Business recommendation method, device, computer equipment and storage medium

Abstract

The present application relates to a service recommendation method, apparatus, computer equipment, and storage medium. The method includes: obtaining at least two pieces of historical account attribute information; classifying the accounts corresponding to the at least two pieces of historical account attribute information to obtain a classified account set, calculating service preference information based on the classified account set, and obtaining account set preference information corresponding to each classified account set; obtaining new account attribute information, comparing the new account attribute information with the attribute information corresponding to each classified account set for similarity, and obtaining a target account set; adjusting the preference information in an initial service recommendation list based on the attribute information corresponding to the target account set and the account set preference information corresponding to the classified account set, and obtaining the new account's preference information for each target service in the initial service recommendation list; and obtaining a target service recommendation list based on the preference information of each target service. This method can improve the accuracy of service recommendations for new accounts.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a service recommendation method, apparatus, computer device, and storage medium. Background Art

[0002] With the development of computer technology, business recommendation algorithms have emerged and have been applied to websites in various fields, including books, music, videos, news, movies, maps, and more. With the increasing popularity of e-commerce in recent years, major e-commerce websites have adopted e-commerce-related recommendation algorithms. Recommendation algorithms not only bring significant additional benefits to these internet merchants, but also improve user satisfaction and increase user stickiness.

[0003] Traditionally, recommendation algorithms can help businesses effectively filter information and recommend suitable services to customers. However, as both customers and services rapidly grow alongside information overload, new customers face a cold start problem. This means recommending services to new customers without any behavioral records, resulting in poor recommendations and low efficiency. Summary of the Invention

[0004] Based on this, it is necessary to provide a business recommendation method, apparatus, computer device, computer-readable storage medium and computer program product that can reorder the business recommendation list according to account attribute information to address the above technical problems.

[0005] In a first aspect, the present application provides a service recommendation method. The method comprises: obtaining at least two pieces of historical account attribute information, wherein the historical account attribute information is inherent information corresponding to accounts that have performed service operations; classifying the accounts corresponding to the at least two pieces of historical account attribute information to obtain classified account sets, calculating service preference information based on the classified account sets to obtain account set preference information corresponding to each of the classified account sets; obtaining new account attribute information, comparing the new account attribute information with the attribute information corresponding to each of the classified account sets to obtain a target account set; the similarity between the attribute information corresponding to the target account set and the new account attribute information satisfies a preset condition; adjusting the preference information in an initial service recommendation list based on the attribute information corresponding to the target account set and the account set preference information corresponding to the classified account set to obtain the preference information of the new account for each target service in the initial service recommendation list; and recommending a target service list corresponding to the new account based on the preference information of each target service; the target service recommendation list is used to make service recommendations to the new account.

[0006] In one embodiment, the preference information in the initial business recommendation list is adjusted based on the attribute information corresponding to the target account set and the account set preference information corresponding to the classified account set to obtain the preference information of the new account for each target business in the initial business recommendation list, including: matching the attribute information corresponding to the target account set with the attribute information corresponding to each classified account set to obtain each attribute information matching result; multiplying each attribute information matching result with the account set preference information corresponding to the matching classified account set to obtain each intermediate preference information; summing each intermediate preference information to obtain the preference information of the new account for each target business in the initial business recommendation list.

[0007] In one embodiment, the attribute information corresponding to the target account set is matched with the attribute information corresponding to each of the classified account sets to obtain the attribute information matching results, including: calculating the intersection of the attribute information corresponding to the target account set and the attribute information corresponding to each of the classified account sets and taking the absolute value to obtain the attribute information intersection; calculating the union of the attribute information corresponding to the target account set and the attribute information corresponding to each of the classified account sets and taking the absolute value to obtain the attribute information union; performing division calculation on the attribute information intersection and the corresponding attribute information union to obtain the attribute information matching results.

[0008] In one embodiment, the business preference information is calculated based on the classified account sets to obtain the account set preference information corresponding to each of the classified account sets, including: summing the behavior attribute information corresponding to each of the classified account sets to obtain the number of historical business operations corresponding to each of the classified account sets, where the behavior attribute information is the historical account attribute information corresponding to the business operations on the historical accounts; dividing the number of historical business operations corresponding to each of the classified account sets by the number of historical operation accounts corresponding to each of the classified account sets to obtain the account set preference information corresponding to each of the classified account sets; the number of historical operation accounts is the accounts in each of the classified account sets that have performed the corresponding business operations.

[0009] In one embodiment, the method further includes: establishing an initial business recommendation matrix by using the account information in the historical account attribute information as the rows of the matrix and the business information as the columns of the matrix; and obtaining the initial business recommendation list by multiplying the account information vector corresponding to each element in the initial business recommendation matrix with the business information vector.

[0010] In one embodiment, the method of multiplying the account information vector and the business information vector corresponding to each element in the initial business recommendation matrix to obtain the initial business recommendation list includes: performing modulus calculation on the account information vector and the business information vector corresponding to each element in the initial business recommendation matrix to obtain the modulus of each account information vector and the modulus of the business information vector; obtaining the angle between the account information vector and the business information vector corresponding to each element in the initial business recommendation matrix; and multiplying the product of the modulus of each account information vector and the modulus of the business information vector by the cosine value corresponding to the angle to obtain the initial business recommendation list.

[0011] In a second aspect, the present application also provides a service recommendation device. The device includes: a historical account attribute information acquisition module for acquiring at least two historical account attribute information, wherein the historical account attribute information is inherent information corresponding to an account that has performed a business operation; an account set preference information acquisition module for classifying the accounts corresponding to the at least two historical account attribute information to obtain classified account sets, and calculating business preference information based on the classified account sets to obtain account set preference information corresponding to each classified account set; a target account set acquisition module for acquiring new account attribute information, and performing similarity comparison between the new account attribute information and the attribute information corresponding to each classified account set to obtain a target account set; the similarity between the attribute information corresponding to the target account set and the new account attribute information satisfies a preset condition; a target business preference information acquisition module for adjusting preference information in an initial business recommendation list based on the attribute information corresponding to the target account set and the account set preference information corresponding to the classified account set to obtain preference information of the new account for each target business in the initial business recommendation list; a target business recommendation module for recommending a target business list corresponding to the new account based on the preference information of each target business; the target business recommendation list is used to make business recommendations to the new account.

[0012] In one embodiment, the module for obtaining the preference information of the target business is further used to match the attribute information corresponding to the target account set with the attribute information corresponding to each classified account set to obtain each attribute information matching result; multiply each attribute information matching result by the account set preference information corresponding to the matching classified account set to obtain each intermediate preference information; and sum each intermediate preference information to obtain the preference information of the new account for each target business in the initial business recommendation list.

[0013] In one embodiment, the module for obtaining the preference information of the target business is further used to calculate the intersection of the attribute information corresponding to the target account set and the attribute information corresponding to each of the classified account sets, and then take the absolute value to obtain the intersection of the attribute information; calculate the union of the attribute information corresponding to the target account set and the attribute information corresponding to each of the classified account sets, and then take the absolute value to obtain the union of the attribute information; perform division calculation on the intersection of the attribute information and the corresponding union of the attribute information to obtain the matching results of each of the attribute information.

[0014] In one embodiment, the account set preference information obtaining module is further used to sum up the behavioral attribute information corresponding to each of the classified account sets to obtain the number of historical business operations corresponding to each of the classified account sets, where the behavioral attribute information is the historical account attribute information corresponding to the business operations on the historical accounts; divide the number of historical business operations corresponding to each of the classified account sets by the number of historical operation accounts corresponding to each of the classified account sets to obtain the account set preference information corresponding to each of the classified account sets; the number of historical operation accounts is the number of accounts in each of the classified account sets that have performed the corresponding business operations.

[0015] In one embodiment, the module for obtaining the initial business recommendation list is further used to establish an initial business recommendation matrix by using the account information in the historical account attribute information as the rows of the matrix and the business information as the columns of the matrix; and to obtain the initial business recommendation list by multiplying the account information vector corresponding to each element in the initial business recommendation matrix with the business information vector.

[0016] In one embodiment, the module for obtaining the initial business recommendation list is further used to perform modulus calculations on the account information vector and the business information vector corresponding to each element in the initial business recommendation matrix to obtain the modulus of each account information vector and the modulus of the business information vector; obtain the angle between the account information vector and the business information vector corresponding to each element in the initial business recommendation matrix; and multiply the product of the modulus of each account information vector and the modulus of the business information vector by the cosine value corresponding to the angle to obtain the initial business recommendation list.

[0017] In a third aspect, the present application further provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor, when executing the computer program, performs the following steps: obtaining at least two pieces of historical account attribute information, the historical account attribute information being inherent information corresponding to accounts that have performed business operations; classifying the accounts corresponding to the at least two pieces of historical account attribute information to obtain classified account sets, calculating business preference information based on the classified account sets to obtain account set preference information corresponding to each of the classified account sets; obtaining new account attribute information, comparing the new account attribute information with the attribute information corresponding to each of the classified account sets to obtain a target account set; the similarity between the attribute information corresponding to the target account set and the new account attribute information satisfies a preset condition; adjusting preference information in an initial business recommendation list based on the attribute information corresponding to the target account set and the account set preference information corresponding to the classified account set to obtain preference information of the new account for each target business in the initial business recommendation list; and recommending a target business list corresponding to the new account based on the preference information of each target business; the target business recommendation list is used to make business recommendations to the new account.

[0018] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the following steps: obtaining at least two pieces of historical account attribute information, wherein the historical account attribute information is inherent information corresponding to accounts that have performed business operations; classifying the accounts corresponding to the at least two pieces of historical account attribute information to obtain classified account sets, calculating business preference information based on the classified account sets to obtain account set preference information corresponding to each of the classified account sets; obtaining new account attribute information, comparing the new account attribute information with the attribute information corresponding to each of the classified account sets to obtain a target account set; the similarity between the attribute information corresponding to the target account set and the new account attribute information satisfies a preset condition; adjusting the preference information in an initial business recommendation list based on the attribute information corresponding to the target account set and the account set preference information corresponding to the classified account set to obtain preference information of the new account for each target business in the initial business recommendation list; and recommending a target business list corresponding to the new account based on the preference information of each target business; the target business recommendation list is used to make business recommendations to the new account.

[0019] In a fifth aspect, the present application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the following steps: obtaining at least two pieces of historical account attribute information, wherein the historical account attribute information is inherent information corresponding to accounts that have performed business operations; classifying the accounts corresponding to the at least two pieces of historical account attribute information to obtain classified account sets, calculating business preference information based on the classified account sets to obtain account set preference information corresponding to each of the classified account sets; obtaining new account attribute information, comparing the new account attribute information with the attribute information corresponding to each of the classified account sets to obtain a target account set; the similarity between the attribute information corresponding to the target account set and the new account attribute information satisfies a preset condition; adjusting the preference information in an initial business recommendation list based on the attribute information corresponding to the target account set and the account set preference information corresponding to the classified account set to obtain preference information of the new account for each target business in the initial business recommendation list; and, based on the preference information of each target business, recommending a target business list corresponding to the new account; the target business recommendation list is used to make business recommendations to the new account.

[0020] The above-mentioned business recommendation method, device, computer equipment, storage medium and computer program product obtain at least two historical account attribute information, where the historical account attribute information is inherent information corresponding to the account that has performed business operations; classify the accounts corresponding to the at least two historical account attribute information to obtain the classified account set, calculate the business preference information based on the classified account set, and obtain the account set preference information corresponding to each classified account set; obtain new account attribute information, compare the new account attribute information with the attribute information corresponding to each classified account set for similarity, and obtain the target account set; the similarity between the attribute information corresponding to the target account set and the new account attribute information meets the preset conditions; adjust the preference information in the initial business recommendation list based on the attribute information corresponding to the target account set and the account set preference information corresponding to the classified account set, and obtain the preference information of the new account for each target business in the initial business recommendation list; recommend a target business list corresponding to the new account based on the preference information of each target business; the target business recommendation list is used to make business recommendations to the new account.

[0021] This approach proposes an optimized recommendation algorithm for new accounts during the re-ranking phase. It first extracts historical account attribute information and classifies the accounts accordingly. It then extracts the correlation features between the historical accounts and services. When recommending services for new accounts, the algorithm calculates a predicted score list for the new account and then re-ranks the initial service recommendation list based on the new account's attribute information. Based on the correlation between the new account's attribute information and services, the algorithm can filter out at least one target service from the initial service recommendation list that better matches the new account's preferences, thereby improving the accuracy and efficiency of service recommendations for new accounts. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is an application environment diagram of a service recommendation method in one embodiment;

[0023] Figure 2 Schematic diagram of a process of a service recommendation method in one embodiment;

[0024] Figure 3 FIG. 1 is a flow chart of a method for obtaining preference information of a target service in one embodiment;

[0025] Figure 4 Schematic diagram of a flow chart of a method for obtaining attribute information matching results in one embodiment;

[0026] Figure 5 1 is a flow chart of a method for obtaining account set preference information in one embodiment;

[0027] Figure 6 Schematic diagram of a flow chart of a method for obtaining an initial service recommendation list in one embodiment;

[0028] Figure 7 A flowchart of a method for obtaining an initial service recommendation list in another embodiment;

[0029] Figure 8 is a schematic diagram of an initial service recommendation matrix in one embodiment;

[0030] Figure 9 is a structural block diagram of a service recommendation device in one embodiment;

[0031] Figure 10 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0033] The service recommendation method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown, terminal 102 acquires data, and server 104 receives the data from terminal 102 in response to instructions from terminal 102 and performs calculations on the acquired data. Server 104 transmits the calculation results back to terminal 102, which then displays them. Terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated with server 104 or placed in the cloud or on other network servers. The server 104 obtains at least two pieces of historical account attribute information from the terminal 102. The historical account attribute information is inherent information corresponding to accounts that have performed business operations. The server 104 classifies the accounts corresponding to the at least two pieces of historical account attribute information to obtain classified account sets. The server 104 calculates business preference information based on the classified account sets to obtain account set preference information corresponding to each classified account set. The server 104 obtains new account attribute information and compares the new account attribute information with the attribute information corresponding to each classified account set to obtain a target account set. The similarity between the attribute information corresponding to the target account set and the new account attribute information satisfies a preset condition. The server 104 adjusts the preference information in the initial business recommendation list based on the attribute information corresponding to the target account set and the account set preference information corresponding to the classified account set to obtain the preference information of the new account for each target business in the initial business recommendation list. The server 104 recommends a target business list corresponding to the new account based on the preference information of each target business. The target business recommendation list is used to make business recommendations to the new account. The terminal 102 may be, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. The IoT devices may be smart speakers, smart TVs, smart air conditioners, smart car devices, etc. The portable wearable device may be a smart watch, a smart bracelet, a head-mounted device, etc. The server 104 may be implemented as an independent server or a server cluster consisting of multiple servers.

[0034] In one embodiment, Figure 2 As shown, a business recommendation method is provided, which is applied to Figure 1 The following steps are used as an example to illustrate the server in the example:

[0035] Step 202: Obtain at least two pieces of historical account attribute information.

[0036] Among them, the historical account attribute information can be the inherent information corresponding to the account that has performed business operations, wherein the attribute information can be but is not limited to purchase, share, follow or collect, like and view details, etc.

[0037] Specifically, the server responds to a command from the terminal, obtains at least two pieces of historical account attribute information from the terminal, and stores the obtained historical account attribute information in a storage unit. When the server needs to process a data record corresponding to any inherent information in the historical account attribute information, it retrieves the data record from the storage unit and transfers it to a volatile storage resource for calculation by the central processing unit. The data record corresponding to any inherent information can be a single piece of data or multiple pieces of data input simultaneously.

[0038] For example, the server 104 responds to the instruction of the terminal 102, obtains at least two historical account attribute information from the terminal 102, and stores them in the storage unit of the server 104, wherein the inherent information obtained by the server 104 corresponds to 10 data records, and multiple data can be input at the same time.

[0039] Step 204 , classify the accounts corresponding to at least two pieces of historical account attribute information to obtain classified account sets, calculate business preference information based on the classified account sets, and obtain account set preference information corresponding to each classified account set.

[0040] The classified account set can be formed by classifying historical accounts with common attribute information in multiple historical accounts. For example: Classified account set 1: Gender: Male, Elderly, Undergraduate; Classified account set 2: Gender: Male, Elderly, Postgraduate; Classified account set 3: Gender: Female, Middle-aged, Undergraduate; Classified account set 4: Gender: Female, Middle-aged, Postgraduate.

[0041] The account set preference information may be a business preference degree corresponding to each classified account set obtained after performing business preference calculation on any classified account set. The business preference degree may reflect the correspondence between the same type of attribute information and the business contained in the corresponding classified account set.

[0042] Specifically, the accounts corresponding to at least two historical account attribute information are classified according to the common manifestations, such as gender, age, education, occupation and other attribute information, to obtain multiple classified account sets. Assume that there are N historical accounts in the i-th category. i Attribute information, the historical accounts can be divided into There are different sets of accounts after classification. Each set of accounts after classification is represented by S j , The following formula is used to calculate the account set preference information of the business corresponding to the account set after the i-th classification:

[0043]

[0044] in Indicates the number of operations on the business by the i-th historical account. x This represents the total number of historical accounts that have performed operations on this business. To more prominently display historical account attribute information, a preference overlay method is used to overlay and normalize the preferences of historical accounts with similar tags. Business preference information is calculated based on each categorized account set to obtain the account set preference information corresponding to each categorized account set.

[0045] For example, suppose a customer interested in business A has K = 3 types of historical account attributes: gender, age, and education level. The first historical account, gender, has N1 = 2 sub-attributes: male and female, with gender being the characteristic attribute. The second historical account, age, has N2 = 2 attributes: middle-aged and elderly. The third historical account, education level, has N3 = 2 attributes: graduate and undergraduate. Thus, customers can be divided into N1 * N2 * N3 = 8 different customer groups: Classified Account Set 1: Male, middle-aged, undergraduate; Classified Account Set 2: Male, middle-aged, graduate; Classified Account Set 3: Male, elderly, undergraduate; Classified Account Set 4: Male, elderly, graduate; Classified Account Set 5: Female, middle-aged, undergraduate; Classified Account Set 6: Female, middle-aged, graduate; Classified Account Set 7: Female, elderly, undergraduate; Classified Account Set 8: Female, elderly, graduate.

[0046] The operation U in the formula represents the union of the attribute features of all historical accounts that have engaged in this product, representing the attribute features of the i-th classified account set. Assume that the i-th classified account set S1 is: male, middle-aged, and undergraduate. There are five historical accounts that have engaged in product A, as follows: Historical Account 1: Belongs to classified account set 1 (male, middle-aged, and undergraduate), purchased twice; Historical Account 2: Belongs to classified account set 3 (male, elderly, and undergraduate), shared twice; Historical Account 3: Belongs to classified account set 4 (male, elderly, and graduate student), liked once; Historical Account 4: Belongs to classified account set 2 (male, middle-aged, and graduate student), shared five times; Historical Account 5: Belongs to classified account set 5 (female, middle-aged, and undergraduate), purchased twice. For U in the formula, only historical Account 1, which purchased this product, falls into category S1.

[0047] Step 206 , obtaining new account attribute information, performing similarity comparison between the new account attribute information and the attribute information corresponding to each classified account set, and obtaining a target account set.

[0048] The new account attribute information may be inherent information corresponding to an account that has not performed any business operations.

[0049] The target account set may be a classified account set selected when the similarity between the attribute information corresponding to the set and the attribute information of the new account meets a preset condition.

[0050] Specifically, the system obtains attribute information corresponding to the new account, then traverses each of the classified account sets based on the attribute information compared to the attribute information of multiple historical accounts. The system then outputs the similarity between the new account attribute information and the attribute information corresponding to each classified account set. The classified account set with the highest similarity after comparison is selected as the target account set.

[0051] For example, the attribute information corresponding to the new account obtained is N. Based on the similarity comparison between this attribute information and the attribute information corresponding to each classified account set, the corresponding similarity X is output for each comparison result, and the classified account set with the highest similarity in attribute information between the two is selected as the target account set.

[0052] Step 208 : Adjust the preference information in the initial service recommendation list based on the attribute information corresponding to the target account set and the account set preference information corresponding to the classified account set to obtain the preference information of the new account for each target service in the initial service recommendation list.

[0053] The initial service recommendation list can be a table of historical accounts and service recommendations, compiled based on historical account attribute information. The table maps historical accounts to service recommendations. The initial service recommendation list will adjust the order of service preferences based on the new account attribute information to facilitate service recommendations for new accounts.

[0054] The preference information of the target business may be the preference degree of the new account for each business in the initial business recommendation table obtained through calculation. The preference degree may be expressed as a percentage or as a level, and the expression method may be adjusted according to the actual situation.

[0055] Specifically, the preference information of a new account for any target business is calculated by first finding a set of classified accounts with similar attribute information to the new account, then analyzing the businesses with higher preference information corresponding to historical accounts, and finally predicting the new account's preference information for the business. After the attribute information is divided, if any new account belongs to one of the classified account sets, the classified account set will be used as the target account set S j , the calculation formula for the new account's preference information for the business is as follows:

[0056]

[0057] in P i That is, the preference of the i-th customer group for the product, S j ∩S i Represents the classified account set S j and the classified account set S i The intersection of attribute information, S j ∪S i Represents the classified account set S j and the classified account set S i For example, the classified account set S i :Gender: Male, middle-aged, undergraduate, account set after classification: S j : Gender: Female, middle-aged, undergraduate, then after calculating the intersection and union, the resulting deviation is 2 / 4 = 0.5. After calculating the preference information for each target service in the initial service recommendation list, the preference information in the initial service recommendation list is adjusted to obtain the new account's preference information for each target service in the initial service recommendation list.

[0058] For example, for the M target services in the initial service recommendation list, the above formula is used to calculate the preference information of the target services for the new account, and the preference information Y1, Y2, ..., Y corresponding to the M target services is obtained. M . And based on Y1, Y2...Y M The preference information of each target business in the initial business recommendation list for the new account is adjusted and updated.

[0059] Step 210 : recommending a list of target services corresponding to the new account based on the preference information of each target service.

[0060] The target service recommendation list may be a list obtained by reordering the target services in the initial service recommendation list according to the calculated preference information of the corresponding target services. The target service recommendation list may be used to recommend services to new accounts.

[0061] Specifically, according to the preference information of each target service, the order of each target service in the initial service recommendation list is adjusted accordingly with respect to the attribute information of the new account, thereby obtaining a target service recommendation list for service recommendation to the new account.

[0062] For example, based on the preference information 1-10 of the target business, the order of the corresponding 10 target businesses in the initial business list is re-sorted from high to low recommendation degree according to the attribute information of the new account, and a target business recommendation list including the preference information 1-10 of the target business is obtained and used for business recommendation for the new account.

[0063] In the above-mentioned business recommendation method, at least two historical account attribute information are obtained, and the historical account attribute information is the inherent information corresponding to the account that has performed business operations; the accounts corresponding to the at least two historical account attribute information are classified to obtain the classified account set, and the business preference information is calculated based on the classified account set to obtain the account set preference information corresponding to each classified account set; new account attribute information is obtained, and the new account attribute information is compared with the attribute information corresponding to each classified account set for similarity to obtain the target account set; the similarity between the attribute information corresponding to the target account set and the new account attribute information meets the preset conditions; the preference information in the initial business recommendation list is adjusted based on the attribute information corresponding to the target account set and the account set preference information corresponding to the classified account set to obtain the preference information of the new account for each target business in the initial business recommendation list; according to the preference information of each target business, a target business recommendation list corresponding to the new account is prepared; the target business recommendation list is used to make business recommendations for the new account.

[0064] This approach proposes an optimized recommendation algorithm for new accounts during the re-ranking phase. It first extracts historical account attribute information and classifies the accounts accordingly. It then extracts the correlation features between the historical accounts and services. When recommending services for new accounts, the algorithm calculates a predicted score list for the new account and then re-ranks the initial service recommendation list based on the new account's attribute information. Based on the correlation between the new account's attribute information and services, the algorithm can filter out at least one target service from the initial service recommendation list that better matches the new account's preferences, thereby improving the accuracy and efficiency of service recommendations for new accounts.

[0065] In one embodiment, Figure 3 As shown, based on the attribute information corresponding to the target account set and the account set preference information corresponding to the classified account set, the preference information in the initial service recommendation list is adjusted to obtain the preference information of the new account for each target service in the initial service recommendation list, including:

[0066] Step 302 : Match the attribute information corresponding to the target account set with the attribute information corresponding to each classified account set to obtain attribute information matching results.

[0067] The attribute information matching result may be the result of calculating the union and intersection of the attribute information corresponding to the target account set and the attribute information corresponding to each classified account set, followed by a division operation. The attribute information matching result may be expressed in various forms, the specific form of expression being determined based on business needs, such as the degree of overlap of attribute information.

[0068] Specifically, the attribute information corresponding to the target account set and the attribute information corresponding to each classified account set are calculated as the union and intersection, and the calculated union is divided by the corresponding intersection to obtain the matching result of each attribute information. After the attribute information is divided, if any new account belongs to one of the classified account sets, the classified account set will be used as the target account set S j , the calculation formula for the new account's preference information for the business is as follows:

[0069]

[0070] in P i That is, the preference of the i-th customer group for the product, S j ∩S i Represents the classified account set S j and the classified account set S i The intersection of attribute information, S j ∪S i Represents the classified account set S j and the classified account set S i The attribute information of the union. Among them, |S j ∩S i | / |S j ∪S i |, it is any attribute information matching result.

[0071] For example, the union and intersection of the attribute information corresponding to the target account set and the attribute information corresponding to the 10 classified account sets are calculated to obtain the union 1-10 and the intersection 1-10. The calculated union 1-10 is divided by the corresponding intersection 1-10 to obtain the attribute information matching result 1-10.

[0072] Step 304 : Multiply each attribute information matching result by the account set preference information corresponding to the matching classified account set to obtain each intermediate preference information.

[0073] The intermediate preference information may be a product of the attribute information matching result and the account set preference information corresponding to each classified account set, and each intermediate preference information is a sub-information constituting the preference information of the target business.

[0074] Specifically, each attribute information matching result is multiplied by the account set preference information corresponding to the matched classified account set to obtain the intermediate preference information corresponding to each attribute information matching result. The specific expression is shown in the formula of step 302 (|S j ∩S i | / |S j ∪Si |)×P i , the product is the intermediate preference information.

[0075] For example, the attribute information matching results 1-10 are calculated and multiplied with the account set preference information 1-10 respectively to obtain the intermediate preference information 1-10. The specific calculation formula is shown in the formula of step 302 (|S j ∩S i | / |S j ∪S i |)×P i .

[0076] Step 306 : performing a sum operation on each intermediate preference information to obtain the preference information of the new account for each target service in the initial service recommendation list.

[0077] Specifically, the sum of each intermediate preference information obtained by multiplying the attribute information matching result with the corresponding account set preference information is calculated, and the sum obtained is the preference information of the new account for each target service in the initial service recommendation table. The specific expression is shown in step 302, where each intermediate preference information is represented by (|S j ∩S i | / |S j ∪S i |)×P i Calculate and obtain, for the expression The summation symbol in is to perform the summation operation on each intermediate preference information in this step.

[0078] For example, the attribute information matching results 1-10 are calculated, and multiplication operations are performed on the account set preference information 1-10 respectively to obtain the intermediate preference information 1-10. The summation operation is performed on the obtained intermediate preference information 1-10 to obtain the preference information of the new account for each target business in the initial business recommendation list.

[0079] In this embodiment, by calculating the product of the matching results of each attribute information and the account set preference information corresponding to the classified account set, it is possible to take into account the introduction of historical account attribute information when calculating the preference information of each target business, thereby improving the accuracy of the algorithm's business recommendations for new accounts.

[0080] In one embodiment, Figure 4 As shown, the attribute information corresponding to the target account set is matched with the attribute information corresponding to each classified account set to obtain the attribute information matching results, including:

[0081] Step 402 : Calculate the intersection of the attribute information corresponding to the target account set and the attribute information corresponding to each classified account set, and take the absolute value to obtain the intersection of the attribute information.

[0082] The attribute information intersection may be information where the attribute information corresponding to the target account set and the attribute information corresponding to each classified account set intersect.

[0083] Specifically, the attribute information corresponding to the target account set is compared with the attribute information corresponding to each classified account set, and then the same parts of the attribute information corresponding to the target account set and the attribute information corresponding to each classified account set are extracted, and the absolute value of the extracted attribute information is removed to obtain the attribute information intersection. Specifically, as shown in the expression in step 302, where |S j ∩S i | is to calculate the intersection of attribute information.

[0084] For example, the attribute information corresponding to the target account set is 1-10, and the attribute information corresponding to the classified account set is 6-15. Then, the same parts of the attribute information corresponding to the target account set and the attribute information corresponding to each classified account set are extracted, and the absolute values ​​of the extracted attribute information are removed to obtain the attribute information intersection |6-10|.

[0085] Step 404 : Calculate the union of the attribute information corresponding to the target account set and the attribute information corresponding to each classified account set, and take the absolute value to obtain the attribute information union.

[0086] The attribute information union may be information that a union exists between the attribute information corresponding to the target account set and the attribute information corresponding to each classified account set.

[0087] Specifically, the attribute information corresponding to the target account set is compared with the attribute information corresponding to each classified account set, and then the attribute information corresponding to the target account set and the attribute information corresponding to each classified account set are merged, and the absolute value of the merged attribute information is removed to obtain the attribute information union. Specifically, as shown in the expression in step 302, where |S j ∪S i | is to calculate the union of attribute information.

[0088] For example, the attribute information corresponding to the target account set is 1-10, and the attribute information corresponding to the classified account set is 6-15. Then, the attribute information corresponding to the target account set is merged with the attribute information corresponding to each classified account set, and the absolute value of the merged attribute information is removed to obtain the attribute information intersection |1-15|.

[0089] Step 406: Perform division calculation on the intersection of the attribute information and the union of the corresponding attribute information to obtain matching results of each attribute information.

[0090] Specifically, the intersection of the attribute information is used as the dividend, and the corresponding union of the attribute information is used as the divisor, and then the two are divided to obtain the matching results of each attribute information. Specifically, as shown in the expression in step 302, (|S j ∩S i | / |S j ∪S i |) is the result of calculating the attribute information matching.

[0091] For example, after calculation, the attribute information intersection |6-10| is obtained, and after calculation, the attribute information intersection |1-15| is obtained, then the attribute information matching result is (|6-10| / |1-15|).

[0092] In this embodiment, by introducing the intersection of attribute information and the union of attribute information and performing a division operation on the intersection of attribute information and the union of attribute information, the calculation accuracy of the attribute information matching result can be improved and the deviation can be reduced.

[0093] In one embodiment, Figure 5 As shown, the business preference information is calculated based on the classified account sets to obtain the account set preference information corresponding to each classified account set, including:

[0094] Step 502 : summing up the behavior attribute information corresponding to each classified account set to obtain the number of historical business operations corresponding to each classified account set.

[0095] The behavior attribute information may be historical account attribute information corresponding to business operations on historical accounts.

[0096] The number of historical business operations may be the union of historical account attribute information in each classified account set that has performed behavioral operations on the business.

[0097] Specifically, the historical account attribute information corresponding to the accounts that have had business operations in each classified account set is summed up, and the sum obtained is the number of historical business operations corresponding to each classified account set. The specific calculation formula is as follows: Assume that there are N historical accounts in the i-th category. i Attribute information, the historical accounts can be divided into There are different sets of accounts after classification. Each set of accounts after classification is represented by S j , The following formula is used to calculate the account set preference information of the business corresponding to the account set after the i-th classification:

[0098]

[0099] in Indicates the number of historical business operations performed by the i-th category historical account on the business, that is, the number of historical business operations corresponding to each classified account set. x Indicates the total number of historical accounts that have performed operations on this business.

[0100] For example, the operation U in the formula represents taking the union of the attribute features of all historical accounts that have engaged in this product as the attribute features of the i-th classified account set. Assume that the i-th classified account set S1 is: male, middle-aged, and undergraduate. There are five historical accounts that have engaged in product A, as follows: Historical Account 1: Belongs to classified account set 1 (male, middle-aged, and undergraduate), purchased twice; Historical Account 2: Belongs to classified account set 3 (male, elderly, and undergraduate), shared twice; Historical Account 3: Belongs to classified account set 4 (male, elderly, and graduate student), liked once; Historical Account 4: Belongs to classified account set 2 (male, middle-aged, and graduate student), shared five times; Historical Account 5: Belongs to classified account set 5 (female, middle-aged, and undergraduate), purchased twice. For U in the formula, among these customers, only Historical Account 1's purchase behavior for this product falls into category S1.

[0101] Step 504 : Perform a division operation on the number of historical business operations corresponding to each classified account set and the number of historical operation accounts corresponding to each classified account set to obtain account set preference information corresponding to each classified account set.

[0102] The number of historical operation accounts may be historical accounts in the classified account set that have performed the corresponding business operations.

[0103] Specifically, statistics are collected on the historical accounts that have performed the corresponding business operations in the classified account set to obtain the number of historical operation accounts. A division operation is performed based on the number of historical business operations corresponding to the classified account set and the number of historical operation accounts corresponding to each classified account set, where the number of historical business operations is used as the dividend and the number of historical operation accounts is used as the divisor. After calculation, the account set preference information corresponding to each classified account set is obtained. The specific calculation formula is shown in step 502. In the formula, P i That is, the account set preference information corresponding to the classified account set.

[0104] For example, the number of historical business operations corresponding to the classified account set is a, and the number of historical operation accounts corresponding to the classified account set is b. The account set preference information is calculated according to the calculation formula in step 502 to obtain the account set preference information P corresponding to the classified account set. x =a / b.

[0105] In this embodiment, by summing the historical account attribute information corresponding to each classified account set and then dividing the sum by the number of historical operation accounts, the weight of historical accounts that have performed business operations can be fully reflected, the impact of historical accounts that have not performed business operations can be reduced, and the accuracy of account set preference information can be improved.

[0106] In one embodiment, Figure 6 As shown, the method further includes:

[0107] Step 602: Create an initial business recommendation matrix by using the account information in the historical account attribute information as the rows of the matrix and the business information as the columns of the matrix.

[0108] The account information may be basic information used to represent an account in historical account attribute information, and may be the holder of the historical account, the holder's personal information, and the like.

[0109] The business information may be business information corresponding to the business that has been handled in the historical account.

[0110] The initial business recommendation matrix may be a matrix established by using account information as rows and business information as columns to represent the relationship between account information and business information.

[0111] Specifically, historical account attribute information is obtained. For example, the source of historical account attribute information can be attribute data of the business (seven-day annualized rate of return, income per 10,000 shares, etc.), the customer portrait corresponding to the account, and the customer behavior information corresponding to the account (collection, like, purchase, sharing, etc.). The historical account attribute information is pre-processed to clean the dirty data in the historical account attribute information, and the format of the historical account attribute information is unified. It is scored according to different behavior scoring standards to obtain the initial business recommendation matrix. Different behaviors of the same account will have different weights. For example, if an account has purchased a product and liked another product, it is assumed that the interest in the purchased product will be greater. Therefore, the following weight score rules are formulated: 6 points for purchase, 3 points for sharing, 2 points for following or collecting, 2 points for likes, and 1 point for viewing details.

[0112] The two dimensions of the rows and columns in the initial business recommendation matrix are account information dimension and business information dimension respectively. The account information dimension includes M accounts and the business information dimension includes N businesses. Figure 8 shown. Figure 8 In [1], the M accounts are specifically account information C-1 to CM, the N services are specifically service information P-1 to PM, and a_ij (i=1 to M, j=1 to N) are elements in the initial service recommendation matrix. Figure 8 The rows in correspond to the account information dimension in the initial business recommendation matrix, and the columns correspond to the business information dimension in the initial business recommendation matrix.

[0113] For example, account information 1-10 is used as the account information dimension in the initial business recommendation matrix, and business information 1-20 is used as the business information dimension in the initial business recommendation matrix, to obtain the initial business recommendation matrix composed of account information 1-10 and business information 1-20.

[0114] Step 604 : Multiply the account information vector and the business information vector corresponding to each element in the initial business recommendation matrix to obtain an initial business recommendation list.

[0115] The account information vector is a vector formed by account information in the initial business recommendation matrix.

[0116] The business information vector is a vector formed by business information in the initial business recommendation matrix.

[0117] Specifically, we take the inner product of each account's customer information vector and the business information vectors of N businesses, and obtain N predicted score values ​​corresponding to each of the N businesses. The inner product of the vectors is proportional to the cosine of the adjacent angles. The smaller the angle, the larger the inner product of the vectors. Therefore, the inner product of the vectors can be used to evaluate the correlation between the account information vector and the business information vector. The following is an example of the calculation formula for the predicted score value:

[0118]

[0119] Among them, m i Represents the account information vector of the i-th customer, n j represents the business information vector of the j-th business, It represents the inner product of the account information vector of the i-th account and the business information vector of the j-th business, that is, the predicted score value for the j-th business handled by the i-th account.

[0120] An initial business recommendation matrix can be obtained by taking the inner product of the account information vectors of each of the M accounts and the business information vectors of the N businesses, where the values ​​in the dimension corresponding to each account in the initial business recommendation matrix are N predicted score values. For example, the account information vectors of each of the M accounts are combined as row vectors into one matrix, while the business information vectors of the N businesses are combined as column vectors into another matrix. Then, the two matrices are subjected to a dot product operation to obtain the initial business recommendation matrix. In this way, the predicted scores corresponding to the M accounts and N businesses can be calculated and restored. Based on the ranking of the N predicted score values, the recommended businesses are determined from the N businesses.

[0121] In this embodiment, by establishing a matrix of historical account attribute information and further obtaining an initial business recommendation list based on the matrix, the relationship between the account information and the business information in the historical account attribute information can be taken into account, so that the correspondence between the scores in the initial business recommendation list and the historical accounts is accurate.

[0122] In one embodiment, Figure 7 As shown, the account information vector corresponding to each element in the initial business recommendation matrix is ​​multiplied by the business information vector to obtain the initial business recommendation list, including:

[0123] Step 702 : performing a modulus calculation on the account information vector and the business information vector corresponding to each element in the initial business recommendation matrix to obtain the modulus of each account information vector and the modulus of the business information vector.

[0124] The modulus of a vector may be the size of the vector, that is, the length of the vector. The modulus of an account information vector is the length of the account information vector, and the modulus of a business information vector is the length of the business information vector.

[0125] Specifically, the account information vector and the business information vector corresponding to each element in the initial business recommendation matrix are modulo calculated to obtain the vector length corresponding to each account information vector and business information vector, that is, the modulus corresponding to each vector.

[0126] For example, the account information vector x and the business information vector y in the initial business recommendation matrix are modulo calculated to obtain the account information vector length |x| corresponding to the account information vector x and the business information vector length |y| corresponding to the business information vector y.

[0127] Step 704: Obtain the angle between the account information vector and the business information vector corresponding to each element in the initial business recommendation matrix.

[0128] Specifically, the angle between any account information vector and the business information vector in the initial business recommendation matrix is ​​obtained, wherein the angle is an angle less than 180 degrees between the two vectors after the starting points of the two vectors are connected.

[0129] For example, the angle between any account information vector x and the business information vector y in the initial business recommendation matrix is ​​obtained, and the angle is 90 degrees.

[0130] Step 706 : Multiply the product of the modulus of each account information vector and the modulus of the service information vector by the cosine value corresponding to the included angle to obtain an initial service recommendation list.

[0131] Specifically, the cosine value of the angle between any account information vector and the business information vector is calculated, and then the product of the modulus of each account information vector and the modulus of the business information vector is multiplied by the cosine value of the angle to obtain the initial business recommendation list.

[0132] For example, based on the inner product of the account information vectors of each of the M accounts and the business information vectors of the N businesses, an initial business recommendation matrix is ​​obtained, where the values ​​in the dimension corresponding to each account in the initial business recommendation matrix are N predicted rating values. For example, the account information vectors of each of the M accounts are combined as row vectors into a matrix, and at the same time, the business information vectors of the N businesses are combined as column vectors into another matrix, and then the two matrices are subjected to a dot product operation to obtain the initial business recommendation matrix. In this way, the predicted rating values ​​corresponding to the M accounts and N businesses can be summarized and calculated to restore the matrix. Based on the ranking of the N predicted rating values, the recommended businesses are determined from the N businesses.

[0133] In this embodiment, by calculating the inner product between the account information vector of each historical account and the business information vector of each business, the predicted score value corresponding to each business in the initial business recommendation list can be calculated more accurately.

[0134] It should be understood that, although the steps in the flowcharts of the above embodiments are shown in sequence as indicated by the arrows, these steps are not necessarily performed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be performed in other orders. Moreover, at least a portion of the steps in the flowcharts of the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily performed at the same time, but can be performed at different times. The execution order of these steps or stages is not necessarily to be performed in sequence, but can be performed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0135] Based on the same inventive concept, embodiments of the present application also provide a service recommendation device for implementing the aforementioned service recommendation method. The solution provided by this device is similar to the solution described in the aforementioned method. Therefore, the specific limitations in one or more service recommendation device embodiments provided below can be found in the aforementioned limitations on the service recommendation method and will not be further elaborated here.

[0136] In one embodiment, Figure 9 As shown, a service recommendation device is provided, comprising: a historical account attribute information acquisition module, an account set preference information acquisition module, a target account set acquisition module, a target service preference information acquisition module, and a target service recommendation module, wherein:

[0137] A historical account attribute information acquisition module 902 is used to acquire at least two pieces of historical account attribute information. The historical account attribute information is inherent information corresponding to an account that has performed business operations.

[0138] An account set preference information obtaining module 904 is configured to classify accounts corresponding to at least two pieces of historical account attribute information to obtain classified account sets, calculate business preference information based on the classified account sets, and obtain account set preference information corresponding to each classified account set;

[0139] The target account set acquisition module 906 is configured to obtain new account attribute information and compare the new account attribute information with the attribute information corresponding to each classified account set to obtain the target account set. The similarity between the attribute information corresponding to the target account set and the new account attribute information satisfies a preset condition.

[0140] The target service preference information obtaining module 908 is configured to adjust the preference information in the initial service recommendation list based on the attribute information corresponding to the target account set and the account set preference information corresponding to the classified account set, thereby obtaining the preference information of the new account for each target service in the initial service recommendation list;

[0141] The target service recommendation module 910 is used to recommend a target service list corresponding to the new account based on the preference information of each target service; the target service recommendation list is used to recommend services to the new account.

[0142] In one embodiment, the module for obtaining the preference information of the target business is further used to match the attribute information corresponding to the target account set with the attribute information corresponding to each classified account set to obtain each attribute information matching result; multiply each attribute information matching result with the account set preference information corresponding to the matching classified account set to obtain each intermediate preference information; and sum up each intermediate preference information to obtain the preference information of the new account for each target business in the initial business recommendation list.

[0143] In one embodiment, the module for obtaining the preference information of the target business is further used to calculate the intersection of the attribute information corresponding to the target account set and the attribute information corresponding to each classified account set, and then take the absolute value to obtain the intersection of the attribute information; calculate the union of the attribute information corresponding to the target account set and the attribute information corresponding to each classified account set, and then take the absolute value to obtain the union of the attribute information; perform division calculation on the intersection of the attribute information and the corresponding union of the attribute information to obtain the matching results of each attribute information.

[0144] In one embodiment, the account set preference information obtaining module is further used to obtain the number of historical business operations corresponding to each classified account set by summing the behavioral attribute information corresponding to each classified account set, where the behavioral attribute information is the historical account attribute information corresponding to the business operations on the historical accounts; the number of historical business operations corresponding to each classified account set is divided by the number of historical operation accounts corresponding to each classified account set to obtain the account set preference information corresponding to each classified account set; the number of historical operation accounts is the number of accounts in each classified account set that have performed the corresponding business operation.

[0145] In one embodiment, the module for obtaining the initial business recommendation list is also used to establish an initial business recommendation matrix by using the account information in the historical account attribute information as the rows of the matrix and the business information as the columns of the matrix; and to obtain the initial business recommendation list by multiplying the account information vector corresponding to each element in the initial business recommendation matrix with the business information vector.

[0146] In one embodiment, the module for obtaining the initial business recommendation list is further used to perform modulus calculations on the account information vector and the business information vector corresponding to each element in the initial business recommendation matrix to obtain the modulus of each account information vector and the modulus of the business information vector; obtain the angle between the account information vector and the business information vector corresponding to each element in the initial business recommendation matrix; multiply the product of the modulus of each account information vector and the modulus of the business information vector by the cosine value corresponding to the angle to obtain the initial business recommendation list.

[0147] Each module in the aforementioned business recommendation device may be implemented in whole or in part through software, hardware, or a combination thereof. Each module may be embedded in or independent of a processor in a computer device in the form of hardware, or may be stored in a memory in the computer device in the form of software, so that the processor can call and execute the corresponding operations of each module.

[0148] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 10As shown. The computer device includes a processor, a memory, and a network interface connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store server data. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a service recommendation method is implemented.

[0149] Those skilled in the art will understand that Figure 10 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0150] In one embodiment, a computer device is further provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps in the above method embodiments when executing the computer program.

[0151] In one embodiment, a computer-readable storage medium is provided, storing a computer program, which implements the steps in the above-mentioned method embodiments when executed by a processor.

[0152] In one embodiment, a computer program product or computer program is provided, the computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the steps of each of the above-described method embodiments.

[0153] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0154] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processor involved in the various embodiments provided herein may be, but are not limited to, a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, and the like.

[0155] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0156] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A business recommendation method, characterized in that: The method comprises: Obtain at least two pieces of historical account attribute information, where the historical account attribute information is inherent information corresponding to an account that has performed business operations; Classifying the accounts corresponding to at least two pieces of the historical account attribute information to obtain a classified account set; Based on the historical business operation data corresponding to each classified account set, the account set preference information of each classified account set for each business is calculated. The account set preference information is calculated using the following formula: in, Indicates the number of operations on the business by the i-th historical account. Indicates the total number of historical accounts that have performed operations on this business; Acquire new account attribute information, compare the new account attribute information with the attribute information corresponding to each of the classified account sets to obtain a target account set; the similarity between the attribute information corresponding to the target account set and the new account attribute information meets a preset condition; Matching the attribute information corresponding to the target account set with the attribute information corresponding to each of the classified account sets to obtain attribute information matching results; Multiplying each attribute information matching result by the account set preference information corresponding to the matched classified account set to obtain each intermediate preference information; performing a sum operation on each of the intermediate preference information to obtain the preference information of the new account for each target service in the initial service recommendation list; A target service recommendation list corresponding to the new account is prepared based on the preference information of each target service; the target service recommendation list is used to recommend services to the new account.

2. The method according to claim 1, characterized in that The matching of the attribute information corresponding to the target account set with the attribute information corresponding to each of the classified account sets to obtain attribute information matching results includes: Calculating the intersection of the attribute information corresponding to the target account set and the attribute information corresponding to each of the classified account sets and taking the absolute value to obtain the attribute information intersection; Calculating a union of the attribute information corresponding to the target account set and the attribute information corresponding to each of the classified account sets, and taking the absolute value to obtain a union of the attribute information; A division calculation is performed on the intersection of the attribute information and the corresponding union of the attribute information to obtain the attribute information matching results.

3. The method according to claim 1, characterized in that The calculating of the business preference information based on the classified account sets to obtain the account set preference information corresponding to each of the classified account sets includes: Summing the behavior attribute information corresponding to each of the classified account sets, obtaining the number of historical business operations corresponding to each of the classified account sets, wherein the behavior attribute information is the historical account attribute information corresponding to the business operations on the historical accounts; A division operation is performed on the number of historical business operations corresponding to each of the classified account sets and the number of historical operation accounts corresponding to each of the classified account sets to obtain account set preference information corresponding to each of the classified account sets; the number of historical operation accounts is the number of accounts in each of the classified account sets that have performed the corresponding business operation.

4. The method according to any one of claims 1 to 3, characterized in that The method further comprises: Using the account information in the historical account attribute information as the rows of a matrix and the business information as the columns of the matrix to establish an initial business recommendation matrix; The initial service recommendation list is obtained by multiplying the account information vector and the service information vector corresponding to each element in the initial service recommendation matrix.

5. The method according to claim 4, characterized in that The multiplication of the account information vector corresponding to each element in the initial service recommendation matrix and the service information vector to obtain the initial service recommendation list includes: Performing a modulus calculation on the account information vector and the business information vector corresponding to each element in the initial business recommendation matrix to obtain the modulus of each account information vector and the modulus of the business information vector; Obtaining the angle between the account information vector and the business information vector corresponding to each element in the initial business recommendation matrix; The product of the module of each of the account information vectors and the module of the service information vector is multiplied by the cosine value corresponding to the angle to obtain the initial service recommendation list.

6. A business recommendation device, characterized in that: The device comprises: A historical account attribute information acquisition module is used to acquire at least two pieces of historical account attribute information, wherein the historical account attribute information is inherent information corresponding to an account that has performed business operations; An account set preference information obtaining module, configured to classify the accounts corresponding to at least two of the historical account attribute information to obtain a classified account set; Based on the historical business operation data corresponding to each classified account set, the account set preference information of each classified account set for each business is calculated. The account set preference information is calculated using the following formula: in, Indicates the number of operations on the business by the i-th historical account. Indicates the total number of historical accounts that have performed operations on this business; a target account set obtaining module, configured to obtain new account attribute information, compare the new account attribute information with the attribute information corresponding to each of the classified account sets to obtain a target account set; and wherein the similarity between the attribute information corresponding to the target account set and the new account attribute information satisfies a preset condition; a module for obtaining target service preference information, configured to match the attribute information corresponding to the target account set with the attribute information corresponding to each of the classified account sets to obtain attribute information matching results; multiply each of the attribute information matching results by the account set preference information corresponding to the matching classified account set to obtain intermediate preference information; and sum each of the intermediate preference information to obtain the new account's preference information for each target service in the initial service recommendation list; The target service recommendation module is configured to create a target service recommendation list corresponding to the new account based on the preference information of each target service; the target service recommendation list is used to recommend services to the new account.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

9. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images