Social information pushing method, related device, equipment and storage medium

Abstract

The invention discloses a social information pushing method, a related device, equipment and a storage medium. The method comprises the steps that in response to a payment transaction request, a consumption feature vector corresponding to a target user account is acquired, and the payment transaction request carries the target user account and equipment position information; determining M user accounts according to the equipment position information and the consumption feature vector corresponding to the target user account; determining N pieces of recommended place information according to the equipment position information; the social information is pushed to at least two terminals, the at least two terminals comprise a target terminal and at least one terminal, the target terminal has logged in the target user account, each terminal in the at least one terminal has logged in one user account in the M user accounts, and the social information comprises N pieces of recommended place information. According to the invention, the social cost of the user is reduced, and the accuracy of social matching can be improved.

Description

Technical Field

[0001] This application relates to the field of Internet technology, and in particular to a method, related apparatus, device, and storage medium for pushing social information. Background Technology

[0002] Social interaction is an essential need in people's daily lives. With the advent of the internet age, people's social interactions have continuously changed, and online social networking has become one of the important ways to fulfill these needs. Users can build friendships with other strangers online, expanding their networks and social connections.

[0003] Currently, users can connect with strangers through online social applications. First, users register an account on the application and fill in personal information (such as interests, location, etc.). Then, the application uses matching algorithms to find potential matches based on this information. After a successful match, users can communicate with their matched counterparts through the application and then choose to meet in real life.

[0004] The inventors have discovered at least the following problems with current solutions: First, the current method relies heavily on user-provided personal information for matching, which limits matching accuracy. Second, users need to communicate to some extent before arranging a meeting, resulting in high communication costs for them. Therefore, an effective method is urgently needed to address these issues. Summary of the Invention

[0005] This application provides a method, related apparatus, device, and storage medium for pushing social information, which not only reduces the social costs for users but also improves the accuracy of social matching.

[0006] In view of this, this application provides a method for pushing social information, including:

[0007] In response to a payment transaction request, the consumption feature vector corresponding to the target user account is obtained, wherein the payment transaction request carries the target user account and device location information;

[0008] Based on the device location information and the consumption feature vector corresponding to the target user account, determine M user accounts, where M is an integer greater than or equal to 1;

[0009] N recommended locations are determined based on the device location information, where N is an integer greater than or equal to 1;

[0010] Social information is pushed to at least two terminals, including a target terminal and at least one other terminal. The target terminal is logged into a target user account, and each of the at least one terminal is logged into one of M user accounts. The social information includes N recommended location information.

[0011] This application also provides a social information push device, including...

[0012] The acquisition module is used to obtain the consumption feature vector corresponding to the target user account in response to the payment transaction request. The payment transaction request carries the target user account and device location information.

[0013] The determination module is used to determine M user accounts based on the device location information and the consumption feature vector corresponding to the target user account, where M is an integer greater than or equal to 1;

[0014] The determination module is also used to determine N recommended location information based on the device location information, where N is an integer greater than or equal to 1;

[0015] The push module is used to push social information to at least two terminals, wherein the at least two terminals include a target terminal and at least one terminal. The target terminal is logged into a target user account, and each of the at least one terminal is logged into one of M user accounts. The social information includes N recommended location information.

[0016] In one possible design, in another implementation of another aspect of the embodiments of this application,

[0017] The acquisition module is specifically used to respond to payment transaction requests sent through the POS device, wherein the device location information carried in the payment transaction request is the current location information of the payment interaction device;

[0018] or,

[0019] In response to a payment transaction request sent by the target terminal, wherein the device location information carried in the payment transaction request is the current location information of the target terminal.

[0020] In one possible design, in another implementation of another aspect of the embodiments of this application,

[0021] The acquisition module is specifically used to query the consumption feature vector corresponding to the target user account from the database;

[0022] The database is used to store the consumption feature vector corresponding to at least one user account. The consumption feature vector is constructed based on the historical consumption data corresponding to the user account.

[0023] In one possible design, in another implementation of another aspect of the embodiments of this application, the social information push device further includes a processing module;

[0024] The acquisition module is also used to acquire the order data set corresponding to the target user account before querying the consumption feature vector corresponding to the target user account from the database. The order data set includes order data from various historical periods.

[0025] The processing module is used to perform statistics on the order data set corresponding to the target user account to obtain the historical consumption data corresponding to the target user account;

[0026] The processing module is also used to standardize the historical consumption data corresponding to the target user account to obtain the consumption feature vector corresponding to the target user account.

[0027] In one possible design, in another implementation of another aspect of the embodiments of this application, the payment transaction request also carries target order data;

[0028] The acquisition module is specifically used to acquire the order data set corresponding to the target user account. The order data set includes order data from various historical time periods.

[0029] By statistically analyzing the order data set corresponding to the target user account and the target order data, the real-time consumption data corresponding to the target user account can be obtained.

[0030] The real-time consumption data corresponding to the target user account is standardized to obtain the consumption feature vector corresponding to the target user account.

[0031] In one possible design, in another implementation of another aspect of the embodiments of this application, the payment transaction request also carries the target consumption time;

[0032] The determination module is specifically used to determine the associated area range based on the device location information;

[0033] Determine the relevant time range based on the target consumption time;

[0034] From the historical consumption data corresponding to each user account, obtain the location and time of the most recent consumption for each user account;

[0035] The candidate user account set includes all user accounts whose most recent purchase location is within the relevant area and whose most recent purchase time is within the relevant time range.

[0036] Select M user accounts from the candidate user account set.

[0037] In one possible design, in another implementation of another aspect of the embodiments of this application, the payment transaction request also carries the target consumption time;

[0038] The determination module is specifically used to obtain the cluster label corresponding to the target user account, wherein the cluster label is used to indicate the clustering attribute information corresponding to the user account;

[0039] Determine the associated area range based on the device location information;

[0040] Determine the relevant time range based on the target consumption time;

[0041] From the historical consumption data corresponding to each user account, obtain the location and time of the most recent consumption for each user account;

[0042] The candidate user account set includes all user accounts whose most recent consumption location is within the associated region, whose most recent consumption time is within the associated time, and whose cluster label is the same as the cluster label corresponding to the target user account.

[0043] Select M user accounts from the candidate user account set.

[0044] In one possible design, in another implementation of another aspect of the embodiments of this application,

[0045] The determination module is specifically used to obtain the historical consumption data corresponding to each user account;

[0046] The historical consumption data corresponding to each user account is standardized to obtain the consumption feature vector corresponding to each user account;

[0047] Based on the consumption feature vector corresponding to each user account, cluster label corresponding to each user account is obtained through cluster analysis;

[0048] Based on the cluster label corresponding to each user account, determine the cluster label corresponding to the target user account.

[0049] In one possible design, in another implementation of another aspect of the embodiments of this application,

[0050] The determination module is specifically used to calculate the similarity between the target user account and each user account in the candidate user account set based on the consumption feature vector corresponding to the target user account and the consumption feature vector corresponding to each user account in the candidate user account set.

[0051] Based on the similarity between the target user account and each user account in the candidate user account set, determine M user accounts.

[0052] In one possible design, in another implementation of another aspect of the embodiments of this application, the social information push device further includes a receiving module;

[0053] The determination module is specifically used to obtain M user accounts through a recommendation model based on the consumption feature vector corresponding to the target user account and the consumption feature vector corresponding to each user account in the candidate user account set.

[0054] The receiving module is used to receive recommendation ratings sent by the target terminal after pushing social information to at least two terminals;

[0055] The processing module is also used to update the model parameters of the recommendation model based on the recommendation score.

[0056] In one possible design, in another implementation of another aspect of the embodiments of this application,

[0057] The determination module is specifically used to determine a set of candidate locations based on the device location information and the most recent consumption location corresponding to each of the M user accounts, wherein the set of candidate locations includes at least one candidate location;

[0058] Obtain the consumption rating corresponding to each candidate location in the candidate location set;

[0059] Based on the consumption rating corresponding to each candidate location, N recommended location information are determined.

[0060] In one possible design, in another implementation of another aspect of the embodiments of this application, N equals 1;

[0061] Social messaging devices also include a sending module;

[0062] The sending module is used to send a social success message to at least two terminals after pushing social information to at least two terminals, if at least two terminals send social consent requests respectively.

[0063] The sending module is also configured to send a social failure message to at least two terminals respectively if at least one of the at least two terminals sends a social rejection request.

[0064] In one possible design, in another implementation of another aspect of the embodiments of this application, N is greater than 1;

[0065] The receiving module is also configured to receive a location selection request sent by each of the at least two terminals after pushing social information to at least two terminals, wherein the location selection request carries any one of N recommended location information;

[0066] The sending module is also used to send a social success message to at least two terminals if the recommended location information carried in each location selection request is the same.

[0067] The sending module is also used to send a social failure message to at least two terminals if the recommended location information carried in each location selection request is not completely the same.

[0068] In one possible design, in another implementation of another aspect of the embodiments of this application, the target terminal is used to display the received social information;

[0069] The target terminal is also used to provide a session control, wherein, in response to a selection operation on the session control, the target terminal displays the session interface corresponding to the target group, the target group including the target user account and M user accounts;

[0070] The target terminal is also used to provide a reservation control; wherein, in response to a selection operation on the reservation control, the target terminal displays a reservation interface corresponding to the recommended location information.

[0071] In another aspect, this application provides a computer device including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the methods described above.

[0072] Another aspect of this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the methods described above.

[0073] Another aspect of this application provides a computer program product, including a computer program that, when executed by a processor, implements the methods described above.

[0074] As can be seen from the above technical solutions, the embodiments of this application have the following advantages:

[0075] This application provides a method for pushing social information. The server obtains the consumption feature vector corresponding to the user who triggered the payment request based on the payment transaction request, and then combines this with device location information for social matching. This allows for a more accurate reflection of the user's consumption habits and preferences, improving matching accuracy. If a match is successful, the device location information is used again to recommend nearby social locations to facilitate offline meetings. This approach not only reduces the user's social costs but also improves the accuracy of social matching. Attached Figure Description

[0076] Figure 1 This is a schematic diagram of an implementation environment for the social information push method in this application embodiment;

[0077] Figure 2 This is a schematic diagram of another implementation environment for the social information push method in this application;

[0078] Figure 3 This is a flowchart illustrating a social information push method in an embodiment of this application.

[0079] Figure 4 This is a schematic diagram illustrating the matching of user accounts in an embodiment of this application;

[0080] Figure 5 This is another schematic diagram illustrating the matching of user accounts in an embodiment of this application;

[0081] Figure 6 This is a schematic diagram illustrating the determination of the surrounding area in an embodiment of this application;

[0082] Figure 7 This is another schematic diagram illustrating the determination of the surrounding area in an embodiment of this application;

[0083] Figure 8 This is another schematic diagram illustrating the determination of the surrounding area in an embodiment of this application;

[0084] Figure 9 This is a schematic diagram of an interface for displaying social information based on different user accounts in an embodiment of this application;

[0085] Figure 10 This is a schematic diagram of a communication architecture for a payment interaction device in an embodiment of this application;

[0086] Figure 11 This is a schematic diagram of an interface for user authorization to report terminal location in an embodiment of this application;

[0087] Figure 12 This is a schematic diagram illustrating the determination of a candidate user account set in an embodiment of this application;

[0088] Figure 13 This is a schematic diagram of a recommendation model trained based on reinforcement learning in an embodiment of this application;

[0089] Figure 14 This is a schematic diagram illustrating the recommendation of social locations based on consumer ratings in an embodiment of this application;

[0090] Figure 15 This is a schematic diagram illustrating a single recommended location information in an embodiment of this application;

[0091] Figure 16 This is a schematic diagram illustrating multiple recommended location information in an embodiment of this application;

[0092] Figure 17 This is a schematic diagram illustrating the implementation of a session in an embodiment of this application;

[0093] Figure 18 This is a schematic diagram illustrating the implementation of a reservation system in an embodiment of this application;

[0094] Figure 19 This is a schematic diagram of the overall process of the social information push method in the embodiments of this application;

[0095] Figure 20 This is a schematic diagram of a social information push device in an embodiment of this application;

[0096] Figure 21 This is a schematic diagram of the structure of a computer device in an embodiment of this application. Detailed Implementation

[0097] This application provides a method, related apparatus, device, and storage medium for pushing social information, which not only reduces the social costs for users but also improves the accuracy of social matching.

[0098] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a particular order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented, for example, in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “corresponding,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0099] It is understood that, in the specific embodiments of this application, when data related to biometrics (e.g., facial images, palm print images, fingerprint images, voiceprint data, etc.), device location information, order data, consumption data, etc., are used in specific products or technologies, user permission or consent is required. That is, before collecting user data, users can be notified through prompts, pop-up windows, or voice prompts. The process of collecting user data only begins after user permission or consent has been obtained. In other words, all user data collected in this application is collected with the user's consent and authorization, and the collection, use, and processing of related data must comply with the relevant laws, regulations, and standards of the relevant countries and regions.

[0100] With the development of stranger-based social networking, finding more accurate and faster matchmaking has become a social need for young people. Traditional stranger-based social networking is mainly based on online applications (APPs). However, users need to communicate with their matched partners within the APP to confirm each other's interests and backgrounds, which consumes users' time and energy, resulting in high communication costs. Furthermore, since online communication cannot provide a complete understanding of the other person, users may have certain psychological barriers and concerns when deciding to meet offline. At the same time, relying on the personal information provided by users as the matching basis may not accurately reflect their actual preferences.

[0101] Based on this, this application provides a method for social matching based on payment, consumption profiles, and geolocation information. By obtaining user identity information and combining it with user consumption data to construct a user consumption profile, and using geolocation information for real-time matching, the method recommends nearby social locations after a successful match to facilitate offline meetings, thereby reducing users' social costs and achieving an efficient, real-time, and convenient method for socializing with strangers.

[0102] Before introducing the specific methods of this application, the application scenarios of this application will be illustrated by example in conjunction with the implementation environment. It should be understood that the following application scenarios are merely illustrative and are not limited to these examples in practice.

[0103] Implementation Environment 1: Biometric Payment;

[0104] The method provided in this application can be applied to... Figure 1 The implementation environment shown includes a payment interaction device 101, a cash register device 102, a terminal 103, a network 104, a server 105, and a database 106.

[0105] Payment interaction device 101 includes, but is not limited to, point-of-sale (POS) terminals, QR code scanners, biometric devices, and near-field communication (NFC) devices. Biometric devices include, but are not limited to, facial recognition devices, fingerprint recognition devices, iris recognition devices, voiceprint recognition devices, and palm-swipe payment devices. This application uses a palm-swipe payment device as an example for illustration.

[0106] The POS device 102 is used for receiving payments, recording sales data, and managing inventory during transactions. The POS device 102 and the payment interaction device 101 can be connected via a universal serial bus (USB).

[0107] Terminal 103 includes, but is not limited to, mobile phones, tablets, laptops, desktop computers, smart voice payment interaction devices, virtual reality devices, smart home appliances, vehicle terminals, and aircraft. Client 1031 is deployed on terminal 103. Client 1031 can run on terminal 103 via a browser, or via a standalone app or mini-program.

[0108] Network 104 uses standard communication technologies and / or protocols, typically the Internet, but can also be any network, including but not limited to Bluetooth, local area network (LAN), metropolitan area network (MAN), wide area network (WAN), mobile, private network, or any combination of virtual private network. In some embodiments, custom or dedicated data communication technologies may be used to replace or supplement the aforementioned data communication technologies.

[0109] Server 105 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence (AI) platforms.

[0110] Database 106 is used to store the consumption feature vectors corresponding to user accounts for use by server 105.

[0111] Based on the above implementation environment, let's take the example of target users making payments by swiping their palms at offline consumption venues.

[0112] In step A1, the target user provides biometric data (e.g., palm print image) to the payment interaction device 101 (e.g., palm payment device) and executes the palm payment process.

[0113] In step A2, the payment interaction device 101 sends the target user account and device location information to the cashier device 102. The target user account (e.g., a user account nicknamed "User A") is the target user's account, and the device location information is the location information of the payment interaction device 101. Based on this, the cashier device 102 sends a payment transaction request carrying the target user account and device location information to the server 105 via network 104.

[0114] In step A3, server 105 responds to the payment transaction request by retrieving the consumption feature vector corresponding to the target user account from database 106.

[0115] In step A4, server 105 matches the relevant information of several user accounts (e.g., user account nicknamed "User B") to be pushed to the target user based on device location information and consumption feature vector corresponding to the target user account.

[0116] In step A5, server 105 obtains several recommended location information (e.g., bookstore, park, restaurant, etc.) based on device location information.

[0117] In step A6, server 105 pushes social information to the terminal used by the target user and the terminal used by user B, respectively. The social information pushed to the target user includes user B's relevant information (e.g., interests, personal signature, etc.) and recommended location information. The social information pushed to user B includes the target user's relevant information (e.g., interests, personal signature, etc.) and recommended location information.

[0118] Implementation Environment 2: QR code payment;

[0119] The method provided in this application can be applied to... Figure 2 The implementation environment shown includes a terminal 201, a network 202, a server 203, and a database 204. It is understood that details regarding the terminal 201, network 202, server 203, and database 204 can be found in the foregoing embodiments and will not be repeated here.

[0120] Based on the above implementation environment, let's take the example of target users making payments by swiping their palms at offline consumption venues.

[0121] In step B1, the target user uses terminal 201 to scan the QR code provided by the merchant and executes the QR code payment process. Terminal 201 sends a payment transaction request carrying the target user's account and device location information to server 203 via network 202. The target user's account (e.g., a user account nicknamed "User A") is the target user's account, and the device location information is the location information of terminal 201.

[0122] In step B2, server 203 responds to the payment transaction request by retrieving the consumption feature vector corresponding to the target user account from database 204.

[0123] In step B3, server 203 matches the relevant information of several user accounts (e.g., user account nicknamed "User B") to be pushed to the target user based on device location information and consumption feature vector corresponding to the target user account.

[0124] In step B4, server 203 obtains several recommended location information (e.g., bookstore, park, restaurant, etc.) based on device location information.

[0125] In step B5, server 203 pushes social information to the terminal used by the target user and the terminal used by user B, respectively. The social information pushed to the target user includes relevant information about user B and recommended location information. The social information pushed to user B includes relevant information about the target user and recommended location information.

[0126] Based on the above introduction, the social information push method in this application will be described below. Please refer to [link / reference]. Figure 3 The social information push method in this application embodiment can be completed independently by the server or in cooperation with the terminal. The method provided in this application includes:

[0127] S301. In response to the payment transaction request, obtain the consumption feature vector corresponding to the target user account, wherein the payment transaction request carries the target user account and device location information;

[0128] In one or more embodiments, after receiving a payment transaction request, the server can obtain its corresponding consumption feature vector based on the target user account and device location information carried in the payment transaction request. Here, the target user account refers to the account used by the target user, and the target user is the user who triggered the payment transaction request.

[0129] Specifically, the server constructs a consumption feature vector corresponding to the target user's account based on the target user's consumption data. For example, consumption data includes, but is not limited to, consumption habits (e.g., consumption frequency, consumption time), consumption preferences (e.g., consumed goods, consumption locations, consumption merchants, etc.), and consumption capacity (consumption amount, number of purchases, maximum amount of a single purchase, etc.).

[0130] It should be noted that the term "in response to" in this application refers to the conditions or states upon which the execution of an operation depends, and one or more operations that can be executed when certain conditions or states are met. These operations can be real-time or have a certain delay.

[0131] S302. Based on the device location information and the consumption feature vector corresponding to the target user account, determine M user accounts, where M is an integer greater than or equal to 1.

[0132] In one or more embodiments, the server obtains a set of candidate user accounts based on device location information and the consumption feature vector corresponding to the target user account. The device location information narrows the matching range, and the consumption feature vector is used to filter users with similar consumption habits and interests to form the candidate user account set. Then, the server determines M user accounts from the candidate user account set to be pushed to the target user. The value of M can be customized by the target user or follow the system default setting. If the target user wants more social opportunities, they can set the value of M larger; if the target user values ​​more precise matching, they can set the value of M smaller.

[0133] The following diagrams illustrate how user accounts are pushed to target users based on the same location and the surrounding area. Let's take M = 1 as an example, and the target user account as "Account A".

[0134] For example, please refer to Figure 4 , Figure 4 This is a schematic diagram illustrating user account matching in an embodiment of this application. As shown in the figure, after a target user makes a purchase at a coffee shop, they trigger a payment transaction request by swiping their palm. The backend then records the purchase history corresponding to the target user account (i.e., "Account A"). Other users who also made purchases at the same coffee shop within one hour prior to the target user's purchase are selected as candidate users, resulting in a corresponding set of candidate user accounts. Next, the similarity between the purchase feature vector of each user account in the candidate user account set and the purchase feature vector of "Account A" is calculated. The user account with the highest similarity (e.g., "Account B") is then selected as the user account recommended to the target user.

[0135] For example, please refer to Figure 5 , Figure 5 This is another schematic diagram illustrating user account matching in this application embodiment. As shown in the figure, after a target user makes a purchase at a coffee shop, they trigger a payment transaction request by swiping their palm. The backend records the purchase history corresponding to the target user account (i.e., "Account A") and, based on the coffee shop's geographical location, determines that the associated area also includes hot pot restaurants, convenience stores, bookstores, and barbecue restaurants. Other users who made purchases within the associated area within one hour prior to the target user's purchase are selected as candidate users, resulting in a corresponding set of candidate user accounts. Next, the similarity between the purchase feature vector of each user account in the candidate user account set and the purchase feature vector of "Account A" is calculated, thereby selecting the user account with the highest similarity (e.g., "Account B") as the user account recommended to the target user.

[0136] S303. Determine N recommended locations based on the device location information, where N is an integer greater than or equal to 1;

[0137] In one or more embodiments, the server determines N recommended locations based on device location information. The recommended location information includes, but is not limited to, the address, location name, and map location of the recommended location; these are not limited here.

[0138] The following sections, with illustrations, will explain how to determine N recommended locations based on the surrounding area.

[0139] For example, please refer to Figure 6 , Figure 6 This is a schematic diagram illustrating the determination of the nearby area in this embodiment of the application. As shown in the figure, assuming that based on the background consumption records, the device location information corresponding to the target user account and the most recent consumption location corresponding to "Account B" are both "coffee shop", the nearby area can be constructed with the location of "coffee shop" as the center (e.g., an area with a radius of 500 meters). Therefore, several locations (e.g., convenience stores, bookstores, etc.) can be selected from the nearby area as recommended locations.

[0140] For example, let's take the user account "Account B" as the one being pushed to the target user. Please refer to [link / reference]. Figure 7 , Figure 7 As shown in the figure, another schematic diagram illustrating the determination of the nearby area range in this application embodiment is provided. Assuming that based on background consumption records, the device location information corresponding to the target user account is "coffee shop," and the most recent consumption location corresponding to "account B" is "hot pot restaurant," the nearby area range (e.g., a radius of 500 meters) can be constructed using the center location between the "coffee shop" and the "hot pot restaurant" as the center. Therefore, several locations (e.g., convenience stores, cinemas, etc.) can be selected from the nearby area range as recommended locations.

[0141] For example, consider a scenario where the user accounts pushed to the target user include "Account B", "Account C", and "Account D". Please refer to [link / reference]. Figure 8 , Figure 8 As shown in the figure, another schematic diagram illustrating the determination of the nearby area range in this application embodiment is provided. Assuming that based on background consumption records, the device location information corresponding to the target user account is "coffee shop," the most recent consumption location corresponding to "account B" is "hot pot restaurant," the most recent consumption location corresponding to "account C" is "barbecue restaurant," and the most recent consumption location corresponding to "account D" is "bookstore," then the nearby area range (e.g., a radius of 500 meters) can be constructed using the center location between "coffee shop," "hot pot restaurant," "barbecue restaurant," and "bookstore" as the center. From this, several locations (e.g., convenience stores) can be selected as recommended locations from the nearby area range.

[0142] S304. Push social information to at least two terminals, wherein the at least two terminals include a target terminal and at least one terminal, the target terminal is logged into a target user account, and each of the at least one terminal is logged into one of M user accounts, and the social information includes N recommended location information.

[0143] In one or more embodiments, the server pushes social information to the target terminal used by the target user based on M user accounts and N recommended location information. It should be noted that the server will also push social information to each of the M user accounts separately; the difference is that the social information pushed to other user accounts must include the target user account.

[0144] Specifically, taking the user account pushed to the target user as "Account B" as an example, assuming the user nickname corresponding to "Account B" is "User B", and the user nickname corresponding to the target user account is "User A". Based on this, please refer to... Figure 9 , Figure 9 This is a schematic diagram of an interface for displaying social information based on different user accounts in an embodiment of this application, wherein, Figure 9 The diagram in Figure (A) shows the social information recommended to the target user's account. This social information includes information related to "Account B", such as the nickname, gender, age, hobbies, and personal signature of "Account B". Figure 9 The diagram in Figure (B) shows the social information recommended to "Account B". This social information includes details related to the target user's account, such as the target user's nickname, gender, age, interests, and personal signature. In addition, the server can also provide advertising information to the target terminal, such as recommending products of interest based on the target user's consumption profile, thereby improving the shopping experience and conversion rate.

[0145] It should be noted that social information also needs to include information on N recommended locations (e.g., the location information of "XXX Park") and the social time, etc., which are not limited here.

[0146] This application provides a method for pushing social information. This method focuses more on users' actual lifestyles and interests, thereby improving the quality of social matching. Simultaneously, selecting offline meeting locations based on device location information increases the likelihood of offline social interaction.

[0147] Optionally, in the above Figure 3 Based on one or more corresponding embodiments, in another optional embodiment provided by this application, responding to a payment transaction request may specifically include:

[0148] In response to a payment transaction request sent through the POS device, wherein the device location information carried in the payment transaction request is the current location information of the payment interaction device;

[0149] or,

[0150] In response to a payment transaction request sent by the target terminal, wherein the device location information carried in the payment transaction request is the current location information of the target terminal.

[0151] In one or more embodiments, two methods for triggering payment transaction requests by devices are described. As can be seen from the foregoing embodiments, the device location information can be the current location information of the payment interaction device or the current location information of the target terminal. The target terminal is the terminal used by the target user.

[0152] The following examples will illustrate how different devices report payment transaction requests to the server.

[0153] 1. Report via POS device;

[0154] Merchants typically place payment interaction devices within their stores, and the location of these devices serves as location information, indicating the merchant's location. Based on this location information, merchants can identify potential social contacts by matching them with users who have made purchases at the same or nearby stores, and this also helps recommend suitable nearby social locations.

[0155] Specifically, for ease of understanding, please refer to Figure 10 , Figure 10 This is a schematic diagram of the communication architecture of a payment interaction device in an embodiment of this application. Taking a palm-scanning payment device as an example, the target user scans their palm on the payment interaction device, and the device uploads the captured palmprint image to the server. The server identifies the target user, generates a payment code, and then sends the payment code to the payment interaction device. The payment interaction device connects to the merchant's POS device via USB and transmits the payment code and device location information to the POS device, which then places an order using the payment code. The POS device sends a payment transaction request to the server, which includes device location information, order data, and the target user's account.

[0156] 2. Reporting via the target terminal;

[0157] Users typically make purchases at designated locations (e.g., inside a store), and the device's location, as location information, can indicate where the user made the purchase. Based on this location information, users who have made purchases at the same or nearby stores can be identified as potential social contacts, and this also helps recommend suitable nearby social locations.

[0158] It is worth noting that the terminal needs to obtain user authorization before reporting location information. For example, to obtain authorization from the target user, please refer to [link / reference needed]. Figure 11 , Figure 11 This is a schematic diagram of an interface for authorizing the reporting of terminal location in an embodiment of this application. Figure 11 Figure (A) shows the authorization interface. 1101 indicates the authorization control, and 1102 indicates the cancellation control. When the target user clicks the authorization control indicated by 1101, a request is triggered for the target terminal to report its device location information to the server, resulting in the following display: Figure 11 The authorization success screen is shown in Figure (B). If the target user clicks the cancel control indicated by 1102, the target terminal will be prevented from reporting device location information to the server.

[0159] Secondly, this application embodiment provides two methods for triggering payment transaction requests using devices. Through the above methods, not only does biometric payment support reporting device location information, but QR code payment also supports reporting device location information. This improves the flexibility and feasibility of information reporting, enabling the backend to obtain more accurate device location information.

[0160] Optionally, in the above Figure 3 Based on one or more corresponding embodiments, in another optional embodiment provided by this application, obtaining the consumption feature vector corresponding to the target user account may specifically include:

[0161] Retrieve the consumption feature vector corresponding to the target user account from the database;

[0162] The database is used to store the consumption feature vector corresponding to at least one user account. The consumption feature vector is constructed based on the historical consumption data corresponding to the user account.

[0163] In one or more embodiments, a method for querying consumption feature vectors based on a database is described. As can be seen from the foregoing embodiments, the consumption feature vectors corresponding to each user account can be stored in a database. Therefore, the server can obtain the consumption feature vector corresponding to a target user account by querying the database.

[0164] Specifically, the server first needs to collect order data for each target user account within a historical time period. This order data includes the target user's spending records at partner merchants (e.g., shopping malls, restaurants) and their spending behavior on partner platforms. Historical consumption data is then statistically analyzed based on the order data corresponding to each target user account. Based on this, a corresponding consumption feature vector is constructed according to the historical consumption data corresponding to the target user account. This historical consumption data includes, but is not limited to, total consumption amount, total number of transactions, average consumption amount, maximum consumption amount, time of the most recent transaction, location of the most recent transaction, and the percentage of goods purchased. The location of the most recent transaction is determined based on device location information.

[0165] Understandably, historical consumption data can comprehensively reflect users' consumption habits and preferences. For example, total consumption amount and total number of purchases can reflect users' consumption activity level, average consumption amount and maximum consumption amount can reflect users' spending power and consumption tendency, and the proportion of consumption categories can reflect users' preference for different goods or services.

[0166] It should be noted that in order to improve the accuracy, reliability and completeness of historical consumption data, the collected order data needs to be cleaned first, that is, to remove duplicate, erroneous and invalid data. Data preprocessing techniques can be used, such as missing value imputation and outlier handling.

[0167] Secondly, this application provides a method for querying consumption feature vectors based on a database. This method pre-constructs consumption feature vectors corresponding to user accounts offline, facilitating rapid user matching and social location matching, thereby improving the efficiency of information recommendation.

[0168] Optionally, in the above Figure 3 Based on one or more corresponding embodiments, in another optional embodiment provided by this application, before querying the consumption feature vector corresponding to the target user account from the database, the following may be included:

[0169] Obtain the order data set corresponding to the target user account, where the order data set includes order data from various historical time periods;

[0170] Statistical analysis is performed on the order data set corresponding to the target user account to obtain the historical consumption data corresponding to the target user account;

[0171] The historical consumption data corresponding to the target user account is standardized to obtain the consumption feature vector corresponding to the target user account.

[0172] In one or more embodiments, a method for constructing a consumption feature vector is described. As can be seen from the foregoing embodiments, based on background consumption records, it is possible to obtain various order data of a target user within a historical period (e.g., the past month), wherein the order data includes, but is not limited to, consumption amount, consumption time, and consumption category.

[0173] The following will use the target user account "Account A" as an example for introduction.

[0174] Specifically, suppose we obtain order data sets corresponding to "Account A", "Account B", and "Account C" from the backend consumption records, where the order data sets include order data from the past month. Based on this, we can separately obtain the historical consumption data corresponding to "Account A", "Account B", and "Account C". Taking historical consumption data including total consumption amount, total number of transactions, average consumption amount, and maximum consumption amount as an example, please refer to Table 1 for easier understanding. Table 1 is an example of historical consumption data corresponding to each user account.

[0175] Table 1

[0176]

[0177]

[0178] To facilitate subsequent data analysis and mining, these historical consumption data need to be standardized. For example, standardization can be performed using the min-max normalization method or the Z-score. The following example illustrates standardization using min-max normalization, where each feature's value is mapped to the interval [0,1] as follows:

[0179] S_normalization=(S_value-S_min) / (S_max-S_min); Formula (1)

[0180] Where S_normalization represents the normalized value. S_value represents the original value before normalization. S_min represents the minimum value for the same attribute. S_max represents the maximum value for the same attribute.

[0181] Taking the normalized total consumption amount corresponding to "Account C" as an example, its original value S_value is 3000, the minimum value of the attribute "total consumption amount" is 500, and the maximum value of the attribute "total consumption amount" is 4000. Based on this, the normalized value is calculated to be 0.714 using formula (1).

[0182] Similarly, the historical consumption data corresponding to each user account is normalized, resulting in the historical consumption data normalization results shown in Table 2.

[0183] Table 2

[0184]

[0185] Based on Table 2, the consumption feature vector corresponding to "Account A (i.e., the target user account)" is (1,0.5,1,1), the consumption feature vector corresponding to "Account B" is (0,0,0,0), and the consumption feature vector corresponding to "Account C" is (0.714,1,0.33,0.4615).

[0186] It should be noted that historical consumption data may also include the time and location of the most recent purchase, and the percentage of products purchased. The location of the most recent purchase can be represented by latitude and longitude. In practical applications, this data also needs to be standardized.

[0187] Furthermore, this application provides a method for constructing a consumption feature vector. By utilizing historical order data corresponding to a user account to construct the consumption feature vector, the user's consumption habits can be fully reflected, which is beneficial for improving the accuracy of social matching. In addition, by standardizing the data, the influence of data units and numerical ranges can be eliminated, facilitating subsequent data analysis and mining.

[0188] Optionally, in the above Figure 3 In addition to one or more corresponding embodiments, in another optional embodiment provided by this application, the payment transaction request also carries target order data;

[0189] Obtain the consumption feature vector corresponding to the target user account, which may specifically include:

[0190] Obtain the order data set corresponding to the target user account, where the order data set includes order data from various historical time periods;

[0191] By statistically analyzing the order data set corresponding to the target user account and the target order data, the real-time consumption data corresponding to the target user account can be obtained.

[0192] The real-time consumption data corresponding to the target user account is standardized to obtain the consumption feature vector corresponding to the target user account.

[0193] In one or more embodiments, a method for constructing a consumption feature vector in real time is described. As can be seen from the foregoing embodiments, after a payment transaction request is triggered, the server can obtain the order data of the target user's current consumption (i.e., the target order data). Therefore, the server combines the set of order data corresponding to the target user's account in the background consumption records (e.g., all order data within the past month) with the target order data to construct a consumption feature vector.

[0194] The following will use the target user account "Account A" as an example for introduction.

[0195] Specifically, order data sets corresponding to "Account A", "Account B", and "Account C" are obtained from the backend consumption records, and historical consumption data as shown in Table 1 is obtained. Assuming that the target user's consumption amount this time is 2600, based on this, the real-time consumption data of "Account A" is statistically analyzed in conjunction with the target order data. For ease of understanding, please refer to Table 3, which is an example of the real-time consumption data corresponding to each user account.

[0196] Table 3

[0197]

[0198] To facilitate subsequent data analysis and mining, these real-time consumption data need to be standardized. Taking the min-max normalization method as an example, the normalization results of the real-time consumption data can be obtained as shown in Table 4 by combining formula (1).

[0199] Table 4

[0200]

[0201] Based on Table 4, the consumption feature vector corresponding to "Account A (i.e., the target user account)" is (1,0.6,1,1), the consumption feature vector corresponding to "Account B" is (0,0,0,0), and the consumption feature vector corresponding to "Account C" is (0.41,1,0.2,0.25).

[0202] Secondly, this application provides a method for constructing a consumption feature vector in real time. By combining the user's current consumption behavior with historical order data, the consumption feature vector not only reflects the user's long-term consumption habits but also embodies the user's latest consumption characteristics, thereby improving the accuracy and reliability of matching.

[0203] Optionally, in the above Figure 3 In addition to one or more corresponding embodiments, in another optional embodiment provided by this application, the payment transaction request also carries the target consumption time;

[0204] Based on the device location information and the consumption feature vector corresponding to the target user account, M user accounts are determined, which may specifically include:

[0205] Determine the associated area range based on the device location information;

[0206] Determine the relevant time range based on the target consumption time;

[0207] From the historical consumption data corresponding to each user account, obtain the location and time of the most recent consumption for each user account;

[0208] The candidate user account set includes all user accounts whose most recent purchase location is within the relevant area and whose most recent purchase time is within the relevant time range.

[0209] Select M user accounts from the candidate user account set.

[0210] In one or more embodiments, a method for obtaining a set of candidate user accounts is described. As can be seen from the foregoing embodiments, after a target user completes a payment, the server can determine the target user's consumption time (i.e., the target consumption time) based on the payment transaction request, and combine this with device location information to filter out a portion of user accounts as a set of candidate user accounts.

[0211] Specifically, the server constructs an associated region range based on device location information. For example, using the location indicated by the device location information as the center, the range within a radius of 200 meters is defined as the associated region range. The server determines the associated time range based on the target consumption time. For example, if the target consumption time is 10:00:00 on October 20, 2024, then the range within approximately one hour can be defined as the associated time range (9:00:00 on October 20, 2024 to 10:00:00 on October 20, 2024).

[0212] Therefore, for ease of understanding, please refer to Figure 12 , Figure 12 This is a schematic diagram illustrating the determination of a candidate user account set in an embodiment of this application. As shown in the figure, 1201 is used to indicate the most recent consumption location (i.e., device location information) corresponding to "Account A (i.e., the target user account)," and an associated area range is constructed based on the location of "Account A." 1202 is used to indicate the most recent consumption location corresponding to "Account B," 1203 is used to indicate the most recent consumption location corresponding to "Account C," 1204 is used to indicate the most recent consumption location corresponding to "Account D," 1205 is used to indicate the most recent consumption location corresponding to "Account E," and 1206 is used to indicate the most recent consumption location corresponding to "Account F."

[0213] Please refer to Table 5, which shows the most recent consumption time for each user account.

[0214] Table 5

[0215] User account Last purchase time Account A October 20, 2024, 10:00:00 Account B October 20, 2024, 9:15:00 Account C October 20, 2024, 9:55:00 Account D October 19, 2024, 17:05:00 Account E October 20, 2024, 9:40:00 Account F October 18, 2024, 10:00:00

[0216] Depend on Figure 12 It can be seen that the most recent consumption locations corresponding to "Account B", "Account C", and "Account D" are within the associated area. Table 5 shows that the most recent consumption times corresponding to "Account B", "Account C", and "Account E" are within the associated time range. Therefore, the intersection of "Account B" and "Account C" is taken as the candidate user account set. Then, M user accounts can be determined from the candidate user account set.

[0217] Secondly, this application provides a method for obtaining a set of candidate user accounts. This method requires filtering not only the associated regional range based on device location information but also the associated time range based on the target user's consumption time. Therefore, combining the associated regional range and the associated time range for matching can narrow down the matching scope. This not only improves social matching efficiency but also ensures that recommended social partners are geographically close, facilitating offline meetings.

[0218] Optionally, in the above Figure 3 In addition to one or more corresponding embodiments, in another optional embodiment provided by this application, the payment transaction request also carries the target consumption time;

[0219] Based on the device location information and the consumption feature vector corresponding to the target user account, M user accounts are determined, which may specifically include:

[0220] Obtain the cluster label corresponding to the target user account, where the cluster label is used to indicate the clustering attribute information corresponding to the user account;

[0221] Determine the associated area range based on the device location information;

[0222] Determine the relevant time range based on the target consumption time;

[0223] From the historical consumption data corresponding to each user account, obtain the location and time of the most recent consumption for each user account;

[0224] The candidate user account set includes all user accounts whose most recent consumption location is within the associated region, whose most recent consumption time is within the associated time, and whose cluster label is the same as the cluster label corresponding to the target user account.

[0225] Select M user accounts from the candidate user account set.

[0226] In one or more embodiments, another method for obtaining a candidate user account set is described. As can be seen from the foregoing embodiments, after a target user completes a payment, the server can determine the target user's consumption time (i.e., the target consumption time) based on the payment transaction request, and combine this with device location information and cluster tags to filter out a subset of user accounts as a candidate user account set. The cluster tag of a user account is used to characterize the clustering attribute information corresponding to that user account.

[0227] It should be noted that cluster tags can be obtained manually. For example, total spending can be divided into different tiers: spending greater than or equal to 5000 belongs to the first tier; spending greater than or equal to 1000 but less than 5000 belongs to the second tier; and spending less than 1000 belongs to the third tier. Similarly, the total number of transactions can be divided into different tiers: spending greater than or equal to 100 belongs to the first tier; spending greater than or equal to 10 but less than 100 belongs to the second tier; and spending less than 10 belongs to the third tier. The maximum spending amount can also be divided into different tiers: maximum spending greater than or equal to 1000 belongs to the first tier; maximum spending greater than or equal to 100 but less than 1000 belongs to the second tier; and maximum spending less than 100 belongs to the third tier. Thus, corresponding cluster tags can be generated by combining different attribute levels. Cluster tags can also be obtained through clustering algorithms; that is, users belonging to the same cluster have the same cluster tag.

[0228] Specifically, the server constructs an associated region range based on device location information. For example, using the location indicated by the device location information as the center, the range within a radius of 200 meters is defined as the associated region range. The server determines the associated time range based on the target consumption time. For example, if the target consumption time is 10:00:00 on October 20, 2024, then the range within approximately one hour can be defined as the associated time range.

[0229] Therefore, for better understanding, please refer to [the relevant documentation / reference]. Figure 12 Among them, "Account A" is the target user account, and "Account B", "Account C" and "Account D" are respectively located in the relevant area for their most recent consumption.

[0230] It should be noted that the cluster label can be represented as "1", "2", "3", etc. For easier understanding, please refer to Table 6, which is an illustration of the most recent consumption time and cluster label for each user account.

[0231] Table 6

[0232]

[0233]

[0234] Table 6 shows that the most recent consumption times for "Account B", "Account C", and "Account E" are within the relevant time range. "Account B", "Account C", and "Account F" all share the same cluster label as "Account A". Therefore, the intersection of "Account B" and "Account C" is used as the candidate user account set. M user accounts can then be determined from this candidate user account set.

[0235] Secondly, this application provides another method for obtaining a set of candidate user accounts. This method filters related regional ranges based on device location information and related time ranges based on the target user's consumption time. Furthermore, it combines cluster tags to filter users. Therefore, matching based on related regional and time ranges narrows the matching scope. This not only improves social matching efficiency but also ensures that recommended social partners are geographically close, facilitating offline meetings. Furthermore, combining cluster tags further narrows the matching scope and identifies users with similar consumption patterns, thereby improving the accuracy of social matching.

[0236] Optionally, in the above Figure 3 Based on one or more corresponding embodiments, in another optional embodiment provided by this application, the cluster tag corresponding to the target user account is obtained, wherein the cluster tag is used to indicate the clustering attribute information corresponding to the user account, and may specifically include:

[0237] Obtain historical consumption data for each user account;

[0238] The historical consumption data corresponding to each user account is standardized to obtain the consumption feature vector corresponding to each user account;

[0239] Based on the consumption feature vector corresponding to each user account, cluster label corresponding to each user account is obtained through cluster analysis;

[0240] Based on the cluster label corresponding to each user account, determine the cluster label corresponding to the target user account.

[0241] In one or more embodiments, a method for generating cluster labels based on a clustering algorithm is described. As can be seen from the foregoing embodiments, cluster labels can be automatically generated based on the consumption feature vector corresponding to a user account using a clustering algorithm. The process of generating cluster labels will be illustrated below using the K-means clustering algorithm as an example.

[0242] Specifically, let's assume the number of clusters is set to 9.

[0243] from sklearn.cluster import KMeans / / Import the "K-means" package for clustering calculations

[0244] from sklearn.preprocessing import StandardScaler / / Import the "sklearn.preprocessing" package for data preprocessing.

[0245] import pandas as pd / / Import the "pandas" package for tabular data processing

[0246] #Assume we have a DataFrame df containing users' historical consumption data.

[0247] df = pd.DataFrame({

[0248] 'user_id':['user1','user2','user3','user4','user5'], / / user account

[0249] 'consumption_amount':[1000,500,1500,800,1200], / / Total consumption amount

[0250] 'consumption_times':[10,5,15,8,12], / / Total number of consumptions

[0251] 'consumption_average':[10,5,15,8,12], / / Average spending

[0252] 'consumption_max':[10,5,15,8,12], / / Maximum consumption amount

[0253] / / A two-dimensional table containing historical consumption data corresponding to user accounts

[0254] #Select the features to be clustered

[0255] features = ['consumption_amount', 'consumption_times', 'consumption_average', 'consumption_max'] / / The selected clustering features include total consumption amount, total number of consumptions, average consumption amount, and maximum consumption amount.

[0256] data = df[features] / / Use the values ​​corresponding to the clustering features as historical consumption data to be processed.

[0257] #Data Standardization

[0258] scaler = StandardScaler()

[0259] data = scaler.fit_transform(data) / / Standardize the data and obtain the consumption feature vector

[0260] # Cluster analysis using the K-means clustering algorithm

[0261] kmeans=KMeans(n_clusters=9,random_state=0)

[0262] kmeans.fit(data)

[0263] # Output the cluster label for each user account

[0264] df['cluster_label']=kmeans.labels

[0265] print(df)

[0266] This yields nine clusters, each with the same cluster label. Therefore, based on the cluster in which the target user account belongs, its corresponding cluster label can be determined.

[0267] Furthermore, this application provides a method for generating cluster labels based on a clustering algorithm. Compared to manual labeling, using a clustering algorithm for labeling not only improves labeling efficiency but also saves on manual processing costs.

[0268] Optionally, in the above Figure 3 Based on one or more corresponding embodiments, in another optional embodiment provided by this application, determining M user accounts from the candidate user account set may specifically include:

[0269] Based on the consumption feature vector corresponding to the target user account and the consumption feature vector corresponding to each user account in the candidate user account set, calculate the similarity between the target user account and each user account in the candidate user account set;

[0270] Based on the similarity between the target user account and each user account in the candidate user account set, determine M user accounts.

[0271] In one or more embodiments, a method for determining user accounts to be pushed to based on similarity is described. As can be seen from the foregoing embodiments, after obtaining a set of candidate user accounts, the similarity between the consumption feature vector corresponding to the target user account and the consumption feature vectors corresponding to each user account in the candidate user account set is calculated.

[0272] It should be noted that similarity calculation methods include, but are not limited to, calculating the Euclidean distance or cosine similarity between two consumer feature vectors. Euclidean distance maps to the interval [0,1], with smaller Euclidean distances indicating greater similarity. Cosine similarity maps to the interval [-1,1], with larger cosine similarities indicating greater similarity.

[0273] The following section will use the calculation of Euclidean distance and cosine similarity as examples. Assume the consumption feature vector includes the standardized total consumption amount and the total number of transactions, and assume the candidate user account set includes "Account B".

[0274] I. Euclidean distance;

[0275] Specifically, the Euclidean distance between consumption feature vectors can be calculated as follows:

[0276]

[0277] Where A represents the consumption feature vector corresponding to the target user account. B represents the consumption feature vector corresponding to "Account B". d(A,B) represents the Euclidean distance between the consumption feature vector corresponding to the target user account and the consumption feature vector corresponding to "Account B". x1 represents the first eigenvalue (e.g., total consumption amount) of the consumption feature vector corresponding to the target user account, and y1 represents the first eigenvalue of the consumption feature vector corresponding to "Account B". n y represents the nth feature value of the consumption feature vector corresponding to the target user account. n This represents the nth feature value of the consumption feature vector corresponding to "Account B".

[0278] Based on formula (2), the Euclidean distance between the target user account and each user account in the candidate user account set can be calculated. Then, the user accounts in the candidate user account set are sorted in ascending order of Euclidean distance, and the top M user accounts are selected as recommendations.

[0279] Optionally, in practical applications, a distance threshold can be further set (e.g., 0.3), meaning that only user accounts whose Euclidean distance is less than or equal to the distance threshold are sorted. If the Euclidean distance of each user account in the candidate user account set is greater than the distance threshold, then no account recommendation is performed.

[0280] II. Cosine similarity;

[0281] Specifically, the cosine similarity between consumption feature vectors can be calculated as follows:

[0282]

[0283] Where A represents the consumption feature vector corresponding to the target user account. B represents the consumption feature vector corresponding to "Account B". D(A,B) represents the cosine similarity between the consumption feature vector corresponding to the target user account and the consumption feature vector corresponding to "Account B". A·B represents the dot product of A and B. ∥A∥ represents the magnitude of A, and ∥B∥ represents the magnitude of B.

[0284] Based on formula (3), the cosine similarity between the target user account and each user account in the candidate user account set can be calculated. Then, the user accounts in the candidate user account set are sorted in descending order of cosine similarity, and the top M user accounts are selected as recommendations.

[0285] Optionally, in practical applications, a similarity threshold can be further set (e.g., 0.8), meaning that only user accounts with a cosine similarity greater than or equal to the similarity threshold are sorted. If the similarity threshold corresponding to each user account in the candidate user account set is less than the similarity threshold, then no account recommendation is made.

[0286] Furthermore, this application provides a method for determining the user account to be pushed to based on similarity. By using the consumption feature vectors between different user accounts to calculate similarity, the accuracy and efficiency of matching can be improved, and false matches can be reduced, thereby enhancing overall accuracy and user experience.

[0287] Optionally, in the above Figure 3 Based on one or more corresponding embodiments, in another optional embodiment provided by this application, determining M user accounts from the candidate user account set may specifically include:

[0288] Based on the consumption feature vector corresponding to the target user account and the consumption feature vector corresponding to each user account in the candidate user account set, M user accounts are obtained through the recommendation model.

[0289] After pushing social information to at least two devices, it may also include:

[0290] Receive recommendation ratings sent by the target terminal;

[0291] The model parameters of the recommendation model are updated based on the recommendation score.

[0292] In one or more embodiments, a method for determining user accounts to be pushed to based on a recommendation model is introduced. As can be seen from the foregoing embodiments, after obtaining a set of candidate user accounts, matching can be performed not only through similarity calculation but also through AI models. The following will introduce a matching method based on an AI model (i.e., a recommendation model).

[0293] Specifically, assume the candidate user account set includes "Account B". Based on this, the consumption feature vectors corresponding to the target user account and "Account B" are used as input to the recommendation model. The recommendation model outputs a similarity score, which can be set between 0 and 1, with a higher score indicating higher similarity. Therefore, the user accounts in the candidate user account set are sorted in descending order of similarity score, and the top M user accounts are selected for recommendation. Optionally, a score threshold can be further set (e.g., 0.8), meaning only user accounts with similarity scores greater than or equal to the threshold are sorted. If the similarity score of each user account in the candidate user account set is less than the similarity threshold, no account recommendation is made.

[0294] It should be noted that the recommendation model can use deep structured semantic models (DSSM), convolutional neural networks (CNN)-DSSM, long short-term memory networks (LSTM)-DSSM, or vector space models, etc., without any limitation here.

[0295] Furthermore, after pushing social information to at least two terminals, the server can also receive recommendation ratings from the terminals. For example, taking the recommendation ratings provided by the target user through the target terminal, assume that a higher recommendation rating indicates a higher user satisfaction with the recommended social information. Therefore, based on the recommendation ratings, the server trains the recommendation model using reinforcement learning methods (e.g., Q-learning, Deep Q-Network).

[0296] For easier understanding, please refer to Figure 13 , Figure 13 This is a schematic diagram of a recommendation model trained based on reinforcement learning in an embodiment of this application, as shown in the figure. t Let a represent the consumption feature vector of each user account in the database at time t. tThis indicates that the similarity score output by the recommendation model represents the M user accounts selected by the target user. After the server recommends social information to the target user and the users corresponding to the M user accounts, it obtains the recommendation scores from these users, i.e., r. t Therefore, the Q value can be updated in the following way:

[0297]

[0298] Among them, Q(S) t ,a t Q(S) represents the Q value at time t. t+1 ,a t+1 Let α represent the Q-value at time t+1. Let α represent the learning rate. Let γ be the discount factor.

[0299] The initial Q-value is a small random value or 0. By iteratively updating the Q-value, a converged Q-value table can be obtained, allowing the selection of the action with the maximum Q-value, thereby achieving the learning of the optimal policy.

[0300] Furthermore, this application provides a method for determining user accounts to be recommended based on a recommendation model. Through this method, reinforcement learning can automatically learn matching strategies, thereby providing users with more accurate real-time matching recommendations and social location recommendations. In this process, by continuously collecting user feedback, the reinforcement learning algorithm can self-adjust to improve matching accuracy and user satisfaction.

[0301] Optionally, in the above Figure 3 Based on one or more corresponding embodiments, in another optional embodiment provided by this application, N recommended location information are determined according to the device location information, which may specifically include:

[0302] Based on the device location information and the most recent consumption location corresponding to each of the M user accounts, a set of candidate locations is determined, wherein the set of candidate locations includes at least one candidate location;

[0303] Obtain the consumption rating corresponding to each candidate location in the candidate location set;

[0304] Based on the consumption rating corresponding to each candidate location, N recommended location information are determined.

[0305] In one or more embodiments, a method for determining recommended location information is described. As can be seen from the foregoing embodiments, based on device location information and the most recent consumption location corresponding to each of the M user accounts, a subset of locations can be selected as a candidate location set. Then, the application programming interface (API) of a review app is further called to obtain the consumption rating of each location, which serves as the basis for selecting recommended locations.

[0306] Understandably, consumer ratings can come from user reviews, data analysis results from review apps, or the system's internal rating mechanism. Consumer ratings reflect user satisfaction with a venue, their experience, and their ability to recommend it, helping the system to identify more popular locations.

[0307] Specifically, let's take the user account "Account B" as the one pushed to the target user as an example. For easier understanding, please refer to [link to relevant documentation]. Figure 14 , Figure 14 This is a schematic diagram illustrating the recommendation of social locations based on consumption ratings in an embodiment of this application. As shown in the figure, assuming that based on backend consumption records, the device location information corresponding to the target user account is "coffee shop," and the most recent consumption location corresponding to "account B" is "hot pot restaurant," a nearby area (e.g., an area with a radius of 500 meters) can be constructed using the center location between "coffee shop" and "hot pot restaurant" as the center. Therefore, a set of candidate locations is first obtained from this nearby area. For example, a candidate location set including "cinema" and "arcade" is used. Next, the API of the review app is called to obtain the consumption rating of "cinema" as 4.2 and the consumption rating of "arcade" as 3.8. Based on this, taking N=1 as an example, the "cinema," with the higher consumption rating, is prioritized as the recommended location, and the recommended location information corresponding to "cinema" is fed back.

[0308] It should be noted that in practical applications, recommended locations can also be filtered based on user interests. For example, if the target user likes food, the restaurant's consumption rating will be multiplied by a weight greater than 1 (e.g., 1.2). Alternatively, if both the target user and the user being recommended to like food, the restaurant's consumption rating will be multiplied by another weight greater than 1 (e.g., 1.5). This will be used as the basis for ranking candidate locations.

[0309] Secondly, this application provides a method for determining recommended location information. By combining device location information with the consumption locations of potential social partners to determine candidate locations, this method can better align with the user's actual activity range, increasing the likelihood of offline interaction. Furthermore, filtering recommended locations based on consumption ratings can provide users with high-quality social venues, improving the accuracy and reliability of recommended locations, thereby enhancing the user's social experience.

[0310] Optionally, in the above Figure 3 In another optional embodiment provided in this application, based on one or more corresponding embodiments, N equals 1;

[0311] After pushing social information to at least two devices, it may also include:

[0312] If at least two terminals send social consent requests respectively, then send a social success message to at least two terminals respectively.

[0313] If at least one of the at least two terminals sends a social rejection request, then a social failure message is sent to each of the at least two terminals.

[0314] In one or more embodiments, a method for providing social feedback based on a single recommended location is described. As described in the foregoing embodiments, the server can push a social location to at least two successfully matched users. That is, at least two successfully matched users can view the social information through their respective terminals. The following description will take the perspective of the target user (e.g., user A) as an example.

[0315] Specifically, for ease of understanding, please refer to Figure 15 , Figure 15 This is a schematic diagram illustrating a single recommended location information in an embodiment of this application, such as... Figure 15 As shown in Figure (A), 1501 indicates the social information pushed to the target terminal, which includes the user account corresponding to the recommended "User B", related information, recommended location information, social time, etc. 1502 indicates the consent to meet control. 1503 indicates the refuse to meet control.

[0316] For example, if the target user clicks the consent control indicated by 1502, a social consent request is triggered. Assume the server also receives the social consent request triggered by user B. Therefore, the target terminal displays as follows: Figure 15 The interface shown in Figure (B) is labeled 1504, indicating a successful social interaction message. Accordingly, the terminal used by User B will also display a successful social interaction message (i.e., "User A has confirmed that they will meet you at XXX Park before 16:00:00 on October 20, 2024").

[0317] For example, if the target user clicks the "Refuse Meeting" control indicated by 1503, a social rejection request is triggered. Consequently, the target terminal displays as follows: Figure 15 The interface shown in Figure (C) is marked with 1505, indicating a failed social interaction message. Correspondingly, the terminal used by User B will also display a failed social interaction message (i.e., "User A has refused to meet").

[0318] Furthermore, the target user can click the feedback control indicated by 1506, thereby displaying the following on the target terminal: Figure 15 The interface shown in Figure (D) is as follows. 1507 indicates the feedback option, meaning the target user can also report the reason for refusing an offline meeting, thus optimizing the recommendation results.

[0319] Secondly, this application provides a method for social feedback based on a single recommended location. Through this method, users can choose whether to arrange an offline meeting with recommended users based on the pushed social information. Thus, through real-time matching recommendations, users can easily find nearby individuals with similar consumption characteristics, further promoting offline social interaction. Simultaneously, recommending suitable social locations improves social convenience, helps expand users' social circles, and increases offline social activity.

[0320] Optionally, in the above Figure 3 Based on one or more corresponding embodiments, in another optional embodiment provided by the present application, N is greater than 1;

[0321] After pushing social information to at least two devices, it may also include:

[0322] Receive a location selection request sent by each of at least two terminals, wherein the location selection request carries any one of N recommended location information;

[0323] If the recommended location information carried in each location selection request is the same, then a social success message is sent to at least two terminals respectively;

[0324] If the recommended location information carried in each location selection request is not exactly the same, a social failure message is sent to at least two terminals respectively.

[0325] In one or more embodiments, a method for providing social feedback based on multiple recommended location information is described. As can be seen from the foregoing embodiments, the server can push at least two social locations to at least two successfully matched users. That is, at least two successfully matched users can view the social information through their respective terminals. The following description will take the perspective of the target user (e.g., user A) as an example.

[0326] Specifically, for ease of understanding, please refer to Figure 16 , Figure 16 This is a schematic diagram illustrating multiple recommended location information in an embodiment of this application, such as... Figure 16 As shown in Figure (A), 1601 indicates the social information pushed to the target terminal, which includes the user account corresponding to the recommended "User B", related information, multiple recommended location information, social time, etc. 1602 indicates the recommended location information selected by the target user (i.e., XXX Park). 1603 indicates the consent to meet control. 1604 indicates the refuse to meet control.

[0327] For example, the target user selects the recommended location information indicated by 1602 and clicks the agreement / meeting control indicated by 1603, thus triggering a location selection request. Assume the server also receives a location selection request triggered by user B, and this request also carries the recommended location information "XXX Park". Therefore, the target terminal displays as follows: Figure 16 The interface shown in Figure (B) is marked with 1605, indicating a successful social interaction message. Accordingly, the terminal used by User B will also display a successful social interaction message (i.e., "User A has confirmed that they will meet you at XXX Park before 16:00:00 on October 20, 2024").

[0328] For example, if the target user clicks the "Refuse Meeting" control indicated by 1604, a social rejection request is triggered. Consequently, the target terminal displays as follows: Figure 16 The interface shown in Figure (C) is marked with 1606, indicating a failed social interaction message. Correspondingly, the terminal used by User B will also display a failed social interaction message (i.e., "User A has refused to meet").

[0329] For example, the target user selects the recommended location information indicated by 1602 and clicks the agreement / meeting control indicated by 1603, thus triggering a location selection request. Assume the server receives the location selection request triggered by user B, and this request carries the recommended location information "XXX Restaurant". Therefore, the target terminal displays as follows: Figure 16 The interface shown in Figure (D) is marked with 1607, indicating a failed social interaction message. Accordingly, the terminal used by User B will also display a failed social interaction message (i.e., "The meeting location selected by you and User A is different; this meeting has failed").

[0330] Secondly, this application embodiment provides a method for social feedback based on multiple recommended location information. Through this method, users can choose one of the multiple recommended locations as a meeting place; the appointment is only considered successful if all users choose the same location. This not only increases the flexibility of location selection but also better tests the tacit understanding between users, thereby enhancing the fun of the activity.

[0331] Optionally, in the above Figure 3 Based on one or more corresponding embodiments, in another optional embodiment provided by this application, the target terminal is used to display the received social information;

[0332] The target terminal is also used to provide a session control, wherein, in response to a selection operation on the session control, the target terminal displays the session interface corresponding to the target group, the target group including the target user account and M user accounts;

[0333] The target terminal is also used to provide a reservation control; wherein, in response to a selection operation on the reservation control, the target terminal displays a reservation interface corresponding to the recommended location information.

[0334] In one or more embodiments, two methods for supporting further interactions between users are described. As can be seen from the foregoing embodiments, upon receiving social information, users can further choose to communicate online with recommended individuals. If the recommended social location is a restaurant, cinema, etc., users can also make reservations or purchase tickets in advance to facilitate offline activities. The methods for implementing conversations and reservations will be described below with reference to the illustrations.

[0335] 1. Provide conversation functionality;

[0336] Specifically, for ease of understanding, please refer to Figure 17 , Figure 17 This is a schematic diagram illustrating the implementation of a session in an embodiment of this application. Figure 17 The diagram in Figure (A) shows the social information pushed by the server to the target terminal, which includes information related to "User B". 1701 is used to indicate a conversation control. When the target user clicks the conversation control indicated by 1701, a selection operation for the conversation control is triggered, and the target terminal displays the following... Figure 17 The diagram in Figure (B) shows the session interface corresponding to the target group. The target group includes the target user account and M user accounts.

[0337] It should be noted that when M equals 1, the target group includes only two users. When M is greater than 1, the target group includes multiple users. Figure 17 Figure (B) in the example uses a target group that includes only two users (i.e., user A and user B), but should not be construed as limiting this application.

[0338] 2. Provide reservation function;

[0339] Specifically, for ease of understanding, please refer to Figure 18 , Figure 18 This is a schematic diagram illustrating the implementation of a reservation system in an embodiment of this application. Figure 18The diagram in Figure (A) shows the social information pushed by the server to the target terminal. This social information includes recommended location information (i.e., XXX restaurant). 1801 is used to indicate a reservation control. When the target user clicks the reservation control indicated by 1801, a selection operation for the reservation control is triggered, and the target terminal displays the following... Figure 18 The reservation interface corresponding to the recommended location information shown in Figure (B) is shown in the middle. Through this reservation interface, you can select reservation information for "XXX Restaurant". The reservation information includes the number of people (e.g., 2 people), date (e.g., October 20, 2024), time (e.g., 12:00) and dining location (e.g., the main hall).

[0340] Secondly, this application provides two methods to support other interactions between users. Through these methods, in addition to pushing social information, users can also engage in temporary conversations, thereby facilitating communication and further increasing the success rate of social activities. Simultaneously, reservations can be made for certain locations (e.g., restaurants), which not only makes it easier for users to meet but also helps create a better meeting environment.

[0341] Based on the above embodiments, the overall process of the social information push method is described below. Please refer to... Figure 19 , Figure 19 The figure shows an overall flowchart of the social information push method in this application embodiment. Specifically:

[0342] In step 1901, the user registers and logs in. That is, the user can register and log in to the product platform using a mobile phone number or other identity authentication methods. The user must authorize the product platform to obtain consumption data and geolocation information, and authorize the product platform to obtain information such as the user's avatar, nickname, interest description, and personal signature.

[0343] In step 1902, payment interaction devices are integrated. That is, payment interaction devices are deployed in partner offline locations (e.g., shopping malls, restaurants, etc.). Users in these locations use these payment interaction devices for identity verification and upload their real-time location information to the product platform. Based on this, taking the palm-swipe payment device as an example, the palm-swipe payment device uses palm biometric technology to obtain the user's identity information and upload real-time location information and consumption information to the product platform.

[0344] In step 1903, a consumption feature vector is constructed. That is, the product platform collects user consumption data and device location information, first cleaning, integrating, and standardizing the data to provide a data foundation for the construction of consumption profiles and matching algorithms. Based on this, combined with the identity information obtained from the payment interaction device, a user consumption profile is constructed. This consumption profile includes, but is not limited to, consumption habits, consumption preferences, and spending power. Through the user's consumption profile, their needs and habits can be understood more accurately, thereby achieving precise matching.

[0345] In step 1904, real-time matching and recommendation are performed. That is, the product platform performs real-time matching based on the user's consumption profile and device location information, recommending other users with similar consumption habits and preferences. Simultaneously, based on the geographical location information of both parties, nearby social locations (e.g., cafes, restaurants, etc.) are recommended, enabling matching between geographically proximate users and improving the real-time nature and convenience of the matching process.

[0346] In step 1905, users engage in offline social interaction. That is, successfully matched users can go to recommended social locations to meet and interact offline, reducing social costs and psychological barriers.

[0347] The social messaging device in this application is described in detail below. Please refer to [link / reference]. Figure 20 , Figure 20 This is a schematic diagram of one embodiment of the social information push device in this application. The social information push device 200 includes:

[0348] The acquisition module 2001 is used to obtain the consumption feature vector corresponding to the target user account in response to the payment transaction request, wherein the payment transaction request carries the target user account and device location information;

[0349] The determination module 2002 is used to determine M user accounts based on the device location information and the consumption feature vector corresponding to the target user account, where M is an integer greater than or equal to 1;

[0350] The determination module 2002 is also used to determine N recommended location information based on the device location information, where N is an integer greater than or equal to 1;

[0351] The push module 2003 is used to push social information to at least two terminals, wherein the at least two terminals include a target terminal and at least one terminal, the target terminal is logged into a target user account, and each of the at least one terminal is logged into one of M user accounts, and the social information includes N recommended location information.

[0352] This application provides a social information push device. Using this device, greater emphasis can be placed on users' actual lifestyles and interests, thereby improving the quality of social matching. Simultaneously, selecting offline meeting locations based on device location information helps increase the likelihood of offline social interaction.

[0353] Optionally, in the above Figure 20 Based on the corresponding embodiments, in another embodiment of the social information push device 200 provided in this application,

[0354] The acquisition module 2001 is specifically used to respond to a payment transaction request sent through the cash register device, wherein the device location information carried in the payment transaction request is the current location information of the payment interaction device;

[0355] or,

[0356] In response to a payment transaction request sent by the target terminal, wherein the device location information carried in the payment transaction request is the current location information of the target terminal.

[0357] This application provides a social information push device. Using this device, not only can biometric payment methods report device location information, but also QR code payment methods can report device location information. This improves the flexibility and feasibility of information reporting, enabling the backend to obtain more accurate device location information.

[0358] Optionally, in the above Figure 20 Based on the corresponding embodiments, in another embodiment of the social information push device 200 provided in this application,

[0359] The acquisition module 2001 is specifically used to query the consumption feature vector corresponding to the target user account from the database;

[0360] The database is used to store the consumption feature vector corresponding to at least one user account. The consumption feature vector is constructed based on the historical consumption data corresponding to the user account.

[0361] This application provides a social information push device. Using this device, consumption feature vectors corresponding to user accounts are pre-constructed offline, facilitating rapid user matching and social location matching, thereby improving the efficiency of information recommendation.

[0362] Optionally, in the above Figure 20 Based on the corresponding embodiments, in another embodiment of the social information push device 200 provided in this application, the social information push device 200 further includes a processing module 2004;

[0363] The acquisition module 2001 is also used to acquire the order data set corresponding to the target user account before querying the consumption feature vector corresponding to the target user account from the database. The order data set includes order data from each historical period.

[0364] Processing module 2004 is used to perform statistics on the order data set corresponding to the target user account to obtain the historical consumption data corresponding to the target user account;

[0365] The processing module 2004 is also used to standardize the historical consumption data corresponding to the target user account to obtain the consumption feature vector corresponding to the target user account.

[0366] This application provides a social information push device. Using this device, a consumption feature vector is constructed using historical order data corresponding to a user account, which can fully reflect the user's consumption habits and improve the accuracy of social matching. Furthermore, by standardizing the data, the influence of data units and numerical ranges can be eliminated, facilitating subsequent data analysis and mining.

[0367] Optionally, in the above Figure 20 Based on the corresponding embodiments, in another embodiment of the social information push device 200 provided in this application, the payment transaction request also carries target order data;

[0368] The acquisition module 2001 is specifically used to acquire the order data set corresponding to the target user account, wherein the order data set includes order data from various historical time periods;

[0369] By statistically analyzing the order data set corresponding to the target user account and the target order data, the real-time consumption data corresponding to the target user account can be obtained.

[0370] The real-time consumption data corresponding to the target user account is standardized to obtain the consumption feature vector corresponding to the target user account.

[0371] This application provides a social information push device. Using this device, the consumption feature vector can not only reflect a user's long-term consumption habits but also their latest consumption characteristics, thereby improving the accuracy and reliability of matching.

[0372] Optionally, in the above Figure 20 Based on the corresponding embodiments, in another embodiment of the social information push device 200 provided in this application, the payment transaction request also carries the target consumption time;

[0373] The determination module 2002 is specifically used to determine the associated area range based on the device location information;

[0374] Determine the relevant time range based on the target consumption time;

[0375] From the historical consumption data corresponding to each user account, obtain the location and time of the most recent consumption for each user account;

[0376] The candidate user account set includes all user accounts whose most recent purchase location is within the relevant area and whose most recent purchase time is within the relevant time range.

[0377] Select M user accounts from the candidate user account set.

[0378] This application provides a social information push device. By combining the associated regional range and associated time range for matching, the matching range can be narrowed. This not only improves social matching efficiency but also ensures that recommended social contacts are geographically close, facilitating offline meetings.

[0379] Optionally, in the above Figure 20 Based on the corresponding embodiments, in another embodiment of the social information push device 200 provided in this application, the payment transaction request also carries the target consumption time;

[0380] The determination module 2002 is specifically used to obtain the cluster label corresponding to the target user account, wherein the cluster label is used to indicate the clustering attribute information corresponding to the user account;

[0381] Determine the associated area range based on the device location information;

[0382] Determine the relevant time range based on the target consumption time;

[0383] From the historical consumption data corresponding to each user account, obtain the location and time of the most recent consumption for each user account;

[0384] The candidate user account set includes all user accounts whose most recent consumption location is within the associated region, whose most recent consumption time is within the associated time, and whose cluster label is the same as the cluster label corresponding to the target user account.

[0385] Select M user accounts from the candidate user account set.

[0386] This application provides a social information push device. By combining the associated regional range and associated time range for matching, the matching scope can be narrowed. This not only improves the efficiency of social matching but also ensures that recommended social partners are geographically close, facilitating offline meetings. Furthermore, combining cluster tags can further narrow the matching scope and match users with similar consumption patterns, thereby improving the accuracy of social matching.

[0387] Optionally, in the above Figure 20 Based on the corresponding embodiments, in another embodiment of the social information push device 200 provided in this application,

[0388] Module 2002 is specifically used to obtain the historical consumption data corresponding to each user account;

[0389] The historical consumption data corresponding to each user account is standardized to obtain the consumption feature vector corresponding to each user account;

[0390] Based on the consumption feature vector corresponding to each user account, cluster label corresponding to each user account is obtained through cluster analysis;

[0391] Based on the cluster label corresponding to each user account, determine the cluster label corresponding to the target user account.

[0392] This application provides a social information push device. Compared to manual labeling, using clustering algorithms for labeling with this device not only improves labeling efficiency but also saves on manual processing costs.

[0393] Optionally, in the above Figure 20 Based on the corresponding embodiments, in another embodiment of the social information push device 200 provided in this application,

[0394] The determination module 2002 is specifically used to calculate the similarity between the target user account and each user account in the candidate user account set based on the consumption feature vector corresponding to the target user account and the consumption feature vector corresponding to each user account in the candidate user account set.

[0395] Based on the similarity between the target user account and each user account in the candidate user account set, determine M user accounts.

[0396] This application provides a social information push device. Using this device can improve the accuracy and efficiency of matching, reduce false matches, and thus improve overall accuracy and user experience.

[0397] Optionally, in the above Figure 20Based on the corresponding embodiments, in another embodiment of the social information push device 200 provided in this application, the social information push device 200 further includes a receiving module 2005;

[0398] The determination module 2002 is specifically used to obtain M user accounts through a recommendation model based on the consumption feature vector corresponding to the target user account and the consumption feature vector corresponding to each user account in the candidate user account set.

[0399] The receiving module 2005 is used to receive the recommendation rating sent by the target terminal after pushing social information to at least two terminals;

[0400] The processing module 2004 is also used to update the model parameters of the recommendation model based on the recommendation score.

[0401] This application provides a social information push device. Using this device, the reinforcement learning algorithm can self-adjust by continuously collecting user feedback during the process, thereby improving matching accuracy and user satisfaction.

[0402] Optionally, in the above Figure 20 Based on the corresponding embodiments, in another embodiment of the social information push device 200 provided in this application,

[0403] The determination module 2002 is specifically used to determine a set of candidate locations based on the device location information and the most recent consumption location corresponding to each of the M user accounts, wherein the set of candidate locations includes at least one candidate location;

[0404] Obtain the consumption rating corresponding to each candidate location in the candidate location set;

[0405] Based on the consumption rating corresponding to each candidate location, N recommended location information are determined.

[0406] This application provides a social information push device. By combining device location information with the consumption locations of potential social partners to determine candidate locations, the device can better align with the user's actual activity range, increasing the likelihood of offline interaction. Furthermore, filtering recommended locations based on consumption ratings can provide users with high-quality social venues, improving the accuracy and reliability of recommended locations and thus enhancing the user's social experience.

[0407] Optionally, in the above Figure 20 Based on the corresponding embodiments, in another embodiment of the social information push device 200 provided in this application, N equals 1;

[0408] The social messaging device 200 also includes a sending module 2006;

[0409] The sending module 2006 is used to send a social success message to at least two terminals after pushing social information to at least two terminals, if at least two terminals send social consent requests respectively.

[0410] The sending module 2006 is also configured to send a social failure message to each of the at least two terminals if at least one of the at least two terminals sends a social rejection request.

[0411] This application provides a social information push device. Using this device, through real-time matching and recommendations, users can easily find nearby individuals with similar consumption characteristics, further promoting offline social interaction. Simultaneously, by recommending suitable social locations, it improves social convenience, helps expand users' social circles, and increases offline social activity.

[0412] Optionally, in the above Figure 20 Based on the corresponding embodiments, in another embodiment of the social information push device 200 provided in this application, N is greater than 1;

[0413] The receiving module 2005 is also configured to receive a location selection request sent by each of the at least two terminals after pushing social information to at least two terminals, wherein the location selection request carries any one of N recommended location information;

[0414] The sending module 2006 is also used to send a social success message to at least two terminals if the recommended location information carried in each location selection request is the same.

[0415] The sending module 2006 is also used to send a social failure message to at least two terminals if the recommended location information carried in each location selection request is not completely the same.

[0416] This application provides a social messaging device. Using this device, an appointment is considered successful only if all users choose the same location to meet. This not only increases the flexibility of location selection but also better tests the tacit understanding between users, thereby enhancing the fun of the activity.

[0417] Optionally, in the above Figure 20 Based on the corresponding embodiments, in another embodiment of the social information push device 200 provided in this application, the target terminal is used to display the received social information;

[0418] The target terminal is also used to provide a session control, wherein, in response to a selection operation on the session control, the target terminal displays the session interface corresponding to the target group, the target group including the target user account and M user accounts;

[0419] The target terminal is also used to provide a reservation control; wherein, in response to a selection operation on the reservation control, the target terminal displays a reservation interface corresponding to the recommended location information.

[0420] This application provides a social information push device. Using this device, users can easily communicate and exchange ideas based on temporary sessions, further increasing the success rate of social activities. Pre-booking not only facilitates meeting users but also helps create a better meeting environment.

[0421] Figure 21 This is a schematic diagram of a computer device structure provided in an embodiment of this application. The computer device 2100 can vary significantly due to different configurations or performance. It may include one or more central processing units (CPUs) 2122 (e.g., one or more processors) and a memory 2132, and one or more storage media 2130 (e.g., one or more mass storage devices) for storing application programs 2142 or data 2144. The memory 2132 and storage media 2130 can be temporary or persistent storage. The program stored in the storage media 2130 may include one or more modules (not shown in the diagram), each module including a series of instruction operations on the computer device. Furthermore, the CPU 2122 may be configured to communicate with the storage media 2130 and execute the series of instruction operations in the storage media 2130 on the computer device 2100.

[0422] Computer device 2100 may also include one or more power supplies 2126, one or more wired or wireless network interfaces 2150, one or more input / output interfaces 2158, and / or one or more operating systems 2141, such as Windows Server. TM Mac OS X TM Unix TM Linux TM FreeBSD TM etc.

[0423] The steps performed by the computer device in the above embodiments can be based on this Figure 21 The computer device structure shown.

[0424] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the methods described in the foregoing embodiments.

[0425] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the methods described in the foregoing embodiments.

[0426] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0427] In this application embodiment, the terms "module" or "unit" refer to a computer program or part of a computer program that has a predetermined function and works with other related parts to achieve a predetermined goal, and can be implemented wholly or partially using software, hardware (such as processing circuitry or memory), or a combination thereof. Similarly, a processor (or multiple processors or memory) can be used to implement one or more modules or units. Furthermore, each module or unit can be part of an overall module or unit that includes the functionality of that module or unit.

[0428] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between apparatuses or units through some interfaces, and may be electrical, mechanical, or other forms.

[0429] The units described as separate components may or may not be physically separate. The 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 the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0430] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0431] If the integrated unit is implemented as 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 this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a server or terminal device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing computer programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0432] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for pushing social information, characterized in that, include: In response to a payment transaction request, a consumption feature vector corresponding to the target user account is obtained, wherein the payment transaction request carries the target user account and device location information; Based on the device location information and the consumption feature vector corresponding to the target user account, M user accounts are determined, where M is an integer greater than or equal to 1; Based on the device location information, N recommended location information are determined, where N is an integer greater than or equal to 1; Social information is pushed to at least two terminals, wherein the at least two terminals include a target terminal and at least one terminal, the target terminal is logged into the target user account, and each of the at least one terminal is logged into one of the M user accounts, and the social information includes the N recommended location information.

2. The push method according to claim 1, characterized in that, The response to the payment transaction request includes: In response to the payment transaction request sent through the POS device, wherein the device location information carried in the payment transaction request is the current location information of the payment interaction device; or, In response to the payment transaction request sent by the target terminal, wherein the device location information carried in the payment transaction request is the current location information of the target terminal.

3. The push method according to claim 1 or 2, characterized in that, The step of obtaining the consumption feature vector corresponding to the target user account includes: Retrieve the consumption feature vector corresponding to the target user account from the database; The database is used to store consumption feature vectors corresponding to at least one user account, and the consumption feature vectors are constructed based on the historical consumption data corresponding to the user account.

4. The push method according to claim 3, characterized in that, Before retrieving the consumption feature vector corresponding to the target user account from the database, the method further includes: Obtain the order data set corresponding to the target user account, wherein the order data set includes order data from various historical time periods; The order data set corresponding to the target user account is statistically analyzed to obtain the historical consumption data corresponding to the target user account; The historical consumption data corresponding to the target user account is standardized to obtain the consumption feature vector corresponding to the target user account.

5. The push method according to claim 1 or 2, characterized in that, The payment transaction request also carries target order data; The step of obtaining the consumption feature vector corresponding to the target user account includes: Obtain the order data set corresponding to the target user account, wherein the order data set includes order data from various historical time periods; By statistically analyzing the order data set corresponding to the target user account and the target order data, the real-time consumption data corresponding to the target user account is obtained. The real-time consumption data corresponding to the target user account is standardized to obtain the consumption feature vector corresponding to the target user account.

6. The push method according to any one of claims 1 to 5, characterized in that, The payment transaction request also carries the target consumption time; The step of determining M user accounts based on the device location information and the consumption feature vector corresponding to the target user account includes: Based on the device location information, the associated area range is determined; Determine the associated time range based on the target consumption time; From the historical consumption data corresponding to each user account, obtain the location and time of the most recent consumption for each user account; The candidate user account set includes all user accounts whose most recent consumption location is within the associated region and whose most recent consumption time is within the associated time range. The M user accounts are determined from the set of candidate user accounts.

7. The push method according to any one of claims 1 to 5, characterized in that, The payment transaction request also carries the target consumption time; The step of determining M user accounts based on the device location information and the consumption feature vector corresponding to the target user account includes: Obtain the cluster label corresponding to the target user account, wherein the cluster label is used to indicate the clustering attribute information corresponding to the user account; Based on the device location information, the associated area range is determined; Determine the associated time range based on the target consumption time; From the historical consumption data corresponding to each user account, obtain the location and time of the most recent consumption for each user account; The candidate user account set includes all user accounts whose most recent consumption location is within the associated region, whose most recent consumption time is within the associated time, and whose cluster label is the same as the cluster label corresponding to the target user account. The M user accounts are determined from the set of candidate user accounts.

8. The push method according to claim 7, characterized in that, The step of obtaining the cluster tag corresponding to the target user account includes: Obtain historical consumption data for each user account; The historical consumption data corresponding to each user account is standardized to obtain the consumption feature vector corresponding to each user account; Based on the consumption feature vector corresponding to each user account, cluster label corresponding to each user account is obtained through cluster analysis; Based on the cluster label corresponding to each user account, determine the cluster label corresponding to the target user account.

9. The push method according to any one of claims 6 to 8, characterized in that, The step of determining the M user accounts from the candidate user account set includes: Based on the consumption feature vector corresponding to the target user account and the consumption feature vector corresponding to each user account in the candidate user account set, calculate the similarity between the target user account and each user account in the candidate user account set; The M user accounts are determined based on the similarity between the target user account and each user account in the candidate user account set.

10. The push method according to any one of claims 6 to 8, characterized in that, The step of determining the M user accounts from the candidate user account set includes: Based on the consumption feature vector corresponding to the target user account and the consumption feature vector corresponding to each user account in the candidate user account set, the M user accounts are obtained through a recommendation model; After pushing social information to at least two terminals, the method further includes: Receive the recommendation rating sent by the target terminal; The model parameters of the recommendation model are updated based on the recommendation score.

11. The push method according to any one of claims 1 to 10, characterized in that, The step of determining N recommended location information based on the device location information includes: Based on the device location information and the most recent consumption location corresponding to each of the M user accounts, a candidate location set is determined, wherein the candidate location set includes at least one candidate location; Obtain the consumption rating corresponding to each candidate location in the candidate location set; Based on the consumption rating corresponding to each candidate location, the information of the N recommended locations is determined.

12. The push method according to any one of claims 1 to 11, characterized in that, The N is equal to 1; After pushing social information to at least two terminals, the method further includes: If the at least two terminals send social consent requests respectively, then send a social success message to the at least two terminals respectively; If at least one of the at least two terminals sends a social rejection request, then a social failure message is sent to each of the at least two terminals.

13. The push method according to any one of claims 1 to 11, characterized in that, The N is greater than 1; After pushing social information to at least two terminals, the method further includes: Receive a location selection request sent by each of the at least two terminals, wherein the location selection request carries any one of the N recommended location information; If the recommended location information carried in each of the location selection requests is the same, then a social success message is sent to each of the at least two terminals respectively; If the recommended location information carried in each location selection request is not exactly the same, a social failure message is sent to each of the at least two terminals respectively.

14. The push method according to any one of claims 1 to 13, characterized in that, The target terminal is used to display the received social information; The target terminal is also used to provide a session control, wherein, in response to a selection operation on the session control, the target terminal displays a session interface corresponding to the target group, the target group including the target user account and the M user accounts; The target terminal is also used to provide a reservation control; wherein, in response to a selection operation on the reservation control, the target terminal displays a reservation interface corresponding to the recommended location information.

15. A social information push device, characterized in that, include: The acquisition module is used to acquire the consumption feature vector corresponding to the target user account in response to a payment transaction request, wherein the payment transaction request carries the target user account and device location information; The determination module is used to determine M user accounts based on the device location information and the consumption feature vector corresponding to the target user account, wherein M is an integer greater than or equal to 1; The determining module is further configured to determine N recommended location information based on the device location information, wherein N is an integer greater than or equal to 1; The push module is used to push social information to at least two terminals, wherein the at least two terminals include a target terminal and at least one terminal, the target terminal is logged into the target user account, and each of the at least one terminal is logged into one of the M user accounts, and the social information includes the N recommended location information.

16. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the push method according to any one of claims 1 to 14.

17. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the push method according to any one of claims 1 to 14.

18. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the push method as described in any one of claims 1 to 14.

Images