Product combination recommendation generation method and device, electronic device, and storage medium

Abstract

This disclosure provides a method, apparatus, electronic device, and storage medium for generating recommended product portfolios, applicable to both big data and financial technology fields. The method includes: reading historical resource consumption records of a target object within a predetermined historical time period from a database, and the target object's planned total resource consumption for multiple predetermined product types; inputting the historical resource consumption records and planned total resource consumption into a ratio calculation model to calculate a recommended liquidity ratio; obtaining a list of products to be recommended, including multiple target products to be recommended to the target object; reading the minimum resource consumption and product type of each of the multiple target products from the database; and generating multiple recommended product portfolios based on the recommended liquidity ratio, the planned total resource consumption, the minimum resource consumption and product type of each of the multiple target products.

Description

Technical Field

[0001] This disclosure relates to the field of big data technology, and specifically to a method, apparatus, device, medium, and program product for generating recommended product combinations. Background Technology

[0002] With the development of the national economy, various financial products have gradually entered users' lives. Users are no longer satisfied with holding a single financial product. The rapid development of mobile internet and artificial intelligence technologies has made it possible to recommend combined financial products. How to recommend suitable combined financial products to users has become an urgent problem to be solved.

[0003] In realizing the concept disclosed herein, the inventors discovered at least the following problems in the related technologies: Currently, for scenarios involving recommended product combinations, existing recommendation methods have some shortcomings. For example, current manual recommendation methods often rely on business personnel combining their work experience with extensive analysis of historical data from various products to arrive at the recommendation results. This method consumes a large amount of manpower, and the results are often unsatisfactory due to the interference of subjective human factors. Another example is current recommendation methods that combine computer technology, which often involve calculating the returns of a large number of product combinations using pre-set algorithms, and then determining the recommendation result based on the return magnitude. This method requires a large amount of computing power to calculate the returns for each combination separately, resulting in significant computer resource consumption and high requirements for the computer's internal performance. Summary of the Invention

[0004] In view of the above problems, this disclosure provides a method, apparatus, device, medium and program product for generating recommended product portfolios.

[0005] One aspect of this disclosure provides a method for generating recommended product portfolios, comprising:

[0006] Read the historical resource consumption records of the target object within a predetermined historical time period from the database, as well as the total planned resource consumption of the target object for multiple predetermined product types;

[0007] Historical resource consumption records and planned total resource consumption are input into the ratio calculation model to calculate the recommended liquidity ratio. The recommended liquidity ratio is the ratio of the recommended resource consumption of multiple single-category products associated with each of the multiple pre-defined product types.

[0008] Obtain a list of products to be recommended, which includes multiple target products to be recommended to the target audience, and the target products belong to the pre-defined product type;

[0009] Read the minimum resource consumption and product type of each of the multiple target products from the database;

[0010] Based on the recommended liquidity ratio, the planned total resource consumption, the minimum resource consumption of each individual target product, and the product type, multiple recommended product portfolios are generated. Each recommended product portfolio includes at least one recommended product and the recommended resource consumption of each individual recommended product. The recommended products are derived from multiple target products.

[0011] According to embodiments of this disclosure, generating multiple recommended product portfolios based on recommended liquidity ratios, planned total resource consumption, minimum resource consumption per item for each of the multiple target products, and product type includes:

[0012] Based on the minimum resource consumption and product type of each of the multiple target products, multiple initial product combinations are generated. Each initial product combination includes at least one initial product and the initial resource consumption of each initial product. The initial products are derived from multiple target products, and the initial resource consumption of each product is greater than or equal to the minimum resource consumption of each product.

[0013] Multiple recommended product portfolios are generated based on the initial product portfolios, recommended liquidity ratios, and planned total resource consumption.

[0014] According to embodiments of this disclosure, generating multiple recommended product portfolios based on multiple initial product portfolios, recommended liquidity ratios, and planned total resource consumption includes:

[0015] The first constraint is established based on the recommended liquidity ratio. The first constraint includes: in any initial product portfolio, the ratio of the initial resource consumption of multiple single categories associated with each of the multiple predetermined product types is equal to the recommended liquidity ratio. The initial resource consumption of a single category is calculated based on the initial resource consumption of each single item corresponding to at least one initial product.

[0016] A second constraint is established based on the total planned resource consumption. The second constraint includes: in any initial product combination, the sum of the initial resource consumption of each of the at least one initial product is less than or equal to the total planned resource consumption.

[0017] Based on the first and second constraints, multiple recommended product combinations are determined from multiple initial product combinations.

[0018] According to embodiments of this disclosure, the multiple predetermined product types include liquidity product types, periodic product types, and closed-end product types, and the historical resource consumption records include multiple historical resource expenditure data, each of which includes the amount of resource expenditure and the expenditure time.

[0019] The recommended liquidity ratio is calculated as follows:

[0020] Based on the amount and time of resource expenditure in historical resource expenditure data, multiple liquidity consumption data and multiple periodic consumption data are determined from multiple historical resource expenditure data. Among them, the multiple expenditure times corresponding to multiple periodic consumption data have a periodic distribution pattern, and the difference between the multiple resource expenditure amounts corresponding to multiple periodic consumption data is less than a preset difference threshold.

[0021] The liquidity resource consumption limit associated with the type of liquidity product is calculated based on multiple liquidity consumption data.

[0022] The periodic resource consumption limits associated with periodic product types are calculated based on multiple periodic consumption data.

[0023] The closed-end resource consumption limit associated with the closed-end product type is calculated based on the planned total resource consumption, the limit for liquidity resource consumption, and the limit for cyclical resource consumption.

[0024] Calculate the ratios of liquidity resource consumption limits, cyclical resource consumption limits, and closed-end resource consumption limits to obtain the recommended liquidity ratio.

[0025] According to embodiments of this disclosure, calculating the periodic resource consumption limit associated with a periodic product type based on multiple periodic consumption data includes:

[0026] Based on the expenditure time, multiple periodic consumption data are classified into special expenditure datasets and non-special expenditure datasets;

[0027] Using a predetermined regression algorithm, based on the resource expenditure corresponding to each periodic consumption data in the non-specific expenditure dataset, the expected non-specific resource expenditure within a predetermined future time period is calculated.

[0028] Retrieve the planned specific resource expenditure amount for the target object within a predetermined future time period from the database;

[0029] Based on the estimated non-specific resource expenditure and the planned specific resource expenditure, the periodic resource consumption limit is calculated.

[0030] According to embodiments of this disclosure, classifying multiple periodic consumption data into specific expenditure datasets and non-specific expenditure datasets based on expenditure time includes:

[0031] Based on the expenditure time, multiple periodic consumption data are clustered using a predetermined clustering algorithm to obtain multiple categorized datasets;

[0032] Label the resource consumption categories corresponding to each of the multiple classification datasets;

[0033] Based on resource consumption categories, multiple categorized datasets are divided into special expenditure datasets and non-special expenditure datasets.

[0034] According to embodiments of this disclosure, obtaining the proposed list of recommended products includes:

[0035] Based on the customer risk level of the target audience and the product risk level of multiple pre-selected products, a list of recommended products is generated after determining multiple target products from the multiple pre-selected products.

[0036] According to embodiments of this disclosure, the above method further includes:

[0037] Determine the priority level for each of the multiple recommended product combinations;

[0038] The target recommended product portfolio is recommended to the target audience based on the priority of each of the multiple recommended product portfolios.

[0039] According to embodiments of this disclosure, determining the priority level of each of the multiple recommended product combinations includes:

[0040] Based on the revenue type of at least one recommended product in each recommended product portfolio, the predicted resource revenue for each recommended product portfolio is calculated.

[0041] Based on the predicted resource returns corresponding to each recommended product portfolio, the priority levels of each recommended product portfolio are determined.

[0042] Another aspect of this disclosure provides a recommended product portfolio generation apparatus, including a first reading module, a calculation module, an acquisition module, a second reading module, and a generation module.

[0043] The first reading module is used to read the historical resource consumption records of the target object within a predetermined historical time period from the database, as well as the total planned resource consumption of the target object for multiple predetermined product types.

[0044] The calculation module is used to input historical resource consumption records and planned total resource consumption into the ratio calculation model to calculate the recommended liquidity ratio. The recommended liquidity ratio is the ratio of the recommended resource consumption of multiple single-category products associated with each of the multiple pre-defined product types.

[0045] The acquisition module is used to acquire a list of products to be recommended. The list of products to be recommended includes multiple target products to be recommended to the target audience. The target products belong to the pre-defined product type.

[0046] The second reading module is used to read the minimum resource consumption and product type of each of the multiple target products from the database.

[0047] The generation module is used to generate multiple recommended product combinations based on the recommended liquidity ratio, the total planned resource consumption, the minimum resource consumption of each individual target product, and the product type. Each recommended product combination includes at least one recommended product and the recommended resource consumption of each individual recommended product. The recommended products are derived from multiple target products.

[0048] According to embodiments of this disclosure, the generation module includes a first generation submodule and a second generation submodule.

[0049] The first generation submodule is used to generate multiple preliminary product combinations based on the minimum resource consumption and product type of each of the multiple target products. Each preliminary product combination includes at least one preliminary product and the initial resource consumption of each preliminary product. The preliminary products are derived from multiple target products, and the initial resource consumption of each product is greater than or equal to the minimum resource consumption of the product. The second generation submodule is used to generate multiple recommended product combinations based on the multiple preliminary product combinations, the recommended liquidity ratio, and the planned total resource consumption.

[0050] According to embodiments of this disclosure, the second generation submodule includes a first establishment unit, a second establishment unit, and a determination unit.

[0051] The system comprises: a first establishing unit for establishing a first constraint based on the recommended liquidity ratio; the first constraint includes: in any initially selected product portfolio, the ratio of the initial resource consumption of multiple single-category products associated with each of the multiple predetermined product types is equal to the recommended liquidity ratio; wherein the initial resource consumption of a single-category product is calculated based on the initial resource consumption of each of the at least one initially selected product; a second establishing unit for establishing a second constraint based on the planned total resource consumption; the second constraint includes: in any initially selected product portfolio, the sum of the initial resource consumption of each of the at least one initially selected product is less than or equal to the planned total resource consumption; and a determining unit for determining multiple recommended product portfolios from the multiple initially selected product portfolios based on the first and second constraints.

[0052] According to embodiments of this disclosure, the multiple predetermined product types include liquid product types, periodic product types, and closed product types, and the historical resource consumption records include multiple historical resource expenditure data, each of which includes the amount of resource expenditure and the expenditure time.

[0053] The calculation module includes a first determination submodule, a first calculation submodule, a second calculation submodule, a third calculation submodule, and a fourth calculation submodule.

[0054] The system comprises four sub-modules: a first determination sub-module, used to determine multiple liquidity consumption data and multiple periodic consumption data from historical resource expenditure data based on the resource expenditure amount and expenditure time of historical resource expenditure data, wherein the multiple expenditure times corresponding to the multiple periodic consumption data have a periodic distribution pattern, and the difference between the multiple resource expenditure amounts corresponding to the multiple periodic consumption data is less than a preset difference threshold; a first calculation sub-module, used to calculate the liquidity resource consumption limit associated with the liquidity product type based on the multiple liquidity consumption data; a second calculation sub-module, used to calculate the periodic resource consumption limit associated with the periodic product type based on the multiple periodic consumption data; a third calculation sub-module, used to calculate the closed-end resource consumption limit associated with the closed-end product type based on the planned total resource consumption, the liquidity resource consumption limit, and the periodic resource consumption limit; and a fourth calculation sub-module, used to calculate the ratio of the liquidity resource consumption limit, the periodic resource consumption limit, and the closed-end resource consumption limit to obtain the recommended liquidity ratio.

[0055] According to embodiments of this disclosure, the second calculation submodule includes a classification unit, a first calculation unit, a reading unit, and a second calculation unit.

[0056] The system includes a classification unit for classifying multiple periodic consumption data into special expenditure datasets and non-special expenditure datasets based on expenditure time; a first calculation unit for calculating the expected non-special resource expenditure amount within a predetermined time period using a predetermined regression algorithm, based on the resource expenditure amount corresponding to each periodic consumption data in the non-special expenditure dataset; a reading unit for reading the planned special resource expenditure amount of the target object within a predetermined time period from the database; and a second calculation unit for calculating the periodic resource consumption limit based on the expected non-special resource expenditure amount and the planned special resource expenditure amount.

[0057] According to embodiments of this disclosure, the classification unit includes a clustering subunit, a labeling subunit, and a classification subunit.

[0058] The clustering subunit uses a predetermined clustering algorithm to cluster multiple periodic consumption data based on expenditure time to obtain multiple categorized datasets; the labeling subunit is used to label the resource consumption categories corresponding to each of the multiple categorized datasets; and the classification subunit is used to divide the multiple categorized datasets into special expenditure datasets and non-special expenditure datasets according to resource consumption categories.

[0059] According to an embodiment of this disclosure, the acquisition module includes a third generation submodule, which is used to generate a list of products to be recommended after determining multiple target products from multiple pre-selected products based on the customer risk level of the target object and the product risk level of multiple pre-selected products.

[0060] According to embodiments of this disclosure, the above-described apparatus further includes a determining module and a recommending module.

[0061] The module is used to determine the priority level of each of the multiple recommended product combinations; the module is used to recommend the target recommended product combination to the target object according to the priority level of each of the multiple recommended product combinations.

[0062] According to embodiments of this disclosure, the determining module includes a fifth calculation submodule and a second determining submodule.

[0063] The fifth calculation submodule is used to calculate the predicted resource returns corresponding to each recommended product portfolio based on the return type of at least one recommended product in each recommended product portfolio; the second determination submodule is used to determine the priority level of each of the multiple recommended product portfolios based on the predicted resource returns corresponding to each recommended product portfolio.

[0064] Another aspect of this disclosure provides an electronic device comprising: one or more processors; and a memory for storing one or more programs, wherein, when the one or more programs are executed by the one or more processors, the one or more processors perform the above-described recommended product portfolio generation method.

[0065] Another aspect of this disclosure provides a computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, cause the processor to perform the above-described method for generating the recommended product combination.

[0066] Another aspect of this disclosure provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method for generating the recommended product portfolio.

[0067] According to the embodiments of this disclosure, in order to address the shortcomings of product portfolio recommendation in the prior art, the above-described method of the embodiments of this disclosure, on the one hand, realizes the automated execution of product portfolio recommendation by automatically performing data reading, calling the calculation model for calculation, and generating results by computer, thereby freeing up manpower and improving the work efficiency of business personnel.

[0068] On the other hand, the method described in the embodiments of this disclosure first calculates the recommended liquidity ratio, and then automatically generates multiple recommended product portfolios based on the recommended liquidity ratio and other few basic parameters read. During the execution of the above method by the computer, only a few key parameters of each product in this scenario need to be obtained, and relatively simple correlation calculations are performed based on these key parameters. The demand for computing resources and computing performance is low, and there is no need to perform profit calculations for a large number of multiple product portfolios as in related technologies. This greatly reduces the consumption of computer resources and the requirements for the internal performance of the computer.

[0069] On the other hand, for the application scenario of product combination recommendation, the recommendation liquidity ratio (the ratio of resource consumption of different types of products) is a key parameter in this scenario. The method described in the embodiments of this disclosure uses the technical means of calculating the recommendation liquidity ratio by combining the historical resource consumption records of the target object. That is, by utilizing the inherent correlation between the user's historical resource consumption and the user's preference for selecting product combinations, a method for recommending product combinations is formed. The recommendation results obtained are in line with the user's behavioral habits, the recommendation results have good reference value, meet the user's need to hold diversified products, and improve the user experience. Attached Figure Description

[0070] The foregoing contents, as well as other objects, features, and advantages of this disclosure, will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:

[0071] Figure 1 The illustration schematically depicts application scenarios of the recommended product portfolio generation method, apparatus, device, medium, and program product according to embodiments of this disclosure;

[0072] Figure 2 A flowchart illustrating a recommended product portfolio generation method according to an embodiment of the present disclosure is shown schematically.

[0073] Figure 3 A flowchart illustrating the calculation of the recommended liquidity ratio according to an embodiment of this disclosure is shown schematically.

[0074] Figure 4 A schematic diagram illustrating a recommended product portfolio generation apparatus according to embodiments of the present disclosure is shown; and

[0075] Figure 5 A block diagram schematically illustrates an electronic device suitable for implementing a recommended product portfolio generation method according to an embodiment of the present disclosure. Detailed Implementation

[0076] The embodiments of the present disclosure will now 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 disclosure. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of the present disclosure for ease of explanation. However, it will be apparent that one or more embodiments may be practiced without these specific details. Furthermore, descriptions of well-known structures and techniques are omitted in the following description to avoid unnecessarily obscuring the concepts of the present disclosure.

[0077] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit this disclosure. The terms “comprising,” “including,” etc., as used herein indicate the presence of the stated features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0078] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0079] When using expressions such as "at least one of A, B, and C", they should generally be interpreted in accordance with the meaning that is commonly understood by a person skilled in the art (e.g., "a system having at least one of A, B, and C" should include, but is not limited to, a system having A alone, a system having B alone, a system having C alone, a system having A and B, a system having A and C, a system having B and C, and / or a system having A, B, and C, etc.).

[0080] In the embodiments disclosed herein, the collection, updating, analysis, processing, use, transmission, provision, disclosure, and storage of data (e.g., including but not limited to user personal information) comply with relevant laws and regulations, are used for legitimate purposes, and do not violate public order and good morals. In particular, necessary measures have been taken to prevent unauthorized access to user personal information data and to safeguard user personal information security, network security, and national security.

[0081] In the embodiments disclosed herein, user authorization or consent is obtained before acquiring or collecting user personal information.

[0082] It should be noted that the recommended product combination generation method, apparatus, device, medium and program product of this disclosure can be applied to the field of big data technology, the field of financial technology, or any field other than the field of big data technology and the field of financial technology. This disclosure does not limit the application field of the above-mentioned recommended product combination generation method, apparatus, device, medium and program product.

[0083] With the development of the national economy, various financial products have gradually entered users' lives. Users are no longer satisfied with holding a single financial product. The rapid development of mobile internet and artificial intelligence technologies has made it possible to recommend combined financial products. How to recommend suitable combined financial products to users has become an urgent problem to solve. Existing product recommendation technologies mostly focus on single products, failing to meet users' needs for diversified products. Furthermore, current recommendation methods for combined product scenarios have some shortcomings. For example, current manual recommendation methods rely on business personnel combining their work experience and extensive analysis of historical data from various products to arrive at recommendations. This method consumes a lot of manpower, and the results are often unsatisfactory due to the interference of subjective human factors. Another example is the current method of combining computer technology for recommendations, which often involves calculating the returns of a large number of product combinations using pre-set algorithms and then determining the recommendation based on the returns. This method requires a large amount of computing power to calculate each combination separately, consuming significant computer resources and placing high demands on the computer's internal performance.

[0084] In view of this, embodiments of this disclosure provide a method for generating recommended product portfolios, including:

[0085] The system retrieves historical resource consumption records of the target object within a predetermined historical time period from the database, as well as the total planned resource consumption of the target object for products under multiple predetermined product types. It then inputs these records and total planned resource consumption into a ratio calculation model to calculate the recommended liquidity ratio, which is the ratio of the resource consumption of multiple single-category recommendations associated with each of the multiple predetermined product types. A list of products to be recommended is obtained, including multiple target products to be recommended to the target object, each belonging to a predetermined product type. The system retrieves the minimum resource consumption and product type of each of the multiple target products from the database. Based on the recommended liquidity ratio, the total planned resource consumption, the minimum resource consumption and product type of each of the multiple target products, multiple recommended product combinations are generated. Each recommended product combination includes at least one recommended product and the corresponding single-item recommended resource consumption of at least one recommended product. The recommended products originate from multiple target products.

[0086] Figure 1 The illustration schematically depicts application scenarios of the recommended product portfolio generation method, apparatus, device, medium, and program product according to embodiments of the present disclosure.

[0087] like Figure 1As shown, application scenario 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 serves as a medium for providing a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.

[0088] Users can interact with server 105 via network 104 using at least one of the first terminal device 101, second terminal device 102, and third terminal device 103 to receive or send messages, etc. Various communication client applications can be installed on the first terminal device 101, second terminal device 102, and third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social media platform software, etc. (for example only).

[0089] The first terminal device 101, the second terminal device 102, and the third terminal device 103 can be various electronic devices with displays and support web browsing, including but not limited to smartphones, tablets, laptops, and desktop computers.

[0090] Server 105 can be a server that provides various services, such as a backend management server that supports websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103 (this is just an example). The backend management server can analyze and process data such as received user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal devices.

[0091] In the application scenario of this disclosure embodiment, business personnel of financial institutions can initiate a request to server 105 to obtain multiple financial product combination recommendation schemes through first terminal device 101, second terminal device 102, and third terminal device 103. In response to the user request, server 105 can execute the recommended product combination generation method of this disclosure embodiment. For example, it can read the historical resource consumption records of the target object in a predetermined historical time period from the database, calculate the recommended liquidity ratio based on the historical resource consumption records using a ratio calculation model, and further generate multiple recommended product combinations based on the recommended liquidity ratio and other basic data obtained (such as the planned total resource consumption, the minimum resource consumption of a single product, etc.), and return them to the business personnel of financial institutions through first terminal device 101, second terminal device 102, and third terminal device 103.

[0092] It should be noted that the recommended product combination generation method provided in this disclosure embodiment can generally be executed by server 105. Correspondingly, the recommended product combination generation apparatus provided in this disclosure embodiment can generally be located in server 105. The recommended product combination generation method provided in this disclosure embodiment can also be executed by a server or server cluster that is different from server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or server 105. Correspondingly, the recommended product combination generation apparatus provided in this disclosure embodiment can also be located in a server or server cluster that is different from server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or server 105.

[0093] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0094] The following will be based on Figure 1 The described scene, through Figures 2-5 The method for generating recommended product combinations according to the disclosed embodiments is described in detail.

[0095] Figure 2 A flowchart illustrating a recommended product portfolio generation method according to an embodiment of the present disclosure is shown schematically.

[0096] like Figure 2 As shown, the recommended product combination generation method of this embodiment includes operations S210 to S250.

[0097] In operation S210, the historical resource consumption records of the target object within a predetermined historical time period are read from the database, as well as the total planned resource consumption of the target object for multiple predetermined product types.

[0098] The recommended product portfolio generation method of this disclosure can be used in scenarios where financial product portfolio purchase plans are recommended to target users. The target user can be the intended target user, and the target user's historical resource consumption records within a predetermined historical time period can be the target user's historical consumption details, including consumption amount and consumption time. The target user's planned total resource consumption for multiple predetermined product types can be the target user's purchase demand for a specific type of financial product, such as the user's planned expenditure on a specific type of financial product.

[0099] It should be noted that, in the embodiments of this disclosure, user consent or authorization is required before obtaining the historical resource consumption records and planned total resource consumption of the target object. For example, a request to obtain user information may be sent to the user before operation S210. Operation S210 is executed only if the user consents or authorizes the acquisition of user information.

[0100] In operation S220, historical resource consumption records and planned total resource consumption are input into the ratio calculation model to calculate the recommended liquidity ratio. The recommended liquidity ratio is the ratio of the recommended resource consumption of multiple single-category products associated with each of the multiple pre-defined product types.

[0101] The proportional calculation model is used to calculate the recommended liquidity ratio based on historical resource consumption records and planned total resource consumption. The recommended liquidity ratio can be, for example, the proportion of purchase expenses recommended to target users for different types of financial products. For instance, it could be the proportion of purchase expenses recommended to users for cash management financial products, periodic development financial products, and closed-end financial products.

[0102] In operation S230, a list of products to be recommended is obtained. This list includes multiple target products intended for recommendation to the target audience, and these target products belong to a predetermined product type. For example, the target products in the list could be financial products that have been screened and deemed suitable for recommendation to the target users.

[0103] In operation S240, the minimum resource consumption per unit and the product type of each of the multiple target products are read from the database. The minimum resource consumption per unit of the target product can be, for example, the minimum investment amount for each financial product. The product type of the target product can be one of the following: cash management wealth management product, periodic development wealth management product, or closed-end wealth management product.

[0104] In operation S250, based on the recommended liquidity ratio, the total planned resource consumption, the minimum resource consumption of each individual target product, and the product type, multiple recommended product portfolios are generated. Each recommended product portfolio includes at least one recommended product and the recommended resource consumption of each individual recommended product. The recommended products are derived from multiple target products.

[0105] Based on the calculated recommended liquidity ratio, this operation generates multiple recommended product combinations using the obtained recommended liquidity ratio and other basic parameters (such as the total planned resource consumption, the minimum resource consumption of each target product, and the product type). Any recommended product combination may include, for example, multiple financial products recommended to the target user, and the recommended purchase amount (resource consumption) for each of the recommended financial products. For instance, one recommended product combination might include 5 financial products: 1 cash management product, 2 periodic investment products, and 2 closed-end products, with recommended purchase amounts for each of these 10 financial products.

[0106] According to the embodiments of this disclosure, in order to address the shortcomings of product portfolio recommendation in the prior art, the above-described method of the embodiments of this disclosure, on the one hand, realizes the automated execution of product portfolio recommendation by automatically performing data reading, calling the calculation model for calculation, and generating results by computer, thereby freeing up manpower and improving the work efficiency of business personnel.

[0107] On the other hand, the method described in the embodiments of this disclosure first calculates the recommended liquidity ratio, and then automatically generates multiple recommended product portfolios based on the recommended liquidity ratio and other few basic parameters read. During the execution of the above method by the computer, only a few key parameters of each product in this scenario need to be obtained, and relatively simple correlation calculations are performed based on these key parameters. The demand for computing resources and computing performance is low, and there is no need to perform profit calculations for a large number of multiple product portfolios as in related technologies. This greatly reduces the consumption of computer resources and the requirements for the internal performance of the computer.

[0108] On the other hand, for the application scenario of product combination recommendation, the recommendation liquidity ratio (the ratio of resource consumption of different types of products) is a key parameter in this scenario. The method described in the embodiments of this disclosure uses the technical means of calculating the recommendation liquidity ratio by combining the historical resource consumption records of the target object. That is, by utilizing the inherent correlation between the user's historical resource consumption and the user's preference for selecting product combinations, a method for recommending product combinations is formed. The recommendation results obtained are in line with the user's behavioral habits, have good reference value, and high recommendation accuracy, which meets the user's need to hold diversified products and improves the user experience.

[0109] According to embodiments of this disclosure, the target products in the proposed product list can be financial products that have been screened and are suitable for recommendation to the target users. Furthermore, obtaining the proposed product list can be done by: determining multiple target products from multiple pre-selected products based on the customer risk level of the target object and the product risk level of multiple pre-selected products, and then generating the proposed product list.

[0110] Specifically, this can involve retrieving the product risk level of each pre-selected product and the target customer's risk level from a database. A higher product risk level indicates a higher level of risk, while the target customer's risk level represents the target user's risk tolerance; a higher target customer risk level indicates a higher risk tolerance. Based on the target customer's risk level and the product risk levels of multiple pre-selected products, multiple target products are determined from these pre-selected products. This can be achieved by filtering out products whose risk level is less than or equal to the customer's risk level—that is, selecting products whose risk level is lower than the customer's risk tolerance level—and then generating a list of proposed recommended products. For example, based on the customer's risk level and the product risk level, cash management products, periodically open-ended products, and closed-end products that meet the customer's risk level can be selected as target products in the recommended product list.

[0111] According to embodiments of this disclosure, by obtaining a list of products to be recommended, multiple recommended product combinations can be generated based on the product attribute data (minimum resource consumption and product type of each product) and the pre-calculated recommended liquidity ratio of the target products to be recommended. This algorithm comprehensively considers the customer's liquidity preference and the customer's risk tolerance. The recommended product combinations take into account the customer's risk preference, thus being more referential, more accurate, and having a better product recommendation effect.

[0112] According to embodiments of this disclosure, after calculating the recommended liquidity ratio through the historical resource consumption records of the target object, multiple recommended product combinations can be generated based on the recommended liquidity ratio and other basic parameters read, as detailed below.

[0113] Specifically, based on the recommended liquidity ratio, the total planned resource consumption, the minimum resource consumption of each of the multiple target products, and the product type, multiple recommended product portfolios are generated, including:

[0114] First, based on the minimum resource consumption and product type of each of the multiple target products, multiple preliminary product combinations are generated. Each preliminary product combination includes at least one preliminary product and the initial resource consumption of each of the at least one preliminary product. The preliminary products are derived from multiple target products, and the initial resource consumption of each product is greater than or equal to the minimum resource consumption of the product.

[0115] For example, a preliminary product portfolio can be formed by selecting some initial products from multiple target products using permutation and combination methods. Each preliminary product portfolio can include at least one product type; for instance, the selected products may belong to one or more of the following categories: cash management financial products, periodic development financial products, and closed-end financial products. Furthermore, the initial resource consumption of each preliminary product must be greater than or equal to its minimum resource consumption; for example, the recommended purchase amount for each preliminary financial product must be greater than or equal to its corresponding minimum investment amount.

[0116] Then, based on the multiple initial product portfolios, recommended liquidity ratios, and planned total resource consumption, multiple recommended product portfolios are generated. That is, constraints are established based on the recommended liquidity ratios and planned total resource consumption, and the portfolios that meet these constraints are selected from the multiple initial product portfolios as the final recommended product portfolios.

[0117] According to embodiments of this disclosure, further, generating multiple recommended product portfolios based on multiple initial product portfolios, recommended liquidity ratios, and planned total resource consumption includes:

[0118] First, a first constraint is established based on the recommended liquidity ratio. The first constraint includes: in any initially selected product portfolio, the ratio of the initial resource consumption of multiple single categories associated with each of the multiple predetermined product types is equal to the recommended liquidity ratio. The initial resource consumption of a single category is calculated based on the initial resource consumption of each of the at least one initially selected product.

[0119] Taking financial products as an example, for instance, in any initial product portfolio, the types of products selected belong to one or more of the following: cash management wealth management products, periodic development wealth management products, and closed-end wealth management products. It is necessary to calculate the initial resource consumption of multiple individual product types associated with each of the multiple predetermined product types.

[0120] Specifically, for cash management products, assuming there are n cash management products initially selected in this portfolio, and the initial resource consumption (purchase amount) of each product is denoted as m. i Then the initial resource consumption of a single type (i.e., the total purchase amount) Q m The calculation is shown in the following formula (1):

[0121] Q m =m1+m2+……m n ------Equation (1)

[0122] For periodically available products, the first step is to read the opening period and corresponding purchase amount (initial resource consumption per item) for each product. Then, match the available purchase period with the product's opening period. Assume there are n products of this type with an initial selection period of t in this combination, and let z denote the initial resource consumption (purchase amount) per item for each product. i Then the initial resource consumption of a single type (i.e., the total purchase amount) Q z The calculation is shown in the following formula (2):

[0123]

[0124] For closed-type products, assuming there are n products of this type initially selected in the portfolio, let f be the initial resource consumption (purchase amount) of each product. i Then the initial resource consumption of a single type (i.e., the total purchase amount) Q f The calculation is shown in the following formula (3):

[0125] Q f =f1+f2+……f n ------Equation (3)

[0126] After calculating the initial resource consumption of each individual product type associated with each of the multiple pre-selected product types, a first constraint is established based on the recommended liquidity ratio. In any initially selected product portfolio, the ratio of the initial resource consumption of each individual product type associated with each of the multiple pre-selected product types is equal to the recommended liquidity ratio, such as the cash management wealth management product Q mentioned above. m Regularly Open-Ended Wealth Management Products Q z Recommended purchase amount Q for closed-end wealth management products f The ratio must satisfy the constraint that it is equal to the recommended liquidity ratio. For example, the first constraint is shown in equation (4) below:

[0127] Q m Q z Q f =Recommended liquidity ratio------Equation (4)

[0128] Subsequently, a second constraint is established based on the total planned resource consumption. This second constraint includes the following: in any initially selected product combination, the sum of the initial resource consumption of each individual item corresponding to at least one initially selected product is less than or equal to the total planned resource consumption. Taking the aforementioned financial combination product as an example, the second constraint could be, for example, as shown in equation (5):

[0129] Q m +Q z +Q f ≤ Total planned resource consumption ------ Equation (5)

[0130] Finally, based on the first and second constraints, multiple recommended product combinations that simultaneously satisfy both constraints are determined from the multiple initially selected product combinations.

[0131] The method described in this embodiment first establishes constraints and then selects product combinations that meet the constraints as the final recommended product combinations. This method of determining recommended product combinations can produce better recommendation results with only a simple constraint algorithm, without having to perform revenue calculations for each of the large number of product combinations as in related technologies. This further reduces the consumption of computer resources and the requirements on the internal performance of the computer.

[0132] According to embodiments of this disclosure, in the application scenarios of these embodiments, the product is used to recommend a combination of products under multiple predetermined product types to a target audience. These multiple predetermined product types may include liquidity product types, cyclical product types, and closed-end product types. Taking financial products as an example, liquidity product types may be cash management wealth management products, cyclical product types may be periodically developed wealth management products, and closed-end product types may be closed-end wealth management products.

[0133] After retrieving the historical resource consumption records of the target object within a predetermined historical time period from the database, as well as the total planned resource consumption of the target object for multiple predetermined product types, the recommended liquidity ratio can be further calculated based on the retrieved data.

[0134] Figure 3 A flowchart illustrating the calculation of the recommended liquidity ratio according to an embodiment of this disclosure is shown schematically.

[0135] like Figure 3 As shown, the method for calculating the recommended liquidity ratio in this embodiment includes operations S310 to S350.

[0136] In operation S310, based on the resource expenditure amount and expenditure time of historical resource expenditure data, multiple liquidity consumption data and multiple periodic consumption data are determined from multiple historical resource expenditure data. Among them, the multiple expenditure times corresponding to multiple periodic consumption data have a periodic distribution pattern, and the difference between the multiple resource expenditure amounts corresponding to multiple periodic consumption data is less than a preset difference threshold. For example, in the customer's historical expenditure account details, data that have the same amount or similar amount spent at the same time interval are periodic consumption data. Similar amount means that the amount fluctuates within a certain percentage, such as 50% fluctuation range.

[0137] The target entity's historical resource consumption records include multiple historical resource expenditure data points. Each historical resource expenditure data point includes the amount of resource expenditure and the expenditure time. For example, historical resource consumption records can be the target customer's historical expenditure details, with each expenditure detail including the expenditure amount, expenditure time, and other information. The target entity's planned total resource consumption for products under multiple pre-ordered product types can be the target customer's expected total purchase amount.

[0138] In the above operation S310, based on the resource expenditure amount and expenditure time of historical resource expenditure data, multiple liquidity consumption data and multiple periodic consumption data are determined from multiple historical resource expenditure data. For example, it can be to mark the expenditure type of the target customer's historical expenditure details. Data of daily living expenditure type is marked as "liquidity consumption data". Daily living expenditure type can include, but is not limited to, "food and beverage", "transportation", "utility payment", "cash withdrawal", "consumption", "ETC", etc.; expenditure type of expenditure in historical expenditure details with the same or similar amount of expenditure at the same time interval is marked as "periodic consumption data". Periodic consumption data can include, but is not limited to, "loan repayment", "social security payment", "credit card repayment", "transfer and remittance", etc.

[0139] In operation S320, the liquidity resource consumption limit associated with the liquidity product type is calculated based on multiple liquidity consumption data in the historical resource expenditure data.

[0140] Specifically, for example, the expenditure amounts marked as "liquidity consumption data" in each period could be summed up, and a machine learning linear regression algorithm could be used to predict the customer's liquidity resource expenditure (e.g., liquidity expenditure amount) in the next period. This prediction could then be used as the liquidity resource consumption limit for purchasing liquidity-type products, such as the amount for purchasing cash management financial products. The period can be set by the customer on the terminal interface or configured by parameters.

[0141] In operation S330, a periodic resource consumption limit associated with a periodic product type is calculated based on multiple periodic consumption data in historical resource expenditure data. For example, a machine learning linear regression algorithm can be used to predict the amount of periodic resource expenditure (such as the amount of periodic expenditure) in the next period based on historical periodic consumption data, and use this as the periodic resource consumption limit for purchasing periodic products, such as the amount for purchasing periodically developed financial products.

[0142] In operation S340, the closed-end resource consumption limit associated with the closed-end product type is calculated based on the planned total resource consumption, the liquidity resource consumption limit, and the periodic resource consumption limit. For example, it could be: subtracting the calculated liquidity resource consumption limit and periodic resource consumption limit from the planned total resource consumption to obtain the closed-end resource consumption limit associated with the closed-end product type. For instance, subtracting the purchase amount of cash management products and periodic open-end products from the target customer's expected total purchase amount yields the purchase amount for the closed-end wealth management product.

[0143] In operation S350, the ratio of liquidity resource consumption limit, periodic resource consumption limit, and closed-end resource consumption limit is calculated to obtain the recommended liquidity ratio.

[0144] According to embodiments of this disclosure, the above method categorizes historical resource expenditure data and predicts future resource expenditure amounts for corresponding types based on historical expenditure data of different types, serving as resource consumption limits for those types, and then calculates the recommended liquidity ratio. By mining the intrinsic correlation between users' historical resource consumption and their preferences for selecting product combinations, the method achieves prediction of users' resource consumption, thereby determining the key parameter (recommendation liquidity ratio) in the product combination recommendation scenario. The resulting recommendation results align with users' behavioral habits, and the recommendation results have good reference value.

[0145] According to embodiments of this disclosure, in processing periodic consumption data, calculating a periodic resource consumption limit associated with a periodic product type based on the characteristics of the periodic consumption data itself may include the following operations:

[0146] Operation 1: Based on the expenditure time, classify multiple periodic consumption data into special expenditure datasets and non-special expenditure datasets.

[0147] Specifically, this operation can be as follows: First, based on the attribute of the expenditure time, a predetermined clustering algorithm is used to cluster multiple periodic consumption data. Data with similar expenditure times are grouped into the same category to obtain multiple categorized datasets. For example, after clustering, multiple datasets are obtained, where the data in the first dataset corresponds to expenditure times around the 15th of each month, the data in the second dataset corresponds to expenditure times around the 25th of each month, and so on.

[0148] Next, the resource consumption categories corresponding to each of the multiple categorized datasets are labeled. For example, they can be labeled as "Loan Repayment," "Social Security Payment," "Credit Card Repayment," "Fund Transfer," and so on. The labeling of categories can be based on the correspondence between each data category and the expenditure time. For example, based on a preset payment rule that loan repayments are deducted around the 15th of each month, the dataset with expenditures around the 15th of each month can be labeled as "Loan Repayment." Similarly, the category of each dataset is labeled sequentially.

[0149] Then, the multiple categorized datasets are divided into specific expenditure datasets and non-specific expenditure datasets according to resource consumption categories. Classification can be based on the consumption characteristics of each resource consumption category. For example, expenditures such as loan repayments and social security contributions are long-term expenditures with relatively fixed amounts, so datasets for "loan repayments" and "social security contributions" can be classified as specific expenditure datasets. Conversely, expenditures such as credit card repayments and bank transfers are variable expenditures with varying cycles and amounts, so datasets for "credit card repayments" and "bank transfers" can be classified as non-specific expenditure datasets.

[0150] Operation 2: Using a predetermined regression algorithm, calculate the expected non-specific resource expenditure within a predetermined time period based on the resource expenditure corresponding to each periodic consumption data in the non-specific expenditure dataset.

[0151] For example, data such as "credit card repayment" and "money transfer" are obtained as attribute values ​​for each data point in the non-specific expenditure dataset: resource expenditure amount (expenditure amount). Then, for each non-specific expenditure dataset, an ensemble regression algorithm is used to predict the customer's expected non-specific resource expenditure amount (expected expenditure amount) within a future period. The expenditure period refers to the date interval between two periodic expenditure details.

[0152] When performing regression predictions, the following processing is performed for each non-specific expenditure dataset.

[0153] First, the data samples in each dataset are divided into T sample sets, each containing M samples. A base learner is then trained on each sample set. Finally, the base learners are combined to obtain the final regression result.

[0154] Specifically, based on the linear regression algorithm, a linear regression model is obtained after training each sample set for T rounds (training times t = 1, 2...T): h i For each dataset, T linear regression models are obtained; then, the T linear regression models obtained through the base learners are combined to obtain the final regression model H(x) as shown in equation (6):

[0155]

[0156] Using the regression model H(x), we can predict the expected non-specific resource expenditure within a predetermined time period in the future. For example, if a customer remits 10,000 yuan to a designated account on the 10th of each month, the expenditure period is 1 month and the expenditure amount is 10,000 yuan. At the same time, the customer has a phone bill expenditure on the 1st of each month, and the expenditure amount fluctuates. If the amount is predicted to be 80 yuan next month, then the expected expenditure amount in addition to the repayment within the expenditure period of 1 month is 10,080 yuan.

[0157] Operation 3: Retrieve the planned specific resource expenditure of the target object from the database within a predetermined future time period. For example, retrieve the customer's resource expenditure plan for a future period, such as loan repayment plan and social security payment plan, including the expected repayment amount and expenditure cycle for the future period.

[0158] Step 4: Calculate the periodic resource consumption limit based on the estimated non-specific resource expenditure and the planned specific resource expenditure.

[0159] Specifically, this can be achieved by summing the expected non-specific resource expenditure and the planned specific resource expenditure to obtain the periodic resource consumption limit, which can then be used as the periodic resource consumption limit for purchasing periodic products, such as the amount for purchasing periodically developed financial products. Simultaneously, the purchase period needs to be determined. The purchase period can specifically include factors related to the expenditure cycle. For example, if a customer's expenditure amount is 10,000 yuan with a 21-day cycle, then the purchase period can be a factor of 21 days, such as 1 day, 3 days, 7 days, or 21 days.

[0160] According to embodiments of this disclosure, further, in the above-described processing of periodic consumption data, based on expenditure time, multiple periodic consumption data are clustered using a predetermined clustering algorithm to obtain multiple categorized datasets. Specifically, this can be achieved by using the K-means clustering algorithm to group data with similar expenditure times into the same category, thus obtaining multiple categorized datasets. Since the K-means clustering algorithm requires pre-determining the number of clusters K, and the determination of the number of clusters has a significant impact on the clustering results, embodiments of this disclosure incorporate the EM algorithm principle, using the maximum likelihood function to calculate the most suitable K value, and then performing clustering using the clustering algorithm, thereby optimizing and improving the clustering method to enhance clustering accuracy.

[0161] The EM (Expectation-Maximization Algorithm) is an iterative algorithm suitable for solving problems with hidden variables, i.e., unobservable recorded values. The EM algorithm involves two variables: a hidden variable (Z) and model parameters (U). The unknown possible classification results are used as the hidden variable Z, and the centroids U of the clusters are used as the model's predicted parameters. Parameter U and variable Z are dependent; if the optimal parameter U is determined, all classification results can be calculated using clustering algorithms. However, since both the classification results and the centroids U are uncertain, the EM algorithm iteratively solves the problem by assigning initial values ​​to the variables and estimating them until the model converges, thus finding the optimal cluster K.

[0162] According to embodiments of this disclosure, the specific steps for clustering multiple periodic consumption data are as follows:

[0163] Operation 1: For the sample set D = {x1, x2, x3, ..., x...} n}, initialize the cluster size parameter K, and randomly select the centroid u of each cluster. k This yields multiple initial centroids: U = {u1, u2, ... u} k}, each sample x i This includes the amount of resources spent and the timing of those expenditures, such as the amount and timing of each historical transaction in the periodic consumption data.

[0164] Operation 2: Using the K-means algorithm, cluster multiple periodic consumption data based on the initial parameter U. The clustering result is multiple initial clusters, C = {c1, c2, c3, ..., ck}.

[0165] Operation 3: Update sample labels based on clustering results: If the clustering result is "Loan Repayment", "Social Security Payment", "Credit Card Repayment", or "Fund Transfer", then these are the new labels. Other classification results are defined as hidden attributes Z. According to the embodiments of this disclosure, since some samples belonging to the same cluster have unclear categories, and the accuracy of classifying them into the same category is difficult to estimate, such fund expenditure categories are defined as hidden attributes, i.e., hidden variables Z.

[0166] Operation 4: Solve for the expected value Q of the log-likelihood function based on the hidden variable Z. Specifically, initialize the centroid U = {u1, u2, ... u} based on the parameters. k}, calculate the expected value Q of the log-likelihood function of the hidden variable Z, as shown in equation (7):

[0167]

[0168] Operation 5: Solve for the optimal core U′={u1′,u2′,...u k Specifically, the centroid that maximizes the expected value Q of the log-likelihood function is found to be the optimal centroid U′, where U′ is calculated as shown in equation (8):

[0169]

[0170] Step 6: After obtaining the optimal centroid U′, use U′ as the centroid of the next cluster in the next iteration. Repeat steps 1-5 to iterate until the maximum likelihood function converges. The calculated K value is the optimal K value (number of clusters). For example, Jeson's inequality can be used to solve for the extrema of function Q using L(k) >= J(Z, Q), and the optimal K value can be output.

[0171] Step 6: Based on the optimal K value, re-cluster the multiple periodic consumption data using the K-means algorithm to obtain multiple classification datasets.

[0172] According to embodiments of this disclosure, in clustering processing, it is necessary to initialize the number of clusters and the centroid of each cluster. Ideally, samples within the same cluster should be as close to the centroid as possible. Embodiments of this disclosure transform the problem of optimizing the number of clusters into the problem of optimizing the centroids. Once the centroids are determined, the hidden classification results can be obtained using clustering algorithms. Embodiments of this disclosure optimize and improve the clustering method by combining the EM algorithm and utilizing the maximum likelihood function to calculate the most suitable number of clusters, thereby improving the accuracy of clustering.

[0173] According to embodiments of this disclosure, after generating multiple recommended product portfolios based on recommended liquidity ratios, planned total resource consumption, minimum resource consumption per item for each of the multiple target products, and product type, the recommended product portfolio generation method of this disclosure further includes: determining the priority level of each of the multiple recommended product portfolios; and recommending target recommended product portfolios to target objects according to the priority levels of the multiple recommended product portfolios. For example, product portfolios with higher priority levels are preferentially recommended to customers.

[0174] Determining the priority of multiple recommended product portfolios can be achieved by calculating the predicted returns of each portfolio. Specifically, this can include:

[0175] Operation 1: Based on the revenue type of at least one recommended product in each recommended product portfolio, calculate the predicted resource revenue for each recommended product portfolio.

[0176] Before performing calculations, it is necessary to determine the return type of each product. For financial products, this could specifically be one of the following: cash management products, fixed-income products, or net asset value (NAV) products.

[0177] Furthermore, for products with different revenue types, the predicted resource revenue x per unit of each product type is calculated based on different prediction algorithms. i The predicted resource returns for each recommended product portfolio are summed by combining the number of units held in each product, resulting in the predicted resource returns for each portfolio as follows: Σx i ·N i , where N i The number of shares held is calculated as follows: For financial products, the number of shares held = purchase amount per product / product price on that day.

[0178] For example, for cash management products, the predicted resource returns for each product can be calculated as follows: The system constructs a return prediction model, uses the historical daily returns of each product as training data, and employs machine learning algorithms to predict the returns of the client's holdings over a future period. The period can be set by the client on the terminal interface or configured by parameters.

[0179] For example, for fixed-income products, the projected return for each product can be calculated by reading the product's annualized return and calculating the return for a future period. Projected customer return = Customer share * Issue price * Annualized return of the corresponding tranche * (Registration date - Value date + 1) / Number of days. For example, if the annualized return for customers is 4.5%, then return = Customer share * Issue price * 4.5% * Number of days held / 365.

[0180] For example, for net asset value (NAV) products, the projected resource returns for each product can be calculated by forecasting the prices of the assets invested in each product, thereby calculating the future returns of the product.

[0181] For equity assets, we obtain the corresponding stock portfolio for the product. We acquire the latest macroeconomic indicators, financial metrics of the invested listed companies, and market sentiment. Using machine learning algorithms, we learn stock prices and predict client returns based on the projected stock price growth. If the investment includes a stock index, we obtain the predicted stock index price by weighting the predicted values ​​of the constituent stocks. Market sentiment can be predicted using a deep learning recurrent neural network (RNN) model.

[0182] For bond assets, bond returns include coupon payments and capital gains. Coupon income per period = face value of the bond * coupon rate; capital gains = present value of the bond - purchase price of the bond. The present value of a bond is the discounted price of future cash flows, including coupon payments and principal, based on the yield. The present value of a bond can be predicted by forecasting its yield. This can be achieved by inputting yield curve data, macroeconomic indicators, and corporate credit risk indicators as variables and using a machine learning linear regression algorithm to train a bond yield model and predict the bond price for a future period.

[0183] Step 2: Based on the predicted resource returns corresponding to each recommended product portfolio, determine the priority level of each portfolio. Taking financial products as an example, the system can generate a product portfolio recommendation list by predicting the customer returns corresponding to each portfolio, sorting them from highest to lowest return. The system can then push product portfolio purchase plans to users. After receiving the recommendations, users can adjust their purchase plans accordingly and execute the purchase transactions.

[0184] According to embodiments of this disclosure, the method first generates a small number of recommended product combinations that meet preset constraints, and then determines the priority of recommendations to users by calculating the predicted resource returns corresponding to each recommended product combination. Since the recommended product combinations are a small number of combinations that meet preset constraints selected from a large number of product combinations, further revenue calculations on them do not require significant computational resources. Compared to related technologies that perform revenue calculations on a large number of different product combinations, this method avoids wasting computational resources by calculating the returns of a large number of product combinations that are not relevant to recommendations (combinations that do not meet the constraints), thus saving computational resources.

[0185] According to embodiments of this disclosure, the predicted resource returns corresponding to each recommended product portfolio are calculated, and the priority of the recommended product portfolios is determined based on the predicted resource returns corresponding to each recommended product portfolio. The generation of the recommendation results takes into account the impact of customer return factors, and is used to recommend financial product portfolios that maximize customer returns. This achieves the goal of maximizing customer returns, and the recommendation results are more likely to be accepted by customers, resulting in better recommendation effects and improved user experience.

[0186] Based on the above-described method for generating recommended product combinations, this disclosure also provides an apparatus for generating recommended product combinations. The following will be combined with... Figure 4 The device is described in detail.

[0187] Figure 4 A schematic block diagram of a recommended product portfolio generation apparatus according to an embodiment of the present disclosure is shown.

[0188] like Figure 4As shown, the recommended product combination generation device 400 of this embodiment includes a first reading module 410, a calculation module 420, an acquisition module 430, a second reading module 440, and a generation module 450.

[0189] The first reading module 410 is used to read the historical resource consumption records of the target object within a predetermined historical time period from the database, as well as the total planned resource consumption of the target object for multiple predetermined product types.

[0190] The calculation module 420 is used to input historical resource consumption records and planned total resource consumption into the ratio calculation model to calculate the recommended liquidity ratio through the ratio calculation model. The recommended liquidity ratio is the ratio of the recommended resource consumption of multiple single-category products associated with each of the multiple pre-defined product types.

[0191] The acquisition module 430 is used to acquire a list of products to be recommended, wherein the list of products to be recommended includes multiple target products to be recommended to the target object, and the target products belong to a predetermined product type;

[0192] The second reading module 440 is used to read the minimum resource consumption and product type of each of the multiple target products from the database.

[0193] The generation module 450 is used to generate multiple recommended product combinations based on the recommended liquidity ratio, the total planned resource consumption, the minimum resource consumption of each individual target product, and the product type. Each recommended product combination includes at least one recommended product and the recommended resource consumption of each individual recommended product. The recommended products are derived from multiple target products.

[0194] According to the embodiments of this disclosure, the aforementioned recommended product portfolio generation device 400, through a first reading module 410, a calculation module 420, an acquisition module 430, a second reading module 440, and a generation module 450, automatically performs operations such as data reading, calling the calculation model for calculation, and result generation by a computer, thereby realizing the automated execution of product portfolio recommendations, freeing up manpower, and improving the work efficiency of business personnel. On the other hand, the first reading module 410 and the calculation module 420 first calculate the recommended liquidity ratio, and then the generation module 450 automatically generates multiple recommended product portfolios based on the recommended liquidity ratio and other few basic parameters read. It only needs to obtain a few key parameters of each product in this scenario and perform relatively simple correlation calculations based on these key parameters. The demand for computing resources and computing performance is low, and there is no need to perform revenue calculations for a large number of multiple product portfolios separately as in related technologies, which greatly reduces the consumption of computer resources and the requirements for the internal performance of the computer. On the other hand, for the application scenario of product combination recommendation, the recommendation liquidity ratio (the ratio of resource consumption of different types of products) is a key parameter in this scenario. The embodiments of this disclosure use the calculation module 420 to calculate the recommendation liquidity ratio by combining the historical resource consumption records of the target object. That is, by utilizing the inherent correlation between the user's historical resource consumption and the user's preference for selecting product combinations, the recommendation results are consistent with the user's behavioral habits, the recommendation results are more reliable, meet the user's need to hold diversified products, and improve the user experience.

[0195] According to embodiments of this disclosure, the generation module includes a first generation submodule and a second generation submodule.

[0196] The first generation submodule is used to generate multiple preliminary product combinations based on the minimum resource consumption and product type of each of the multiple target products. Each preliminary product combination includes at least one preliminary product and the initial resource consumption of each preliminary product. The preliminary products are derived from multiple target products, and the initial resource consumption of each product is greater than or equal to the minimum resource consumption of the product. The second generation submodule is used to generate multiple recommended product combinations based on the multiple preliminary product combinations, the recommended liquidity ratio, and the planned total resource consumption.

[0197] According to embodiments of this disclosure, the second generation submodule includes a first establishment unit, a second establishment unit, and a determination unit.

[0198] The system comprises: a first establishing unit for establishing a first constraint based on the recommended liquidity ratio; the first constraint includes: in any initially selected product portfolio, the ratio of the initial resource consumption of multiple single-category products associated with each of the multiple predetermined product types is equal to the recommended liquidity ratio; wherein the initial resource consumption of a single-category product is calculated based on the initial resource consumption of each of the at least one initially selected product; a second establishing unit for establishing a second constraint based on the planned total resource consumption; the second constraint includes: in any initially selected product portfolio, the sum of the initial resource consumption of each of the at least one initially selected product is less than or equal to the planned total resource consumption; and a determining unit for determining multiple recommended product portfolios from the multiple initially selected product portfolios based on the first and second constraints.

[0199] According to embodiments of this disclosure, the multiple predetermined product types include liquid product types, periodic product types, and closed product types, and the historical resource consumption records include multiple historical resource expenditure data, each of which includes the amount of resource expenditure and the expenditure time.

[0200] The calculation module includes a first determination submodule, a first calculation submodule, a second calculation submodule, a third calculation submodule, and a fourth calculation submodule.

[0201] The system comprises four sub-modules: a first determination sub-module, used to determine multiple liquidity consumption data and multiple periodic consumption data from historical resource expenditure data based on the resource expenditure amount and expenditure time of historical resource expenditure data, wherein the multiple expenditure times corresponding to the multiple periodic consumption data have a periodic distribution pattern, and the difference between the multiple resource expenditure amounts corresponding to the multiple periodic consumption data is less than a preset difference threshold; a first calculation sub-module, used to calculate the liquidity resource consumption limit associated with the liquidity product type based on the multiple liquidity consumption data; a second calculation sub-module, used to calculate the periodic resource consumption limit associated with the periodic product type based on the multiple periodic consumption data; a third calculation sub-module, used to calculate the closed-end resource consumption limit associated with the closed-end product type based on the planned total resource consumption, the liquidity resource consumption limit, and the periodic resource consumption limit; and a fourth calculation sub-module, used to calculate the ratio of the liquidity resource consumption limit, the periodic resource consumption limit, and the closed-end resource consumption limit to obtain the recommended liquidity ratio.

[0202] According to embodiments of this disclosure, the second calculation submodule includes a classification unit, a first calculation unit, a reading unit, and a second calculation unit.

[0203] The system includes a classification unit for classifying multiple periodic consumption data into special expenditure datasets and non-special expenditure datasets based on expenditure time; a first calculation unit for calculating the expected non-special resource expenditure amount within a predetermined time period using a predetermined regression algorithm, based on the resource expenditure amount corresponding to each periodic consumption data in the non-special expenditure dataset; a reading unit for reading the planned special resource expenditure amount of the target object within a predetermined time period from the database; and a second calculation unit for calculating the periodic resource consumption limit based on the expected non-special resource expenditure amount and the planned special resource expenditure amount.

[0204] According to embodiments of this disclosure, the classification unit includes a clustering subunit, a labeling subunit, and a classification subunit.

[0205] The clustering subunit uses a predetermined clustering algorithm to cluster multiple periodic consumption data based on expenditure time to obtain multiple categorized datasets; the labeling subunit is used to label the resource consumption categories corresponding to each of the multiple categorized datasets; and the classification subunit is used to divide the multiple categorized datasets into special expenditure datasets and non-special expenditure datasets according to resource consumption categories.

[0206] According to an embodiment of this disclosure, the acquisition module includes a third generation submodule, which is used to generate a list of products to be recommended after determining multiple target products from multiple pre-selected products based on the customer risk level of the target object and the product risk level of multiple pre-selected products.

[0207] According to embodiments of this disclosure, the above-described apparatus further includes a determining module and a recommending module.

[0208] The module is used to determine the priority level of each of the multiple recommended product combinations; the module is used to recommend the target recommended product combination to the target object according to the priority level of each of the multiple recommended product combinations.

[0209] According to embodiments of this disclosure, the determining module includes a fifth calculation submodule and a second determining submodule.

[0210] The fifth calculation submodule is used to calculate the predicted resource returns corresponding to each recommended product portfolio based on the return type of at least one recommended product in each recommended product portfolio; the second determination submodule is used to determine the priority level of each of the multiple recommended product portfolios based on the predicted resource returns corresponding to each recommended product portfolio.

[0211] According to embodiments of this disclosure, any multiple modules among the first reading module 410, calculation module 420, acquisition module 430, second reading module 440, and generation module 450 can be combined into one module, or any one of these modules can be split into multiple modules. Alternatively, at least some of the functions of one or more of these modules can be combined with at least some of the functions of other modules and implemented in one module. According to embodiments of this disclosure, at least one of the first reading module 410, calculation module 420, acquisition module 430, second reading module 440, and generation module 450 can be at least partially implemented as hardware circuitry, 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-package, an application-specific integrated circuit (ASIC), or implemented in hardware or firmware by any other reasonable means of integrating or packaging the circuitry, or implemented in any one of software, hardware, and firmware methods, or in a suitable combination of any of these methods. Alternatively, at least one of the first reading module 410, the calculation module 420, the acquisition module 430, the second reading module 440, and the generation module 450 can be at least partially implemented as a computer program module, which can perform corresponding functions when the computer program module is run.

[0212] Figure 5 A block diagram schematically illustrates an electronic device suitable for implementing a recommended product portfolio generation method according to an embodiment of the present disclosure.

[0213] like Figure 5 As shown, an electronic device 500 according to an embodiment of the present disclosure includes a processor 501, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 502 or a program loaded from a storage portion 508 into a random access memory (RAM) 503. The processor 501 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 501 may also include onboard memory for caching purposes. The processor 501 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.

[0214] RAM 503 stores various programs and data required for the operation of electronic device 500. Processor 501, ROM 502, and RAM 503 are interconnected via bus 504. Processor 501 performs various operations of the method flow according to embodiments of the present disclosure by executing programs in ROM 502 and / or RAM 503. It should be noted that the programs may also be stored in one or more memories other than ROM 502 and RAM 503. Processor 501 may also perform various operations of the method flow according to embodiments of the present disclosure by executing programs stored in said one or more memories.

[0215] According to embodiments of this disclosure, the electronic device 500 may further include an input / output (I / O) interface 505, which is also connected to a bus 504. The electronic device 500 may also include one or more of the following components connected to the I / O interface 505: an input section 506 including a keyboard, mouse, etc.; an output section 507 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 508 including a hard disk, etc.; and a communication section 509 including a network interface card such as a LAN card, modem, etc. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to the I / O interface 505 as needed. A removable medium 511, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 510 as needed so that computer programs read from it can be installed into the storage section 508 as needed.

[0216] This disclosure also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The computer-readable storage medium carries one or more programs that, when executed, implement the method according to the embodiments of this disclosure.

[0217] According to embodiments of this disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, such as including, but not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this disclosure, the computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, according to embodiments of this disclosure, the computer-readable storage medium may include ROM 502 and / or RAM 503 and / or one or more memories other than ROM 502 and RAM 503 described above.

[0218] Embodiments of this disclosure also include a computer program product comprising a computer program containing program code for performing the methods shown in the flowchart. When the computer program product is run on a computer system, the program code is used to cause the computer system to implement the recommended product portfolio generation method provided in embodiments of this disclosure.

[0219] When the computer program is executed by the processor 501, it performs the functions defined in the system / apparatus of this disclosure embodiments. According to embodiments of this disclosure, the systems, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0220] In one embodiment, the computer program may rely on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may also be transmitted and distributed in the form of signals over a network medium, and may be downloaded and installed via the communication section 509, and / or installed from a removable medium 511. The program code contained in the computer program can be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination thereof.

[0221] In such an embodiment, the computer program can be downloaded and installed from a network via communication section 509, and / or installed from removable medium 511. When the computer program is executed by processor 501, it performs the functions defined in the system of this disclosure embodiment. According to embodiments of this disclosure, the systems, devices, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0222] According to embodiments of this disclosure, program code for executing the computer programs provided in embodiments of this disclosure can be written in any combination of one or more programming languages. Specifically, these computational programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, languages ​​such as Java, C++, Python, "C", or similar programming languages. The program code can execute entirely on the user's computing device, partially on the user's device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0223] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0224] Those skilled in the art will understand that the features described in the various embodiments and / or claims of this disclosure can be combined or combined in various ways, even if such combinations or combinations are not explicitly described in this disclosure. In particular, the features described in the various embodiments and / or claims of this disclosure can be combined or combined in various ways without departing from the spirit and teachings of this disclosure. All such combinations and / or combinations fall within the scope of this disclosure.

[0225] The embodiments of this disclosure have been described above. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of this disclosure. Although various embodiments have been described above, this does not mean that the measures in the various embodiments cannot be used advantageously in combination. The scope of this disclosure is defined by the appended claims and their equivalents. Various substitutions and modifications can be made by those skilled in the art without departing from the scope of this disclosure, and all such substitutions and modifications should fall within the scope of this disclosure.

Claims

1. A method for generating recommended product combinations, comprising: The database is used to read the historical resource consumption records of the target object within a predetermined historical time period, as well as the total planned resource consumption of the target object for multiple predetermined product types. The multiple predetermined product types include liquidity product types, periodic product types, and closed-end product types. The historical resource consumption records include multiple historical resource expenditure data, each of which includes the amount of resource expenditure and the expenditure time. The historical resource consumption records and the planned total resource consumption are input into a ratio calculation model to calculate a recommended liquidity ratio. The recommended liquidity ratio is the ratio of multiple single-category recommended resource consumption amounts associated with each of the multiple predetermined product types. Calculating the recommended liquidity ratio includes: determining multiple liquidity consumption data and multiple periodic consumption data from the historical resource expenditure data based on the resource expenditure amount and expenditure time; calculating a liquidity resource consumption limit associated with the liquidity product type based on the multiple liquidity consumption data; calculating a periodic resource consumption limit associated with the periodic product type based on the multiple periodic consumption data, where the periodic consumption data also includes a periodic factor; calculating a closed-end resource consumption limit associated with the closed-end product type based on the planned total resource consumption, the liquidity resource consumption limit, and the periodic resource consumption limit; and calculating the ratio of the liquidity resource consumption limit, the periodic resource consumption limit, and the closed-end resource consumption limit to obtain the recommended liquidity ratio. Obtaining the proposed recommended product list includes: based on the customer risk level of the target object and the product risk level of multiple pre-selected products, determining multiple target products from the multiple pre-selected products and generating the proposed recommended product list; wherein, the proposed recommended product list includes multiple target products to be recommended to the target object, and the target products belong to the predetermined product type; Read the minimum resource consumption and product type of each of the multiple target products from the database; Based on the recommended liquidity ratio, the planned total resource consumption, the minimum resource consumption of each of the multiple target products, and the product type, multiple recommended product combinations are generated. Each of the recommended product combinations includes at least one recommended product and the recommended resource consumption of each of the at least one recommended product. The recommended products are derived from the multiple target products. Determine the priority level of each of the multiple recommended product combinations; recommend the target recommended product combination to the target object according to the priority level of each of the multiple recommended product combinations.

2. The method according to claim 1, wherein, Based on the recommended liquidity ratio, the planned total resource consumption, the minimum resource consumption of each of the multiple target products, and the product type, multiple recommended product combinations are generated, including: Based on the minimum resource consumption and product type of each of the multiple target products, multiple preliminary product combinations are generated, wherein each preliminary product combination includes at least one preliminary product and the initial resource consumption of each of the at least one preliminary product. The preliminary products are derived from the multiple target products, and the initial resource consumption of each product is greater than or equal to the minimum resource consumption of each product. Multiple recommended product combinations are generated based on the multiple initially selected product combinations, the recommended liquidity ratio, and the total planned resource consumption.

3. The method according to claim 2, wherein, Based on the multiple initially selected product portfolios, the recommended liquidity ratio, and the planned total resource consumption, multiple recommended product portfolios are generated, including: A first constraint is established based on the recommended liquidity ratio, wherein the first constraint includes: in any of the initial product combinations, the ratio of the initial resource consumption of multiple single categories associated with each of the multiple predetermined product types is equal to the recommended liquidity ratio, wherein the initial resource consumption of a single category is calculated based on the initial resource consumption of a single item corresponding to each of the at least one initial product. A second constraint is established based on the total planned resource consumption, wherein the second constraint includes: in any of the initial product combinations, the sum of the initial resource consumption of each of the at least one initial product is less than or equal to the total planned resource consumption. Based on the first constraint and the second constraint, multiple recommended product combinations are determined from the multiple initially selected product combinations.

4. According to the method of claim 1, the multiple expenditure times corresponding to the multiple periodic consumption data have a periodic distribution pattern, and the difference between the multiple resource expenditures corresponding to the multiple periodic consumption data is less than a preset difference threshold.

5. The method according to claim 1, wherein, The periodic resource consumption limits associated with the periodic product type, calculated based on the multiple periodic consumption data, include: Based on the expenditure time, the multiple periodic consumption data are classified into special expenditure datasets and non-special expenditure datasets; Using a predetermined regression algorithm, based on the resource expenditure corresponding to each periodic consumption data in the non-specific expenditure data set, the expected non-specific resource expenditure in the future predetermined time period is calculated. Read the planned specific resource expenditure of the target object from the database for a predetermined future time period; The periodic resource consumption limit is calculated based on the estimated non-specific resource expenditure and the planned specific resource expenditure.

6. The method according to claim 5, wherein, Based on the expenditure time, the multiple periodic consumption data are classified into specific expenditure datasets and non-specific expenditure datasets, including: Based on the expenditure time, the multiple periodic consumption data are clustered using a predetermined clustering algorithm to obtain multiple categorized datasets; Label the resource consumption category corresponding to each of the multiple classification datasets; The multiple categorized datasets are divided into special expenditure datasets and non-special expenditure datasets according to the resource consumption categories.

7. The method according to claim 1, wherein, Determining the priority level of each of the multiple recommended product combinations includes: Based on the revenue type of each of the at least one recommended product in each of the recommended product combinations, the predicted resource revenue corresponding to each of the recommended product combinations is calculated. Based on the predicted resource returns corresponding to each of the recommended product portfolios, the priority level of each of the multiple recommended product portfolios is determined.

8. A recommended product portfolio generation apparatus, comprising: The first reading module is used to read the historical resource consumption records of the target object within a predetermined historical time period from the database, as well as the total planned resource consumption of the target object for multiple predetermined product types. The multiple predetermined product types include liquidity product types, periodic product types, and closed product types. The historical resource consumption records include multiple historical resource expenditure data, and each historical resource expenditure data includes the amount of resource expenditure and the expenditure time. The calculation module is used to input the historical resource consumption records and the planned total resource consumption into a ratio calculation model to calculate a recommended liquidity ratio. The recommended liquidity ratio is the ratio of multiple single-category recommended resource consumption amounts associated with each of the multiple predetermined product types. Calculating the recommended liquidity ratio includes: determining multiple liquidity consumption data and multiple periodic consumption data from the historical resource expenditure data based on the resource expenditure amount and expenditure time; calculating a liquidity resource consumption limit associated with the liquidity product type based on the multiple liquidity consumption data; calculating a periodic resource consumption limit associated with the periodic product type based on the multiple periodic consumption data, wherein the periodic consumption data also includes a periodic factor; calculating a closed-end resource consumption limit associated with the closed-end product type based on the planned total resource consumption, the liquidity resource consumption limit, and the periodic resource consumption limit; and calculating the ratio of the liquidity resource consumption limit, the periodic resource consumption limit, and the closed-end resource consumption limit to obtain the recommended liquidity ratio. The acquisition module is used to acquire a list of products to be recommended, including: based on the customer risk level of the target object and the product risk level of multiple pre-selected products, determining multiple target products from the multiple pre-selected products and generating the list of products to be recommended, wherein the list of products to be recommended includes multiple target products to be recommended to the target object, and the target products belong to the predetermined product type; The second reading module is used to read the minimum resource consumption and product type of each of the multiple target products from the database. The generation module is used to generate multiple recommended product combinations based on the recommended liquidity ratio, the total planned resource consumption, the minimum resource consumption of each of the multiple target products, and the product type. Each of the recommended product combinations includes at least one recommended product and the recommended resource consumption of each of the at least one recommended product. The recommended products are derived from the multiple target products. The determining module is used to determine the priority level of each of the multiple recommended product combinations; and to recommend the target recommended product combination to the target object according to the priority level of each of the multiple recommended product combinations.

9. An electronic device, comprising: One or more processors; Storage device for storing one or more programs. Wherein, when the one or more programs are executed by the one or more processors, the one or more processors perform the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, cause the processor to perform the method according to any one of claims 1 to 7.

11. A computer program product comprising a computer program that, when executed by a processor, implements the method according to any one of claims 1 to 7.

Images