Product pushing material generation method and device, equipment, storage medium and product

Abstract

The invention provides a product pushing material generation method and device, equipment, a storage medium and a product. Relates to the technical field of data processing. The method comprises the steps of obtaining user behavior data and user basic data, obtaining user portrait data from multiple dimensions based on the user behavior data and the user basic data, and ensuring that the user portrait data is adaptive to user requirements; by constructing the multi-dimensional vector of the user portrait and the multi-dimensional vector of each piece of product information, the to-be-recommended product information for the current user is determined, associated data of the user and the product is formed, it is further ensured that the to-be-recommended product information meets user requirements, and the user experience feeling is improved. And generating a product pushing material according to the user behavior data and the to-be-recommended product information. The whole process does not need manual intervention, and the generation efficiency of the product pushing material is improved.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method, apparatus, device, storage medium, and product for generating product push materials. Background Technology

[0002] With the rapid development of fintech, banks need to present their diverse financial products to customers in an efficient and intelligent way to improve customer experience and push efficiency.

[0003] Currently, the traditional process of producing push notification materials relies on manual design and content editing, which is not only time-consuming and labor-intensive, but also difficult to respond to user needs, resulting in reduced efficiency in generating product push notification materials and a poor user experience. Summary of the Invention

[0004] This application provides a method, apparatus, device, storage medium, and product for generating product push materials, in order to improve the efficiency of product push material generation and user experience.

[0005] In a first aspect, embodiments of this application provide a method for generating product push materials, comprising: responding to user interaction operations on an interactive interface, collecting user behavior data on the interactive interface; receiving user basic data sent by a data terminal; obtaining user profile data of the current user based on the user behavior data and the user basic data; constructing a multi-dimensional vector of the user profile based on the user profile data; constructing a multi-dimensional vector of each product information based on each product information in a product database; determining the product information to be recommended for the current user based on the multi-dimensional vector of the user profile and the multi-dimensional vector of each product information; generating product push materials based on the product information to be recommended and the user behavior data; and displaying the product push materials on the interactive interface.

[0006] In one possible implementation, collecting user behavior data on the interactive interface includes: obtaining display platform information of the interactive interface; obtaining tracking technology adapted to the display platform information; embedding collection nodes in the interactive interface through tracking technology; and activating the collection nodes when the user performs interactive operations on the interactive interface, so that the collection nodes collect user behavior data of the interactive operations.

[0007] In one possible implementation, user profile data for the current user is obtained based on user behavior data and user basic data, including: extracting user behavior features based on user behavior data; obtaining user preference tags for the current user based on user behavior features using a clustering algorithm; obtaining potential user demand tags for the current user based on user behavior data using an association rule mining algorithm; obtaining basic user attribute tags for the current user based on user basic data according to preset rules; and determining the user preference tags, potential user demand tags, and basic user attribute tags as user profile data.

[0008] In one possible implementation, product information to be recommended for the current user is determined based on the multidimensional vector of the user profile and the multidimensional vector of each product information, including: obtaining the similarity value between each product information and the user profile data based on the multidimensional vector of the user profile and the multidimensional vector of each product information; and identifying product information with similarity values ​​exceeding a preset similarity threshold as recommended product information for the current user.

[0009] In one possible implementation, product push materials are generated based on the product information to be recommended and user behavior data, including: obtaining the product display format selected by the user from the user behavior data; wherein the product display format includes a product display template; and generating product push materials based on the product information to be recommended according to the product display template.

[0010] In one possible implementation, the method further includes: collecting historical user behavior data and historical user basic data of multiple users, as well as historical product information to be recommended and historical product information purchased by each user; obtaining the behavior similarity value between user behavior data and historical user behavior data of each user; obtaining the basic data similarity value between user basic data and historical user basic data of each user; if the behavior similarity value of any user exceeds a preset behavior similarity threshold, and the basic data similarity value exceeds a preset basic data similarity threshold, then the historical product information to be recommended and historical product information purchased by any user are determined as the product information to be recommended for the current user.

[0011] Secondly, embodiments of this application provide a device for generating product push materials, comprising: a collection module for collecting user behavior data on the interactive interface in response to user interaction operations; a receiving module for receiving basic user data sent by a data terminal; an acquisition module for acquiring user profile data of the current user based on the user behavior data and the basic user data; a first construction module for constructing a multi-dimensional vector of the user profile based on the user profile data; a second construction module for constructing a multi-dimensional vector of each product information based on each product information in a product database; a determination module for determining the product information to be recommended for the current user based on the multi-dimensional vector of the user profile and the multi-dimensional vector of each product information; a generation module for generating product push materials based on the product information to be recommended and the user behavior data; and a display module for displaying the product push materials on the interactive interface.

[0012] Thirdly, embodiments of this application provide an apparatus, including: a memory and a processor;

[0013] The memory stores the instructions that the computer executes;

[0014] The processor executes computer execution instructions stored in memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0015] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible implementations of the first aspect.

[0016] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0017] The product push material generation method, apparatus, device, storage medium, and product provided in this application embodiment respond to user interaction operations on the interactive interface, collect user behavior data, and receive user basic data sent by the data terminal; based on the user behavior data and user basic data, obtain the current user's user profile data; starting from the user behavior data and user basic data, obtain user profile data from multiple dimensions to ensure that the product push materials are adapted to user needs; by constructing a multi-dimensional vector of the user profile and a multi-dimensional vector of each product information, determine the product information to be recommended for the current user, forming user-product association data, further ensuring that the product information to be recommended meets user needs and improving user experience. Product push materials are generated based on user behavior data and the product information to be recommended. The entire process requires no manual intervention, improving the generation efficiency of product push materials. This method obtains the current user's user profile data through user behavior data and user basic data, realizing a data-driven automated generation process for product push materials, solving the problems of traditional push material production relying on manual labor and low response efficiency. It improves the generation efficiency of product push materials and user experience. Attached Figure Description

[0018] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0019] Figure 1 A schematic diagram illustrating a scenario for the method of generating product push materials provided in an embodiment of this application;

[0020] Figure 2 A flowchart illustrating the method for generating product promotion materials provided in this application embodiment. Figure 1 ;

[0021] Figure 3 A flowchart illustrating the method for generating product promotion materials provided in this application embodiment. Figure 2 ;

[0022] Figure 4 This is a schematic diagram of the structure of the product delivery material generation device provided in the embodiments of this application;

[0023] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0024] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0025] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0026] The collection, storage, use, processing, transmission, provision, and disclosure of financial data or user data involved in the technical solution of this application all comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0027] It should be noted that in the embodiments of this application, certain software, components, models and other existing solutions in the industry may be mentioned. These should be regarded as exemplary and are only intended to illustrate the feasibility of implementing the technical solution of this application. However, it does not mean that the applicant has used or necessarily used the solution.

[0028] Figure 1 This is a schematic diagram illustrating a scenario for the method of generating product push materials provided in an embodiment of this application, such as... Figure 1 As shown, it includes: a receiving device 101, a processing device 102, and a display device 103.

[0029] It is understood that the structure illustrated in the embodiments of this application does not constitute a specific limitation on the method for generating product push materials. In other feasible embodiments of this application, the above architecture may include more or fewer components than illustrated, or combine some components, or split some components, or arrange different components, which can be determined according to the actual application scenario and is not limited here. Figure 1 The components shown can be implemented in hardware, software, or a combination of both.

[0030] In the specific implementation process, the receiving device 101 can be an input / output interface or a communication interface, which can collect user behavior data on the interactive interface and receive user basic data sent by the data terminal.

[0031] The processing device 102 can obtain user profile data of the current user based on user behavior data and user basic data; construct a multi-dimensional vector of the user profile based on the user profile data; construct a multi-dimensional vector of each product information based on the product information in the product database; determine the product information to be recommended for the current user based on the multi-dimensional vector of the user profile and the multi-dimensional vector of each product information; and generate product push materials based on the product information to be recommended and the user behavior data.

[0032] The display device 103 can be used to display the materials pushed to the product.

[0033] It should be understood that the aforementioned processor can be implemented by reading instructions from memory and executing those instructions, or it can be implemented through chip circuitry.

[0034] Furthermore, the network architecture and business scenarios described in the embodiments of this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided in the embodiments of this application. As those skilled in the art will know, with the evolution of network architecture and the emergence of new business scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0035] To address the aforementioned technical problems, this application proposes the following technical concept: The inventors have conceived of a data-driven automated generation process for product push materials. Responding to user interactions on the interface, user behavior data is collected; basic user data is received from the data terminal; based on the user behavior data and basic user data, user profile data for the current user is obtained; user profile data is acquired from multiple dimensions to ensure it aligns with user needs; by constructing multi-dimensional vectors of the user profile and multi-dimensional vectors of various product information, the product information to be recommended for the current user is determined, forming a user-product association, further ensuring that the recommended product information meets user needs and improving user experience. Product push materials are generated based on user behavior data and the product information to be recommended. The entire process requires no manual intervention, improving the efficiency of product push material generation.

[0036] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0037] Figure 2 A flowchart illustrating the method for generating product promotion materials provided in this application embodiment. Figure 1 ,like Figure 2 As shown, the method includes:

[0038] S201: In response to user interaction on the interactive interface, collect user behavior data on the interactive interface.

[0039] Specifically, the process involves: acquiring information about the display platform of the interactive interface; acquiring tracking technologies that are compatible with the display platform information; embedding data collection nodes into the interactive interface using tracking technologies; and activating the data collection nodes when the user interacts with the interactive interface, so that the data collection nodes collect user behavior data of the interactive operations.

[0040] In this embodiment, the platform information includes mobile applications, mini-programs, websites, and touchscreen devices.

[0041] In this embodiment, the display platform information to which the interactive interface belongs is obtained, and the tracking technology is matched according to the operating environment, technology stack characteristics and interaction scenarios of the display platform information.

[0042] Optionally, data collection nodes can be deployed at page access nodes, click operation nodes, browsing behavior nodes, and input interaction nodes, etc.

[0043] In this embodiment, user interaction operations include behavioral data such as product browsing history, product browsing order, button click operations, product viewing operations, and duration of time spent on the interactive interface.

[0044] S202: Receive user basic data sent by the data terminal.

[0045] User basic data refers to basic information that users fill in or are authorized to obtain when registering in the system, and that is not easily changed with interactive behavior, including age, occupation, contact information, region, and education.

[0046] S203: Obtain the current user profile data based on user behavior data and basic user data.

[0047] The user profile data includes user preference tags, user potential need tags, and user basic attribute tags.

[0048] Specifically, user behavior features are extracted based on user behavior data; user preference tags are obtained based on user behavior features using a clustering algorithm; potential user need tags are obtained based on user behavior data using an association rule mining algorithm; basic user attribute tags are obtained based on user basic data according to preset rules; and user preference tags, potential user need tags, and basic user attribute tags are used to determine user profile data.

[0049] In this embodiment, a clustering algorithm is used, with behavioral preference features as the clustering target. Users with similar behavioral preference features are grouped into one group, and user preference labels for the same group are obtained. The essence of the clustering algorithm is to group similar features into one cluster. A single user cannot be clustered with other users. Therefore, before executing step S203, historical user behavior data from multiple users is collected, and the behavioral preference features of each user's behavior data are extracted.

[0050] Specifically, an association rule mining algorithm is used to analyze the correlations between user behavior data, such as browsing mortgage products and clicking on home renovation loans, or checking fixed deposits and paying attention to children's education and financial management. If a user's behavior satisfies the condition that browsing product A and operating on product B has a support rate greater than 20% and a confidence rate greater than 60%, then an associated demand is determined, and a potential demand tag is generated. For example, if a user browses mortgage products multiple times and then frequently checks the home renovation loan calculator, a tag indicating potential home renovation loan demand is generated.

[0051] Optionally, the association rule mining algorithm is the Apriori algorithm.

[0052] Optionally, both user preference tags and user potential demand tags can reveal potential business opportunity clues.

[0053] In this embodiment, user basic attribute tags are directly mapped to the current user based on user basic data according to preset rules. For example, age 20-35 years old is mapped to youth tag, 36-55 years old is mapped to middle-aged tag, and 56 years and above is mapped to elderly tag; occupation of corporate management is mapped to corporate executive tag, etc.

[0054] S204: Construct a multi-dimensional vector of the user profile based on the user profile data.

[0055] In this embodiment, user profile data includes user preference tags, user potential need tags, and user basic attribute tags.

[0056] In this embodiment, a multi-dimensional vector of user profile is constructed based on user preference tags, user potential demand tags, and user basic attribute tags. This vector includes three dimensions: product benefit sensitivity, risk preference fit, and term preference fit.

[0057] For example, regarding the product revenue sensitivity corresponding to user preference tags: if the frequency of clicking on revenue details is greater than or equal to the high-frequency click threshold, the product revenue sensitivity corresponding to the user preference tag is 1; if the frequency of clicking on revenue details is greater than or equal to the medium-frequency click threshold and less than the high-frequency click threshold, the product revenue sensitivity corresponding to the user preference tag is 0.7; if the frequency of clicking on revenue details is less than the medium-frequency click threshold, the product revenue sensitivity corresponding to the user preference tag is 0.4. The high-frequency click threshold and the medium-frequency click threshold are preset and can be adjusted according to actual conditions.

[0058] For example, regarding the sensitivity of product benefits corresponding to user potential demand tags: if the user potential demand tags include retirement needs and children's education needs, the sensitivity of product benefits corresponding to the user potential demand tags is 0.9; if the user potential demand tags do not include retirement needs and children's education needs, the sensitivity of product benefits corresponding to the user potential demand tags is 0.2.

[0059] For example, regarding the product revenue sensitivity corresponding to user basic attribute tags: if the asset size in the user's basic attribute tags is greater than or equal to the high asset threshold, the product revenue sensitivity corresponding to the user's basic attribute tags is 1; if the asset size is greater than or equal to the medium asset threshold but less than the high asset threshold, the product revenue sensitivity corresponding to the user's basic attribute tags is 0.8; if the asset size is less than the medium asset threshold, the product revenue sensitivity corresponding to the user's basic attribute tags is 0.5. The high asset threshold and the medium asset threshold are preset and can be adjusted according to actual circumstances.

[0060] For example, regarding product benefit sensitivity, the preset weights for user preference tags are 0.5, for user potential demand tags are 0.2, and for user basic attribute tags are 0.3. These preset weights can be adjusted according to actual circumstances. The product benefit sensitivity value is obtained by weighted summation of the product benefit sensitivity corresponding to user preference tags, user potential demand tags, and user basic attribute tags.

[0061] For example, regarding the risk preference fit corresponding to a user preference tag: if the frequency of purchasing high-risk products is greater than or equal to the high-frequency purchase threshold, the risk preference fit corresponding to the user preference tag is 1; if the frequency of purchasing high-risk products is greater than or equal to the medium-frequency purchase threshold but less than the high-frequency purchase threshold, the risk preference fit corresponding to the user preference tag is 0.7; if the frequency of purchasing high-risk products is less than the medium-frequency purchase threshold, the risk preference fit corresponding to the user preference tag is 0.4. The high-frequency purchase threshold and the medium-frequency purchase threshold are preset and can be adjusted according to actual circumstances.

[0062] For example, regarding the risk preference fit of user potential demand tags: if the user potential demand tag has investment appreciation demand, the risk preference fit of the user potential demand tag is 0.9; if the user potential demand tag has housing demand or consumption demand, the risk preference fit of the user potential demand tag is 0.4, and the risk preference fit of the remaining user potential demand tags is 0.5.

[0063] For example, regarding the risk preference fit of the user's basic attribute tags: if it is a youth tag, the risk preference fit of the user preference tag is 0.8; if it is a middle-aged tag, the risk preference fit of the user preference tag is 0.6; if it is an elderly tag, the risk preference fit of the user preference tag is 0.3.

[0064] For example, regarding risk preference fit, the preset weights for user preference tags are 0.5, for user potential need tags are 0.2, and for user basic attribute tags are 0.3. These preset weights can be adjusted according to actual circumstances. The risk preference fit value is obtained by weighted summation of the risk preference fit values ​​corresponding to user preference tags, user potential need tags, and user basic attribute tags.

[0065] For example, regarding the time-limited preference fit for user preference tags: if the frequency of browsing long-term products is greater than or equal to the high-frequency browsing threshold, the time-limited preference fit for the user preference tag is 1; if the frequency of browsing medium-term products is greater than or equal to the high-frequency browsing threshold, the time-limited preference fit for the user preference tag is 0.7; if the frequency of browsing short-term products is greater than or equal to the high-frequency browsing threshold, the time-limited preference fit for the user preference tag is 0.4. The high-frequency browsing threshold is preset and can be adjusted according to actual conditions.

[0066] For example, regarding the term preference fit of a user's potential demand tag: if the user's potential demand tag includes retirement needs or children's education needs, the term preference fit of the user's potential demand tag is 0.9; if the user's potential demand tag includes housing needs, the term preference fit of the user's potential demand tag is 0.7; and the risk preference fit of the remaining term preference fits is 0.4.

[0067] For example, regarding the time-limit preference fit of the user's basic attribute tags: if it is a youth tag, the time-limit preference fit of the user's preference tag is 0.3, which belongs to short-term preference; if it is a middle-aged tag, the time-limit preference fit of the user's preference tag is 0.7, which belongs to medium-term preference; if it is an elderly tag, the time-limit preference fit of the user's preference tag is 0.9, which belongs to long-term preference.

[0068] For example, regarding term preference fit, the preset weights for user preference tags are 0.5, for user potential demand tags are 0.2, and for user basic attribute tags are 0.3. These preset weights can be adjusted according to actual circumstances. The term preference fit value is obtained by weighted summation of the term preference fit values ​​corresponding to user preference tags, user potential demand tags, and user basic attribute tags.

[0069] For example, the multidimensional vector of the user profile is [0.6, 0.9, 0.3].

[0070] S205: Based on the product information in the product database, construct a multi-dimensional vector of each product information.

[0071] In this embodiment, the product database aggregates bank product content, covering product images, video materials, detailed descriptions, and related information such as interest rates.

[0072] In this embodiment, the multi-dimensional vector of the user profile includes three dimensions: product return sensitivity, risk preference fit, and term preference fit. Product return sensitivity represents the product return range, risk preference fit represents the product risk level, and term preference fit represents the product term.

[0073] In this embodiment, quantitative values ​​for different dimensions are determined based on the product attributes of each product information. These quantitative values ​​are determined and adjusted according to actual circumstances. For example, different product return ranges correspond to different product return sensitivities, with the product return range increasing from low to high, and the corresponding product return sensitivities also increasing accordingly. Similarly, product risk levels are divided into five levels, from low to high risk: R1, R2, R3, R4, and R5, with corresponding risk preference fits of 1, 0.8, 0.6, 0.4, and 0.2, respectively. Likewise, different product terms correspond to different term preference fits, with the product term increasing from low to high, and the corresponding term preference fits also increasing accordingly.

[0074] For example, the multidimensional vector of any product information is [0.6, 0.6, 0.8].

[0075] S206: Based on the multidimensional vector of the user profile and the multidimensional vector of each product information, determine the product information to be recommended for the current user.

[0076] Specifically, based on the multidimensional vectors of user profiles and the multidimensional vectors of each product information, the similarity value between each product information and the user profile data is obtained; product information with similarity values ​​exceeding a preset similarity threshold is identified as recommended product information for the current user.

[0077] In this embodiment, a cosine similarity algorithm can be used to obtain the similarity value between each product information and the user profile data.

[0078] In this embodiment, the preset similarity threshold is pre-set, which can be 0.8, or it can be adjusted according to the actual situation.

[0079] In this embodiment, the product information to be recommended can be one or multiple. As long as the similarity value exceeds the preset similarity threshold, it is determined to be the product information to be recommended.

[0080] S207: Generate product push materials based on the product information to be recommended and user behavior data.

[0081] Specifically, the product display format selected by the user is obtained from user behavior data; the product display format includes a product display template; the product information to be recommended is used to generate product push materials according to the product display template.

[0082] In this embodiment, the product display format selected by the user refers to the product display format selected by the user through a click operation on the interactive interface.

[0083] In this embodiment, the intelligent building engine is invoked to generate product recommendation materials based on the selected product template from the product information to be recommended. The intelligent building engine is the underlying code logic that generates product recommendation materials from the selected product template.

[0084] Optionally, if the user does not select a product display format in the interactive interface, the default product display format will be used.

[0085] S208: Display product promotion materials on the interactive interface.

[0086] In this embodiment, the display channel module supports display on various online or offline channels, including online mobile applications, mini-programs, and touchscreen devices. The displayed content can be optimized according to the characteristics of different display channels to maximize the display effect.

[0087] Optionally, natural language processing and speech synthesis technologies can be used to provide digital virtual human-based broadcast guidance services. The digital virtual human can flexibly adjust its broadcast content, speaking speed, and style based on user feedback, thus providing a personalized service experience. For example, it can broadcast product promotion materials.

[0088] In summary, this method responds to user interactions on the interface, collects user behavior data, and receives basic user data from the data terminal. Based on the user behavior data and basic user data, it obtains the current user's profile data. Starting with the user behavior data and basic user data, it acquires user profile data from multiple dimensions to ensure that product push materials are adapted to user needs. By constructing multi-dimensional vectors of the user profile and various product information, it determines the product information to be recommended for the current user, forming user-product association data, further ensuring that the recommended product information meets user needs and improving user experience. Based on the user behavior data and the product information to be recommended, product push materials are generated. The entire process requires no manual intervention, improving the efficiency of product push material generation. This method obtains the current user's profile data through user behavior data and basic user data, realizing a data-driven automated process for generating product push materials, solving the problems of traditional push material production relying on manual labor and low response efficiency. It improves the efficiency of product push material generation and user experience.

[0089] refer to Figure 3 , Figure 3 A flowchart illustrating the method for generating product promotion materials provided in this application embodiment. Figure 2 Based on the above embodiments, this embodiment describes another process for determining product information to be recommended, detailed below:

[0090] S301: Collect historical user behavior data and historical user basic data from multiple users, as well as historical product information to be recommended and historical product information purchased by each user.

[0091] In this embodiment, the collected data covers historical user behavior data and historical user basic data of multiple users. The system logs are used to associate the product information that each user has been recommended with, as well as the product information that each user has purchased in the past after the product information was recommended to each user, forming a three-dimensional mapping library of user behavior data, user basic data, recommended product information, and purchased product information.

[0092] S302: Obtain the behavioral similarity value between user behavior data and each user's historical user behavior data.

[0093] In this embodiment, a current user behavior vector is constructed based on current user behavior data, and a user behavior vector for each user is constructed based on each user's historical user behavior data. For example, browsing the credit card page twice, clicking on the consumer loan page once, and staying on the financial management page for 3 minutes. For example, the credit card interaction score is 0.4, the consumer loan interaction score is 0.5, the financial management interaction score is 0.3, and the current user behavior vector = [0.4, 0.5, 0.3].

[0094] In this embodiment, the Pearson correlation coefficient algorithm is used to calculate the similarity value between the historical user behavior data of each user and the current user behavior data. The value ranges from -1 to 1, and the closer it is to 1, the more similar the behavior is.

[0095] S303: Obtain the basic data similarity value between the user's basic data and the historical user basic data of each user.

[0096] In this embodiment, a vector of user basic data is constructed. User basic data, such as age 35, enterprise personnel, medium to high assets, and married with children, is standardized according to historical data format. Discrete attributes are encoded as 0-1 values, such as enterprise personnel = 1 and married with children = 1. Continuous attributes are normalized, such as age 35 being normalized to 0.42 in the 20-60 age range. For example, the vector of user basic data is [0.42, 1, 1].

[0097] Optionally, a combined similarity algorithm is used to adapt the basic data of discrete and continuous attributes. The similarity values ​​of discrete attributes, the similarity values ​​of continuous attributes, the preset weights of discrete attributes, and the preset weights of continuous attributes are weighted and summed to obtain the similarity value of the basic data.

[0098] The preset weights for discrete attributes and continuous attributes are pre-set and can be adjusted according to actual conditions.

[0099] S304: If the behavior similarity value of any user exceeds the preset behavior similarity threshold, and the basic data similarity value exceeds the preset basic data similarity threshold, then the historical product information to be recommended and the historical purchased product information corresponding to any user will be determined as the product information to be recommended for the current user.

[0100] In this embodiment, both the preset behavior similarity threshold and the preset basic data similarity threshold are set to 0.8. However, they can be adjusted depending on the bank recommendation scenario.

[0101] In this embodiment, related products are extracted: from the three-dimensional mapping library of user behavior data and user basic data-recommended product information-purchased product information, the historical recommended product information and historical purchased product information of users whose behavior similarity value exceeds a preset behavior similarity threshold and whose basic data similarity value exceeds a preset basic data similarity threshold are extracted; if the historical recommended product information and the historical purchased product information are consistent, it means that the historical product information to be recommended meets the user's purchase needs, and the historical recommended product information is then used as the product information to be recommended.

[0102] In this embodiment, the purpose of steps S301 to S304 is to develop a collaborative filtering recommendation scheme based on the similarity of user behavior data and the similarity of user basic data. The core idea is to find historical user behavior data and user basic data with similarity and reuse their highly adaptable product information, which is suitable for scenarios such as new user cold start and insufficient user profile dimensions.

[0103] In summary, collecting historical user behavior data and basic user data, along with historical product information for each user and their purchased products, transforms scattered historical behavior data into reusable recommendation resources. By calculating behavioral similarity values ​​and filtering based on preset behavioral similarity thresholds, as well as calculating user basic data similarity values ​​and filtering based on preset user basic data similarity thresholds, the product information to be recommended is determined. This allows new users to obtain recommended product information by associating with historical user behavior data with minimal interaction, improving the applicability and accuracy of product recommendations, thereby increasing user satisfaction.

[0104] Figure 4This is a schematic diagram of the structure of the product delivery material generation device provided in the embodiments of this application, as shown below. Figure 4 As shown, the product push material generation device provided in this embodiment includes: a collection module 401, a receiving module 402, an acquisition module 403, a first construction module 404, a second construction module 405, a determination module 406, a generation module 407, and a display module 408.

[0105] The data acquisition module 401 is used to collect user behavior data in response to user interaction operations on the interactive interface.

[0106] Receiver module 402 is used to receive user basic data sent by the data terminal;

[0107] The acquisition module 403 is used to acquire the user profile data of the current user based on user behavior data and user basic data;

[0108] The first construction module 404 is used to construct a multi-dimensional vector of the user profile based on the user profile data;

[0109] The second construction module 405 is used to construct a multi-dimensional vector of product information based on the product information in the product database.

[0110] The determination module 406 is used to determine the product information to be recommended for the current user based on the multi-dimensional vector of the user profile and the multi-dimensional vector of each product information;

[0111] The generation module 407 is used to generate product push materials based on the product information to be recommended and user behavior data.

[0112] The display module 408 is used to display product promotion materials on the interactive interface.

[0113] In one possible implementation, the acquisition module 401 is specifically used for: acquiring display platform information of the interactive interface; acquiring tracking technology adapted to the display platform information; embedding acquisition nodes in the interactive interface through tracking technology; and activating the acquisition nodes when the user performs interactive operations on the interactive interface, so that the acquisition nodes collect user behavior data of the interactive operations.

[0114] In one possible implementation, the acquisition module 403: extracts user behavior features based on user behavior data; obtains the current user's user preference tags based on the user behavior features using a clustering algorithm; obtains the current user's potential demand tags based on the user behavior data using an association rule mining algorithm; obtains the current user's basic attribute tags based on the user's basic data according to preset rules; and determines the user preference tags, potential demand tags, and basic attribute tags as user profile data.

[0115] In one possible implementation, the determining module 406 is specifically used to: obtain the similarity value between each product information and the user profile data based on the multidimensional vector of the user profile and the multidimensional vector of each product information; and determine the product information whose similarity value exceeds a preset similarity threshold as the recommended product information of the current user.

[0116] In one possible implementation, the generation module 407 is specifically used to: obtain the product display format selected by the user from user behavior data; wherein the product display format includes a product display template; and generate product push materials based on the product display template and the product information to be recommended.

[0117] In one possible implementation, the product recommendation material generation device further includes a judgment module. Specifically, the judgment module is used to: collect historical user behavior data and historical user basic data from multiple users, as well as each user's historical product information to be recommended and historical product purchase information; obtain the behavioral similarity value between the user behavior data and each user's historical user behavior data; obtain the basic data similarity value between the user's basic data and each user's historical user basic data; if the behavioral similarity value of any user exceeds a preset behavioral similarity threshold, and the basic data similarity value exceeds a preset basic data similarity threshold, then the historical product information to be recommended and the historical product purchase information corresponding to that user are determined as the current user's product information to be recommended.

[0118] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 5 As shown, the electronic device provided in this embodiment includes at least one processor 501 and a memory 502. Optionally, the electronic device further includes a communication component 503. The processor 501, memory 502, and communication component 503 are connected via a bus.

[0119] In a specific implementation, at least one processor 501 executes computer execution instructions stored in memory 502, causing at least one processor 501 to perform the above-described method.

[0120] The specific implementation process of processor 501 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0121] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0122] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0123] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0124] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0125] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0126] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0127] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0128] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0129] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0130] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0131] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0132] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0133] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A method for generating product promotion materials, characterized in that, Applied to electronic devices, including: Responding to user interactions on the interactive interface, collecting user behavior data on the interactive interface; Receive basic user data sent by the data terminal; Based on the user behavior data and the user basic data, obtain the current user's user profile data; Based on the user profile data, construct a multi-dimensional vector of the user profile; Based on the product information in the product database, construct a multi-dimensional vector for each product information; Based on the multidimensional vector of the user profile and the multidimensional vector of each product information, determine the product information to be recommended for the current user; Based on the product information to be recommended and the user behavior data, product push materials are generated; The product promotion materials are displayed on the interactive interface.

2. The method according to claim 1, characterized in that, The collection of user behavior data on the interactive interface includes: Obtain the display platform information of the interactive interface; Acquire tracking technologies that are compatible with the information of the display platform; By using the aforementioned data collection technology, data acquisition nodes are embedded in the interactive interface; When a user interacts with the interface, the data collection node is activated, enabling the data collection node to collect user behavior data related to the interaction.

3. The method according to claim 1, characterized in that, The step of obtaining the current user's user profile data based on the user behavior data and the user's basic data includes: Based on the user behavior data, extract user behavior features; Based on the user behavior characteristics, the user preference tags of the current user are obtained through clustering algorithms. Based on the user behavior data, the potential user demand tags of the current user are obtained through association rule mining algorithms. Based on the user's basic data, obtain the current user's basic attribute tags according to preset rules; The user preference tags, the user potential need tags, and the user basic attribute tags are identified as user profile data.

4. The method according to claim 1, characterized in that, The step of determining the product information to be recommended for the current user based on the multidimensional vector of the user profile and the multidimensional vector of each product information includes: Based on the multidimensional vector of the user profile and the multidimensional vector of each product information, obtain the similarity value between each product information and the user profile data; Product information with similarity values ​​exceeding a preset similarity threshold is identified as recommended product information for the current user.

5. The method according to claim 1, characterized in that, The step of generating product push materials based on the product information to be recommended and the user behavior data includes: The user's selected product display format is obtained from the user behavior data; wherein the product display format includes a product display template. The product information to be recommended is used to generate product promotion materials according to the product display template.

6. The method according to any one of claims 1-5, characterized in that, The method further includes: Collect historical user behavior data and historical user basic data from multiple users, as well as historical product information to be recommended and historical product information purchased by each user. Obtain the behavior similarity value between the user behavior data and the historical user behavior data of each user; Obtain the basic data similarity value between the user's basic data and each user's historical user basic data; If the behavioral similarity value of any user exceeds a preset behavioral similarity threshold, and the basic data similarity value exceeds a preset basic data similarity threshold, then the historical product information to be recommended and the historical purchased product information corresponding to any user will be determined as the product information to be recommended for the current user.

7. A device for generating product delivery material, characterized in that, Applied to electronic devices, including: The data collection module is used to respond to user interactions on the interactive interface and collect user behavior data on the interactive interface. The receiving module is used to receive basic user data sent by the data terminal; The acquisition module is used to acquire the current user's user profile data based on the user behavior data and the user basic data; The first construction module is used to construct a multi-dimensional vector of the user profile based on the user profile data; The second construction module is used to construct a multi-dimensional vector of product information based on the product information in the product database. The determination module is used to determine the product information to be recommended for the current user based on the multi-dimensional vector of the user profile and the multi-dimensional vector of each product information; The generation module is used to generate product push materials based on the product information to be recommended and the user behavior data; The display module is used to display the product promotion materials on the interactive interface.

8. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1-6.

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

10. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method of any one of claims 1-6.

Images