlets for the target users is calculated.

[0224] The expected value of each target user is sorted in descending order based on the target bank branches to obtain a second preference list of the target bank branches.

[0225] In a possible implementation, the second processing module 603 is further configured to:

[0226] The first preference list and the second preference list are feasibly paired to obtain a first feasible pairing set.

[0227] Based on the user set, the bank branch set and the first feasible pairing set, a set of under-capacity bank branches is obtained.

[0228] For each undercrowded bank branch in the set of undercrowded bank branches, the first probability that each user in the user set forms a first stable pairing with each undercrowded bank branch in the current space-time period is calculated, and the second probability that each user in the user set forms a second stable pairing with each undercrowded bank branch in the next space-time period is calculated.

[0229] The first user benefit and the first network benefit of the first stable pairing in the current space-time period are calculated based on the first probability, and the second user benefit and the second network benefit of the second stable pairing in the next space-time period are calculated based on the second probability.

[0230] A first quotient is obtained based on a ratio of the first user's revenue to the second user's revenue, and a second quotient is obtained based on a ratio of the first network's revenue to the second network's revenue.

[0231] When both the first quotient value and the second quotient value are greater than a preset threshold, the first stable pairing is added to the stable pairing set.

[0232] When the first quotient value or the second quotient value is less than the preset threshold value, the stable pairing calculation of the next spatiotemporal period is entered.

[0233] In a possible implementation, the second processing module 603 is further configured to:

[0234] Match users and bank branches in the current time and space period to obtain multiple feasible pairs.

[0235] Calculate the priority corresponding to each feasible pairing.

[0236] The feasible pairings are sorted in descending order of priority to obtain a first feasible pairing set.

[0237] In a possible implementation, the second processing module 603 is further configured to:

[0238] Based on the user set and the bank branch set, a stable pairing set is determined from the first feasible pairing set.

[0239] Based on the stable pairing set, the bank branches that are not full and still in business hours are screened out from the bank branch set to obtain the set of not full bank branches.

[0240] In a possible implementation, the second processing module 603 is further configured to:

[0241] The first feasible pairing set is traversed, and the feasible pairing with the highest priority in the first feasible pairing set is added to the stable pairing set as a stable pairing.

[0242] From the first set of feasible pairings, other feasible pairings associated with users in the stable pairings are deleted.

[0243] When the number of stable pairings of bank branches reaches the expected number of bank branches, other feasible pairings related to the bank branches are deleted from the first feasible pairing set.

[0244] After the traversal is completed, a stable pairing set and an updated first feasible pairing set are obtained.

[0245] In a possible implementation, the second processing module 603 is further configured to:

[0246] Based on the stable pairing set, the number of stable pairings corresponding to each bank branch in the bank branch set is determined.

[0247] When the number of stable pairings reaches the maximum user limit of the corresponding bank branch, or when the bank branch in the stable pairing is not in business hours, the bank branch is deleted from the bank branch set to obtain a set of under-capacity bank branches.

[0248] The device provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effects are similar and will not be described in detail in this embodiment.

[0249] Figure 7 This is a schematic diagram of the structure of the electronic device provided in this application. Figure 7 As shown, the electronic device provided in this embodiment includes: at least one processor 701 and a memory 702. Optionally, the device further includes a communication component 703. The processor 701, the memory 702 and the communication component 703 are connected via a bus 704.

[0250] In a specific implementation process, at least one processor 701 executes computer-executable instructions stored in the memory 702 , so that at least one processor 701 executes the above-mentioned bank branch recommendation method or methods.

[0251] The specific implementation process of the processor 701 can be found in the above method embodiment. Its implementation principle and technical effects are similar and will not be repeated here in this embodiment.

[0252] In the above embodiments, it should be understood that the processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. A general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the present invention may be directly executed by a hardware processor or by a combination of hardware and software modules in the processor.

[0253] The memory may include a high-speed memory (Random Access Memory, RAM), and may also include a non-volatile memory (NVM), such as at least one disk memory.

[0254] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. Buses can be classified into address buses, data buses, and control buses. For ease of illustration, the buses in the drawings of this application are not limited to just one bus or just one type of bus.

[0255] The present application also provides a computer program product, including a computer program, which implements the above method when executed by a processor.

[0256] The present application also provides a computer-readable storage medium, in which computer-executable instructions are stored. When a processor executes the computer-executable instructions, the above method is implemented.

[0257] The above-mentioned readable storage medium can be implemented by any type of volatile or non-volatile memory device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The readable storage medium can be any available medium that can be accessed by a general-purpose or special-purpose computer.

[0258] An exemplary readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be an integral part of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the readable storage medium can also exist in the device as discrete components.

[0259] The division of units is merely a logical functional division; actual implementations may employ alternative divisions, such as combining or integrating multiple units or components into another system, or omitting or disabling certain features. Furthermore, any coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between devices or units, whether electrical, mechanical, or otherwise, through some interface.

[0260] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0261] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0262] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program codes.

[0263] Those skilled in the art will appreciate that all or part of the steps in the above-described method embodiments can be implemented using hardware associated with program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0264] Finally, it should be noted that those skilled in the art will readily identify other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses, or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary techniques in the art not disclosed herein. The present invention is not limited to the precise structure described above and illustrated in the accompanying drawings, and various modifications and variations may be made without departing from the scope thereof. The scope of the present invention is limited solely by the appended claims.

Claims

1. A bank branch recommendation method, characterized in that: include: Get the user set and bank branch set of the current time and space period; The current space-time period is a time interval of the business hours of multiple bank outlets in the bank outlet set, and a space-time unit formed by the combination of the locations of the bank outlets; Calculating, based on the user set and the bank branch set, a first preference list corresponding to each user and a second preference list corresponding to each bank branch; Pairing users and bank branches based on the first preference list, the second preference list, the current spatiotemporal period, and the spatiotemporal period next to the current spatiotemporal period to obtain a stable pairing set; Based on the stable pairing set, bank branch recommendation information is generated.

2. The method according to claim 1, characterized in that The step of calculating, based on the user set and the bank branch set, a first preference list corresponding to each user and a second preference list corresponding to each bank branch includes: Calculating the target user's expected value for each bank branch in the bank branch set to obtain a first preference list of the target user; the target user is any user in the user set; calculating the congestion degree of the target bank branch based on an initial preference list corresponding to the target bank branch, the number of users initially located within the target bank branch, and the maximum customer flow corresponding to the target bank branch; the target bank branch being any bank branch in the set of bank branches; Calculating the target bank branch's expected value for the target user based on the target user's initial satisfaction list and the target bank branch's congestion level; The expected value of each target user is sorted in descending order based on the target bank outlets to obtain a second preference list of the target bank outlets.

3. The method according to claim 1, characterized in that The pairing of users and bank branches based on the first preference list, the second preference list, the current spatiotemporal period, and the spatiotemporal period next to the current spatiotemporal period to obtain a stable pairing set includes: Performing feasible pairing on the first preference list and the second preference list to obtain a first feasible pairing set; Obtaining a set of under-capacity bank branches based on the user set, the bank branch set, and the first feasible pairing set; For each under-capacity bank branch in the set of under-capacity bank branches, calculating a first probability that each user in the user set forms a first stable pairing with each under-capacity bank branch in the current spatiotemporal period, and calculating a second probability that each user in the user set forms a second stable pairing with each under-capacity bank branch in the next spatiotemporal period; Calculate the first user revenue and the first network revenue of the first stable pairing in the current space-time period based on the first probability, and calculate the second user revenue and the second network revenue of the second stable pairing in the next space-time period based on the second probability; Obtaining a first quotient based on a ratio of the first user's revenue to the second user's revenue, and obtaining a second quotient based on a ratio of the first network point's revenue to the second network point's revenue; When both the first quotient value and the second quotient value are greater than a preset threshold, adding the first stable pairing to the stable pairing set; When the first quotient value or the second quotient value is smaller than the preset threshold value, the stable pairing calculation of the next spatiotemporal period is entered.

4. The method according to claim 3, characterized in that The performing feasible pairing on the first preference list and the second preference list to obtain a first feasible pairing set includes: Matching users and bank branches in the current time-space period to obtain multiple feasible pairs; Calculating the priority corresponding to each of the feasible pairings; The feasible pairings are sorted in descending order of priority to obtain the first feasible pairing set.

5. The method according to claim 4, characterized in that The obtaining of a set of under-capacity bank branches based on the user set, the bank branch set, and the first feasible pairing set includes: Determining a stable pairing set from the first feasible pairing set based on the user set and the bank branch set; Based on the stable pairing set, bank outlets that are not fully staffed and are still in business hours are screened out from the bank outlet set to obtain the not fully staffed bank outlet set.

6. The method according to claim 5, characterized in that The determining of a stable pairing set from the first feasible pairing set based on the user set and the bank branch set includes: Traversing the first feasible pairing set, and adding the feasible pairing with the highest priority in the first feasible pairing set as a stable pairing to the stable pairing set; Deleting other feasible pairings related to the user in the stable pairing from the first feasible pairing set; When the number of stable pairs of the bank branch reaches the expected number of the bank branch, deleting other feasible pairs related to the bank branch from the first set of feasible pairs; After the traversal is completed, the stable pairing set and the updated first feasible pairing set are obtained.

7. The method according to claim 5, characterized in that The step of screening out bank outlets that are not fully staffed and are still in business hours from the set of bank outlets based on the stable pairing set, to obtain the set of bank outlets that are not fully staffed, includes: Determining, based on the stable pairing set, the number of stable pairings corresponding to each bank branch in the bank branch set; When the number of stable pairings reaches the maximum user limit of the corresponding bank branch, or when the bank branch in the stable pairing is not in business hours, the bank branch is deleted from the bank branch set to obtain the under-capacity bank branch set.

8. A bank branch recommendation device, characterized in that: include: An acquisition module is used to obtain the user set and bank branch set of the current time and space period; The current space-time period is a time interval of the business hours of multiple bank outlets in the bank outlet set, and a space-time unit formed by the combination of the locations of the bank outlets; A first processing module is configured to calculate, based on the user set and the bank branch set, a first preference list corresponding to each user and a second preference list corresponding to each bank branch; a second processing module, configured to pair users and bank branches based on the first preference list, the second preference list, the current spatiotemporal period, and the spatiotemporal period next to the current spatiotemporal period to obtain a stable pairing set; The third processing module is configured to generate bank branch recommendation information based on the stable pairing set.

9. An electronic device, characterized in that: include: Memory, processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, so that the processor performs the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 7 when executed by a processor.

Images