Merchant recommendation method, device, equipment, medium and product

Abstract

The application discloses a merchant recommendation method and device, equipment, medium and product, and belongs to the technical field of computer data processing. The method comprises the following steps: acquiring target data corresponding to M users and N merchants, wherein the target data comprises first transaction data between the M users and the N merchants, and first attribute data of the M users and the N merchants; calculating a customer group matching degree and a preference matching degree between each user and each merchant based on the first transaction data and the first attribute data; determining an interaction feature based on the first transaction data, the customer group matching degree and the preference matching degree between each user and each merchant; and inputting the interaction feature into a preset merchant recommendation model to output a recommendation result. According to the embodiment of the application, the accuracy of merchant recommendation can be improved, and the problem that the accuracy of merchant recommendation is not high can be solved.

Description

Technical Field

[0001] This application belongs to the field of computer data processing technology, and in particular relates to a merchant recommendation method, apparatus, equipment, medium and product. Background Technology

[0002] With the development and application of internet technology, the amount of information available to users has increased significantly. In order to promote user conversion, the platform can predict each user's personalized merchant preferences within a certain range of users and merchants, and recommend different merchants to different users through differentiated display on the front-end page, thereby improving the user experience.

[0003] In related technologies, when recommending merchants to users, the main basis is the distance between the user and the merchant, as well as a list of popular merchants. Although this solution can achieve the function of merchant recommendation to a certain extent, the recommendation method is relatively crude, it does not fully explore the user's personalized preferences, and the accuracy of merchant recommendation is not high. Summary of the Invention

[0004] This application provides a merchant recommendation method, apparatus, equipment, medium, and product that can improve the problem of low accuracy in merchant recommendations.

[0005] In a first aspect, embodiments of this application provide a merchant recommendation method, the method comprising:

[0006] Obtain target data corresponding to M users and N merchants. The target data includes the first transaction data between the M users and N merchants, as well as the first attribute data of the M users and N merchants. The first attribute data includes the user's demographic information attributes and consumption preference attributes, as well as the merchant's basic attributes and customer demographic information attributes.

[0007] Based on the first transaction data and the first attribute data, calculate the customer group matching degree and preference matching degree between each user and each merchant;

[0008] Based on the first transaction data, as well as the customer group matching degree and preference matching degree between each user and each merchant, the interaction characteristics are determined;

[0009] Input interaction features into the preset merchant recommendation model and output recommendation results.

[0010] Secondly, embodiments of this application provide a merchant recommendation device, the device comprising:

[0011] The acquisition module is used to acquire target data corresponding to M users and N merchants. The target data includes the first transaction data between the M users and N merchants, as well as the first attribute data of the M users and N merchants. The first attribute data includes the user's demographic information attributes and consumption preference attributes, as well as the merchant's basic attributes and customer demographic information attributes.

[0012] The calculation module is used to calculate the customer group matching degree and preference matching degree between each user and each merchant based on the first transaction data and the first attribute data;

[0013] The determination module is used to determine interaction characteristics based on the first transaction data, as well as the customer group matching degree and preference matching degree between each user and each merchant;

[0014] The recommendation module is used to input interaction features into a preset merchant recommendation model and output recommendation results.

[0015] Thirdly, embodiments of this application provide an electronic device, including: a processor and a memory storing computer program instructions; the processor executes the computer program instructions to implement the steps of the merchant recommendation method shown in the first aspect.

[0016] Fourthly, embodiments of this application provide a computer-readable storage medium on which a program or instructions are stored, which, when executed by a processor, implement the steps of the merchant recommendation method as described in the first aspect.

[0017] Fifthly, embodiments of this application provide a computer program product stored in a non-volatile storage medium, which, when executed by at least one processor, implements the steps of the merchant recommendation method as described in the first aspect.

[0018] In a sixth aspect, embodiments of this application provide a chip including a processor and a communication interface, the communication interface being coupled to the processor, the processor being used to run programs or instructions to implement the steps of the merchant recommendation method as described in the first aspect.

[0019] This application provides a merchant recommendation method, apparatus, device, medium, and product. In a scenario where merchant recommendations are made for M users, first transaction data and first attribute data between the M users and all merchants to be recommended (i.e., N merchants) can be obtained. Based on this, customer group matching degree and preference matching degree between each user and each merchant are calculated using the first transaction data and the first attribute data. Interaction features are then determined based on the first transaction data and the customer group matching degree and preference matching degree between each user and each merchant. Since the first transaction data reflects the historical transaction behavior between users and merchants, the customer group matching degree reflects the degree of matching between the user and the merchant's customer group, and the preference matching degree reflects the degree of interest each user has in each merchant, the interaction features determined based on the first transaction data, customer group matching degree, and preference matching degree can accurately reflect the user's preference for N merchants. Using these interaction features to recommend merchants to M users effectively improves the accuracy of merchant recommendations. Attached Figure Description

[0020] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 A flowchart of an embodiment of the merchant recommendation method provided in the first aspect of this application;

[0022] Figure 2 A flowchart of another embodiment of the merchant recommendation method provided in the first aspect of this application;

[0023] Figure 3 A flowchart of yet another embodiment of the merchant recommendation method provided in the first aspect of this application;

[0024] Figure 4 A flowchart of yet another embodiment of the merchant recommendation method provided in the first aspect of this application;

[0025] Figure 5 A schematic diagram of the structure of an embodiment of the merchant recommendation device provided in the second aspect of this application;

[0026] Figure 6 A schematic diagram of the structure of an embodiment of the electronic device provided in the third aspect of this application. Detailed Implementation

[0027] The features and exemplary embodiments of various aspects of this application will be described in detail below. To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain this application and not to limit it. For those skilled in the art, this application can be implemented without some of these specific details. The following description of the embodiments is merely to provide a better understanding of this application by illustrating examples.

[0028] With the development and application of internet technology, the amount of information available to users has increased significantly. To promote user conversion, platforms can predict each user's personalized merchant preferences within a certain range of users and merchants, and recommend different merchants to different users through differentiated display on the front-end page, thereby improving the user experience. In related technologies, merchant recommendations are mainly based on the distance between the user and the merchant, as well as a list of popular merchants. While this solution can achieve the function of merchant recommendation to a certain extent, the recommendation method is relatively crude, lacks sufficient understanding of users' personalized preferences, and the accuracy of merchant recommendations is not high.

[0029] To address the aforementioned problems, this application provides a merchant recommendation method, apparatus, device, medium, and product. In a scenario where merchant recommendations are made for M users, it acquires first attribute data and first transaction data reflecting historical transaction behavior between users and merchants. Based on the first transaction data and the first attribute data, it calculates a customer group matching degree reflecting the degree of matching between users and merchant customer groups, and a preference matching degree reflecting each user's interest in each merchant. Furthermore, the interaction features determined based on the first transaction data, customer group matching degree, and preference matching degree accurately reflect the user's preference for N merchants. Utilizing these interaction features to recommend merchants to M users effectively improves the accuracy of merchant recommendations.

[0030] The merchant recommendation method in this application embodiment can be applied to merchant recommendation scenarios. The following, in conjunction with the accompanying drawings, will provide a detailed description of the merchant recommendation method provided in this application embodiment through specific embodiments.

[0031] The first aspect of this application provides a merchant recommendation method that can be applied to electronic devices, meaning that the merchant recommendation method can be executed by electronic devices. It should be noted that the aforementioned executing entity does not constitute a limitation on this application.

[0032] Figure 1 A flowchart illustrating an embodiment of the merchant recommendation method provided in the first aspect of this application. (See flowchart for example.) Figure 1 As shown, the merchant recommendation method may include steps 110-140.

[0033] Step 110: Obtain target data corresponding to M users and N merchants, wherein the target data includes the first transaction data between the M users and N merchants, and the first attribute data of the M users and N merchants.

[0034] The first attribute data includes users' demographic information attributes and consumption preference attributes, as well as merchants' basic attributes and customer demographic information attributes.

[0035] Step 120: Based on the first transaction data and the first attribute data, calculate the customer group matching degree and preference matching degree between each user and each merchant.

[0036] Step 130: Based on the first transaction data, and the customer group matching degree and preference matching degree between each user and each merchant, determine the interaction characteristics.

[0037] Step 140: Input interaction features into the preset merchant recommendation model and output the recommendation results.

[0038] The merchant recommendation method provided in this application, in a scenario of recommending merchants to M users, can obtain first transaction data and first attribute data between the M users and all merchants to be recommended, i.e., N merchants. Based on this, the user-merchant matching degree and preference matching degree are calculated for each user and each merchant, and interaction features are determined based on the first transaction data and the user-merchant matching degree and preference matching degree. Since the first transaction data reflects the historical transaction behavior between users and merchants, the user-merchant matching degree reflects the degree of matching between the user and the merchant's customer base, and the preference matching degree reflects the user's interest in each merchant, the interaction features determined based on the first transaction data, user-merchant matching degree, and preference matching degree can accurately reflect the user's preference for N merchants. Using these interaction features to recommend merchants to M users effectively improves the accuracy of merchant recommendations.

[0039] The specific implementation of the above steps will be described in detail below with reference to the embodiments.

[0040] Step 110 involves obtaining target data corresponding to M users and N merchants, wherein the target data includes the first transaction data between the M users and N merchants, and the first attribute data of the M users and N merchants.

[0041] Among them, M users can be the recommended objects, and N merchants can be the merchants to be recommended to M users; the first transaction data is the historical transaction behavior data between M users and N merchants, which may include transaction time, transaction amount, number of transactions, etc.

[0042] In one example, M users can represent all users in region 1, and N merchants can represent merchants in region 1 whose merchant category is food and beverage.

[0043] In another example, M users can be all users in region 1, and N merchants can be all merchants in region 2. Region 1 can include region 2.

[0044] In one embodiment, in the first attribute data, the user's demographic information attributes may include at least one of gender, age, income, and spending power; the user's consumption preference attributes may include at least one of merchant category, peak consumption time, and preferred consumption amount; the merchant's customer demographic information attributes may include at least one of the main customer group's gender, main customer group's age, main customer group's income, and main customer group's spending power; and the merchant's basic merchant attributes may include at least one of merchant category, business hours, and average transaction amount.

[0045] It should be noted that the user's demographic information attributes can correspond to the attributes of the merchant's customer demographic information attributes, and the user's consumption preference attributes can correspond to the attributes of the merchant's basic attributes.

[0046] Step 120 involves calculating the customer group matching degree and preference matching degree between each user and each merchant based on the first transaction data and the first attribute data.

[0047] In some embodiments of this application, in order to calculate the customer group matching degree between each user and each merchant, step 120 may specifically include the following steps: dividing the M users into multiple customer groups based on the demographic information attributes of the M users; calculating the total number of transactions between each merchant and each customer group based on the first transaction data; and calculating the customer group matching degree between each user and each merchant based on the customer group to which each user belongs and the total number of transactions between each merchant and each customer group.

[0048] Specifically, electronic devices can segment M users into different customer groups based on their demographic information attributes; and calculate the customer group matching degree between each user and each merchant based on the customer group to which each user belongs and the total number of transactions between each merchant and each customer group, thus obtaining M*N customer group matching degrees.

[0049] For example, M users can be divided into two customer groups: male and female, based on gender; into multiple customer groups of different age groups, based on age; into multiple customer groups of different consumption levels, based on spending power; and into multiple customer groups of different income levels, based on income.

[0050] In one embodiment, step 120 may specifically include: calculating the customer group matching degree between each user and each merchant based on formula (1).

[0051]

[0052] Among them, C score For customer group matching degree, k is the number of dimensions, L i＝a Let N be the total number of transactions for the merchant when dimension i is a, and N be the total number of transactions for the merchant.

[0053] In one example, the user's demographic attributes include two dimensions: age and gender, k = 2, and user A is 20 years old, belonging to the 18-25 year old young customer group. Then L i＝a This refers to the total number of transactions between the merchant and young customers aged 18-25.

[0054] In this embodiment, after dividing M users into multiple customer groups, the total number of transactions between each merchant and each customer group can be calculated to reflect the matching degree between the merchant and different customer groups. Based on this, the customer group to which each user belongs and the total number of transactions between each merchant and each customer group can be accurately determined, effectively utilizing the merchant's customer demographic information attribute data. By combining the correlation between user demographic information attributes and merchant customer demographic information attributes to calculate the customer group matching degree, the accuracy of the calculation is improved. Furthermore, by adding customer group matching degree to the interaction features, the completeness of the interaction features is enhanced.

[0055] In some embodiments of this application, in order to calculate the preference matching degree between each user and each merchant, step 120 may specifically include the following steps: calculating the preference matching degree between each user and each merchant based on the consumption preference attributes of M users and the basic merchant attributes of N merchants.

[0056] Specifically, electronic devices can determine the similarity between each user's consumption preference attributes and each merchant's basic merchant attributes, i.e., the preference matching degree between each user and each merchant.

[0057] In one embodiment, step 120 may specifically include: calculating the first similarity between the user's consumption merchant category and the merchant's merchant category, the second similarity between the user's consumption concentration time and the merchant's business hours, and the third similarity between the user's consumption amount preference and the merchant's average transaction amount; and calculating the similarity between the user's consumption preference attribute and the merchant's basic merchant attribute based on the first similarity, the second similarity, the third similarity and their corresponding weights.

[0058] For example, if a user's merchant category includes the merchant's merchant category, then the first similarity a1 can be 1; otherwise, it is 0. If the user's consumption is concentrated within the merchant's business hours, then the second similarity a2 can be 1; otherwise, it is 0. The smaller the difference between the user's consumption amount preference and the merchant's average transaction amount, the higher the third similarity a3. a1*p1+a2*p2+a3*p3 is the similarity between the user's consumption preference attribute and the merchant's basic merchant attribute, where p1, p2, and p3 are the corresponding weights.

[0059] In another embodiment, the preference matching degree between each user and each merchant can be calculated based on the consumption preference attributes of M users, the basic merchant attributes of N merchants, and their correlation coefficients.

[0060] In this embodiment, more basic attribute information related to merchants is introduced to enhance the richness of merchant recommendation features. Based on users' consumption preference attributes and merchants' basic attributes, the preference matching degree between each user and each merchant can be calculated simply and effectively. Interaction features are constructed based on this preference matching degree, enabling these features to reflect users' personalized preferences and further improving the accuracy of merchant special offers. When using these interaction features to recommend merchants to users, the probability of pushing merchants to users' interests is increased, achieving precise merchant recommendations.

[0061] Step 130 involves determining the interaction characteristics based on the first transaction data and the customer group matching degree and preference matching degree between each user and each merchant.

[0062] In one embodiment, the first transaction data, customer group matching degree, and preference matching degree between each user and each merchant can be used as the interaction features corresponding to that user and that merchant.

[0063] In some embodiments of this application, if merchant recommendations are based on a list of popular merchants, the recommendations tend to concentrate on top merchants with high transaction volumes, while insufficient recommendations for other lower-ranked merchants result in low accuracy. To address this issue of insufficient recommendations for lower-ranked merchants... Figure 2 A flowchart of another embodiment of the merchant recommendation method provided in the first aspect of this application, step 130 may include... Figure 2 Steps 210-240 are shown.

[0064] Step 210: Based on the first transaction data, determine the adjustment parameters for each user for each of the N merchants.

[0065] Step 220: Based on the first attribute data, cluster the M users and N merchants respectively to obtain the clustering results.

[0066] Specifically, electronic devices can cluster M users based on their demographic information attributes and consumption preference attributes to obtain user clusters with similar attributes; and cluster N merchants based on their basic merchant attributes and customer demographic information attributes to obtain merchant clusters with similar attributes.

[0067] For example, M users include user A, user B, and user C, and N merchants include merchant A, merchant B, and merchant C. Merchants A and B are both restaurants, and merchant C is a convenience store. User A and user B are similar in age, gender, and consumption preferences. Therefore, users A and B can be grouped into the same user cluster, and merchants A and B can be grouped into the same merchant cluster.

[0068] Step 230: Adjust the first transaction data according to the adjustment parameters and clustering results to obtain the second transaction data.

[0069] In some embodiments of this application, for users and merchants who have not engaged in any transactions, in order to fill in the missing consumption data between them, the clustering results may include user clusters and merchant clusters. Figure 3 A flowchart of another embodiment of the merchant recommendation method provided in the first aspect of this application, step 230 may include... Figure 3 Steps 310-340 are shown.

[0070] Step 310: If the number of transactions between the first user among M users and the first merchant among N merchants is zero, determine at least one second user belonging to the same user cluster as the first user, and / or at least one second merchant belonging to the same merchant cluster as the first merchant.

[0071] Among them, there is no historical transaction between the first user and the first merchant, the first user and the second user belong to the same user cluster, and the first merchant and the second merchant belong to the same merchant cluster.

[0072] Step 320: Determine the third transaction data in the first transaction data, which is the target transaction information between the first user and the first merchant. The third transaction data includes transaction data between at least one second user and the first merchant, and / or, transaction data between at least one second merchant and the first user.

[0073] Specifically, the third transaction data may be the first average number of transactions between at least one second user and the first merchant, or the second average number of transactions between at least one second merchant and the first user, or the average of the first average number of transactions and the second average number of transactions.

[0074] Referring to the example above, users A and B are grouped into the same user cluster, and merchants A and B are grouped into the same merchant cluster. If the number of transactions between user B and merchant B is zero, then the transaction data between user A and merchant B, and / or the transaction data between user B and merchant A, can be identified as third transaction data, and this third transaction data can be used as the target transaction information between user B and merchant B.

[0075] Step 330: Add the target transaction information to the first transaction data to obtain the fourth transaction data.

[0076] Step 340: Adjust the fourth transaction data according to the adjustment parameters to obtain the second transaction data.

[0077] In this embodiment, based on the first transaction data, adjustment parameters for each user and each of the N merchants can be determined. Based on the first attribute data, clustering can be performed on the M users and N merchants, grouping users or merchants with similar attributes together to obtain clustering results. The first transaction data is then adjusted according to the adjustment parameters and the clustering results to obtain second transaction data. Thus, compared to related technologies that directly recommend merchants based on their sales volume, this application recommends merchants to M users based on the second transaction data between the M users and the merchants to be recommended. This second transaction data is modified based on adjustment parameters that reflect the true historical transaction situation between the M users and the N merchants. Therefore, this second transaction data can more accurately reflect the merchants that the M users are interested in among the N merchants. Using this second transaction data to recommend merchants to the M users can improve the accuracy of merchant recommendations. At the same time, even if there is no historical transaction between a merchant and a user, the consumption information between the merchant and the user can be supplemented based on the clustering results, thereby improving the feature coverage. This achieves the effect of adjusting the data by combining individual behavior and overall behavior, improving the phenomenon that bottom merchants cannot be recommended because they have few transactions with other users or no historical transaction behavior, and increasing the recommendation strength of bottom merchants.

[0078] In some embodiments of this application, each user can correspond to N adjustment parameters. Step 340 adjusts the fourth transaction data according to the adjustment parameters to obtain the second transaction data, which may specifically include the following steps: constructing a first interaction matrix based on the fourth transaction data, the first interaction matrix including M*N elements, each element representing the number of transactions between the user and the merchant; multiplying the M*N adjustment parameters corresponding to the M users by the M*N elements respectively to obtain the second interaction matrix; determining the second transaction data based on the second interaction matrix, the second transaction data including the adjusted number of transactions between the M users and the N merchants.

[0079] Specifically, after adding target transaction information to the first transaction data, a fourth transaction data can be obtained, which includes transaction data between each user and each merchant; the first interaction matrix can be an M-row N-column matrix, and for each element P in the matrix, it is multiplied by the corresponding adjustment parameter to obtain the second interaction matrix.

[0080] Step 240: Based on the first transaction data, second transaction data, customer group matching degree, and preference matching degree between M users and N merchants, construct interaction features.

[0081] Specifically, the first transaction data, second transaction data, customer group matching degree, and preference matching degree between each user and each merchant can be used as the interaction features corresponding to each user and each merchant.

[0082] Step 140 involves inputting interaction features into a preset merchant recommendation model and outputting recommendation results.

[0083] In one embodiment, the first attribute data may further include a geographic attribute. The method may further include: constructing user characteristics for each user based on the geographic attribute, population information attribute, and consumption preference attribute corresponding to each user; constructing merchant characteristics for each merchant based on the geographic attribute, customer population information attribute, and merchant basic attribute corresponding to each merchant; step 140 inputs interaction features into the preset merchant recommendation model and outputs recommendation results, which may specifically include: inputting interaction features, user features, and merchant features into the preset merchant recommendation model and outputting recommendation results.

[0084] In this embodiment, more attribute information related to users and merchants can be introduced to construct user features and merchant features. When recommending merchants to users, the three main parts of the constructed user features, merchant features, and interaction features are integrated to form a complete merchant recommendation feature system, which improves the richness of merchant recommendation features. Furthermore, feature construction is carried out from the perspective of user preference consistency and customer group matching, which further improves the accuracy of merchant recommendation features and makes the recommendation results more accurate.

[0085] In some embodiments of this application, in order to determine the adjustment parameters for each user for each of the N merchants, the basic attributes of the merchants may include the merchant category. Figure 4 A flowchart of another embodiment of the merchant recommendation method provided in the first aspect of this application, step 210 may include... Figure 4 Steps 410-450 are shown.

[0086] Step 410: Based on the first transaction data and the merchant category of each merchant, determine the first repurchase period corresponding to each merchant category, and the N second repurchase periods corresponding to each merchant.

[0087] The first repurchase period is used to characterize the average interval between every two transactions between each merchant category and M users, and the second repurchase period is used to characterize the average interval between every two transactions between each merchant and M users.

[0088] For example, M users include user A, user B and user C, and N merchants include merchant A, merchant B and merchant C. Merchants A and B are classified as restaurants, and merchant C is classified as a convenience store. Electronic devices can determine the average interval between every two transactions between users A, B, and C and merchants A and B based on the transaction times between users A, B, and C and merchants in the catering category, thus obtaining the first repurchase cycle for catering merchants. Electronic devices can also determine the average interval between every two transactions between users A, B, and C and merchants in the convenience store category based on the transaction times between users A, B, and C and merchants A, B, and C, thus obtaining the first repurchase cycle for convenience store merchants. Furthermore, electronic devices can determine the average interval between every two transactions between users A, B, and C and merchants A, B, and C, respectively, thus obtaining the second repurchase cycle for merchant A; the same applies to merchants B and C.

[0089] Step 420: Based on the first transaction data, calculate the cumulative number of transactions between M users and each category of merchants in the most recent first repurchase cycle, and obtain the first transaction count corresponding to each merchant category.

[0090] Referring to the example above, there are M users, including user A, user B, and user C, and N merchants, including merchant A, merchant B, and merchant C. Merchants A and B are classified as restaurants, and merchant C is classified as a convenience store. If the first repurchase period for the restaurant merchants is 20 days, then the first transaction count for the restaurant merchants is the cumulative number of transactions between users A, B, and C and merchants A and B over those 20 days.

[0091] Step 430: Calculate the number of second transactions between each user and merchants of the same category based on the cumulative number of transactions between each user and N merchants in the most recent N second repurchase cycles and the merchant categories of the N merchants.

[0092] Continuing with the example above, if Merchant A's second repurchase period is 30 days, Merchant B's is 10 days, and Merchant C's is 2 days, then for User A, the cumulative number of transactions (3) between User A and Merchant A in the last 30 days can be added to the cumulative number of transactions (2) between User A and Merchant B in the last 10 days, resulting in the second number of transactions (5) between User A and catering merchants. The cumulative number of transactions (3) between User A and Merchant C in the last 2 days is the second number of transactions between User A and convenience store merchants.

[0093] Step 440: Based on the first transaction data, obtain the first transaction time of the most recent transaction between each user and each merchant, and the second transaction time of the most recent transaction between each user and each category of merchants.

[0094] Step 450: Based on the first number of transactions, the second number of transactions, the first transaction time, and the second transaction time, calculate the adjustment parameters for each user to each merchant, and obtain M*N adjustment parameters for M users.

[0095] In this embodiment of the application, after determining the first repurchase period and the second repurchase period, the cumulative number of transactions between M users and each category of merchants in the most recent first repurchase period, and the cumulative number of transactions between each user and N merchants in the most recent N second repurchase periods are calculated. This takes into account time factors and information on similar consumption scenarios, and obtains adjustment parameters for adjusting consumption intensity based on time periods and similar consumption scenarios, thereby improving the accuracy of data characteristics of the second transaction data.

[0096] In one embodiment, step 450 may include: calculating the adjustment parameter δzj for each user to each merchant based on formula (2).

[0097]

[0098] Where z represents the user, j represents the merchant, i represents the merchant category of merchant j, and T now T represents the current time. zj The first transaction time of the most recent transaction between user z and merchant j. For merchant j, the second repurchase period For the first repurchase period corresponding to merchant category i, T zi Let N be the second transaction time of the most recent transaction between user z and merchant of merchant type i. i C represents the number of the first transaction corresponding to merchant category i. zi The number of the second transactions corresponding to user z and merchant j is ω1 and ω2, which are preset adjustment factors. The values ​​of ω1 and ω2 can be the same or different. They can be set to values ​​greater than zero according to actual needs. This application does not make specific restrictions here.

[0099] For example, ω1 and ω2 can both be set to 0.1, Merchant A belongs to the catering category, and the time T between the first transaction time of User A's most recent transaction with Merchant A and the current time is... now -T zj The value is 80. The time T between the second most recent transaction between User A and a restaurant merchant and the current time is 80. now -T zj The value is 10, which is the second repurchase cycle for Merchant A. The first repurchase cycle for catering merchants is 20, and the first transaction number N for catering merchants is 30. i The number is 500, and the number of the second transactions corresponding to user A and merchant A is C. zi If the value is 5, then the adjustment parameters for user A and merchant A can be (80+0.1) / 30*(10+0.1) / 20*(500-5) / 500=1.33.

[0100] Based on the same inventive concept, the second aspect of this application provides a merchant recommendation device. Figure 5 A schematic diagram of an embodiment of the merchant recommendation device provided in the second aspect of this application.

[0101] like Figure 5 As shown, the merchant recommendation device 500 may specifically include: an acquisition module 510, a calculation module 520, a determination module 530, and a recommendation module 540.

[0102] The acquisition module 510 is used to acquire target data corresponding to M users and N merchants. The target data includes the first transaction data between the M users and N merchants, as well as the first attribute data of the M users and N merchants. The first attribute data includes the user's demographic information attribute and consumption preference attribute, as well as the merchant's basic attribute and customer demographic information attribute.

[0103] The calculation module 520 is used to calculate the customer group matching degree and preference matching degree between each user and each merchant based on the first transaction data and the first attribute data;

[0104] The determination module 530 is used to determine the interaction characteristics based on the first transaction data, as well as the customer group matching degree and preference matching degree between each user and each merchant;

[0105] The recommendation module 540 is used to input interaction features into the preset merchant recommendation model and output recommendation results.

[0106] The merchant recommendation device provided in this application, in a scenario of recommending merchants to M users, can acquire first transaction data and first attribute data between the M users and all merchants to be recommended, i.e., N merchants. Based on this, it calculates the customer group matching degree and preference matching degree between each user and each merchant based on the first transaction data and the customer group matching degree and preference matching degree between each user and each merchant, and determines interaction features based on the first transaction data and the customer group matching degree and preference matching degree between each user and each merchant. Since the first transaction data reflects the historical transaction behavior between users and merchants, the customer group matching degree reflects the degree of matching between the user and the merchant's customer group, and the preference matching degree reflects the degree of interest of each user in each merchant, the interaction features determined based on the first transaction data, customer group matching degree, and preference matching degree can accurately reflect the user's preference for N merchants. Using these interaction features to recommend merchants to M users effectively improves the accuracy of merchant recommendations.

[0107] In some embodiments of this application, the determining module includes: a determining submodule, used to determine the adjustment parameters for each user on each of the N merchants based on the first transaction data; a clustering submodule, used to cluster the M users and N merchants respectively based on the first attribute data to obtain clustering results; an adjusting submodule, used to adjust the first transaction data according to the adjustment parameters and the clustering results to obtain second transaction data; and a constructing submodule, used to construct interaction features based on the first transaction data, the second transaction data, customer group matching degree, and preference matching degree between the M users and the N merchants.

[0108] In some embodiments of this application, the basic attributes of the merchant include the merchant category. Based on the first transaction data, the determining submodule includes: a determining unit, used to determine the first repurchase period corresponding to each merchant category and N second repurchase periods corresponding to each merchant, based on the first transaction data and the merchant category of each merchant, wherein the first repurchase period is used to characterize the average interval between every two transactions between each category of merchant and M users, and the second repurchase period is used to characterize the average interval between every two transactions between each merchant and M users; and a calculation unit, used to calculate the cumulative number of transactions between M users and each category of merchant in the most recent first repurchase period based on the first transaction data. The system includes: a first transaction count for each merchant category; a calculation unit, which calculates the second transaction count between each user and merchants of the same category based on the cumulative number of transactions between each user and N merchants in the most recent N second repurchase cycles and the merchant categories of the N merchants; an acquisition unit, which acquires the first transaction time of each user's most recent transaction with each merchant and the second transaction time of each user's most recent transaction with each category of merchants based on the first transaction data; and a calculation unit, which calculates the adjustment parameters for each user to each merchant based on the first transaction count, the second transaction count, the first transaction time, and the second transaction time, to obtain M*N adjustment parameters for M users.

[0109] In some embodiments of this application, the clustering results include user clusters and merchant clusters. The adjustment submodule includes: a determining unit, configured to determine at least one second user belonging to the same user cluster as the first user, and / or at least one second merchant belonging to the same merchant cluster as the first merchant, when the number of transactions between the first user and the first merchant among M users is zero; the determining unit is further configured to determine third transaction data in the first transaction data, which is the target transaction information between the first user and the first merchant, the third transaction data including transaction data between at least one second user and the first merchant, and / or transaction data between at least one second merchant and the first user; an adding unit, configured to add target transaction information to the first transaction data to obtain fourth transaction data; and an adjusting unit, configured to adjust the fourth transaction data according to adjustment parameters to obtain second transaction data.

[0110] In some embodiments of this application, the adjustment unit is specifically used to: construct a first interaction matrix based on the fourth transaction data, the first interaction matrix including M*N elements, the elements being used to represent the number of transactions between users and merchants; multiply the M*N adjustment parameters corresponding to the M users by the M*N elements respectively to obtain a second interaction matrix; and determine the second transaction data based on the second interaction matrix, the second transaction data including the adjusted number of transactions between the M users and N merchants.

[0111] In some embodiments of this application, the calculation module includes: a segmentation submodule, used to divide the M users into multiple customer groups based on the demographic information attributes of the M users; a calculation submodule, used to calculate the total number of transactions between each merchant and each customer group based on the first transaction data; and the calculation submodule is further used to calculate the customer group matching degree between each user and each merchant based on the customer group to which each user belongs and the total number of transactions between each merchant and each customer group.

[0112] In some embodiments of this application, the calculation submodule is specifically used to: calculate the preference matching degree between each user and each merchant based on the consumption preference attributes of M users and the basic merchant attributes of N merchants.

[0113] In some embodiments of this application, the first attribute data further includes a geographic attribute. The device further includes: a construction module, used to construct user features for each user based on the geographic attribute, population information attribute, and consumption preference attribute corresponding to each user; a construction module, used to construct merchant features for each merchant based on the geographic attribute, customer population information attribute, and merchant basic attribute corresponding to each merchant; and a recommendation module specifically used to: input interaction features, user features, and merchant features into a preset merchant recommendation model, and output recommendation results.

[0114] A third aspect of this application also provides an electronic device. Figure 6 A schematic diagram of the structure of an embodiment of the electronic device provided in the third aspect of this application. (See attached diagram.) Figure 6 As shown, the electronic device 600 includes a memory 601, a processor 602, and a computer program stored in the memory 601 and executable on the processor 602.

[0115] In one example, the processor 602 described above may include a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits that may be configured to implement the embodiments of this application.

[0116] Memory 601 may include read-only memory (ROM), random access memory (RAM), disk storage media device, optical storage media device, flash memory device, electrical, optical, or other physical / tangible memory storage device. Therefore, typically, memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the merchant-recommended method in the embodiments according to the first aspect of this application.

[0117] The processor 602 runs a computer program corresponding to the executable program code by reading the executable program code stored in the memory 601, in order to implement the merchant recommendation method in the embodiment of the first aspect described above.

[0118] In some examples, electronic device 600 may also include communication interface 603 and bus 604. For example, Figure 6 As shown, the memory 601, processor 602, and communication interface 603 are connected through bus 604 and complete communication with each other.

[0119] The communication interface 603 is mainly used to enable communication between various modules, devices, units, and / or equipment in the embodiments of this application. Input devices and / or output devices can also be connected through the communication interface 603.

[0120] Bus 604 includes hardware, software, or both, that couples components of electronic device 600 together. For example, and not limitingly, bus 604 may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-E) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 604 may include one or more buses. Although specific buses are described and illustrated in the embodiments of this application, this application considers any suitable bus or interconnection.

[0121] A fourth aspect of this application provides a computer-readable storage medium storing a program or instructions. When executed by a processor, the program or instructions can implement the merchant recommendation method described in the first aspect and achieve the same technical effect. To avoid repetition, further details are omitted here. The aforementioned computer-readable storage medium may include non-transitory computer-readable storage media, such as read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks, etc., and is not limited thereto.

[0122] The fifth aspect of this application provides a computer program product stored in a non-volatile storage medium. When executed by at least one processor, the computer program product implements the steps of the merchant recommendation method as shown in the first aspect. The specific content of the merchant recommendation method can be found in the relevant descriptions in the above embodiments, and will not be repeated here.

[0123] The sixth aspect of this application provides a chip including a processor and a communication interface, the communication interface and the processor being coupled together. The processor is used to run programs or instructions to implement various processes of the merchant recommendation method embodiments shown in the first aspect, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0124] It should be understood that the chip mentioned in the embodiments of this application may also be referred to as a system-on-a-chip, system chip, chip system, or system-on-a-chip, etc.

[0125] It should be clarified that the various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. For the device embodiments, user terminal embodiments, equipment embodiments, system embodiments, and computer-readable storage medium embodiments, the relevant parts can be referred to the description section of the method embodiments. This application is not limited to the specific steps and structures described above and shown in the figures. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application. Furthermore, for the sake of brevity, detailed descriptions of known methods and techniques are omitted here.

[0126] The aspects of this application have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by dedicated hardware performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0127] Those skilled in the art will understand that the above embodiments are exemplary and not restrictive. Different technical features appearing in different embodiments can be combined to achieve beneficial effects. Based on a study of the drawings, specification, and claims, those skilled in the art should be able to understand and implement other variations of the disclosed embodiments. In the claims, the term "comprising" does not exclude other means or steps; the quantifier "a" does not exclude a plurality; the terms "first" and "second" are used to identify names and not to indicate any particular order. No reference numerals in the claims should be construed as limiting the scope of protection. The functionality of multiple parts appearing in the claims can be implemented by a single hardware or software module. The appearance of certain technical features in different dependent claims does not mean that these technical features cannot be combined to achieve beneficial effects.

Claims

1. A merchant recommendation method, characterized in that, The method includes: Obtain target data corresponding to M users and N merchants. The target data includes first transaction data between the M users and the N merchants, and first attribute data of the M users and N merchants. The first attribute data includes the demographic information attributes and consumption preference attributes of the users, and the basic attributes of the merchants and the demographic information attributes of the customer groups of the merchants. Based on the first transaction data and the first attribute data, calculate the customer group matching degree and preference matching degree between each user and each merchant; Based on the first transaction data, and the customer group matching degree and preference matching degree between each user and each merchant, the interaction characteristics are determined; Input the interaction features into the preset merchant recommendation model and output the recommendation results; The step of determining interaction characteristics based on the first transaction data, and the customer group matching degree and preference matching degree between each user and each merchant, includes: Based on the first transaction data, determine the adjustment parameters for each user for each of the N merchants; Based on the first attribute data, clustering is performed on the M users and the N merchants respectively to obtain clustering results; The first transaction data is adjusted according to the adjustment parameters and the clustering results to obtain the second transaction data; The first transaction data, the second transaction data, the customer group matching degree, and the preference matching degree between each user and each merchant are used as the interaction features corresponding to each user and each merchant. The step of calculating the customer group matching degree and preference matching degree between each user and each merchant based on the first transaction data and the first attribute data includes: Based on the demographic information attributes of the M users, the M users are divided into multiple customer groups; Based on the first transaction data, calculate the total number of transactions between each merchant and each customer group; Based on each user's customer group and the total number of transactions between each merchant and each customer group, the customer group matching degree between each user and each merchant is calculated.

2. The method according to claim 1, characterized in that, The basic attributes of the merchants include merchant categories. The step of determining the adjustment parameters for each user for each of the N merchants based on the first transaction data includes: Based on the first transaction data and the merchant category of each merchant, a first repurchase period corresponding to each merchant category and N second repurchase periods corresponding to each merchant are determined. The first repurchase period is used to characterize the average interval between each two transactions between each merchant category and the M users, and the second repurchase period is used to characterize the average interval between each two transactions between each merchant and the M users. Based on the first transaction data, calculate the cumulative number of transactions between the M users and each category of merchants in the most recent first repurchase cycle, and obtain the first transaction number corresponding to each merchant category; Based on the cumulative number of transactions between each user and N merchants in the most recent N second repurchase cycles, and the merchant categories of the N merchants, calculate the number of second transactions between each user and merchants of the same category; Based on the first transaction data, obtain the first transaction time of the most recent transaction between each user and each merchant, and the second transaction time of the most recent transaction between each user and each category of merchants; Based on the first number of transactions, the second number of transactions, the first transaction time, and the second transaction time, the adjustment parameters for each user to each merchant are calculated, resulting in M*N adjustment parameters for M users.

3. The method according to claim 1, characterized in that, The clustering results include user clusters and merchant clusters. The step of adjusting the first transaction data according to the adjustment parameters and the clustering results to obtain the second transaction data includes: If the number of transactions between the first user among the M users and the first merchant among the N merchants is zero, then determine at least one second user belonging to the same user cluster as the first user, and / or at least one second merchant belonging to the same merchant cluster as the first merchant. The third transaction data in the first transaction data is determined to be the target transaction information between the first user and the first merchant. The third transaction data includes the transaction data between the at least one second user and the first merchant, and / or the transaction data between the at least one second merchant and the first user. Add the target transaction information to the first transaction data to obtain the fourth transaction data; The fourth transaction data is adjusted according to the adjustment parameters to obtain the second transaction data.

4. The method according to claim 3, characterized in that, Each user corresponds to N adjustment parameters. The step of adjusting the fourth transaction data according to the adjustment parameters to obtain the second transaction data includes: A first interaction matrix is ​​constructed based on the fourth transaction data. The first interaction matrix includes M*N elements, which are used to represent the number of transactions between the user and the merchant. Multiply the M*N adjustment parameters corresponding to the M users by M*N elements respectively to obtain the second interaction matrix; Based on the second interaction matrix, the second transaction data is determined, which includes the adjusted number of transactions between M users and N merchants.

5. The method according to claim 1, characterized in that, The step of calculating the customer group matching degree and preference matching degree between each user and each merchant based on the first transaction data and the first attribute data includes: Based on the consumption preference attributes of the M users and the basic merchant attributes of the N merchants, calculate the preference matching degree between each user and each merchant.

6. The method according to claim 1, characterized in that, The first attribute data also includes a geographic attribute, and the method further includes: Based on the geographic attributes, demographic information attributes, and consumption preference attributes corresponding to each user, user characteristics for each user are constructed. Based on the geographical attributes, customer demographic information attributes, and basic merchant attributes corresponding to each merchant, the merchant characteristics of each merchant are constructed. The step of inputting the interaction features into the preset merchant recommendation model and outputting recommendation results includes: Input the interaction features, user features, and merchant features into the preset merchant recommendation model, and output the recommendation result.

7. A merchant recommendation device, characterized in that, include: The acquisition module is used to acquire target data corresponding to M users and N merchants. The target data includes first transaction data between the M users and the N merchants, and first attribute data of the M users and the N merchants. The first attribute data includes the demographic information attributes and consumption preference attributes of the users, and the basic attributes of the merchants and customer demographic information attributes of the merchants. The calculation module is used to calculate the customer group matching degree and preference matching degree between each user and each merchant based on the first transaction data and the first attribute data; The determination module is used to determine interaction characteristics based on the first transaction data and the customer group matching degree and preference matching degree between each user and each merchant; The recommendation module is used to input the interaction features into a preset merchant recommendation model and output recommendation results; The determining module is specifically used for: Based on the first transaction data, determine the adjustment parameters for each user for each of the N merchants; Based on the first attribute data, clustering is performed on the M users and the N merchants respectively to obtain clustering results; The first transaction data is adjusted according to the adjustment parameters and the clustering results to obtain the second transaction data; The first transaction data, the second transaction data, the customer group matching degree, and the preference matching degree between each user and each merchant are used as the interaction features corresponding to each user and each merchant. The calculation module is specifically used for: Based on the demographic information attributes of the M users, the M users are divided into multiple customer groups; Based on the first transaction data, calculate the total number of transactions between each merchant and each customer group; Based on each user's customer group and the total number of transactions between each merchant and each customer group, the customer group matching degree between each user and each merchant is calculated.

8. An electronic device, characterized in that, The device includes: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, it implements the merchant recommendation method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program or instructions that, when executed by a processor, implement the merchant recommendation method as described in any one of claims 1 to 6.

10. A computer program product, characterized in that, The computer program product is stored in a non-volatile storage medium, and when executed by at least one processor, the computer program product implements the merchant recommendation method as described in any one of claims 1 to 6.

Images