Data processing method, device and storage medium for display information

Abstract

The present disclosure relates to a data processing method, device and storage medium for display information, the method includes collecting user behavior data based on the display information, obtaining the user's response result to the display information according to the user behavior data; determining a preset display intervention factor and a user adjustment factor, and constructing an analysis model according to the display intervention factor and the user adjustment factor, the analysis model is used to determine the degree of influence of the display intervention factor on the response result based on the user adjustment factor; in the analysis model, under the control of the value of the user adjustment factor, the display information is input to obtain the effect value of the display information on the response result through the display intervention factor; according to the effect value, the display information is marked with a corresponding influence label; the device and storage medium are used to implement the above method. The present disclosure helps to improve the accuracy of the display information by analyzing and modeling both the display information and the user.

Description

Technical Field

[0001] The present application relates to the field of data analysis, and in particular to a data processing method, device and storage medium for displaying information. Background Art

[0002] With the development of intelligent marketing tools and big data analysis, display information analysis of design marketing has been widely used in retail customer marketing activities of banks. The usual application method is based on user behavior research, using big data insight technology to analyze and find customer preferences and patterns, and conduct targeted precision marketing. The mainstream methods of quantitatively evaluating marketing effects and judging the effectiveness of strategies are mostly based on user behavior data such as click-through rate and page dwell time, using randomized controlled experiments or quasi-experimental research designs to conduct comparative evaluation of effects, and using differential methods, regression methods, etc. to evaluate effects.

[0003] Among them, the correlation between features obtained by collecting user behavior data based on randomized experimental design and applying statistical methods or machine learning algorithms is a common method in marketing activity evaluation. However, the current processing methods have the following problems:

[0004] (1) Marketing is a process of continuous interaction with users, and it applies a lot of marketing psychology and consumer behavior theories and applications. Therefore, in planning and execution, more experience factors based on marketing psychology research are often incorporated. There is no in-depth research on whether each factor has an impact on the outcome variable and the extent of the impact, and there is a lack of data support for strategy optimization.

[0005] (2) Commonly used statistical analysis or algorithms reflect the correlation between features. Correlation cannot replace causality. In some marketing scenarios, it is still necessary to understand more complex causal relationships in order to complete the closed loop from insight to optimization.

[0006] (3) Randomized controlled trials have significant limitations. They take time to design and deploy, and take up sufficient traffic, resulting in high experimental costs. In addition, it is difficult to use very strict screening criteria when selecting research subjects in randomized controlled trials, and their applicability is limited. Some experimental designs may damage the user experience, such as recommending content that does not match the user's interests. When there are too many options for testing, it is difficult to try all of them. Under non-strict conditions, the experimental and control groups often show non-significant differences. Summary of the invention

[0007] Based on this, it is necessary to provide a data processing method, device and storage medium for display information in response to the above technical problems. The technical solution disclosed in the present invention includes:

[0008] A data processing method for displaying information comprises the following steps:

[0009] Collecting user behavior data based on the display information, and obtaining the user's response result to the display information according to the user behavior data;

[0010] Determining a preset display intervention factor and a user adjustment factor, and constructing an analysis model according to the display intervention factor and the user adjustment factor, wherein the analysis model is used to determine the degree of influence of the display intervention factor on the response result based on the user adjustment factor;

[0011] In the analysis model, under the control of the value of the user adjustment factor, the display information is input to obtain the effect value of the display information on the response result through the display intervention factor;

[0012] According to the effect value, a corresponding influence label is marked for the display information.

[0013] In one embodiment, the determining of a preset display intervention factor and a user adjustment factor, constructing an analysis model according to the display intervention factor and the user adjustment factor, wherein the analysis model is used to determine the degree of influence of the display intervention factor on the response result based on the user adjustment factor, comprises:

[0014] Extracting the display intervention factor based on the display information, wherein the display information acts on the response result through the display intervention factor;

[0015] Determining the user adjustment factor, wherein the user adjustment factor acts on the response result by affecting the display intervention factor;

[0016] The analysis model is constructed with the display intervention factor as the mediating variable, the user adjustment factor as the moderating variable, the display information as the independent variable, and the response result as the dependent variable. The analysis model is a moderating mediation effect model.

[0017] In one embodiment, in the analysis model, under the control of the value of the user adjustment factor, inputting the display information, and obtaining the effect value of the display information on the response result through the display intervention factor includes:

[0018] Based on the analysis model, setting the value of the user adjustment factor, inputting the display information, and calculating the individual effect value of each display intervention factor of the display information on the response result;

[0019] The sum of the individual effect values ​​is calculated to obtain the effect value of the display information acting on the response result through the display intervention factor.

[0020] In one embodiment, the determining of the user adjustment factor, wherein the user adjustment factor acts on the response result by influencing the display intervention factor, comprises:

[0021] Acquire historical data of the display information, and filter user samples according to a set filtering rule, wherein the filtering rule includes that push information of the display information is successfully sent to the user within a set period;

[0022] Acquire a user identity in the user sample and some user information corresponding to the user identity according to the user sample; the user information includes one or more of asset level, age, active frequency, and gender;

[0023] Based on the user information, the user adjustment factor is calculated and selected through a logistic regression formula.

[0024] In one embodiment, the method of calculating and selecting the user adjustment factor based on the user information by means of a logistic regression formula includes:

[0025] Extracting feedback data on whether the user responds to the push information according to the user identity in the user sample;

[0026] Performing univariate quality analysis on the user information;

[0027] Discretize the continuous variables in the user information and divide them into several levels according to preset layers;

[0028] Generate a sample data set based on the user information and feedback data, split the sample data set into a training set and a test set, and build a logistic regression model;

[0029] Based on the logistic regression model, the user information is input, and the user adjustment factor is screened according to the output of the logistic regression model.

[0030] In one embodiment, the step of constructing the analysis model comprises:

[0031] The stepwise test method is used to verify whether the mediation effect of the analysis model is significant;

[0032] Using latent moderator rule structural equation method to verify whether the fit of the analysis model meets the preset standard;

[0033] When the verification result of the analysis model does not satisfy the condition that the mediation effect is significant or the degree of fit does not meet the preset standard, the analysis model is reconstructed and verified.

[0034] In one embodiment, the method further comprises:

[0035] According to a set first threshold, determining whether the effect value of the display information is greater than or equal to the first threshold, and when the determination result is that the effect value of the display information is greater than or equal to the first threshold, marking the influence label of the display information as the user adjustment factor corresponding to the effect value;

[0036] The push information of the display information is sent to the user in a targeted manner according to the user adjustment factor in the influence tag.

[0037] In one embodiment, the method further comprises:

[0038] According to the set second threshold, determine whether the individual effect value is greater than or equal to the second threshold. When the judgment result is that the individual effect value is greater than or equal to the second threshold, store the individual effect value and the display intervention factor corresponding to the individual effect value in the database.

[0039] A computer device includes a memory and a processor, wherein the memory stores a computer program, and is characterized in that when the processor executes the computer program, the steps of the above-mentioned data processing method for displaying information are implemented.

[0040] A computer-readable storage medium stores a computer program thereon, characterized in that when the computer program is executed by a processor, the steps of the above-mentioned data processing method for displaying information are implemented.

[0041] The beneficial effects of the above-mentioned data processing method, device and storage medium for display information are as follows:

[0042] By analyzing and modeling both the display information and the user, we can obtain the influence and degree of the display intervention factor on the response results under the influence of the user adjustment factor. The adjustment of the user adjustment factor helps to design the targeted deployment of the display information and avoid the error expansion caused by the simple mediation analysis. At the same time, the display intervention factor is obtained by splitting and extracting the display information, which can perform a causal analysis of the display information on the response results in smaller units, which helps to improve the accuracy of the display information analysis and is more suitable for the environment under actual commercial operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the conventional technology, the drawings required for use in the embodiments or the conventional technology descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0044] Figure 1 A schematic diagram of a flow chart of a data processing method for display information provided in an embodiment;

[0045] Figure 2 A schematic diagram of the structure of an analysis model is provided in one embodiment. DETAILED DESCRIPTION

[0046] In order to facilitate understanding of the present application, the present application will be described more fully below with reference to the relevant drawings. Embodiments of the present application are provided in the drawings. However, the present application can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present application more thorough and comprehensive.

[0047] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein in the specification of this application are only for the purpose of describing specific embodiments and are not intended to limit this application.

[0048] The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims. The terms "comprises," "comprising," or any other variant thereof are intended to cover non-exclusive inclusions, so that a process, method, product, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such a process, method, product, or device. In the absence of further restrictions, it is not excluded that there are other identical or equivalent elements in the process, method, product, or device comprising the elements.

[0049] The data processing method for display information provided by the present disclosure can be applied in the application environment of the marketing optimization strategy perceived by financial retail customers. By analyzing the display information, the causal relationship between the elements in the display information that are helpful to marketing can be quantitatively evaluated, thereby quantifying the impact of marketing elements, making targeted adjustments and optimizations, improving the evaluation framework of financial retail customer marketing, and continuously optimizing the influencing factors to achieve improved marketing results. Display information can be display information, pictures, short links, videos, advertising slogans, etc. Taking display information as an example, display information has been used as a carrier of a network mobile page to display marketing materials that interact with users. The present disclosure can be applied in financial marketing optimization strategies, but it does not exclude that it can also be applied in other occasions, such as the analysis and optimization of e-commerce app display information, or the placement of advertisements in websites or promotional links linked to other domain names. The display information analysis method can be used to evaluate the effect of advertising by counting the number of clicks.

[0050] The implementation scheme of this specification is described below using a specific implementation scenario of page analysis of a data processing method for displaying information in the field of financial sales.

[0051] See also Figure 1 This embodiment provides a data processing method for display information, including the following steps:

[0052] Step S10: Collecting user behavior data based on the display information, and obtaining the user's response result to the display information according to the user behavior data.

[0053] In this embodiment, the displayed information can be understood as an interactive mobile page for marketing to financial retail customers. It applies a large number of perceptual marketing visual designs and elements of advanced user consumption psychology. By displaying product content, it triggers users' recognition and interest in the product, increases clicks and improves conversions.

[0054] User behavior data usually refers to the data of users browsing displayed information, such as user personal information, clicks, favorites, likes, orders, etc.

[0055] The response results include no response and executing a response action. Executing a response action can be understood as the user responding to the product displayed on the page when browsing the displayed information, such as clicking to place an order. By obtaining user behavior data, it is possible to directly or indirectly obtain whether the user clicks in response to the displayed information.

[0056] Step S20: Determine the preset display intervention factor and user adjustment factor, and construct an analysis model based on the display intervention factor and the user adjustment factor. The analysis model is used to determine the influence of the display intervention factor on the response result based on the user adjustment factor.

[0057] Display intervention factors can be understood as elements on the display page that have a positive impact on promoting marketing success, such as holiday elements, image and text ratio, page theme, etc. User adjustment factors can be understood as elements that have a positive impact on promoting marketing success based on the user's own characteristics, such as user age, gender, etc. Combining display intervention factors and user adjustment factors can establish an analysis model from two aspects: display information and users. The analysis model can be used to determine the degree of influence of display intervention factors on response results based on user adjustment factors.

[0058] Step S30: In the analysis model, under the control of the value of the user adjustment factor, the display information is input to obtain the effect value of the display information on the response result through the display intervention factor.

[0059] Effect value usually refers to the numerical value of the intensity of a quantitative phenomenon in statistics. The actual statistical quantity of the effect value may include the correlation between two variables, the regression coefficient in a regression model, the difference in the mean values ​​between different treatments, etc. The larger the absolute value of the effect value, the stronger the effect, that is, the more obvious the impact phenomenon. In this embodiment, the relationship between the displayed information and the response result is measured by the effect value.

[0060] Combined with the constructed analysis model, the effect value of the display information on the response results through the display intervention factor under the control of the user adjustment factor can be calculated. For example, it is possible to obtain whether the display information has a positive or negative impact on the user's response results for users of a certain age group, as well as the quantitative value of the degree of impact.

[0061] The data processing method for display information provided in this embodiment analyzes and models the display information user and both sides of the user to obtain the influence and degree of influence of the display information on the response result through the display intervention factor under the influence of the user adjustment factor. It not only performs causal relationship analysis on the regulated mediating variables, but also avoids the error expansion caused by simple mediating analysis, which helps to improve the accuracy of page analysis and is more suitable for the environment under actual commercial operations.

[0062] In one embodiment, the above step S20 includes:

[0063] Step S22: extracting a display intervention factor based on the display information, and the display information acts on the response result through the display intervention factor.

[0064] In page analysis, the page is often used as a whole to measure the impact on the user response results. This embodiment extracts the display intervention factor by splitting the display information, and when measuring the impact of the display information on the response results, the impact of the display intervention factor of the display information on the response results is obtained.

[0065] Step S24: Determine a user adjustment factor, which acts on the response result by influencing the display intervention factor.

[0066] There are many influencing factors on the user side, such as user age, gender, asset level, active frequency, etc. When building the model, select user factors that have a more obvious impact on the response results.

[0067] Step S26: Construct an analysis model with the display intervention factor as the mediating variable, the user adjustment factor as the moderating variable, the display information as the independent variable, and the response result as the dependent variable. The analysis model is a moderated mediation effect model.

[0068] A mediating variable usually refers to a variable between an independent variable and a dependent variable. It can partially explain the indirect effect of the independent variable on the dependent variable, that is, the effect of the independent variable on the dependent variable can be divided into two parts: direct effect and indirect effect. A moderating variable usually refers to the effect of affecting other variables in the model. In other words, the effect of a specific variable in the model will depend on the value of the moderating variable. The moderating mediation effect usually refers to the influence of the value of the moderating variable on the magnitude of the indirect effect that the mediating variable can explain, and explains the effect of the indirect effect under the condition that the moderating variable takes a specific value.

[0069] In the application scenario of this embodiment, the user adjustment factor is used as an adjustment variable. For example, the influence of the user age on the response result is measured by influencing the display intervention factor, that is, the effect of the display intervention factor depends on the value of the user adjustment factor.

[0070] Combination Figure 2 In this analysis model, the independent variable is denoted as X, the dependent variable is denoted as Y, the moderating variable is denoted as Z, the interaction between the independent variable X and the moderating variable Z is denoted as XZ, and the mediating variable is denoted as M.

[0071] The regression equation of this analysis model is as follows:

[0072] M i =α i0 +α i1 X+α i2 Z+a i3 ZX+ε i ,i=1,2.....n

[0073]

[0074] The regression equation can characterize the relationship between the display intervention factor and the display information, as well as the relationship between the response result and the display intervention factor, the display information and the user adjustment factor. Among them, the adjustment variable affects the relationship between the independent variable and the mediating variable. XZ represents the correlation between X and Z, which is the product term of the independent variable and the adjustment variable. The coefficient α i It refers to the regression coefficient between the independent variable, the moderating variable and the mediating variable. The coefficient β i It refers to the regression coefficient of the mediating variable on the dependent variable. Continuous variables are obtained through regression analysis, and categorical variables are obtained through variance analysis. We will not elaborate on this here. The label i represents the i-th display intervention factor in the display information, and n represents that there are n display intervention factors determined in this model. C 1 is the regression coefficient between X and Y, C 2 is the regression coefficient between M and Y, C 3 is the regression coefficient between XZ and Y; ε is the residual error.

[0075] In the above embodiment, the above step S30 includes:

[0076] Step S32: Based on the analysis model, the value of the user adjustment factor is set, the display information is input, and the individual effect value of each display intervention factor of the display information on the response result is calculated.

[0077] Step S34: Calculate the sum of individual effect values ​​to obtain the effect value of the intervention factor on the response result by displaying the information.

[0078] The display information includes several display intervention factors, and the collection of the effects of each display intervention factor on the response result constitutes the display information through the display intervention factor on the response result. Combined with the regression equation of the constructed analysis model, C 3 is the total regulatory effect of X on Y, C' 3 is the direct regulatory effect of X on Y, M i The individual moderating effects of , the effect relationships are as follows:

[0079]

[0080] According to the above equation, the effect values ​​of all display intervention factors in the display information on the response results under the influence of the user adjustment factor can be calculated, that is, the effect values ​​of the display information on the response results through the display intervention factors.

[0081] In one embodiment, the above step S24 includes:

[0082] Step S242: Acquire historical data of display information, and filter user samples according to set filtering rules, where the filtering rules include that push information of display information is successfully sent to the user within a set period.

[0083] When selecting the user adjustment factor, this embodiment uses historical data before being affected by the display intervention factor. In marketing strategies, page links are often pushed through push information such as text messages, and push information that is not sent due to blocked text messages or system reasons is removed to obtain user samples for which push information has been successfully sent in the past three months.

[0084] Step S244: obtaining the user identity of the user sample and several user factors corresponding to the user identity according to the user sample, wherein the user information includes one or more of asset level, age, active frequency, and gender.

[0085] Step S246: Based on the user information, the user adjustment factor is calculated and screened by a logistic regression formula.

[0086] According to the logistic regression formula, the user adjustment factor is determined from a number of user factors, and generally 3 or more user adjustment factors are selected.

[0087] In the above embodiment, the above step S246 includes:

[0088] Step S2461: Extract feedback data on whether the user responds to the push information based on the user identity in the user sample.

[0089] Feedback data can be understood as whether the user clicks to browse the displayed information after receiving the push information.

[0090] Step S2463: Perform univariate quality analysis on user factors.

[0091] The user factors in the above step S2461 are processed by univariate analysis. The univariate analysis mainly focuses on the description and statistical inference of the univariate. It aims to reflect the basic information contained in a large amount of sample data in the simplest summary form and describe the concentration or discrete trend in the sample data. The univariate statistical inference is to infer the overall situation from the sample data, which mainly includes interval estimation and statistical hypothesis testing.

[0092] When analyzing data quality, missing values ​​are processed using the mean and median, and outliers are removed. Null values ​​can also be manually marked.

[0093] Step S2465: discretize the continuous variables in the user factors and divide them into several levels according to preset levels.

[0094] Transform user factors into numerical values. For example, if age is a continuous variable, stratify the age into age groups, and use equal frequency division to divide the active frequency into three categories: low, medium, and high.

[0095] Step S2467: Generate a sample data set based on user information and feedback data, split the sample data set into a training set and a test set, and build a logistic regression model. Use a random sampling method to divide the training set and the test set into a 4:1 ratio to train the logistic regression model.

[0096] Step S2469: Based on the logistic regression model, user information is input, and user adjustment factors are screened according to the output of the logistic regression model.

[0097] Before using the logistic regression model to screen user adjustment factors, the user factors can be preliminarily screened by calculating the Weight of Evidence (WOE) and Information Value (IV). The WOE value represents the proportion of users in the user sample who respond to push information according to the currently calculated user factors. For example, the proportion of users who respond to push information according to age groups can be specifically obtained. For a group of user factors, the greater the difference between the proportion of this group that responds to push information and the proportion of the user sample that responds and does not respond is and the proportion of the user sample that responds and does not respond as a whole, the greater the IV value. Otherwise, the smaller the IV value, the IV value can be used as an indicator to measure the degree of influence of a certain user factor on the response to push information. Combining the WOE value and the IV value, a preliminary screening can be performed from many user factors, and then the screening can be performed through the logistic regression model.

[0098] The formula for the logistic regression model is as follows:

[0099] y=σ(f(x))=σ(ω T x) = 1 / (1 + e -ωTx )

[0100] Among them, ω T is a set of weights trained in step S2467. Through the formula of the logistic regression model, the probability that a certain user factor causes the user to respond to the push information can be calculated.

[0101] In one embodiment, the above step S20 further includes:

[0102] The stepwise test method is used to verify whether the mediation effect of the analysis model is significant, and the latent moderator rule structural equation method is used to verify whether the fit of the analysis model meets the preset standards.

[0103] When the verification results of the analysis model do not meet the conditions of significant mediation effect or the goodness of fit does not meet the preset standards, the analysis model is rebuilt and verified.

[0104] In this embodiment, the stepwise test method and the latent moderator rule structural equation method are used for dual analysis and verification, with the stepwise test method as a supplement and the latent moderator rule structural equation method as a main method, which can effectively control the measurement error.

[0105] In one embodiment, the method further comprises:

[0106] Step S40: according to the set first threshold, determine whether the effect value of the display information is greater than or equal to the first threshold; when the determination result is that the effect value of the display information is greater than or equal to the first threshold, mark the influence label of the display information as the user adjustment factor corresponding to the effect value.

[0107] For the effect value of the display information output by the analysis model, the first threshold is set to distinguish the degree of influence of the display information on the response result through the display intervention factor. When the judgment result is that the effect value of the display information is greater than or equal to the first threshold, it can be understood that for the current user adjustment factor, the display information has a high degree of influence on the response result through the display intervention factor, and the current user adjustment factor can be used as an example of the influence label of the display information. Taking the display page as an example, for users aged 20-30, a certain display page has a high degree of influence on the response result through the display intervention factor, and the influence label of the display page can be recorded as the age of 20-30.

[0108] Step S50: Send push information of display information to the user in a targeted manner according to the user adjustment factor in the influence tag.

[0109] According to the influence tag of the displayed information, for example, the displayed information with the influence tag of the age between 20 and 30 is targeted and pushed to the users in the age between 20 and 30. Therefore, it can be known that the displayed information pushed to the users in the age between 20 and 30 has a higher probability of getting a click response based on the influence tag.

[0110] In the above embodiment, the method further includes:

[0111] According to the set second threshold, it is determined whether the individual effect value is greater than or equal to the second threshold. When the judgment result is that the individual effect value is greater than or equal to the second threshold, the individual effect value and the display intervention factor corresponding to the individual effect value are stored in the database.

[0112] According to the above analysis model, the degree of influence of each display intervention factor in the display information on the response result can be obtained. When the judgment result is that the individual effect value is greater than or equal to the second threshold, it can be understood that for the current user adjustment factor, the display intervention factor has a higher degree of influence on the response result, and this display intervention factor can be stored in the database. In the subsequent display information configuration process, the required display intervention factor can be called from the database. For example, for the user adjustment factor in the 20-30 age stage, the effect value of the holiday-themed display intervention factor is higher, and this display intervention factor can be stored in the database. When it is necessary to configure the display information for users in the 20-30 age stage in the future, the pre-stored display intervention factor can be called from the database.

[0113] It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1At least part of the steps may include multiple steps or multiple stages. These steps or stages are not necessarily performed at the same time, but can be performed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed in turn or alternately with other steps or at least part of the steps or stages in other steps.

[0114] Based on the description of the above method embodiment, this embodiment provides a computer device, including a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, the steps of a data processing method for displaying information described in the above embodiment are implemented.

[0115] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of a data processing method for displaying information described in the above embodiment are implemented.

[0116] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0117] In the description of this specification, the description with reference to the terms "some embodiments", "other embodiments", "ideal embodiments", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic descriptions of the above terms do not necessarily refer to the same embodiment or example.

[0118] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features of the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0119] In this specification, each embodiment of the above-mentioned method, device and storage medium is described in a progressive manner. The same and similar parts between the embodiments can be referred to each other or to the corresponding method embodiment description. Each embodiment focuses on the differences from other embodiments. For relevant parts, refer to the partial description of the method embodiment. The specific ones can be obtained according to the description of the aforementioned method embodiment, and they should all fall within the scope of implementation protected by this application. No detailed description of the implementation scheme of each embodiment will be made here.

[0120] In the description of this specification, the description with reference to the terms "some embodiments", "other embodiments", "ideal embodiments", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic descriptions of the above terms do not necessarily refer to the same embodiment or example.

[0121] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features of the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0122] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be construed as limiting the scope of the patent application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent application shall be subject to the attached claims.

Claims

1. A data processing method for displaying information, characterized in that: The steps include: Collecting user behavior data based on the display information, and obtaining the user's response result to the display information according to the user behavior data; Extracting a display intervention factor based on the display information, wherein the display information acts on the response result through the display intervention factor; Acquire historical data of the display information, and filter user samples according to a set filtering rule, wherein the filtering rule includes that push information of the display information is successfully sent to the user within a set period; Acquire a user identity in the user sample and some user information corresponding to the user identity according to the user sample; the user information includes one or more of asset level, age, active frequency, and gender; Extracting feedback data on whether the user responds to the push information according to the user identity in the user sample; Performing univariate quality analysis on the user information; Discretize the continuous variables in the user information and divide them into several levels according to preset layers; Generate a sample data set based on the user information and feedback data, split the sample data set into a training set and a test set, and build a logistic regression model; Based on the logistic regression model, the user information is input, and the user adjustment factor is screened according to the output of the logistic regression model; An analysis model is constructed with the display intervention factor as a mediating variable, the user adjustment factor as a moderating variable, the display information as an independent variable, and the response result as a dependent variable. The analysis model is a moderating mediation effect model. Based on the analysis model, setting the value of the user adjustment factor, inputting the display information, and calculating the individual effect value of each display intervention factor of the display information on the response result; Calculating the sum of the individual effect values ​​to obtain the effect value of the display information acting on the response result through the display intervention factor; According to the effect value, a corresponding influence label is marked for the display information.

2. The method according to claim 1, characterized in that The construction of the analysis model then includes: The stepwise test method is used to verify whether the mediation effect of the analysis model is significant; Using latent moderator rule structural equation method to verify whether the fit of the analysis model meets the preset standard; When the verification result of the analysis model does not satisfy the condition that the mediation effect is significant or the degree of fit does not meet the preset standard, the analysis model is reconstructed and verified.

3. The method according to claim 1, characterized in that The method further comprises: According to a set first threshold, determining whether the effect value of the display information is greater than or equal to the first threshold, and when the determination result is that the effect value of the display information is greater than or equal to the first threshold, marking the influence label of the display information as the user adjustment factor corresponding to the effect value; The push information of the display information is sent to the user in a targeted manner according to the user adjustment factor in the influence tag.

4. The method according to claim 1, characterized in that The method further comprises: According to the set second threshold, determine whether the individual effect value is greater than or equal to the second threshold. When the judgment result is that the individual effect value is greater than or equal to the second threshold, store the individual effect value and the display intervention factor corresponding to the individual effect value in the database.

5. The method according to claim 1, characterized in that The response results include no response and executing a response action; the response action includes giving feedback on the product displayed on the page when the user browses the displayed information.

6. The method according to claim 1, characterized in that The method of inputting the user information based on the logistic regression model and screening the user adjustment factor according to the output of the logistic regression model also includes performing a preliminary screening of the user factors by calculating the evidence weight and the information value.

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

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

Images