Marketing data processing method and device, computer equipment and storage medium
By building a marketing prediction model and determining marketing strategies based on the model, the problem that marketing strategy formulation in the existing technology depends on experience and manual analysis is solved, and data-driven marketing strategy formulation is realized, improving the applicability and quality of the strategy and reducing corporate risks.
Patent Information
- Application Number
- CN202411680284.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-22
- Publication Date
- 2025-03-11
AI Technical Summary
In the existing marketing management practice, marketing strategy formulation relies on experience and manual analysis, which leads to a high degree of personalization in the strategy formulation process, affecting the consistency and predictability of marketing effects, and it is difficult to accurately evaluate the return on investment, increasing corporate risks. The marketing strategies formed under traditional methods do not have wide applicability.
By obtaining historical marketing data and user marketing expectations, extracting customer characteristic data, building a marketing prediction model, and determining marketing strategies based on the model and user expectations, getting rid of the influence of human subjective factors, and improving the applicability and quality of the strategy.
The data-driven marketing strategy formulation is realized, which improves the consistency and predictability of marketing strategies, reduces corporate risks, improves the applicability and quality of marketing strategies, and solves the problems of low efficiency in marketing data processing and inaccurate processing results.
Smart Images
Figure CN119599725A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of data processing, and in particular to a marketing data processing method, device, computer equipment and storage medium. Background Art
[0002] In the existing marketing management practice, the formulation of marketing strategies mainly relies on the experience of marketing managers and manual analysis of historical sales data. Although the marketing management system on the market provides a user interface (UI) for marketers to input strategies based on manual analysis, such as preferential and discount policies formulated for specific customer groups, in actual operation, this highly subjective strategy formulation method exposes several problems. First, due to the high degree of personalization in the strategy formulation process, even when facing the same target group, different marketers may propose completely different plans, which affects the consistency and predictability of marketing effects. Secondly, in the absence of systematic quantitative analysis support, it is difficult to accurately evaluate the return on investment (ROI) before marketing activities, which increases the risks faced by enterprises and may cause marketing expenses to exceed expected returns. Finally, the marketing strategies formed under traditional methods often do not have a wide range of applicability. The effects of the same strategy when applied to different scenarios vary significantly, limiting its promotion value. The existence of these problems urgently requires the development of more scientific and reasonable marketing data processing methods. Summary of the invention
[0003] Based on this, it is necessary to provide a marketing data processing method, device, computer equipment and storage medium to address the above technical issues, so as to solve the problems of low marketing data processing efficiency and inaccurate processing results.
[0004] A marketing data processing method, comprising:
[0005] Obtain historical marketing data and user marketing expectations;
[0006] extracting customer characteristic data from the historical marketing data;
[0007] Building a marketing prediction model based on the customer characteristic data;
[0008] A marketing strategy is determined based on the marketing prediction model and the user marketing expectations.
[0009] A marketing data processing device, comprising:
[0010] Acquisition module, used to obtain historical marketing data and user marketing expectations;
[0011] A feature data extraction module, used to extract customer feature data from the historical marketing data;
[0012] A prediction model building module is used to build a marketing prediction model based on the customer feature data;
[0013] The marketing strategy determination module is used to determine the marketing strategy according to the marketing prediction model and the user marketing expectation.
[0014] A computer device comprises a memory, a processor and computer-readable instructions stored in the memory and executable on the processor, wherein the processor implements the above-mentioned marketing data processing method when executing the computer-readable instructions.
[0015] One or more readable storage media storing computer-readable instructions, which, when executed by one or more processors, enable the one or more processors to perform the marketing data processing method as described above.
[0016] The above-mentioned marketing data processing method, device, computer equipment and storage medium obtain marketing strategies based on historical marketing data, get rid of the influence of human subjective factors, and the applicability of marketing strategies is more extensive; they can automatically optimize and learn various marketing parameters to improve the quality of marketing strategies; they determine marketing strategies through marketing prediction models, which solves the problems of low efficiency in marketing data processing and inaccurate processing results. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.
[0018] Figure 1 is a schematic diagram of an application environment of a marketing data processing method in one embodiment of the present invention;
[0019] Figure 2 is a flow chart of a marketing data processing method in one embodiment of the present invention;
[0020] Figure 3 is a structural schematic diagram of a marketing data processing device in one embodiment of the present invention;
[0021] Figure 4 is a schematic diagram of a computer device in one embodiment of the present invention. DETAILED DESCRIPTION
[0022] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0023] The marketing data processing method provided in this embodiment can be applied in Figure 1 In an application environment, a client communicates with a server. The client includes but is not limited to various personal computers, laptops, smart phones, tablet computers, and portable wearable devices. The server can be implemented with an independent server or a server cluster consisting of multiple servers.
[0024] In one embodiment, if Figure 2 As shown, a marketing data processing method is provided, and the method is applied in Figure 1 The server in is taken as an example to illustrate, including the following steps S10-S40.
[0025] S10. Obtain historical marketing data and user marketing expectations.
[0026] Understandably, historical marketing data refers to data related to marketing activities collected through different channels and methods over a period of time in the past, including but not limited to customer age, gender, order data, purchase amount, etc. Historical marketing data can be stored in a user database. User marketing expectations refer to the goals that marketing activities expect to achieve, including but not limited to sales, marketing costs, etc. User marketing expectations can be customized by users.
[0027] S20. Extracting customer characteristic data from the historical marketing data.
[0028] Understandably, customer feature data related to the marketing prediction model can be extracted from historical marketing data. Customer feature data includes multiple different dimensions.
[0029] Optionally, step S20, i.e., extracting customer feature data from the historical marketing data, includes:
[0030] S201, extracting the historical marketing data to obtain initial feature data;
[0031] S202: Clean the initial feature data to obtain the customer feature data; the customer feature data includes customer attribute data, customer order data and customer behavior data.
[0032] Understandably, historical marketing data can be extracted first to obtain initial feature data, such as age, gender, order data, purchase amount, etc., and then these initial feature data can be cleaned, such as supplementing missing values, encoding enumeration data, etc., to form customer feature data.
[0033] Customer feature data includes three types: customer attribute data, customer order data, and customer behavior data. Customer attribute data refers to the basic information of customers, such as gender, age, place of residence / work, education level, etc. Customer order data refers to the historical purchase information of customers, including the number of orders, the number of days since the last purchase, the total amount of order purchases, the maximum amount of purchases at one time, the minimum amount of purchases at one time, the SKUs purchased and their quantities, the SKUs of discounted products purchased by customers, etc. Customer behavior data refers to operational behaviors during the purchase process, such as the product links clicked by customers, the time customers stay on the product details page (after weighted calculation), the SKUs of products that customers have added to the shopping cart but have not checked out, the coupons that customers have received but not used, the coupons that customers have received and used, the traffic platforms from which customers have successfully purchased, and the traffic platforms from which customers have not completed purchases, etc. These three types of data can be added or deleted according to the different industries of users.
[0034] S30: constructing a marketing prediction model according to the customer characteristic data.
[0035] Understandably, a marketing prediction model can be constructed based on the extracted multiple customer feature data. Here, a marketing prediction model can be constructed based on a deep learning algorithm.
[0036] Optionally, step S30, i.e., constructing a marketing prediction model according to the customer feature data, includes:
[0037] S301, constructing an initial marketing prediction model;
[0038] S302, inputting the customer feature data into the initial marketing prediction model to obtain a model error parameter;
[0039] S303, inputting the model error parameter into a gradient descent model to obtain a model correction parameter of the initial marketing prediction model;
[0040] S304, determining a revised marketing prediction model according to the model revision parameters;
[0041] S305: When the model error parameter output by the modified marketing prediction model meets the preset requirements, the modified marketing prediction model is determined as the marketing prediction model.
[0042] Understandably, an initial marketing prediction model can be constructed, customer feature data can be input, an expected value can be obtained, and the error between the expected value and the true value can be calculated, which is the model error parameter (loss value). The model error parameter is input into the gradient descent model to obtain the model correction parameter of the initial marketing prediction model. The initial marketing prediction model is adjusted according to the model correction parameter to obtain a revised marketing prediction model. Customer feature data is input into the revised marketing prediction model to obtain a revised expected value, and the error between the revised expected value and the true value can be calculated, which is the model error parameter output by the revised marketing prediction model. The iterative calculation is performed continuously in this way. When the model error parameter output by the revised marketing prediction model meets the preset requirements, the revised marketing prediction model is determined as the marketing prediction model. Here, the preset requirements can be set according to actual needs, such as the model error parameter can be less than a preset threshold.
[0043] Optionally, the initial marketing prediction model includes:
[0044]
[0045] Among them, x i Represents the customer feature data of the i-th row;
[0046] n is the total number of rows of customer feature data;
[0047] w i Represents the weight coefficient of the i-th row in the weight matrix W;
[0048] ReLU() stands for rectified linear unit;
[0049] expv is the customer's expected purchase volume.
[0050] Understandably, if the customer feature data contains n rows and each row contains m features, the customer feature data can be represented by an n*m matrix. An m*m weight matrix W can be generated by a random function. The customer feature data of each row is selected and multiplied by the weight matrix to obtain the product, and then the product is processed by a rectified linear unit (i.e., an activation function) to obtain the result of applying the ReLU activation function, and finally the sum of the various results is calculated, which is the customer's expected purchase amount. In one example, the customer's expected purchase amount can be the customer's expected purchase amount.
[0051] Optionally, the gradient descent model includes:
[0052]
[0053] in, The partial derivative result calculated for this round of iteration;
[0054] A 1is the estimated value of the previous round in the iterative calculation;
[0055] A 2 It is the estimated value after optimization in this round of iteration;
[0056] Z 2 The difference calculated for the previous iteration;
[0057] a is the correction coefficient randomly generated in each round of iteration calculation;
[0058] Y is the result value calculated based on the historical marketing data.
[0059] It is understandable that in order to minimize the model error parameter, the model correction parameter can be calculated using the gradient descent model. In one example, the model correction parameter includes the partial derivative result of the weight
[0060] After cyclic iterations of the above gradient descent model, a better solution for the weight matrix W can be obtained.
[0061] S40: Determine a marketing strategy according to the marketing prediction model and the user marketing expectation.
[0062] Understandably, after obtaining the marketing prediction model, predictions are made through the marketing prediction model to obtain expected values. When the expected values meet the user's marketing expectations, corresponding marketing parameters can be output, and then marketing strategies can be formulated based on the marketing parameters.
[0063] Optionally, step S40, i.e., determining a marketing strategy according to the marketing prediction model and the user marketing expectation, includes:
[0064] S401, determining marketing parameters according to the marketing prediction model and the user marketing expectation; the marketing parameters include target customer group, target channel, coupon information and / or sales product category;
[0065] S402: Determine the marketing strategy according to the marketing parameters.
[0066] Understandably, after obtaining the marketing prediction model, the marketing prediction model is used to make predictions and obtain the expected value. When the expected value meets the user's marketing expectations, the corresponding marketing parameters can be output. In one example, the marketing parameters include but are not limited to the target channel (on which Internet media should the ad be placed, such as Douyin, Weibo, etc.), the target customer group (age group, gender, amount of money already spent, customer level), coupon information (whether it is issued, discount rate, discount category, applicable product range), sales product category (product category range, SKU quantity), placement amount, and placement timing.
[0067] After obtaining the marketing parameters, a marketing workflow can be formed and a complete marketing strategy can be generated.
[0068] This embodiment obtains marketing strategies based on historical marketing data, gets rid of the influence of human subjective factors, and has a wider applicability of marketing strategies; it can automatically optimize and learn various marketing parameters to improve the quality of marketing strategies; it determines marketing strategies through marketing prediction models to solve the problems of low efficiency in marketing data processing and inaccurate processing results.
[0069] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.
[0070] In one embodiment, a marketing data processing device is provided, and the marketing data processing device corresponds one-to-one to the marketing data processing method in the above embodiment. Figure 3 As shown, the marketing data processing device includes:
[0071] An acquisition module 10 is used to acquire historical marketing data and user marketing expectations;
[0072] A feature data extraction module 20, used to extract customer feature data from the historical marketing data;
[0073] A prediction model building module 30 is used to build a marketing prediction model based on the customer feature data;
[0074] The marketing strategy determination module 40 is used to determine the marketing strategy according to the marketing prediction model and the user marketing expectation.
[0075] Optionally, the feature data extraction module 20 includes:
[0076] Extracting the historical marketing data to obtain initial feature data;
[0077] The initial feature data is cleaned to obtain the customer feature data; the customer feature data includes customer attribute data, customer order data and customer behavior data.
[0078] Optionally, the prediction model building module includes:
[0079] Constructing an initial model unit for constructing an initial marketing prediction model;
[0080] An error parameter acquisition unit, used for inputting the customer feature data into the initial marketing prediction model to obtain a model error parameter;
[0081] A correction parameter acquisition unit, used for inputting the model error parameter into a gradient descent model to obtain a model correction parameter of the initial marketing prediction model;
[0082] A determination and correction model unit, used for determining a correction marketing prediction model according to the model correction parameters;
[0083] The prediction model obtaining unit is used to determine the modified marketing prediction model as the marketing prediction model when the model error parameter output by the modified marketing prediction model meets the preset requirements.
[0084] Optionally, the initial marketing prediction model includes:
[0085]
[0086] Among them, x i Represents the customer feature data of the i-th row;
[0087] n is the total number of rows of customer feature data;
[0088] w i Represents the weight coefficient of the i-th row in the weight matrix W;
[0089] ReLU() stands for rectified linear unit;
[0090] expv is the customer's expected purchase volume.
[0091] Optionally, the gradient descent model includes:
[0092]
[0093] in, The partial derivative result calculated for this round of iteration;
[0094] A 1 is the estimated value of the previous round in the iterative calculation;
[0095] A 2 It is the estimated value after optimization in this round of iteration;
[0096] Z 2 The difference calculated for the previous iteration;
[0097] a is the correction coefficient randomly generated in each round of iteration calculation;
[0098] Y is the result value calculated based on the historical marketing data.
[0099] Optionally, determining the marketing strategy module 40 includes:
[0100] A parameter determination unit, configured to determine marketing parameters according to the marketing prediction model and the user marketing expectation; the marketing parameters include target customer groups, target channels, coupon information and / or sales commodity categories;
[0101] A strategy determination unit is used to determine the marketing strategy according to the marketing parameters.
[0102] The specific definition of the marketing data processing device can be found in the definition of the marketing data processing method above, which will not be repeated here. Each module in the above-mentioned marketing data processing device can be implemented in whole or in part by software, hardware, or a combination thereof. Each of the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.
[0103] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 4 As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a readable storage medium and an internal memory. The readable storage medium stores an operating system, computer-readable instructions and a database. The internal memory provides an environment for the operation of the operating system and computer-readable instructions in the readable storage medium. The database of the computer device is used to store data involved in the marketing data processing method. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer-readable instructions are executed by the processor, a marketing data processing method is implemented. The readable storage medium provided in this embodiment includes a non-volatile readable storage medium and a volatile readable storage medium.
[0104] In one embodiment, a computer device is provided, including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, wherein the processor executes the computer-readable instructions to implement the following steps:
[0105] Obtain historical marketing data and user marketing expectations;
[0106] extracting customer characteristic data from the historical marketing data;
[0107] Building a marketing prediction model based on the customer characteristic data;
[0108] A marketing strategy is determined based on the marketing prediction model and the user marketing expectations.
[0109] In one embodiment, one or more computer-readable storage media storing computer-readable instructions are provided. The readable storage media provided in this embodiment include non-volatile readable storage media and volatile readable storage media. The readable storage media store computer-readable instructions, and when the computer-readable instructions are executed by one or more processors, the following steps are implemented:
[0110] Obtain historical marketing data and user marketing expectations;
[0111] extracting customer characteristic data from the historical marketing data;
[0112] Building a marketing prediction model based on the customer characteristic data;
[0113] A marketing strategy is determined based on the marketing prediction model and the user marketing expectations.
[0114] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing related hardware through computer-readable instructions, and the computer-readable instructions can be stored in a non-volatile readable storage medium or a volatile readable storage medium. When the computer-readable instructions are executed, they may 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 may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.
[0115] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.
[0116] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.
Claims
1. A marketing data processing method, characterized in that: include: Obtain historical marketing data and user marketing expectations; extracting customer characteristic data from the historical marketing data; Building a marketing prediction model based on the customer characteristic data; A marketing strategy is determined based on the marketing prediction model and the user marketing expectations.
2. The marketing data processing method according to claim 1, characterized in that: The extracting customer characteristic data from the historical marketing data comprises: Extracting the historical marketing data to obtain initial feature data; The initial feature data is cleaned to obtain the customer feature data; the customer feature data includes customer attribute data, customer order data and customer behavior data.
3. The marketing data processing method according to claim 1, characterized in that: The step of constructing a marketing prediction model according to the customer feature data includes: Build initial marketing forecasting model; Inputting the customer characteristic data into the initial marketing prediction model to obtain a model error parameter; Inputting the model error parameters into a gradient descent model to obtain model correction parameters of the initial marketing prediction model; Determining a revised marketing forecasting model according to the model revision parameters; When the model error parameter output by the modified marketing prediction model meets the preset requirements, the modified marketing prediction model is determined as the marketing prediction model.
4. The marketing data processing method according to claim 3, characterized in that: The initial marketing forecasting model includes: Among them, x i Represents the customer feature data of the i-th row; n is the total number of rows of customer feature data; w i Represents the weight coefficient of the i-th row in the weight matrix W; ReLU() stands for rectified linear unit; expv is the customer's expected purchase volume.
5. The marketing data processing method according to claim 3, characterized in that: The gradient descent model includes: in, The partial derivative result calculated for this round of iteration; A1 is the estimated value of the previous round in iterative calculation; A2 is the estimated value after optimization in this round of iteration; Z2 is the difference calculated in the previous round of iteration; a is the correction coefficient randomly generated in each round of iterative calculation; Y is the result value calculated based on the historical marketing data.
6. The marketing data processing method according to claim 1, characterized in that: Determining a marketing strategy according to the marketing prediction model and the user marketing expectation includes: Determining marketing parameters according to the marketing prediction model and the user marketing expectations; the marketing parameters include target customer groups, target channels, coupon information and / or sales product categories; The marketing strategy is determined according to the marketing parameters.
7. A marketing data processing device, characterized in that: include: Acquisition module, used to obtain historical marketing data and user marketing expectations; A feature data extraction module, used to extract customer feature data from the historical marketing data; A prediction model building module is used to build a marketing prediction model based on the customer feature data; The marketing strategy determination module is used to determine the marketing strategy according to the marketing prediction model and the user marketing expectation.
8. The marketing data processing device according to claim 7, characterized in that: The prediction model building module comprises: Constructing an initial model unit for constructing an initial marketing prediction model; An error parameter acquisition unit, used for inputting the customer feature data into the initial marketing prediction model to obtain a model error parameter; A correction parameter acquisition unit, used for inputting the model error parameter into a gradient descent model to obtain a model correction parameter of the initial marketing prediction model; A determination and correction model unit, used for determining a correction marketing prediction model according to the model correction parameters; The prediction model obtaining unit is used to determine the modified marketing prediction model as the marketing prediction model when the model error parameter output by the modified marketing prediction model meets the preset requirements.
9. A computer device comprising a memory, a processor, and computer-readable instructions stored in the memory and executed on the processor, characterized in that: When the processor executes the computer-readable instructions, the marketing data processing method according to any one of claims 1 to 6 is implemented.
10. One or more readable storage media storing computer-readable instructions, wherein when the computer-readable instructions are executed by one or more processors, the one or more processors execute the marketing data processing method according to any one of claims 1 to 6.