Information pushing method, apparatus, device, storage medium and product

Abstract

The application provides an information pushing method and device, equipment, storage medium and product, and relates to the field of big data. The method comprises the following steps: obtaining intention feature data of a user to be classified; inputting the intention feature data of the user to be classified into a trained user intention classification model to classify the user to be classified, so as to determine the user intention of the user; determining a user with positive user intention as a target user, and obtaining the viewing frequency of each function page of the target user; determining to push the to-be-pushed information corresponding to the function page on the function page with a higher viewing frequency; and displaying the function page with a higher viewing frequency on the user terminal of the target user. The personalized pushing of the user can be realized, the user experience is improved, the information can be pushed on the function page, the information redundancy of the homepage can be avoided, and the information pushed can be obtained by the user with positive intention.

Description

Technical Field

[0001] This application relates to the field of big data, and in particular to an information push method, apparatus, device, storage medium and product. Background Technology

[0002] To enable users of financial software to understand financial knowledge and information about wealth management products, it is necessary to push relevant information and data to users.

[0003] Currently, when pushing information to users, information is typically displayed on the homepage of financial software to all users. However, some users may not be interested in the pushed information or do not want to know it at the moment. Displaying pushed information on the homepage of financial software on each user's terminal leads to information redundancy on the homepage and affects the user experience. Summary of the Invention

[0004] This application provides an information push method, apparatus, device, storage medium, and product to solve the problem that displaying pushed information on the homepage of financial software on various user terminals leads to redundancy of homepage information and affects customer experience.

[0005] Firstly, this application provides an information push method, including:

[0006] Obtain the intention feature data of users to be classified;

[0007] Input the intention feature data of the users to be classified into the trained user intention classification model to classify the users to be classified in order to determine the users' user intentions;

[0008] Identify users with positive user intent as target users and obtain the frequency with which target users view each functional page;

[0009] The system determines which function pages are most frequently viewed and pushes the corresponding information to be pushed to them; the function pages most frequently viewed are displayed on the target user's user terminal.

[0010] Optionally, the information push method further includes a step of training a user intention classification model; the user intention classification model is a Naive Bayes classifier, and the step of training the user intention classification model includes:

[0011] Obtain sample feature data for each sample user; determine the intention label for each sample user based on the behavioral data of each sample user; train an initial user intention classification model using the sample feature data and intention labels of the sample users to obtain the trained user intention classification model.

[0012] Optionally, the information to be pushed is product information, and the step of determining the sample user's intention tags based on the sample user's behavioral data includes:

[0013] If it is determined that a user views product information more frequently than a preset threshold, or views product information for a longer period than a preset threshold, or the user engages in purchasing, saving, or recommending products, then the sample user's intention tag is determined to be positive.

[0014] Optionally, the user intention classification model is a Naive Bayes classifier. After obtaining the intention feature data of the user to be classified, the method further includes: discretizing the continuous intention feature data in the intention feature data to transform the continuous intention feature data into discrete intention feature data; the method of inputting the intention feature data of the user to be classified into the trained user intention classification model to classify the user to be classified includes: using the trained user intention classification model to classify the user to be classified based on the discrete intention feature data of the user to be classified.

[0015] Optionally, the discretization of continuous intention feature data in the intention feature data includes: discretizing the continuous intention feature data using a quantile discretization algorithm.

[0016] Optionally, obtaining the intention feature data of the users to be classified includes: calculating the income level data of the users to be classified using the income calculation formula and the inflow and outflow data and account balance data; and determining the age data, gender data, financial management experience data and income level data as the intention feature data of the users to be classified.

[0017] Optionally, determining to push product information corresponding to a frequently viewed function page includes: sorting the function pages in descending order of viewing frequency; determining the function pages ranked before the preset frequency as frequently viewed function pages; and if it is determined that there is information to be pushed corresponding to a frequently viewed function page, then determining to push the information to be pushed corresponding to that function page on the frequently viewed function page.

[0018] Optionally, determining to push the information to be pushed to the function page corresponding to the function page that is viewed frequently includes: determining to use digital human technology to push the information to be pushed to the function page corresponding to the function page that is viewed frequently.

[0019] Secondly, this application provides an information push device, comprising:

[0020] The acquisition module is used to acquire the intention feature data of users to be classified;

[0021] The classification module is used to input the intention feature data of the users to be classified into the trained user intention classification model to classify the users to be classified in order to determine the users' user intentions.

[0022] The user identification module is used to identify users with positive user intentions as target users and obtain the viewing frequency of target users on each functional page.

[0023] The page determination module is used to determine which function pages are most frequently viewed and push the corresponding information to be pushed; the function pages most frequently viewed are displayed on the target user's user terminal.

[0024] Thirdly, this application provides an electronic device, including: a processor, and a memory and a transceiver communicatively connected to the processor;

[0025] The memory stores computer-executed instructions; the transceiver is used for sending and receiving data.

[0026] The processor executes computer execution instructions stored in the memory to implement the information push method described in any of the above aspects.

[0027] Fourthly, this application provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the information push method described in any of the above aspects.

[0028] Fifthly, this application provides a computer program product, including computer execution instructions, which, when executed by a processor, implement the information push method described in any of the above aspects.

[0029] This application provides an information push method, apparatus, device, storage medium, and product that acquires the intention feature data of users to be classified; inputs the intention feature data of users to be classified into a trained user intention classification model to classify the users to be classified, thereby determining the users' user intentions; identifies users with positive user intentions as target users, and obtains the viewing frequency of target users on each functional page; determines which functional pages with high viewing frequency will be pushed with the corresponding information; and displays the functional pages with high viewing frequency on the target users' user terminals. This enables personalized push notifications to users, pushing information only to positive users, improving the user experience for users with negative intentions; pushing information on functional pages avoids information redundancy on the homepage, improving the user experience for users with positive intentions; and pushing information on frequently viewed functional pages also makes it easier for users with positive intentions to receive the pushed information. Attached Figure Description

[0030] 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.

[0031] Figure 1 A flowchart illustrating the information push method provided in this application embodiment;

[0032] Figure 2 This is a schematic diagram of the network architecture provided in an embodiment of this application;

[0033] Figure 3 This is a schematic diagram of the structure of the information push device provided in the embodiments of this application;

[0034] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0035] 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

[0036] 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.

[0037] The terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. In the following descriptions of embodiments, "a plurality of" means two or more, unless otherwise explicitly defined.

[0038] First, the prior art involved in this invention will be described and analyzed in detail.

[0039] Currently, when pushing information to customers, information is typically displayed on the homepage of financial software to all customers. However, some users may not be interested in the pushed information or do not want to know it at the moment. Displaying pushed information on the homepage of financial software on each user's terminal leads to information redundancy on the homepage and affects the customer experience.

[0040] The inventors discovered in their research that users can be accurately categorized to determine their interest in receiving targeted information, thereby providing personalized services. Furthermore, financial software homepages typically contain entry points to functional pages such as deposits, transfers, loans, utility payments, wealth management, pensions, credit cards, task centers, and annual statements. After a user enters a functional page through a homepage entry, it can be determined that the user may be interested in the relevant information, allowing for precise information pushes. Therefore, this application provides an information push method that obtains the intention characteristic data of users to be categorized, categorizes users based on this data to determine their user intentions, identifies users with positive intentions as target users, obtains the target user's viewing frequency for each functional page, determines which functional pages have the highest viewing frequency to push the corresponding information, and displays the most frequently viewed functional pages on the target user's device.

[0041] 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, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with relevant laws, regulations and standards, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0042] It should be noted that the information push method, apparatus, device, and storage medium of this application can be used in the field of big data, or in any field other than big data. The application fields of the information push method, apparatus, device, storage medium, and product of this application are not limited.

[0043] 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.

[0044] Figure 1This is a flowchart illustrating an information push method provided in an embodiment of this application. This embodiment addresses the problem that displaying pushed information on the homepage of financial software on various user terminals leads to information redundancy on the homepage, affecting customer experience. The method in this embodiment is applied to an information push device, which can be located in an electronic device. The electronic device can be a digital computer of various forms, such as a laptop computer, desktop computer, workbench, personal digital assistant, server, blade server, mainframe computer, and other suitable computers.

[0045] like Figure 1 As shown, the specific steps of this method are as follows:

[0046] Step S101: Obtain the intention feature data of the users to be classified.

[0047] The intention feature data of the users to be classified is the feature data used to classify user intentions. For example, intention features may include user attribute characteristics, such as user age, user gender, and user region. It may also include user financial characteristics, such as whether they have investment experience and user income.

[0048] In this embodiment of the application, the entity executing the information push method can obtain the intention feature data of the users to be classified.

[0049] In this embodiment, user intent can be positive or negative. Users with positive intent pay attention to the information pushed to them, while users with negative intent do not pay attention to the information pushed to them. Optionally, user intent can also be uncertain.

[0050] Step S102: Input the intention feature data of the users to be classified into the trained user intention classification model to classify the users to be classified in order to determine the users' intentions.

[0051] In this embodiment of the application, the subject executing the information push method can input the intention feature data of the user to be classified into the trained user intention classification model. After the trained user intention classification model classifies the user to be classified, it outputs the user's intention.

[0052] It should be understood that the user intention classification model in the embodiments of this application can be any type of classification model, and the embodiments of this application are not limited thereto. For example, it can be a decision tree model, a support vector machine model, a Naive Bayes model, etc.

[0053] Step S103: Identify users with positive user intentions as target users and obtain the viewing frequency of each functional page by the target users.

[0054] In one optional implementation, the viewing frequency of each functional page by the target user can be the number of times the user clicks the function page open control within a preset time period. The preset time period can be a period of time relative to the current time, for example, a month prior to the current time; the number of clicks by the user on the function page open control within the past month can be determined as the viewing frequency of each functional page by the target user.

[0055] Step S104: Determine which function page to push the corresponding information to be pushed on the function page with the highest viewing frequency.

[0056] Among them, the function pages that are viewed most frequently are displayed on the target user's terminal.

[0057] Optionally, digital human technology can be used to push the corresponding information to be pushed to the function page on the function page that is viewed frequently.

[0058] Currently, financial software often integrates digital human modules as online intelligent financial assistants. These modules can provide professional explanations of financial knowledge, intelligent recommendations for wealth management products, and patient explanations of procedures, thus supporting banks' online digital transformation and driving the development of new products, services, and models through data. Therefore, digital human technology can be used to push relevant information to frequently viewed function pages, improving the efficiency with which users access this information.

[0059] Specifically, after determining that digital human technology will be used to push the information to be pushed to the function page corresponding to the function page that is viewed frequently, in response to the user's viewing operation of the function page that is viewed frequently on the user terminal, a digital human including the information to be pushed is generated on the function page. Through the interaction between the user and the digital human, the user can learn about the information to be pushed.

[0060] Figure 2 The network architecture diagram provided in this application embodiment is shown in the figure. Electronic device 2 can obtain the intention feature data of the user to be classified from database 3, and input the electronic device 2 into the trained user intention classification model to determine that the user intention is a positive target user. Based on the target user's viewing frequency of each functional page, the functional pages with higher viewing frequency are determined. In response to the target user's opening operation of the functional page with higher viewing frequency on user terminal 1, the push information corresponding to the functional page is sent to user terminal 1 so that user terminal 1 pushes the push information corresponding to the functional page on the functional page with higher viewing frequency.

[0061] The information push method provided in this application involves: acquiring the intention feature data of users to be classified; inputting the intention feature data of users to be classified into a trained user intention classification model to classify the users to be classified, thereby determining the users' user intentions; identifying users with positive user intentions as target users and acquiring the viewing frequency of target users for each functional page; determining which functional pages to push corresponding information to be pushed to; and displaying the functional pages with high viewing frequency on the target users' user terminals. This method enables personalized push notifications to users, pushing information only to positive users, thus improving the user experience for users with negative intentions; pushing information on functional pages avoids information redundancy on the homepage, improving the user experience for users with positive intentions; and pushing information on functional pages with high viewing frequency also makes it easier for users with positive intentions to know about the pushed information.

[0062] The information push method provided in this application also includes a step of training a user intention classification model. The step of training the user intention classification model includes:

[0063] Step S201: Obtain sample feature data for each sample user.

[0064] Step S202: Determine the intention tags of each sample user based on the behavioral data of each sample user.

[0065] Step S203: Use the sample feature data and intention labels of sample users to train the initial user intention classification model to obtain the trained user intention classification model.

[0066] In this embodiment, sample feature data corresponds to intention feature data. Specifically, the sample user's sample feature data and the sample feature data are used to train the user classification model.

[0067] Specifically, sample users can be identified from existing users, and information can be pushed to these sample users. Feedback on the pushed information can be collected, and an initial user intention classification model can be trained based on the sample users' behavioral feedback and their characteristic data. The trained user intention classification model can then be obtained. Optionally, middle- and high-income users can be identified as sample users.

[0068] In this embodiment of the application, before obtaining the sample feature data of each sample user, feature selection can be performed to determine the features related to user intent from the available user features. Specifically, data visualization technology can be used to determine the correlation between features and user intent.

[0069] The information push method provided in this application involves acquiring sample feature data of each sample user; determining the intention tags of each sample user based on their behavioral data; and training an initial user intention classification model using the sample feature data and intention tags of the sample users to obtain a trained user intention classification model. This method enables automatic labeling of sample users' intention tags without manual judgment, improving the efficiency of tag labeling and thus enhancing the training efficiency of the user intention classification model.

[0070] In one optional implementation, the behavioral data of the sample users may include: the frequency and duration of viewing product information, as well as purchase behavior, or collection behavior, or recommendation behavior.

[0071] Specifically, if it is determined that a user views product information more frequently than a preset threshold, or views product information for a duration exceeding a preset threshold, or engages in purchasing, saving, or recommending the product, then the sample user's intention label is determined to be positive. If it is determined that a user views product information less frequently than or equal to a preset threshold, views product information for a duration less than or equal to a preset threshold, and does not engage in purchasing, saving, or recommending the product, then the sample user's intention label is determined to be negative.

[0072] This application does not limit the specific values ​​of the preset duration threshold and the preset number of times threshold, and they can be adjusted based on actual needs. Setting the preset duration threshold and the preset number of times threshold higher is beneficial to improving the success rate of precision marketing, but the possibility of potential user churn will also increase accordingly.

[0073] In this embodiment, the user's intention tags are determined based on the frequency and duration of the user's viewing of product information, as well as whether the user has made a purchase, collection, or recommendation of the product, thereby improving the accuracy of the determined user tags.

[0074] The information push method provided in the above embodiments is illustrated below with a specific example. In this embodiment, the user intention classification model is a Naive Bayes classifier, and the information push method may include the following steps:

[0075] Step S301: Obtain the age data, gender data, financial management experience data, inflow data, and account balance data of the users to be categorized.

[0076] Step S302: Calculate the income level data of the users to be classified using the income calculation formula, inflow data, and account balance data.

[0077] Step S303: Determine the age data, gender data, financial management experience data, and income level data as the intentional characteristic data of the users to be classified.

[0078] In this embodiment of the application, the user's age data, gender data, financial experience data, and income level data can be determined as the intentional characteristic data of the user to be classified.

[0079] In this embodiment of the application, the income level can be calculated based on the user's inflow data and account balance data. The income calculation formula can be expressed as I=kΣλ+Φ; where λ represents the annual inflow data, k is the set savings coefficient, and Φ is the account balance data at the end of the month.

[0080] Financial management experience data can be determined based on a user's historical financial management data. This data can indicate whether or not a user has any financial management experience. Gender data and financial management experience data can be represented as Boolean values.

[0081] Age data can be obtained from user's birthday information or ID card information, and is discrete data.

[0082] In this embodiment, since age data, gender data, financial experience data, income level data are highly correlated with user intentions and are relatively easy to obtain, user age data, gender data, financial experience data, and income level data can be identified as the intention feature data of the user to be classified, which can improve the accuracy of the trained user intention classification model in classifying users.

[0083] Step S304: Discretize the continuous intention feature data in the intention feature data to transform the continuous intention feature data into discrete intention feature data.

[0084] It should be understood that the Naive Bayes classifier is a classification method based on Bayes' theorem and the assumption of conditional independence of features.

[0085] In this application, since the Naive Bayes classifier is based on the perfect Bayes theorem, it has strong model representation and learning capabilities, and exhibits high efficiency and high accuracy. It is especially suitable for learning small-scale datasets. Therefore, the user intention classification model can be the Naive Bayes classifier.

[0086] Bayes' theorem is as follows:

[0087] P(X,Y)=P(Y|X)×P(X)=P(X|Y)×P(Y)

[0088]

[0089] The following is the reasoning process from Bayes' theorem to the Naive Bayes classifier:

[0090] Given a training dataset (X,Y), where each sample X includes N-dimensional features, i.e. (X1,X2,...Xn), and a class label set with K classes, i.e. (Y1,Y2,...Yk).

[0091] Naive Bayes classification is as follows: based on a new sample X, calculate P(Y1|X), P(Y2|X)...P(Yk|X), and the class to which sample X belongs is the one with the highest probability.

[0092] According to Bayes' theorem, we can obtain:

[0093]

[0094] P(Yk) can be directly calculated from the sample, and P(X|Yj) = P(X1,X2,...Xn|Yj) can be determined.

[0095] Based on the assumption of conditional independence of features, we can deduce that:

[0096]

[0097] P(Xi|Yj) can also be obtained from the sample data, and P(Yk|X) can be solved as follows:

[0098]

[0099] It should be understood that, since the Naive Bayes classifier requires feature values ​​to be discrete or approximated by a probability distribution function, it is necessary to discretize the continuous intention feature data in the intention feature data to transform it into discrete intention feature data.

[0100] In this embodiment of the application, age data and income level data can be continuous intention feature data. Before classifying users to be classified using age data, gender data, financial experience data, and income level data, the age data and income level data need to be discretized.

[0101] Alternatively, a quantile discretization algorithm can be used to discretize continuous intention feature data.

[0102] For example, the method for discretizing age data can be as follows: the number of segments can be predetermined, the age threshold can be segmented based on the number of segments and the age data of active sample users, and the discretized age data can be determined based on the segmentation of the age threshold.

[0103] In this embodiment, the quantile discretization algorithm has a low grouping degree, which improves the speed of discretization processing of continuous intention feature data; and the discretization effect of the quantile discretization algorithm is good, which can ensure the quality of discretization processing and thus improve the accuracy of the trained user intention classification model.

[0104] Step S305: Use the trained user intention classification model to classify the users to be classified based on their discrete intention feature data.

[0105] Specifically, gender data, financial experience data, and discretized age and income level data can be substituted into the trained user intention classification model to obtain the user intentions output by the trained user intention classification model.

[0106] Step S306: Identify users with positive user intentions as target users and obtain the viewing frequency of each functional page by the target users.

[0107] Step S307: Sort the functional pages in descending order of viewing frequency.

[0108] Step S308: Select the function pages that are ranked before the preset frequency as the function pages with higher viewing frequency.

[0109] Step S309: If it is determined that there is a function page with a high viewing frequency that needs to be pushed, then it is determined to push the function page with the high viewing frequency that needs to be pushed.

[0110] The preset frequency ranking is pre-defined. For example, the preset frequency ranking can be 21, then the function pages with the highest viewing frequency in the top 20 are determined as the function pages with higher viewing frequency.

[0111] In this embodiment, the information to be pushed is preset. It should be understood that not all functional pages have corresponding information to be pushed preset. Therefore, information is pushed only to functional pages where information to be pushed has been preset and the viewing frequency is also specified.

[0112] For example, four categories of push notifications can be pre-set: precious metals, wealth management, credit cards, and seasonal consumption. If it is determined that the functional page corresponding to a certain push notification among the four categories ranks higher than the preset frequency, then the push notification for that specific functional page will be pushed. For instance, if the functional page corresponding to precious metals ranks higher than the preset frequency, then the push notification for precious metals will be pushed on that functional page.

[0113] The information push method provided in this application sorts functional pages in descending order of viewing frequency; functional pages ranked before a preset frequency ranking are identified as frequently viewed functional pages; if it is determined that there is information to be pushed to a frequently viewed functional page, then the information to be pushed to that functional page is pushed to that functional page. By sorting the functional pages to determine the frequently viewed functional pages, it is possible to push the information to be pushed to those frequently viewed functional pages. Furthermore, both the information to be pushed and the preset frequency ranking can be pre-set, improving the flexibility of the information push method.

[0114] Figure 3 This is a schematic diagram of the structure of an information push device provided in an embodiment of this application. The information push device provided in this embodiment can execute the processing flow provided in the information push method embodiment. Figure 3 As shown, the information push device 50 includes: an acquisition module 501, a classification module 502, a user determination module 503, and a page determination module 504.

[0115] Specifically, module 501 is used to acquire the intention feature data of users to be classified.

[0116] The classification module 502 is used to input the intention feature data of the user to be classified into the trained user intention classification model to classify the user to be classified in order to determine the user's intention.

[0117] The user identification module 503 is used to identify users with positive user intentions as target users and obtain the viewing frequency of target users on each functional page.

[0118] The page determination module 504 is used to determine which function pages are most frequently viewed and push the corresponding information to be pushed to them; the function pages most frequently viewed are displayed on the target user's user terminal.

[0119] The apparatus provided in this application embodiment can be specifically used to execute the method embodiment provided in Embodiment 1 above, and the specific functions will not be repeated here.

[0120] Optionally, the information push device 50 further includes: a training module; the training module is used to: acquire sample feature data of each sample user; determine the intention label of each sample user based on the behavioral data of each sample user; and train an initial user intention classification model using the sample feature data and intention labels of the sample users to obtain a trained user intention classification model.

[0121] Optionally, the information to be pushed is product information. The training module is specifically used to: determine that if the frequency of a user viewing product information is greater than a preset number of times threshold, or the duration of a user viewing product information is greater than a preset duration threshold, or the user has a purchase behavior, collection behavior, or recommendation behavior for the product, then determine that the intention tag of the sample user is positive.

[0122] Optionally, the user intention classification model is a Naive Bayes classifier. The acquisition module 501 is further used to: discretize the continuous intention feature data in the intention feature data to transform the continuous intention feature data into discrete intention feature data; the classification module 502 is specifically used to: classify the user to be classified based on the discrete intention feature data of the user to be classified using the trained user intention classification model.

[0123] Optionally, the acquisition module 501 is also specifically used to: discretize the continuous intention feature data using a quantile discretization algorithm.

[0124] Optionally, the acquisition module 501 is specifically used to: calculate the income level data of the user to be classified using the income calculation formula and the inflow and outflow data and account balance data; and determine the age data, gender data, financial management experience data and income level data as the intentional characteristic data of the user to be classified.

[0125] Optionally, the page determination module 504 is specifically used to: sort the functional pages in descending order of viewing frequency; determine the functional pages that are ranked before the preset frequency as functional pages with higher viewing frequency; if it is determined that there is information to be pushed to a functional page with higher viewing frequency, then determine to push the information to be pushed to that functional page on that functional page with higher viewing frequency.

[0126] Optionally, the page determination module 504 is specifically used to: determine whether to use digital human technology to push the information to be pushed to the function page corresponding to the function page that is viewed frequently.

[0127] The apparatus provided in this application embodiment can be specifically used to execute the above method embodiments, and its specific functions will not be described in detail here.

[0128] Figure 4 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application, such as... Figure 4 As shown, this application also provides an electronic device 60, including: a processor 601, and a memory 602 and a transceiver 603 communicatively connected to the processor 601. The memory 602 stores computer-executable instructions; the transceiver 603 is used for sending and receiving data; the processor 601 executes the computer-executable instructions stored in the memory 602 to implement the method provided in any embodiment of this application.

[0129] Specifically, the program may include program code, which includes computer-executable instructions. Memory 602 may include high-speed RAM, and may also include non-volatile memory, such as at least one disk storage device. The computer-executable instructions are stored in memory 602 and configured to be executed by processor 601 to implement the method provided in any embodiment of this application. Related descriptions can be understood by referring to the relevant descriptions and effects corresponding to the steps in the accompanying drawings, and will not be elaborated upon here.

[0130] In this embodiment, the memory 602 and the processor 601 are connected via a bus. 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. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 4 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0131] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the method provided in any embodiment of this application.

[0132] This application also provides a computer program product, including computer execution instructions, which, when executed by a processor, implement the method provided in any embodiment of this application.

[0133] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules 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 through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0134] The modules described as separate components may or may not be physically separate. Similarly, the components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0135] Furthermore, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated module can be implemented in hardware or in a combination of hardware and software functional modules.

[0136] The program code used to implement the methods of this application may be written in any combination of one or more programming languages. This program code may be provided to the processor or controller of a general-purpose computer, special-purpose computer, or other programmable full-path fusion device, such that when executed by the processor or controller, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0137] In the context of this application, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0138] Furthermore, although the operations are described in a specific order, this should be understood as requiring that such operations be performed in the specific order shown or in sequential order, or requiring that all illustrated operations be performed to achieve the desired result. In certain environments, multitasking and parallel processing may be advantageous. Similarly, although several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of this application. Certain features described in the context of individual embodiments may also be implemented in combination in a single implementation. Conversely, various features described in the context of a single implementation may also be implemented individually or in any suitable sub-combination in multiple implementations.

[0139] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.

[0140] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. An information push method, characterized in that, include: Obtain the intention feature data of users to be classified; The intention feature data of the users to be classified is input into the trained user intention classification model to classify the users to determine their user intentions; the user intention is positive or negative, a positive user intention indicates that the user pays attention to the information pushed to them, and a negative user intention indicates that the user does not pay attention to the information pushed to them. Users with positive user intent are identified as target users. Within a preset time period, the number of times the target users click on the open controls of each function page in the financial software is counted, and the number of clicks is taken as the viewing frequency of the target user on the corresponding function page. Sort the functional pages in descending order of viewing frequency; The function pages that are ranked higher than the preset frequency are identified as the function pages that are viewed more frequently. If it is determined that there is a function page with a high viewing frequency that corresponds to information to be pushed, then it is determined to push the information corresponding to that function page to that function page; the information to be pushed is associated with the function of the function page; the function page with a high viewing frequency is displayed on the target user's user terminal.

2. The method according to claim 1, characterized in that, The information push method further includes a step of training a user intention classification model; the user intention classification model is a Naive Bayes classifier, and the step of training the user intention classification model includes: Obtain sample feature data for each sample user; Determine the intention tags of each sample user based on their behavioral data. The initial user intention classification model is trained using the sample feature data and intention labels of the sample users to obtain the trained user intention classification model.

3. The method according to claim 2, characterized in that, The information to be pushed is product information, and the determination of the sample users' intention tags based on the sample users' behavioral data includes: If it is determined that a user views product information more frequently than a preset threshold, or views product information for a longer period than a preset threshold, or the user engages in purchasing, saving, or recommending products, then the sample user's intention tag is determined to be positive.

4. The method according to claim 1, characterized in that, The user intent classification model is a Naive Bayes classifier. After obtaining the intent feature data of the users to be classified, the method further includes: The continuous intention feature data in the intention feature data is discretized to transform the continuous intention feature data into discrete intention feature data; The step of inputting the intention feature data of the users to be classified into the trained user intention classification model to classify the users includes: The trained user intention classification model is used to classify users based on their discrete intention feature data.

5. The method according to claim 4, characterized in that, The discretization process for the continuous intention feature data in the intention feature data includes: The quantile discretization algorithm is used to discretize continuous intention feature data.

6. The method according to any one of claims 1-5, characterized in that, The acquisition of the intention feature data of the users to be classified includes: The income level data of users to be classified is calculated using income calculation formulas, inflow data, and account balance data. Age, gender, financial experience, and income level data were identified as the intended characteristics of users to be categorized.

7. The method according to any one of claims 1-5, characterized in that, The step of determining to push the corresponding information to be pushed to the function page on the function page with high viewing frequency includes: We have determined to use digital human technology to push the corresponding information to be pushed to the function pages that are viewed frequently.

8. An information push device, characterized in that, include: The acquisition module is used to acquire the intention feature data of users to be classified; The classification module is used to input the intention feature data of the user to be classified into the trained user intention classification model, classify the user to be classified, and determine the user's intention; the user intention is positive or negative, a positive user intention indicates that the user pays attention to the information pushed to him, and a negative user intention indicates that the user does not pay attention to the information pushed to him. The user identification module is used to identify users with positive user intentions as target users, count the number of times the target users click on the open controls of each function page in the financial software within a preset time period, and use the number of clicks as the viewing frequency of the target user on the corresponding function page. The page determination module is used to sort the functional pages in descending order of viewing frequency; determine the functional pages that are ranked higher than the preset frequency as functional pages with higher viewing frequency; if it is determined that there is information to be pushed to a functional page with higher viewing frequency, then determine to push the information to be pushed to that functional page with higher viewing frequency; the information to be pushed is associated with the function of the functional page; the functional pages with higher viewing frequency are displayed on the user terminal of the target user.

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

10. 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.

11. A computer program product comprising computer-executable instructions, characterized in that, When the computer execution instructions are executed by the processor, they implement the method as described in any one of claims 1-6.

Images