Product pushing method and device, computer equipment and storage medium

Abstract

The invention relates to the technical field of financial science and technology, in particular to a product pushing method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring key feature data of a target user, wherein the key feature data comprises multi-dimensional key identity feature data and multi-dimensional key behavior feature data; inputting the key feature data of the target user into a target super network model to obtain a target product output by the target super network model; generating a target push path according to the target product and the target browsing habit data of the target user; and pushing the target product to the target user according to the target pushing path.

Description

Technical Field

[0001] The present application relates to the field of financial technology, and in particular to a product push method, device, computer equipment and storage medium. Background Art

[0002] With the rapid development of network information technology, Internet finance is increasingly being used in many scenarios such as work and life. Many banks or financial institutions that provide financial services have emerged in the financial market, and each of them has launched a wide range of financial products to meet the needs of different customers.

[0003] In traditional technology, the bank customer marketing system is relatively monotonous. Under the existing customer big data system, when developing new marketing activities, it is impossible to effectively utilize the existing customer system and conduct precise marketing. Therefore, it is urgently needed to be improved. Summary of the invention

[0004] Based on this, it is necessary to provide a method, device, computer equipment and storage medium that can accurately push financial products to each user in response to the above technical problems.

[0005] In a first aspect, the present application provides a product push method, the method comprising:

[0006] Obtain key feature data of target users, including multi-dimensional key identity feature data and multi-dimensional key behavior feature data;

[0007] Inputting the key feature data of the target user into the target hypernetwork model to obtain the target product output by the target hypernetwork model;

[0008] Generate target push paths based on target products and target users’ target browsing habits data;

[0009] Push target products to target users based on the target push path.

[0010] In one embodiment, the hypernetwork model is trained in the following manner:

[0011] Iteratively train the initial hypernetwork model based on the key feature data of sample users;

[0012] If the accuracy of the initial hypernetwork model meets the requirements, the initial hypernetwork model is used as the target hypernetwork model.

[0013] In one embodiment, the method further comprises:

[0014] Obtaining initial feature data of sample users;

[0015] Extract core feature data from the initial feature data according to the influencing factors of each dimensional feature in the initial feature data;

[0016] Perform dimensionality reduction on the core feature data to obtain the key feature data of the sample users.

[0017] In one embodiment, the target browsing habit data includes first browsing habit data and second browsing habit data; generating a target push path according to the target browsing habit data of each target product and target user includes:

[0018] When the current browsing information of the target user is obtained, the browsing keywords and the current positioning information are determined according to the current browsing information;

[0019] Determine the target user's first browsing habit data based on browsing keywords;

[0020] Determine the second browsing habit data of the target user according to the current positioning information;

[0021] A target push path is generated according to the first browsing habit data, the second browsing habit data and each target product.

[0022] In one embodiment, generating a target push path according to target browsing habit data of each target product and target user includes:

[0023] In the case where the current browsing information of the target user is not obtained, similar users corresponding to the target user are determined;

[0024] Generate a target push path based on the target browsing habit data of each target product and similar users.

[0025] In one embodiment, generating a target push path according to target browsing habit data of each target product and target user includes:

[0026] When a new product is detected, the new product is determined based on the similarity between the new product and the target product;

[0027] Generate target push paths based on target products, new marketing products and target users' target browsing habit data.

[0028] In a second aspect, the present application also provides a product pushing device, the device comprising:

[0029] An acquisition module is used to acquire key feature data of target users, the key feature data including multi-dimensional key identity feature data and multi-dimensional key behavior feature data;

[0030] The product push module is used to input the key feature data of the target user into the target hypernetwork model to obtain the target product output by the target hypernetwork model;

[0031] A path generation module is used to generate a target push path based on the target product and the target user's target browsing habit data;

[0032] The push module is used to push target products to target users according to the target push path.

[0033] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0034] Obtain key feature data of target users, including multi-dimensional key identity feature data and multi-dimensional key behavior feature data;

[0035] Inputting the key feature data of the target user into the target hypernetwork model to obtain the target product output by the target hypernetwork model;

[0036] Generate target push paths based on target products and target users’ target browsing habits data;

[0037] Push target products to target users based on the target push path.

[0038] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the following steps are implemented:

[0039] Obtain key feature data of target users, including multi-dimensional key identity feature data and multi-dimensional key behavior feature data;

[0040] Inputting the key feature data of the target user into the target hypernetwork model to obtain the target product output by the target hypernetwork model;

[0041] Generate target push paths based on target products and target users’ target browsing habits data;

[0042] Push target products to target users based on the target push path.

[0043] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0044] Obtain key feature data of target users, including multi-dimensional key identity feature data and multi-dimensional key behavior feature data;

[0045] Inputting the key feature data of the target user into the target hypernetwork model to obtain the target product output by the target hypernetwork model;

[0046] Generate target push paths based on target products and target users’ target browsing habits data;

[0047] Push target products to target users based on the target push path.

[0048] The above-mentioned product push method, device, computer equipment and storage medium, the target super network model in the present application is different from a simple network structure, and can be used to express the association relationship between multiple nodes; according to the super network, multi-feature fusion analysis is performed on the multi-dimensional key identity feature data and multi-dimensional key behavior feature data of the target user, and experimental modeling can be performed by refining multiple important and unimportant feature dimensions of the customer through the lasso method, and finally the optimal weight combination of each feature is obtained to accurately extract the marketing features of the existing customer data, so as to facilitate the analysis and identification of potential customers with these features, and carry out precision marketing; further, according to the target browsing habit data of the target product and the target user, a target push path is generated; according to the target push path, the target product is pushed to the target user, so that each target user has a corresponding target push path, thereby realizing the targeted push of the target product and further improving the product push efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the drawings required for use in the embodiments or the related technical descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0050] FIG1 is a schematic diagram of a process flow of a product push method in one embodiment;

[0051] FIG2 is a schematic diagram of a process for generating a target push path in one embodiment;

[0052] FIG3 is a structural block diagram of a product pushing device in one embodiment;

[0053] FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0055] The embodiment of the present application provides a product push method, as shown in FIG1 , which is executed by a computer device and includes the following steps 101 to 104. Among them:

[0056] Step 101, obtaining key feature data of a target user, where the key feature data includes multi-dimensional key identity feature data and multi-dimensional key behavior feature data.

[0057] Among them, key identity feature data represents the user's attribute feature information, such as name, age, gender, etc.; key behavior feature data is used to represent the user's behavior information, such as page browsing, page dwelling, product selection, etc.

[0058] Specifically, the user performs relevant financial business operations by selecting a platform such as an application software APP, a small program or a web page on the target terminal. Specifically, the financial business operations may include but are not limited to the following: "clicking" and "sliding pages" in the financial product display page. Among them, the data fields that may be generated by the "click" operation include: click link, click jump, click position, etc., and the data fields that may be generated for the "sliding page" operation include: up, down, left, right, etc. Each business operation will generate interactive data with the system, and what the computer device in this embodiment needs to do is to obtain these interactive data.

[0059] Step 102: input the key feature data of the target user into the target hypernetwork model to obtain the target product output by the target hypernetwork model.

[0060] Among them, supernetworks are divided into two categories, one is a network-based supernetwork, and the other is a hypergraph-based supernetwork. The target supernetwork model in this embodiment is a hypergraph-based supernetwork. Any network that can be described by a hypergraph is a supernetwork, which is called a Hypernetwork, a network that is higher than and beyond the existing network. It has the characteristics of multiple layers, multiple levels, and multiple dimensions. It can be understood that the traditional graph structure of a hypergraph can only express the relationship between two nodes, and the hypergraph can represent the relationship between the combined effects of multiple nodes. The hyperedge is composed of multiple nodes, and the relationship between the hyperedges together constitutes a hypergraph. The use of a hypergraph means that we can display information about relationships characterized by multiple objects.

[0061] Specifically, in this embodiment, the lasso method is used to conduct experimental modeling, and finally the optimal weight combination of each feature (i.e., the key feature data of the target user) is obtained, and then the key feature data of the target user is analyzed using the target hypernetwork model to accurately extract the marketing features of the existing customer data, so as to facilitate the analysis and identification of potential customers with these features and carry out precision marketing.

[0062] Step 103: Generate a target push path based on the target product and the target user's target browsing habit data.

[0063] Among them, the browsing habit data is the browsing habit data determined based on the user's historical browsing situation; illustratively, the browsing habit data may include the page type preferred by the user, the product type preferred by the user, etc.; the target push path refers to using each designated browsing interface as the push path.

[0064] Step 104: Push the target product to the target user according to the target push path.

[0065] Specifically, each page in the target push path is matched with the target product, and according to the matching relationship, the corresponding target product is displayed in each page of the push path to achieve the push of the target product.

[0066] In the above-mentioned product push method, the target super network model in this application is different from a simple network structure and can be used to express the association relationship between multiple nodes; based on the super network, multi-feature fusion analysis is performed on the multi-dimensional key identity feature data and multi-dimensional key behavior feature data of the target user. It is possible to extract multiple important and unimportant feature dimensions of the customer and perform experimental modeling through the lasso method, and finally obtain the optimal weight combination of each feature to accurately extract the marketing features of the existing customer data, so as to facilitate the analysis and identification of potential customers with these features and carry out precision marketing; further, based on the target browsing habit data of the target product and the target user, a target push path is generated; based on the target push path, the target product is pushed to the target user, so that each target user has a corresponding target push path, thereby realizing targeted push of the target product and further improving the product push efficiency.

[0067] In one embodiment, the hypernetwork model is trained in the following manner: the initial hypernetwork model is iteratively trained according to the key feature data of the sample users; if the accuracy of the initial hypernetwork model meets the requirements, the initial hypernetwork model is used as the target hypernetwork model.

[0068] The key feature data of the sample users are obtained in the following manner: obtaining the initial feature data of the sample users; extracting the core feature data from the initial feature data according to the influencing factors of the features of each dimension in the initial feature data; performing dimensionality reduction processing on the core feature data to obtain the key feature data of the sample users.

[0069] Optionally, obtain the initial feature data of the sample users; extract the core feature data from the initial feature data based on the influencing factors of each dimension of the features in the initial feature data, including: extracting features that are important for product push from the initial feature data, and filtering out irrelevant or redundant information. Feature selection methods include filtering methods (such as correlation coefficient method, chi-square test method), wrapping methods (such as recursive feature elimination method), and embedding methods (such as Lasso regression, decision tree algorithm). This step can reduce the data dimension, simplify the computational complexity of the model, and retain the predictive ability.

[0070] Optionally, the core feature data is subjected to dimensionality reduction processing to obtain the core feature data of the sample users, including: after the core feature data is extracted, the number of features is further reduced by dimensionality reduction methods while retaining the main information in the data. Dimensionality reduction methods include linear methods (such as principal component analysis PCA, linear discriminant analysis LDA) and nonlinear methods (such as t-distributed neighborhood embedding t-SNE, local linear embedding LLE). Dimensionality reduction processing helps to improve the operation speed and generalization ability of the model and prevent overfitting.

[0071] Furthermore, the training method of the hypernetwork may include: first constructing a search space of the hypernetwork. The search space of the hypernetwork may be constructed by optional substructures of each layer of the hypernetwork. The number of layers of the hypernetwork may be set, and at least one optional substructure may be configured for each layer, such as a convolutional structure, a fully connected structure, a pooling structure, a normalization module, and the like. Each optional substructure of each layer in the search space may be connected to each optional substructure of an adjacent layer to form a hypernetwork.

[0072] In this embodiment, the initial feature data of the sample users are screened to obtain important influencing features and non-important influencing features, and the important influencing features are determined as the key feature data of the sample users. Based on the key feature data of the sample users, the initial hypernetwork model is trained to obtain the target hypernetwork model, so that the target hypernetwork model has the ability of multi-feature analysis.

[0073] In one embodiment, as shown in FIG. 2 , a target browsing habit data is provided, including first browsing habit data and second browsing habit data; an optional method of generating a target push path according to the target browsing habit data of each target product and target user specifically includes the following process:

[0074] Step 201: after obtaining the current browsing information of the target user, determine the browsing keywords and the current location information according to the current browsing information.

[0075] Among them, the browsing keyword can be a word or a short sentence. Since the key information cannot be used directly to locate the user's browsing habits, the key information can be mapped with the user's browsing habits in advance. For example, the user's usual habits are basically fixed in a certain period of time, but some external factors may cause the user's browsing habits to change. Furthermore, the user's current location information can be obtained by locating the user. If the user uses a desktop computer, the user's current geographic location can be obtained through the user's IP; if the user uses a mobile phone, the user can be located through the GPS global positioning system.

[0076] Step 202: Determine the first browsing habit data of the target user according to the browsing keyword.

[0077] Specifically, after a mapping relationship between key information and user browsing habits is generated, the mapping relationship may be stored in a background database; when a user browses information, key information may be retrieved from the current user's browsing information through a search engine, and the key information may be pre-set, such as pre-defining a key information library. When searching for browsing information, the key information in the key information library is searched to obtain a ranking of the key information, and top-ranked key information is obtained, such as the top three or top five, and then first browsing habit data is obtained based on the key information through a mapping relationship between the key information and the user's browsing habits.

[0078] Step 203: Determine the second browsing habit data of the target user according to the current positioning information.

[0079] Specifically, after obtaining the first browsing habit data and the second browsing habit data, the priority of the first browsing habit data and the second browsing habit data can be set. For example, the priority of the first browsing habit data is set to high priority, and the priority of the second browsing habit data is set to low priority. Then, when pushing information to the user, information is pushed to the user based on the first browsing habit data first, and then target products are pushed to the user based on the second browsing habit data.

[0080] Step 204: Generate a target push path according to the first browsing habit data, the second browsing habit data and each target product.

[0081] Furthermore, in the case where the current browsing information of the target user is not obtained, similar users corresponding to the target user are determined; and a target push path is generated according to the target browsing habit data of each target product and similar users.

[0082] Optionally, the customer group to which the target user belongs may be determined first, and within the customer group, similar users corresponding to the target user may be determined through the cosine similarity between users.

[0083] In one embodiment, a target push path is generated based on the target browsing habit data of each target product and target user, including: when a new product is detected online, a new product is determined based on the similarity between the new product and the target product; and a target push path is generated based on the target browsing habit data of each target product, the new marketing product and the target user.

[0084] In this embodiment, new products are also added to the target push path, so that newly launched products can be pushed in a timely manner.

[0085] It should be understood that, although the steps in the flowcharts involved in the above embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0086] Based on the same inventive concept, the embodiment of the present application also provides a product pushing device for implementing the product pushing method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations of one or more product pushing device embodiments provided below can refer to the limitations of the product pushing method above, and will not be repeated here.

[0087] In an exemplary embodiment, as shown in FIG3 , a product push device 1 is provided, comprising: an acquisition module 11, a recommendation module 12 and a selection module 13, wherein:

[0088] The acquisition module 11 is used to acquire the user information and purchased product set corresponding to the target user;

[0089] The recommendation module 12 is used to input the user information of the target user into the recommendation model to obtain the recommended product set output by the recommendation model, wherein the user information includes at least user identity information, user asset information, user transaction information and user preference information;

[0090] A selection module 13, configured to select a target product corresponding to a target user from the purchased product set and the recommended product set according to the product index corresponding to the purchased product set and the product index corresponding to the recommended product set;

[0091] Among them, product indicators include at least product return dimension indicators, product risk dimension indicators and product capital scale dimension indicators.

[0092] In an exemplary embodiment, the apparatus further includes a construction module, the construction module being used to: obtain user information of a sample user, and a sample product combination selected by the sample user information from a selectable product set, wherein the sample product combination includes products of at least one category;

[0093] The user information corresponding to the sample users and the sample product combination are used as training samples to train the initial model and obtain the recommendation model.

[0094] In an exemplary embodiment, the device further includes a set selection module 13, wherein the set selection module 13 is used to

[0095] Input product information pairs of optional products into the mean square error model to obtain each effective product combination;

[0096] Build optional product sets based on each valid product combination.

[0097] Each module in the above-mentioned product push device can be implemented in whole or in part by software, hardware or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each module.

[0098] In an exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be shown in FIG4. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit, and an input device. The processor, the memory, and the input / output interface are connected via a system bus, and the communication interface, the display unit, and the input device are connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner may be implemented through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a product push method is implemented. The display unit of the computer device is used to form a visually visible picture, which may be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0099] Those skilled in the art will understand that the structure shown in FIG. 4 is merely a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0100] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0101] Obtain key feature data of target users, including multi-dimensional key identity feature data and multi-dimensional key behavior feature data;

[0102] Inputting the key feature data of the target user into the target hypernetwork model to obtain the target product output by the target hypernetwork model;

[0103] Generate target push paths based on target products and target users’ target browsing habits data;

[0104] Push target products to target users based on the target push path.

[0105] In one embodiment, when the processor executes the computer program, the following steps are also implemented: iteratively training the initial hypernetwork model according to the key feature data of the sample users; if the accuracy of the initial hypernetwork model meets the requirements, the initial hypernetwork model is used as the target hypernetwork model.

[0106] In one embodiment, when the processor executes the computer program, the following steps are also implemented: obtaining initial feature data of sample users; extracting core feature data from the initial feature data based on the influencing factors of each dimensional feature in the initial feature data; and performing dimensionality reduction processing on the core feature data to obtain key feature data of the sample users.

[0107] In one embodiment, the processor also implements the following steps when executing the computer program: when obtaining the current browsing information of the target user, determining the browsing keywords and the current positioning information according to the current browsing information; determining the first browsing habit data of the target user according to the browsing keywords; determining the second browsing habit data of the target user according to the current positioning information; and generating a target push path according to the first browsing habit data, the second browsing habit data and each target product.

[0108] In one embodiment, when the processor executes the computer program, the following steps are also implemented: determining similar users corresponding to the target user without obtaining the current browsing information of the target user; generating a target push path based on the target browsing habit data of each target product and similar users.

[0109] In one embodiment, when the processor executes the computer program, the following steps are also implemented: when a new product is detected to be online, a new product is determined based on the similarity between the new product and the target product; and a target push path is generated based on the target browsing habit data of each target product, the new marketing product and the target user.

[0110] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0111] Obtain key feature data of target users, including multi-dimensional key identity feature data and multi-dimensional key behavior feature data;

[0112] Inputting the key feature data of the target user into the target hypernetwork model to obtain the target product output by the target hypernetwork model;

[0113] Generate target push paths based on target products and target users’ target browsing habits data;

[0114] Push target products to target users based on the target push path.

[0115] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: iteratively training the initial hypernetwork model according to the key feature data of the sample users; if the accuracy of the initial hypernetwork model meets the requirements, the initial hypernetwork model is used as the target hypernetwork model.

[0116] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining initial feature data of sample users; extracting core feature data from the initial feature data based on the influencing factors of each dimensional feature in the initial feature data; and performing dimensionality reduction processing on the core feature data to obtain key feature data of the sample users.

[0117] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: when the current browsing information of the target user is obtained, the browsing keywords and the current positioning information are determined according to the current browsing information; the first browsing habit data of the target user is determined according to the browsing keywords; the second browsing habit data of the target user is determined according to the current positioning information; and the target push path is generated according to the first browsing habit data, the second browsing habit data and each target product.

[0118] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: determining similar users corresponding to the target user without obtaining the current browsing information of the target user; generating a target push path based on the target browsing habit data of each target product and similar users.

[0119] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: when a new product is detected online, a new product is determined based on the similarity between the new product and the target product; and a target push path is generated based on the target browsing habit data of each target product, the new marketing product and the target user.

[0120] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps:

[0121] Obtain key feature data of target users, including multi-dimensional key identity feature data and multi-dimensional key behavior feature data;

[0122] Inputting the key feature data of the target user into the target hypernetwork model to obtain the target product output by the target hypernetwork model;

[0123] Generate target push paths based on target products and target users’ target browsing habits data;

[0124] Push target products to target users based on the target push path.

[0125] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: iteratively training the initial hypernetwork model according to the key feature data of the sample users; if the accuracy of the initial hypernetwork model meets the requirements, the initial hypernetwork model is used as the target hypernetwork model.

[0126] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining initial feature data of sample users; extracting core feature data from the initial feature data based on the influencing factors of each dimensional feature in the initial feature data; and performing dimensionality reduction processing on the core feature data to obtain key feature data of the sample users.

[0127] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: when the current browsing information of the target user is obtained, the browsing keywords and the current positioning information are determined according to the current browsing information; the first browsing habit data of the target user is determined according to the browsing keywords; the second browsing habit data of the target user is determined according to the current positioning information; and the target push path is generated according to the first browsing habit data, the second browsing habit data and each target product.

[0128] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: determining similar users corresponding to the target user without obtaining the current browsing information of the target user; generating a target push path based on the target browsing habit data of each target product and similar users.

[0129] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: when a new product is detected online, a new product is determined based on the similarity between the new product and the target product; and a target push path is generated based on the target browsing habit data of each target product, the new marketing product and the target user.

[0130] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0131] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0132] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0133] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A product push method, characterized in that: The method comprises: Acquire key feature data of the target user, wherein the key feature data includes multi-dimensional key identity feature data and multi-dimensional key behavior feature data; Inputting the key feature data of the target user into the target hypernetwork model to obtain the target product output by the target hypernetwork model; Generate a target push path according to the target product and the target browsing habit data of the target user; According to the target push path, the target product is pushed to the target user.

2. The method according to claim 1, characterized in that The hypernetwork model is trained in the following way: Iteratively train the initial hypernetwork model based on the key feature data of sample users; If the accuracy of the initial hypernetwork model meets the requirement, the initial hypernetwork model is used as the target hypernetwork model.

3. The method according to claim 2, characterized in that The method further comprises: Acquiring initial feature data of the sample user; Extracting core feature data from the initial feature data according to the influencing factors of the features of each dimension in the initial feature data; The core feature data is subjected to dimensionality reduction processing to obtain key feature data of the sample user.

4. The method according to claim 1, characterized in that: The target browsing habit data includes first browsing habit data and second browsing habit data; generating a target push path according to the target product and the target browsing habit data of the target user includes: When the current browsing information of the target user is obtained, determining browsing keywords and current location information according to the current browsing information; Determining first browsing habit data of the target user according to the browsing keyword; Determining second browsing habit data of the target user according to the current positioning information; The target push path is generated according to the first browsing habit data, the second browsing habit data and the target product.

5. The method according to claim 1, characterized in that The generating a target push path according to the target product and the target browsing habit data of the target user includes: In the case where the current browsing information of the target user is not obtained, determining similar users corresponding to the target user; A target push path is generated according to each target product and the target browsing habit data of the similar users.

6. The method according to claim 1, characterized in that The generating a target push path according to the target product and the target browsing habit data of the target user includes: When a new product is detected to be online, a new product is determined based on the similarity between the new product and the target product; A target push path is generated based on the target products, new marketing products and the target browsing habit data of the target users.

7. A product pushing device, characterized in that: The device comprises: An acquisition module, used to acquire key feature data of a target user, wherein the key feature data includes multi-dimensional key identity feature data and multi-dimensional key behavior feature data; A product push module, used to input the key feature data of the target user into the target hypernetwork model to obtain the target product output by the target hypernetwork model; A path generation module, used to generate a target push path according to the target product and the target browsing habit data of the target user; A push module is used to push the target product to the target user according to the target push path.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images