Non-resident user product recommendation method and device, electronic equipment and storage medium

Abstract

The invention provides a non-resident user product recommendation method, which can be used in the technical field of big data, and comprises the following steps: generating a non-resident user portrait according to non-resident user information, the non-resident user portrait comprising a non-resident user feature tag; determining a target service according to a statistical result of non-residential user feature tags included in all non-residential user portraits of the first target region; determining a target non-resident user from non-resident users in the first target region according to the service requirement feature information of the target service; determining a target resident user from the resident users in the second target region according to the target non-resident user; inputting the target non-resident user information, the target resident user information and product information in a product library into a mixed recommendation model to generate target product recommendation information; and recommending the target product to the target non-resident user according to the target product recommendation information. The invention further provides a product recommendation device, electronic equipment and a storage medium.

Description

Technical Field

[0001] The present disclosure relates to the field of big data technology, and more specifically, to a product recommendation method, device, electronic device, computer-readable storage medium, and computer program product for non-residential users. Background Art

[0002] The users of financial industries such as banks, securities, and funds are very extensive, including domestic resident users and foreign non-resident users. When foreign non-resident users need to handle related business before entering China, they cannot accurately express the required products due to language barriers and unfamiliarity with the business and products they need to handle. Different foreign non-resident users face many choices and cannot accurately understand them one by one, which leads to users being unable to obtain the most suitable business or product recommendations, resulting in poor user experience, inability to effectively maintain user use, and slow business growth. Summary of the invention

[0003] In view of the above problems, the present disclosure provides a product recommendation method, device, electronic device, readable storage medium and computer program product for non-residential users, which can automatically make accurate service recommendations for non-residential users, meet the needs of different non-residential users, effectively attract users and improve user experience.

[0004] One aspect of the present disclosure provides a product recommendation method for non-residential users, the method including but not limited to: generating a non-residential user portrait based on non-residential user information of non-residential users, the non-residential user portrait including a non-residential user feature tag; obtaining all non-residential user portraits in a first target area, and determining a target business based on statistical results of the non-residential user feature tags included in all non-residential user portraits, the target business having business requirement feature information; determining a target non-residential user from the non-residential users in the first target area based on the business requirement feature information; determining a target resident user from resident users in a second target area based on the target non-residential user; inputting the target non-residential user information of the target non-residential user, the target resident user information of the target resident user, and the product information in a product library into a hybrid recommendation model to generate target product recommendation information, wherein the hybrid recommendation model is trained using the non-residential user information, resident user information, and product information in a product library included in a data set to train the hybrid recommendation model to be trained; and recommending a target product to the target non-residential user based on the target product recommendation information.

[0005] In some embodiments of the present disclosure, a non-resident user portrait is generated based on the non-resident user information of the non-resident user, including: obtaining non-resident attribute information input by the non-resident user, and generating first non-resident user information including a first non-resident tag; obtaining product behavior information of the non-resident user, and generating second non-resident user information including a second non-resident tag; obtaining product association information input by the non-resident user, and generating third non-resident user information including a third non-resident tag; generating a non-resident user portrait based on the first non-resident user information, the second non-resident user information and the third non-resident user information, the user portrait including a non-resident user feature tag generated based on the first non-resident tag, the second non-resident tag and the third non-resident tag.

[0006] In some embodiments of the present disclosure, target non-residential users are determined from the non-residential users in the first target area based on the business requirement characteristic information, including: screening non-residential user characteristic tags included in all non-residential user portraits of the first target area based on characteristic tags corresponding to the business requirement characteristic information; judging whether the characteristic information corresponding to the non-residential user characteristic tags of the screened non-residential users meets the business requirement characteristic information, and if so, determining the non-residential users as target non-residential users.

[0007] In some embodiments of the present disclosure, target resident users are determined from resident users in a second target area based on the target non-resident users, including: matching the non-resident user profile of the target non-resident user with the resident user profile of the resident users in the second target area to generate a matching value; and determining the resident user whose matching value meets a preset matching value as the target resident user.

[0008] In some embodiments of the present disclosure, the hybrid recommendation model is constructed by a user-based collaborative filtering algorithm and a content-based collaborative filtering algorithm.

[0009] In some embodiments of the present disclosure, a hybrid recommendation model to be trained is trained using non-residential user information, resident user information and product information in a product library included in a data set, including: inputting the non-residential user information and resident user information in the data set into a user-based collaborative filtering algorithm in the hybrid recommendation model to be trained to determine first product recommendation information; inputting the non-residential user information in the data set and the product information in the product library into a content-based collaborative filtering algorithm in the hybrid recommendation model to be trained to determine second product recommendation information; generating target product recommendation information based on the first product recommendation information and the second product recommendation information; adjusting the parameters of the user-based collaborative filtering algorithm and the content-based collaborative filtering algorithm according to an evaluation index between the target product recommendation information and preset product recommendation information until a training termination condition is met to obtain the hybrid recommendation model.

[0010] In some embodiments of the present disclosure, the target non-resident user information of the target non-resident user, the target resident user information of the target resident user, and the product information in the product library are input into the hybrid recommendation model to generate target product recommendation information, including: inputting the target non-resident user information of the target non-resident user and the target resident user information of the target resident user into the user-based collaborative filtering algorithm in the hybrid recommendation model, determining target similar resident users from the target resident users, the similarity value between the target similar resident users and the target non-resident users meets a first preset similarity value, and the target similar resident users have associated product information; determining first product recommendation information based on the associated product information; determining product recommendation feature information based on the target non-resident user information of the target non-resident user; inputting the product recommendation feature information and the product information in the product library into the collaborative filtering algorithm of the opportunity product in the hybrid recommendation model, determining target similar products from the product information in the product library, the similarity value between the product feature information of the target similar products and the product recommendation feature information meets a second preset similarity value; determining second product recommendation information based on the target similar products; generating target product recommendation information based on the first product recommendation information and the second product recommendation information.

[0011] In some embodiments of the present disclosure, the method further includes: before determining the first product recommendation information based on the associated product information, determining whether the associated product information meets the target non-residential user usage conditions; if so, determining the first product recommendation information based on the associated product information; if not, obtaining associated product features of the associated product information, and generating update information based on the associated product features, wherein the update information is used to update the product information in the product library.

[0012] In another aspect of the present disclosure, a product recommendation device for non-resident users is provided, comprising: a first generation module, configured to generate a non-resident user portrait based on non-resident user information of the non-resident user, wherein the non-resident user portrait includes a non-resident user feature tag; a first determination module, configured to obtain all non-resident user portraits in a first target area, and determine a target business based on statistical results of the non-resident user feature tags included in all non-resident user portraits, wherein the target business has business requirement feature information; a second determination module, configured to determine a target non-resident user from the non-resident users in the first target area based on the business requirement feature information; and a third determination module, configured to determine a target non-resident user from the non-resident users in the first target area. A determination module is configured to determine target resident users from resident users in a second target area based on the target non-resident users; a second generation module is configured to input target non-resident user information of the target non-resident users, target resident user information of the target resident users and product information in a product library into a hybrid recommendation model to generate target product recommendation information, wherein the hybrid recommendation model is obtained by training the hybrid recommendation model to be trained using the non-resident user information, resident user information and product information in a product library included in a data set; and a recommendation module is configured to recommend a target product to the target non-resident user based on the target product recommendation information.

[0013] In some embodiments of the present disclosure, the first generation module includes a first generation sub-module, which is configured to: obtain non-resident attribute information input by a non-resident user, and generate first non-resident user information including a first non-resident tag; obtain product behavior information of the non-resident user, and generate second non-resident user information including a second non-resident tag; obtain product association information input by the non-resident user, and generate third non-resident user information including a third non-resident tag; generate a non-resident user portrait based on the first non-resident user information, the second non-resident user information and the third non-resident user information, the user portrait including a non-resident user feature tag generated based on the first non-resident tag, the second non-resident tag and the third non-resident tag.

[0014] In some embodiments of the present disclosure, the second determination module includes a second determination submodule, configured to: screen the non-resident user characteristic tags included in all non-resident user portraits of the first target area according to the characteristic tags corresponding to the business requirement characteristic information; determine whether the characteristic information corresponding to the non-resident user characteristic tags of the screened non-resident users meets the business requirement characteristic information, and if so, determine the non-resident users as target non-resident users.

[0015] In some embodiments of the present disclosure, the third determination module includes a third determination sub-module, which is configured to: match the non-resident user profile of the target non-resident user with the resident user profile of the resident user in the second target area to generate a matching value; and determine the resident user whose matching value meets a preset matching value as the target resident user.

[0016] In some embodiments of the present disclosure, the device also includes a training module, configured to: input the non-resident user information and the resident user information in the data set into the user-based collaborative filtering algorithm in the hybrid recommendation model to be trained to determine the first product recommendation information; input the non-resident user information in the data set and the product information in the product library into the content-based collaborative filtering algorithm in the hybrid recommendation model to be trained to determine the second product recommendation information; generate target product recommendation information based on the first product recommendation information and the second product recommendation information; adjust the parameters of the user-based collaborative filtering algorithm and the content-based collaborative filtering algorithm according to the evaluation index between the target product recommendation information and the preset product recommendation information until the training termination condition is met to obtain the hybrid recommendation model.

[0017] In some embodiments of the present disclosure, the second generation module includes a second generation submodule, which is configured to: input the target non-resident user information of the target non-resident user and the target resident user information of the target resident user into the user-based collaborative filtering algorithm in the hybrid recommendation model, determine the target similar resident user from the target resident user, the similarity value between the target similar resident user and the target non-resident user satisfies the first preset similarity value, and the target similar resident user has associated product information; determine the first product recommendation information based on the associated product information; determine the product recommendation feature information based on the target non-resident user information of the target non-resident user; input the product recommendation feature information and the product information in the product library into the collaborative filtering algorithm of the opportunity product in the hybrid recommendation model, determine the target similar product from the product information in the product library, the similarity value between the product feature information of the target similar product and the product recommendation feature information satisfies the second preset similarity value; determine the second product recommendation information based on the target similar product; generate the target product recommendation information based on the first product recommendation information and the second product recommendation information.

[0018] In some embodiments of the present disclosure, the device also includes a judgment module, configured to: before determining the first product recommendation information based on the associated product information, determine whether the associated product information meets the target non-residential user usage conditions; if so, determine the first product recommendation information based on the associated product information; if not, obtain associated product characteristics of the associated product information, and generate update information based on the associated product characteristics, and the update information is used to update the product information in the product library.

[0019] In another aspect of the present disclosure, an electronic device is provided, comprising: one or more processors; a storage device for storing one or more computer programs, wherein the one or more processors execute the one or more computer programs to implement the steps of the method described above.

[0020] In another aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method described above are implemented.

[0021] In another aspect of the present disclosure, a computer program product is provided, comprising a computer program, wherein when the computer program is executed by a processor, the steps of the method described above are implemented.

[0022] According to the embodiments of the present disclosure, by obtaining the portraits of all non-user residents in the first target area and determining the target business, it is possible to more accurately determine the business content based on the non-resident user information, and effectively improve the promotion effect of the business. By determining the target non-resident users according to the business requirement characteristics, the implementation objects of the business content can be accurately screened out, and the precise delivery of the target business can be maximized. Further inputting the target non-resident user information, the target resident user information and the product information into the trained hybrid recommendation model can improve the recommendation accuracy and coverage of the target product, make up for the limitations of a single algorithm, and enhance the ability of personalized recommendation. At the same time, using the target resident user information as the input of the hybrid recommendation model can recommend products to non-resident users in combination with the preferences and behaviors of resident users, covering more recommendation blind spots, and adapting to different needs and scenarios. The product recommendation algorithm for non-resident users of the embodiments of the present disclosure can change the state of blindly recommending products, improve the accuracy and recommendation range of product recommendations, effectively improve the user experience and improve the efficiency of business recommendations. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The above contents and other objects, features and advantages of the present disclosure will become more apparent through the following description of the embodiments of the present disclosure with reference to the accompanying drawings, in which:

[0024] Figure 1A schematic diagram schematically shows a system architecture to which the product recommendation method for non-resident users according to an embodiment of the present disclosure can be applied;

[0025] Figure 2 A flowchart schematically shows a product recommendation method for non-residential users according to an embodiment of the present disclosure;

[0026] Figure 3 A flowchart of a product recommendation method for non-resident users in operation S210 according to an embodiment of the present disclosure is schematically shown;

[0027] Figure 4 A flowchart of a product recommendation method for non-resident users in operation S230 according to an embodiment of the present disclosure is schematically shown;

[0028] Figure 5 A flowchart of a product recommendation method for non-resident users in operation S240 according to an embodiment of the present disclosure is schematically shown;

[0029] Figure 6 A flowchart of a product recommendation method for non-resident users according to an embodiment of the present disclosure in training a hybrid recommendation model is schematically shown;

[0030] Figure 7 A flowchart of a product recommendation method for non-resident users in operation S250 according to an embodiment of the present disclosure is schematically shown;

[0031] Figure 8 A flowchart of a product recommendation method for non-resident users according to an embodiment of the present disclosure before determining first product recommendation information is schematically shown;

[0032] Fig. 9 A block diagram schematically shows a product recommendation device for non-residential users according to an embodiment of the present disclosure; and

[0033] Fig.10 A block diagram of an electronic device according to an embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0034] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present disclosure. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of the embodiments of the present disclosure. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of the concepts of the present disclosure.

[0035] The terms used herein are only for describing specific embodiments and are not intended to limit the present disclosure. The terms "comprise", "include", etc. used herein indicate the existence of features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.

[0036] All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification, and should not be interpreted in an idealized or overly rigid manner.

[0037] When using expressions such as "at least one of A, B or C, etc.", it should generally be interpreted in accordance with the meaning of the expression generally understood by those skilled in the art (for example, "a system having at least one of A, B or C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.). The terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more features.

[0038] In the technical solution of the present disclosure, the user information (including but not limited to user personal information, user image information, user device information, such as location information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved are all information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with relevant laws, regulations and standards, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0039] In the scenario of using personal information for automated decision-making, the methods, devices, and systems provided by the embodiments of the present invention provide users with corresponding operation portals for users to choose to agree or reject the automated decision-making results; if the user chooses to reject, the expert decision-making process will be entered. The expression "automated decision-making" here refers to the activity of automatically analyzing and evaluating an individual's behavioral habits, interests and hobbies, or economic, health, credit status, etc. through computer programs, and making decisions. The expression "expert decision-making" here refers to the activity of making decisions by people who specialize in a certain field, have specialized experience, knowledge and skills, and have reached a certain level of professionalism.

[0040] In this article, the term "non-resident user" refers to a foreign person who does not have resident status. The term "non-resident user information" refers to a variety of information related to non-resident users, such as the attribute information, behavior information, and product-related information of foreign persons. The term "resident user" refers to a domestic person with resident status. The term "resident user information" refers to a variety of information related to resident users, such as the attribute information, behavior information, and product-related information of domestic persons.

[0041] The existing financial industry includes various financial products, one for domestic resident users and the other for overseas non-resident users. Due to the different operational strategies and requirements of domestic resident users and overseas non-resident users, there are certain differences and differences between the products of resident users and non-resident users. Products applied to resident users cannot be directly and completely applied to non-resident users. When non-resident users overseas apply for or choose the required products, due to language barriers, they are unfamiliar with the various contents of the products and cannot accurately express the required products. Due to nationality, language, etc., non-resident users overseas require a large variety of products. Service personnel for non-resident users need to master the characteristics of multiple products and have high professional requirements, which results in non-resident users being unable to accurately obtain the required products. At the same time, the experience of non-resident users is poor, and it is impossible to effectively maintain the long-term use of users, resulting in customer loss, slow business growth, and even business decline.

[0042] In order to solve the above-mentioned problems, the embodiments of the present disclosure provide a product recommendation method, device, electronic device, computer-readable storage medium and computer program product for non-residential users, which can automatically make accurate service recommendations for non-residential users, meet the needs of different non-residential users, effectively attract users and improve user experience.

[0043] The product recommendation method for non-residential users of the disclosed embodiment includes but is not limited to: generating a non-residential user portrait based on non-residential user information of the non-residential user, the non-residential user portrait including a non-residential user feature tag; obtaining all non-residential user portraits in the first target area, and determining a target business based on statistical results of the non-residential user feature tags included in all non-residential user portraits, the target business having business requirement feature information; determining a target non-residential user from non-residential users in the first target area based on the business requirement feature information; determining a target resident user from resident users in the second target area based on the target non-residential user; inputting target non-residential user information of the target non-residential user, target resident user information of the target resident user, and product information in a product library into a hybrid recommendation model to generate target product recommendation information, wherein the hybrid recommendation model is trained using the non-residential user information, resident user information, and product information in a product library included in a data set to train the hybrid recommendation model to be trained; and recommending a target product to the target non-residential user based on the target product recommendation information.

[0044] According to the embodiments of the present disclosure, by obtaining the portraits of all non-user residents in the first target area and determining the target business, it is possible to more accurately determine the business content based on the non-resident user information, and effectively improve the promotion effect of the business. By determining the target non-resident users according to the business requirement characteristics, the implementation objects of the business content can be accurately screened out, and the precise delivery of the target business can be maximized. Further inputting the target non-resident user information, the target resident user information and the product information into the trained hybrid recommendation model can improve the recommendation accuracy and coverage of the target product, make up for the limitations of a single algorithm, and enhance the ability of personalized recommendation. At the same time, using the target resident user information as the input of the hybrid recommendation model can recommend products to non-resident users in combination with the preferences and behaviors of resident users, covering more recommendation blind spots, and adapting to different needs and scenarios. The product recommendation algorithm for non-resident users of the embodiments of the present disclosure can change the state of blindly recommending products, improve the accuracy and recommendation range of product recommendations, effectively improve the user experience and improve the efficiency of business recommendations.

[0045] Figure 1 The schematic diagram schematically shows the system architecture of the product recommendation method for non-resident users of the embodiment of the present disclosure. It should be noted that: Figure 1What is shown is only an example of a system architecture to which the embodiments of the present disclosure can be applied, in order to help those skilled in the art understand the technical content of the present disclosure, but it does not mean that the embodiments of the present disclosure cannot be used in other devices, systems, environments or scenarios. It should be noted that the product recommendation method for non-resident users provided by the embodiments of the present disclosure can be used in the field of big data technology, related aspects of the financial field in the field of big data, and can also be used in any field other than the financial field. The product recommendation method and device for non-resident users provided by the embodiments of the present disclosure do not limit the application field.

[0046] like Figure 1 As shown, an exemplary system architecture 100 to which the product recommendation method for non-residential users can be applied may include terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 is used to provide a medium for a communication link between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0047] Users can use terminal devices 101, 102, 103 to interact with server 105 through network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, 103, such as email client applications, file processing applications, shopping applications, web browser applications, search applications, instant messaging tools, mailbox clients, social platform software, etc. (only as examples).

[0048] The terminal devices 101, 102, 103 may be various electronic devices having display screens and supporting functions such as data input, file sending, data analysis, data processing, web browsing, etc., including but not limited to smart phones, tablet computers, laptop computers, desktop computers, etc.

[0049] The server 105 may be a server that provides various services, such as a background management server that provides support for the data obtained or websites browsed by the user using the terminal devices 101, 102, and 103 (for example only). The background management server may analyze and process the received data such as user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to the user request) to the terminal device. It may also analyze or process the files sent by the user, and control the terminal device based on the processing results, such as limiting the access of the terminal device.

[0050] It should be noted that the product recommendation method for non-residential users provided in the embodiment of the present disclosure can generally be executed by terminal devices 101, 102, 103, and server 105. Accordingly, the product recommendation device for non-residential users provided in the embodiment of the present disclosure can generally be set in terminal devices 101, 102, 103, and server 105. The product recommendation method for non-residential users provided in the embodiment of the present disclosure can also be executed by a terminal device that is different from the terminal devices 101, 102, 103 and can communicate with the terminal devices 101, 102, 103. Accordingly, the product recommendation device for non-residential users provided in the embodiment of the present disclosure can also be set in a terminal device that is different from the terminal devices 101, 102, 103 and can communicate with the terminal devices 101, 102, 103.

[0051] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0052] The following will be passed Figures 2 to 8 The product recommendation method for non-residential users in an embodiment of the present disclosure is described in detail.

[0053] Figure 2 A flowchart of a product recommendation method for non-residential users according to an embodiment of the present disclosure is schematically shown.

[0054] like Figure 2 As shown, process 200 of the product recommendation method for non-residential users includes operations S210 to S260.

[0055] In operation S210, a non-resident user portrait is generated according to the non-resident user information of the non-resident user, and the non-resident user portrait includes a non-resident user feature tag.

[0056] In an embodiment of the present disclosure, the non-residential user information of a non-residential user may include information input by the non-residential user on a terminal device, and may also include information obtained through the operating behavior of the non-residential user when the non-residential user operates the terminal device, such as obtaining the operating behavior of the non-residential user by burying points in the terminal, and may also include information obtained by analyzing information input by the non-residential user on the terminal device.

[0057] Non-resident user portraits include non-resident user feature tags, such as user ID type, user nationality, user behavior, estimated entry time, user risk level, user activity, user income range, user interests and hobbies, user transaction history, etc. Non-resident user feature tags can be used to classify and count non-resident users.

[0058] By processing the non-resident user information and generating a non-resident user portrait, it is convenient to analyze each non-resident user and accurately determine the characteristic label of each non-resident user. In the subsequent product recommendation process, processing based on the non-resident user characteristic label can improve the accuracy of product recommendations on the one hand, so that different product recommendation information can be provided for different non-resident users, effectively improving the experience of non-resident users. On the other hand, based on the non-resident user portrait, the overall situation of the current non-resident users can be accurately analyzed, so as to accurately determine the target business, facilitate recommending products to non-resident users according to the target business, and effectively promote business growth.

[0059] In operation S220, all non-resident user portraits in the first target area are obtained, and a target business is determined based on statistical results of non-resident user feature tags included in all non-resident user portraits, where the target business has business requirement feature information.

[0060] The first target region refers to a region where accurate business analysis needs to be performed, for example, the first target region is the X province region of country A. Another example is the entire territory of country A.

[0061] According to operation S221, each non-resident user has a corresponding non-resident user portrait. By acquiring all non-resident user portraits in the first target area, it is convenient to analyze all non-resident users.

[0062] In an embodiment of the present disclosure, after all non-resident user portraits in the first target area are acquired, statistical analysis is performed on the non-resident user portraits.

[0063] For example, statistics are collected on the distribution and account movement of non-resident users in the first target area, and a visualization is generated. Account movement refers to the operations of non-resident users on their accounts, such as collection and payment. Specifically, based on the statistics, the distribution of non-resident users in each province in the first target area and the distribution of the number of users in each branch institution in the top 10 provinces in terms of the number of non-resident user customers are generated; based on the statistics, the growth and account movement of non-resident users in different time dimensions are generated, for example, the month-on-month growth of non-resident users (corresponding to the number of non-resident users) / month-on-month account movement (corresponding to the number of non-resident users moving) is displayed in the form of a heat map, and the growth and number of non-resident users in three dimensions of day, month and year are provided in the form of data on the data dashboard. The number of existing non-resident users reflects the total number of non-resident users as of the current time (data validity t-2 days).

[0064] For example, a non-resident user growth trend chart is generated to support viewing the growth trend of non-resident users of each institution within up to 12 time units in three dimensions: day, month, and year. The recent growth of non-resident users can be observed through the trend chart, and the cyclical law of customer acquisition can be captured from the trend, and the target business can be determined based on this law.

[0065] For example, a Venn diagram of user movement channels is generated, and the user movement status of non-resident users in each channel is queried according to the three dimensions of the day, month and year (data time limit t-2 days). The overlapping part is the number of non-resident users who have moved their users in two channels at the same time. By tracking the dynamic changes in the proportion of the overlapping part of the Venn diagram, it is possible to preliminarily observe whether the business diversion goal between channels has been successfully achieved, thereby redefining or adjusting the target business.

[0066] For example, a business data statistics table is generated to query the total business volume, total transaction volume and total transaction amount, foreign currency deposit balance and other information of non-resident users by channel according to the three dimensions of the day, month and year (data time limit t-2 days), and display it in multiple currencies.

[0067] In the embodiments of the present disclosure, according to the statistical results of the non-residential user feature labels included in all non-residential user portraits, the target business can be determined, for example, by analyzing the statistical results through an analysis model, or by processing the statistical results through a neural network. In an optional embodiment, the target business can also be manually determined by business personnel based on observations of the statistical results.

[0068] The target business has business requirement characteristic information, which can be used to screen out qualified non-residential users from all non-residential users, accurately locate the scope of product recommendations, improve the accuracy and success rate of product recommendations, and improve efficiency.

[0069] In operation S230, target non-residential users are determined from non-residential users in the first target area according to the service requirement characteristic information.

[0070] In the embodiment of the present disclosure, there are a large number of non-residential users in the first target area. If recommendations are made for all non-residential users, it is impossible to accurately make recommendations for different non-residential users, resulting in poor recommendation effects and poor non-residential user experience.

[0071] Through operation S230, non-residential users who meet the target service recommendation requirements can be accurately determined, thereby improving the accuracy of the recommendation. The process of operation S230 will be described in detail below.

[0072] In operation S240, target residential users are determined from residential users in the second target area according to the target non-residential users.

[0073] In the embodiments of the present disclosure, there are certain differences and similarities between residential users and non-residential users. The resident user information of the residential users can be used as an association for recommending products to non-residential users, which can cover a wider recommendation blind spot and improve the accuracy of recommendations for non-residential users.

[0074] In operation S250, target non-residential user information of target non-residential users, target resident user information of target resident users, and product information in the product library are input into the hybrid recommendation model to generate target product recommendation information, wherein the hybrid recommendation model is trained using the non-residential user information, resident user information, and product information in the product library included in the data set to train the hybrid recommendation model to be trained.

[0075] In the embodiment of the present disclosure, the hybrid recommendation model is constructed by the collaborative filtering algorithm based on users and the collaborative filtering algorithm based on content. The hybrid recommendation model utilizes different types of recommendation algorithms, can combine the advantages of multiple algorithms, make up for the limitations of a single algorithm, and improve the accuracy and coverage of recommendations.

[0076] In operation S260, a target product is recommended to a target non-resident user according to the target product recommendation information.

[0077] After generating the target product recommendation information, the target product is recommended to the target non-residential user. For example, the target product can be a part selected from the target product recommendation information for display, and can be adjusted according to the display interface of the device used by the user, thereby further improving the user experience.

[0078] Combine the following Figures 3 to 8 The specific process of the product recommendation method for non-residential users in the embodiment of the present disclosure is described in detail.

[0079] Figure 3 The flowchart of the product recommendation method for non-residential users in operation S210 according to an embodiment of the present disclosure is schematically shown.

[0080] like Figure 3 As shown, operation S210 includes operations S211 to S214.

[0081] In operation S211, non-resident attribute information input by a non-resident user is obtained, and first non-resident user information including a first non-resident tag is generated.

[0082] Exemplarily, the non-resident attribute information may be the certificate type information input by the non-resident user, or the nationality information of the non-resident user. The first non-resident tag is a tag related to the non-resident attribute information, for example, the first non-resident tag includes tags such as certificate type and nationality.

[0083] In operation S212, product behavior information of the non-resident user is obtained, and second non-resident user information including a second non-resident tag is generated.

[0084] For example, when non-residential users select products, they will check, browse, purchase and perform other operations on the products. The product behavior information of non-residential users on product operations is obtained by embedding points on non-residential user terminals, and the obtained product behavior information is screened, standardized and other processes are performed to generate second non-residential user information.

[0085] In operation S213, the product association information input by the non-resident user is obtained, and third non-resident user information including a third non-resident tag is generated.

[0086] For example, product association information refers to information associated with the product to be selected by the non-resident user, such as the user risk level, that is, the risk level of the product that the non-resident user can accept. Based on the product association information, the product that satisfies the non-resident user can be determined, thereby filtering out other products, which is conducive to improving the accuracy of product recommendations in the subsequent product recommendation process.

[0087] In the embodiments of the present disclosure, the product association information may be directly input by the non-resident user, or may be obtained by analyzing other information input by the non-resident user.

[0088] In operation S214, a non-resident user portrait is generated based on the first non-resident user information, the second non-resident user information, and the third non-resident user information, and the user portrait includes a non-resident user feature tag generated based on the first non-resident tag, the second non-resident tag, and the third non-resident tag.

[0089] By analyzing and processing the first non-resident user information, the second non-resident user information and the third non-resident user information, a non-resident user portrait is generated, and the various characteristics of the non-resident user can be accurately determined through the non-resident user portrait, thereby facilitating the analysis of non-resident users to determine target services and make product recommendations.

[0090] In the embodiments of the present disclosure, authorization of a non-resident user is obtained before obtaining the non-resident user information of the non-resident user, and after obtaining the authorization of the non-resident user, the non-resident user information of the non-resident user is obtained.

[0091] Figure 4 The flowchart of the product recommendation method for non-resident users in operation S230 according to an embodiment of the present disclosure is schematically shown.

[0092] like Figure 4 As shown, operation S230 includes operations S231 to S233.

[0093] In operation S231, non-resident user feature tags included in all non-resident user portraits in the first target area are screened according to feature tags corresponding to the business requirement feature information.

[0094] The business requirement characteristic information is the specific content corresponding to the characteristic tag. For example, if the characteristic tag is the transaction amount, the business requirement characteristic information is that the transaction amount is greater than 500,000.

[0095] In operation S232, it is determined whether the characteristic information corresponding to the non-resident user characteristic tag of the screened non-resident user meets the characteristic information required by the service.

[0096] After screening the non-resident user characteristic tags, it is further determined whether the characteristic information corresponding to the non-resident user characteristic tags meets the business requirement characteristic information. For example, if the business requirement characteristic information is that the transaction amount is greater than 500,000, if the characteristic information corresponding to the non-resident user characteristic tag is that the transaction amount is 300,000, then it is determined that the characteristic information corresponding to the non-resident user characteristic tag does not meet the business requirement characteristic information.

[0097] In operation S233, if the conditions are met, the non-resident user is determined as a target non-resident user.

[0098] According to the embodiments of the present disclosure, target non-resident users are screened out from non-resident users in the first target area, which can improve the accuracy of product recommendations.

[0099] Figure 5 The flowchart of the product recommendation method for non-residential users in operation S240 according to an embodiment of the present disclosure is schematically shown.

[0100] like Figure 5 As shown, operation S240 includes operation S241 to operation S242.

[0101] In operation S241, the non-resident user portrait of the target non-resident user is matched with the resident user portrait of the resident user in the second target area to generate a matching value.

[0102] In operation S242, a residential user whose matching degree value satisfies a preset matching degree value is determined as a target residential user.

[0103] In the embodiment of the present disclosure, the second target area may be the same area as the first target area, or may be a different area from the first target area.

[0104] In this embodiment, the second target area is the same as the first target area. By matching the non-resident user portrait of the target non-resident user with the resident user portrait of the resident users in the area, target resident users who are the same or similar to the non-resident users can be selected from the resident users. Products can be recommended to non-resident users based on the target resident users, which can further improve the accuracy of product recommendations. At the same time, product recommendation information is generated based on the relationship between resident users and non-resident users, which can cover recommendation blind spots.

[0105] Figure 6 The flowchart of the product recommendation method for non-resident users in training the hybrid recommendation model according to an embodiment of the present disclosure is schematically shown.

[0106] In an embodiment of the present disclosure, a hybrid recommendation model is constructed by a user-based collaborative filtering algorithm and a content-based collaborative filtering algorithm. The hybrid recommendation model is obtained by training the hybrid recommendation model to be trained using non-resident user information, resident user information and product information in the product library included in the data set.

[0107] like Figure 6 As shown, the process 300 of training the hybrid recommendation model includes operations S310 to S340.

[0108] In operation S310, non-resident user information and resident user information in the data set are input into a user-based collaborative filtering algorithm in a hybrid recommendation model to be trained to determine first product recommendation information.

[0109] In operation S320, the non-resident user information in the data set and the product information in the product library are input into a content-based collaborative filtering algorithm in the hybrid recommendation model to be trained to determine the second product recommendation information.

[0110] In operation S330, target product recommendation information is generated according to the first product recommendation information and the second product recommendation information.

[0111] In operation S340, parameters of the user-based collaborative filtering algorithm and the content-based collaborative filtering algorithm are adjusted according to the evaluation index between the target product recommendation information and the preset product recommendation information until the training termination condition is met to obtain a hybrid recommendation model.

[0112] In an embodiment of the present disclosure, during the training process of the hybrid recommendation model, the data set includes a training set and a validation set. During the training process, the hybrid model is trained using the non-resident user information, resident user information, and product information in the product library included in the training set to obtain the output of the hybrid model, i.e., the target product recommendation information. The target product recommendation information output by the model is then compared with the preset product recommendation information in the training set, and the parameters of the user-based collaborative filtering algorithm and the content-based collaborative filtering algorithm are adjusted according to an evaluation index (e.g., a loss function) between the two, so that the output result of the hybrid recommendation model is as close to the real output data as possible.

[0113] Exemplarily, the evaluation indicators include accuracy, recall, precision, and F1 score.

[0114] Figure 7 The flowchart of the product recommendation method for non-resident users in operation S250 according to an embodiment of the present disclosure is schematically shown.

[0115] like Figure 7 As shown, operation S250 includes operations S251 to S256.

[0116] In operation S251, the target non-resident user information of the target non-resident user and the target resident user information of the target resident user are input into the user-based collaborative filtering algorithm in the hybrid recommendation model, and the target similar resident user is determined from the target resident users. The similarity value between the target similar resident user and the target non-resident user satisfies the first preset similarity value, and the target similar resident user has associated product information.

[0117] For example, the user-based collaborative filtering algorithm can determine target similar residential users similar to target non-residential users from target residential users, that is, find users similar to target non-residential users from target residential users, thereby recommending products to target non-residential users based on target residential users.

[0118] The target similar residential users have associated product information, and the associated product information refers to the products purchased or used by the target similar residential users.

[0119] In operation S252, first product recommendation information is determined based on the associated product information.

[0120] For example, the first product recommendation information includes product name, recommendation time, etc. The first product recommendation information is determined based on the associated product information of the target similar user, and the first product recommendation information is ultimately used to generate a target recommended product for the target non-resident user.

[0121] In operation S253, product recommendation feature information is determined based on the target non-resident user information of the target non-resident user.

[0122] In the embodiment of the present disclosure, after the target non-residential users are determined, by analyzing the target resident user information of each target resident user, it is possible to determine which characteristics the products required by these target non-residential users have, that is, determine the product recommendation characteristic information. The product recommendation characteristic information is the characteristic information of the products required by the predicted target non-residential users.

[0123] In operation S254, the product recommendation feature information and the product information in the product library are input into the collaborative filtering algorithm of the opportunity product in the hybrid recommendation model, and the target similar product is determined from the product information in the product library. The similarity value between the product feature information of the target similar product and the product recommendation feature information satisfies the second preset similarity value.

[0124] According to the product recommendation feature information, target similar products can be determined from the product library, that is, the products closest to the products required by the target non-residential users can be determined.

[0125] In operation S255, second product recommendation information is determined according to the target similar product.

[0126] In an embodiment of the present disclosure, at least a part of the target similar products is used as the second product recommendation information to finally generate the target product recommendation information.

[0127] In operation S256, target product recommendation information is generated according to the first product recommendation information and the second product recommendation information.

[0128] The first product recommendation information and the second product recommendation information are integrated, such as weighted average, switching integration, etc., to generate target product recommendation information, so that the target non-resident user can obtain more accurate target product recommendation information.

[0129] In addition, the target product recommendation information is recommended to target non-residential users determined based on the target business, which can more accurately improve the conversion effect of the business and improve business efficiency.

[0130] Figure 8 The flowchart of the product recommendation method for non-resident users according to an embodiment of the present disclosure before determining the first product recommendation information is schematically shown.

[0131] like Figure 8 As shown, before determining the first product recommendation information according to the associated product information, a process 400 is also included, and the process 400 includes operation S410 and operation S420.

[0132] In operation S410, it is determined whether the associated product information meets the target non-resident user usage conditions. If so, operation S252 is performed, and if not, operation S420 is performed.

[0133] In an embodiment of the present disclosure, it is determined whether the associated product information satisfies the target non-resident user usage conditions, for example, it is determined whether the associated product information is a product available to the target non-resident user. If the associated product information of the resident user can be used by the target non-resident user, then the target non-resident user usage conditions are met, otherwise, they are not met.

[0134] In operation S420, the associated product features of the associated product information are obtained, and update information is generated according to the associated product features. The update information is used to update the product information in the product library.

[0135] According to operation S420, by generating update information, the product information in the product library can be updated or adjusted according to the needs of the target non-residential users, and ultimately the use experience of the target non-residential users can be improved.

[0136] Fig. 9 A block diagram of a product recommendation device for non-residential users according to an embodiment of the present disclosure is schematically shown.

[0137] In some embodiments of the present disclosure, a product recommendation device for non-resident users is also provided, such as Fig. 9 As shown, the product recommendation device 500 for non-residential users includes: a first generation module 501, a first determination module 502, a second determination module 503, a third determination module 504, a second generation module 505, and a recommendation module 506.

[0138] The first generating module 501 is configured to generate a non-resident user portrait according to the non-resident user information of the non-resident user, wherein the non-resident user portrait includes a non-resident user feature tag. In one embodiment, the first generating module 501 can be used to perform the operation S210 described above, which will not be described in detail here.

[0139] The first determination module 502 is configured to obtain all non-resident user portraits in the first target area, and determine the target business according to the statistical results of the non-resident user feature tags included in all non-resident user portraits, wherein the target business has business requirement feature information. In one embodiment, the first determination module 502 can be used to perform the operation S220 described above, which will not be repeated here.

[0140] The second determination module 503 is configured to determine a target non-resident user from the non-resident users in the first target area according to the service requirement characteristic information. In one embodiment, the second determination module 503 may be used to perform the operation S230 described above, which will not be described in detail herein.

[0141] The third determination module 504 is configured to determine the target residential users from the residential users in the second target area according to the target non-residential users. In one embodiment, the third determination module 504 can be used to perform the operation S240 described above, which will not be described in detail herein.

[0142] The second generation module 505 is configured to input the target non-resident user information of the target non-resident user, the target resident user information of the target resident user, and the product information in the product library into the hybrid recommendation model to generate target product recommendation information, wherein the hybrid recommendation model is obtained by training the hybrid recommendation model to be trained using the non-resident user information, resident user information, and product information in the product library included in the data set. In one embodiment, the second generation module 505 can be used to perform the operation S250 described above, which will not be repeated here.

[0143] The recommendation module 506 is configured to recommend the target product to the target non-resident user according to the target product recommendation information. In one embodiment, the recommendation module 506 can be used to perform the operation S260 described above, which will not be described in detail here.

[0144] In some embodiments of the present disclosure, the first generation module includes a first generation sub-module, which is configured to: obtain non-resident attribute information input by a non-resident user, and generate first non-resident user information including a first non-resident tag; obtain product behavior information of the non-resident user, and generate second non-resident user information including a second non-resident tag; obtain product association information input by the non-resident user, and generate third non-resident user information including a third non-resident tag; generate a non-resident user portrait based on the first non-resident user information, the second non-resident user information and the third non-resident user information, the user portrait including a non-resident user feature tag generated based on the first non-resident tag, the second non-resident tag and the third non-resident tag.

[0145] In some embodiments of the present disclosure, the second determination module includes a second determination submodule, configured to: screen the non-resident user characteristic tags included in all non-resident user portraits of the first target area according to the characteristic tags corresponding to the business requirement characteristic information; determine whether the characteristic information corresponding to the non-resident user characteristic tags of the screened non-resident users meets the business requirement characteristic information, and if so, determine the non-resident users as target non-resident users.

[0146] In some embodiments of the present disclosure, the third determination module includes a third determination sub-module, which is configured to: match the non-resident user profile of the target non-resident user with the resident user profile of the resident user in the second target area to generate a matching value; and determine the resident user whose matching value meets a preset matching value as the target resident user.

[0147] In some embodiments of the present disclosure, the device also includes a training module, configured to: input the non-resident user information and the resident user information in the data set into the user-based collaborative filtering algorithm in the hybrid recommendation model to be trained to determine the first product recommendation information; input the non-resident user information in the data set and the product information in the product library into the content-based collaborative filtering algorithm in the hybrid recommendation model to be trained to determine the second product recommendation information; generate target product recommendation information based on the first product recommendation information and the second product recommendation information; adjust the parameters of the user-based collaborative filtering algorithm and the content-based collaborative filtering algorithm according to the evaluation index between the target product recommendation information and the preset product recommendation information until the training termination condition is met to obtain the hybrid recommendation model.

[0148] In some embodiments of the present disclosure, the second generation module includes a second generation submodule, which is configured to: input the target non-resident user information of the target non-resident user and the target resident user information of the target resident user into the user-based collaborative filtering algorithm in the hybrid recommendation model, determine the target similar resident user from the target resident user, the similarity value between the target similar resident user and the target non-resident user satisfies the first preset similarity value, and the target similar resident user has associated product information; determine the first product recommendation information based on the associated product information; determine the product recommendation feature information based on the target non-resident user information of the target non-resident user; input the product recommendation feature information and the product information in the product library into the collaborative filtering algorithm of the opportunity product in the hybrid recommendation model, determine the target similar product from the product information in the product library, the similarity value between the product feature information of the target similar product and the product recommendation feature information satisfies the second preset similarity value; determine the second product recommendation information based on the target similar product; generate the target product recommendation information based on the first product recommendation information and the second product recommendation information.

[0149] In some embodiments of the present disclosure, the device also includes a judgment module, configured to: before determining the first product recommendation information based on the associated product information, determine whether the associated product information meets the target non-residential user usage conditions; if so, determine the first product recommendation information based on the associated product information; if not, obtain associated product characteristics of the associated product information, and generate update information based on the associated product characteristics, and the update information is used to update the product information in the product library.

[0150] According to an embodiment of the present disclosure, any multiple modules of the first generation module 501, the first determination module 502, the second determination module 503, the third determination module 504, the second generation module 505, and the recommendation module 506 can be combined in one module for implementation, or any one of the modules can be split into multiple modules. Alternatively, at least part of the functions of one or more of these modules can be combined with at least part of the functions of other modules and implemented in one module. According to an embodiment of the present disclosure, at least one of the first generation module 501, the first determination module 502, the second determination module 503, the third determination module 504, the second generation module 505, and the recommendation module 506 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application specific integrated circuit (ASIC), or can be implemented by hardware or firmware such as any other reasonable way of integrating or packaging the circuit, or implemented in any one of the three implementation methods of software, hardware, and firmware, or in any appropriate combination of any of them. Alternatively, at least one of the first generation module 501, the first determination module 502, the second determination module 503, the third determination module 504, the second generation module 505, and the recommendation module 506 can be at least partially implemented as a computer program module, and when the computer program module is executed, the corresponding function can be performed.

[0151] Fig.10 A block diagram of an electronic device according to an embodiment of the present disclosure is schematically shown. Fig.10 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0152] like Fig.10 As shown, the electronic device 600 according to an embodiment of the present disclosure includes a processor 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage part 608 into a random access memory (RAM) 603. The processor 601 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or a related chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 601 may also include an onboard memory for caching purposes. The processor 601 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present disclosure.

[0153] In RAM 603, various programs and data required for the operation of electronic device 600 are stored. Processor 601, ROM 602 and RAM 603 are connected to each other through bus 604. Processor 601 performs various operations of the method flow according to the embodiment of the present disclosure by executing the program in ROM 602 and / or RAM 603. It should be noted that the program can also be stored in one or more memories other than ROM 602 and RAM 603. Processor 601 can also perform various operations of the method flow according to the embodiment of the present disclosure by executing the program stored in the one or more memories.

[0154] According to an embodiment of the present disclosure, the electronic device 600 may further include an input / output (I / O) interface 605, which is also connected to the bus 604. The electronic device 600 may further include one or more of the following components connected to the I / O interface 605: an input portion 606 including a keyboard, a mouse, etc.; an output portion 607 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage portion 608 including a hard disk, etc.; and a communication portion 609 including a network interface card such as a LAN card, a modem, etc. The communication portion 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the I / O interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed, so that a computer program read therefrom is installed into the storage portion 608 as needed.

[0155] The present disclosure also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or may exist independently without being assembled into the device / apparatus / system. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed, the method according to the embodiment of the present disclosure is implemented.

[0156] According to an embodiment of the present disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, for example, may include but is not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, an apparatus or a device. For example, according to an embodiment of the present disclosure, a computer-readable storage medium may include the ROM 602 and / or RAM 603 described above and / or one or more memories other than ROM 602 and RAM 603.

[0157] The embodiment of the present disclosure also includes a computer program product, which includes a computer program, and the computer program contains program code for executing the method shown in the flowchart. When the computer program product is run in a computer system, the program code is used to enable the computer system to implement the method provided by the embodiment of the present disclosure.

[0158] The above functions defined in the system / device of the embodiment of the present disclosure are performed when the computer program is executed by the processor 601. According to the embodiment of the present disclosure, the system, device, module, unit, etc. described above can be implemented by a computer program module.

[0159] In one embodiment, the computer program may rely on tangible storage media such as optical storage devices, magnetic storage devices, etc. In another embodiment, the computer program may also be transmitted and distributed in the form of signals on a network medium, and downloaded and installed through the communication part 609, and / or installed from a removable medium 611. The program code contained in the computer program may be transmitted using any appropriate network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0160] In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 609, and / or installed from the removable medium 611. When the computer program is executed by the processor 601, the above functions defined in the system of the embodiment of the present disclosure are performed. According to the embodiment of the present disclosure, the system, device, apparatus, module, unit, etc. described above can be implemented by a computer program module.

[0161] According to an embodiment of the present disclosure, the program code for executing the computer program provided by the embodiment of the present disclosure can be written in any combination of one or more programming languages. Specifically, these computing programs can be implemented using high-level process and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, Java, C++, python, "C" language or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, partially on the remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, using an Internet service provider to connect through the Internet).

[0162] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0163] It will be appreciated by those skilled in the art that the features described in the various embodiments and / or claims of the present disclosure may be combined and / or combined in a variety of ways, even if such combinations and / or combinations are not explicitly described in the present disclosure. In particular, the features described in the various embodiments and / or claims of the present disclosure may be combined and / or combined in a variety of ways without departing from the spirit and teachings of the present disclosure. All of these combinations and / or combinations fall within the scope of the present disclosure.

[0164] The embodiments of the present disclosure are described above. However, these embodiments are only for illustrative purposes and are not intended to limit the scope of the present disclosure. Although the embodiments are described above separately, this does not mean that the measures in the various embodiments cannot be used in combination to advantage. The scope of the present disclosure is defined by the attached claims and their equivalents. Without departing from the scope of the present disclosure, those skilled in the art may make a variety of substitutions and modifications, which should all fall within the scope of the present disclosure.

Claims

1. A product recommendation method for non-residential users, characterized in that: include: Generate a non-resident user portrait based on the non-resident user information of the non-resident user, wherein the non-resident user portrait includes a non-resident user feature tag; Acquire all non-resident user portraits in the first target area, and determine a target service according to statistical results of non-resident user feature tags included in all non-resident user portraits, wherein the target service has service requirement feature information; determining a target non-resident user from the non-resident users in the first target area according to the service requirement characteristic information; determining target resident users from resident users in a second target area according to the target non-resident users; Inputting the target non-residential user information of the target non-residential user, the target resident user information of the target resident user, and the product information in the product library into the hybrid recommendation model to generate target product recommendation information, wherein the hybrid recommendation model is obtained by training the hybrid recommendation model to be trained using the non-residential user information, the resident user information, and the product information in the product library included in the data set; and Recommending a target product to the target non-resident user according to the target product recommendation information.

2. The method according to claim 1, characterized in that Generate a non-resident user profile based on the non-resident user information of the non-resident user, including: Acquire non-resident attribute information input by a non-resident user, and generate first non-resident user information including a first non-resident tag; Acquire product behavior information of non-resident users and generate second non-resident user information including a second non-resident tag; Acquire product-related information input by the non-resident user, and generate third non-resident user information including a third non-resident tag; A non-resident user portrait is generated based on the first non-resident user information, the second non-resident user information and the third non-resident user information, and the user portrait includes a non-resident user characteristic tag generated based on the first non-resident tag, the second non-resident tag and the third non-resident tag.

3. The method according to claim 1, characterized in that Determining target non-resident users from the non-resident users in the first target area according to the service requirement characteristic information includes: screening non-resident user characteristic tags included in all non-resident user profiles of the first target area according to the characteristic tags corresponding to the business requirement characteristic information; Determine whether the feature information corresponding to the non-resident user feature tag of the non-resident user after screening meets the feature information required by the service, If the conditions are met, the non-resident user is determined as a target non-resident user.

4. The method according to claim 1, characterized in that: Determining target resident users from resident users in the second target area according to the target non-resident users includes: Matching the non-resident user profile of the target non-resident user with the resident user profile of the resident user in the second target area to generate a matching value; The residential user whose matching value meets the preset matching value is determined as the target residential user.

5. The method according to claim 1, characterized in that The hybrid recommendation model is constructed by a user-based collaborative filtering algorithm and a content-based collaborative filtering algorithm.

6. The method according to claim 5, characterized in that The hybrid recommendation model to be trained is trained using the non-resident user information, resident user information and product information in the product library included in the data set, including: Inputting the non-resident user information and the resident user information in the data set into a user-based collaborative filtering algorithm in the hybrid recommendation model to be trained to determine the first product recommendation information; Inputting the non-resident user information in the data set and the product information in the product library into a content-based collaborative filtering algorithm in the hybrid recommendation model to be trained to determine the second product recommendation information; Generate target product recommendation information according to the first product recommendation information and the second product recommendation information; The parameters of the user-based collaborative filtering algorithm and the content-based collaborative filtering algorithm are adjusted according to the evaluation index between the target product recommendation information and the preset product recommendation information until the training termination condition is met to obtain the hybrid recommendation model.

7. The method according to claim 5, characterized in that Inputting the target non-resident user information of the target non-resident user, the target resident user information of the target resident user, and the product information in the product library into the hybrid recommendation model to generate target product recommendation information, including: Inputting the target non-resident user information of the target non-resident user and the target resident user information of the target resident user into the user-based collaborative filtering algorithm in the hybrid recommendation model, determining a target similar resident user from the target resident users, wherein the similarity value between the target similar resident user and the target non-resident user satisfies a first preset similarity value, and the target similar resident user has associated product information; Determining first product recommendation information according to the associated product information; determining product recommendation feature information based on the target non-resident user information of the target non-resident user; Input the product recommendation feature information and the product information in the product library into the collaborative filtering algorithm of the opportunity product in the hybrid recommendation model, determine the target similar product from the product information in the product library, and the similarity value between the product feature information of the target similar product and the product recommendation feature information satisfies a second preset similarity value; Determining second product recommendation information according to the target similar product; Generate target product recommendation information according to the first product recommendation information and the second product recommendation information.

8. The method according to claim 7, characterized in that Also includes: Before determining the first product recommendation information according to the associated product information, Determining whether the associated product information meets the target non-resident user usage conditions; If satisfied, determining first product recommendation information according to the associated product information; If not satisfied, the associated product features of the associated product information are obtained, and update information is generated according to the associated product features, where the update information is used to update the product information in the product library.

9. A product recommendation device for non-residential users, characterized in that: include: A first generating module is configured to generate a non-resident user portrait according to the non-resident user information of the non-resident user, wherein the non-resident user portrait includes a non-resident user feature tag; A first determination module is configured to obtain all non-resident user portraits in a first target area, and determine a target business according to statistical results of non-resident user feature tags included in all non-resident user portraits, wherein the target business has business requirement feature information; A second determination module is configured to determine a target non-resident user from the non-resident users in the first target area according to the service requirement characteristic information; A third determination module is configured to determine target resident users from resident users in a second target area according to the target non-resident users; A second generating module is configured to input the target non-residential user information of the target non-residential user, the target resident user information of the target resident user, and the product information in the product library into the hybrid recommendation model to generate target product recommendation information, wherein the hybrid recommendation model is obtained by training the hybrid recommendation model to be trained using the non-residential user information, the resident user information, and the product information in the product library included in the data set; as well as A recommendation module is configured to recommend a target product to the target non-resident user based on the target product recommendation information.

10. An electronic device, comprising: one or more processors; a storage device for storing one or more computer programs, The method is characterized in that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 8.

11. 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 8 are implemented.

12. 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 8 are implemented.

Images