Advertisement marketing user screening method and system

Abstract

The invention provides an advertisement marketing user screening method and system, and belongs to the technical field of advertisement marketing, and the method comprises the steps: collecting and cleaning user basic information, and constructing a comprehensive and accurate user data basis; user features are further refined through user layering and identification of static and dynamic tags. Through introduction of the dynamic label, an enterprise can more accurately position a target user group and formulate a more effective marketing strategy. And dynamic adjustment of an updating period is introduced, so that an enterprise can flexibly arrange the updating frequency of user data according to actual conditions, user screening is more flexible and efficient, changes of user behaviors can be reflected in time, the timeliness and accuracy of screening are ensured, powerful support is provided for formulating a targeted marketing strategy, and the method is suitable for popularization and application. And an enterprise can allocate marketing resources more reasonably.

Description

Technical Field

[0001] The present invention relates to the technical field of advertising marketing, and particularly to an advertising marketing user screening method and system. Background Art

[0002] In today's highly competitive business environment, advertising marketing, as an important means for enterprises to promote products and services, faces numerous challenges. For example, in order to formulate more precise marketing strategies, enterprises need to conduct a detailed analysis of user characteristics. However, traditional user screening methods often only focus on basic user attributes such as age, gender, and region, while ignoring dynamic characteristics such as user purchase behavior, hobbies, and evaluation tendencies. This makes it difficult for enterprises to deeply explore the potential needs of users, thus limiting the pertinence and effectiveness of marketing strategies. In advertising marketing, accurately positioning the target user group is the key to achieving efficient conversion. However, due to the complexity and diversity of user data, many enterprises find it difficult to accurately identify user groups with potential value. This results in a lack of pertinence and precision in formulating marketing strategies by enterprises, and ineffective utilization of marketing resources. With the continuous change of the market environment and the continuous evolution of user behavior, user data is also constantly updated and changed. However, many enterprises have a lag problem in user data update, resulting in the inability of user screening results to timely reflect changes in user behavior. This makes it difficult for enterprises' marketing strategies to keep up with the market rhythm, thus affecting the marketing effect.

[0003] Therefore, it is necessary to provide an advertising marketing user screening method and system to solve the above technical problems. Summary of the Invention

[0004] To solve the above technical problems, the present invention provides an advertising marketing user screening method and system for solving the problems of insufficient refinement of user characteristics, inaccurate positioning of target users, and lag in user data update in traditional marketing user screening methods.

[0005] An advertising marketing user screening method provided by the present invention, the screening method includes the following steps: Through a network big data platform, obtain the basic user information of each marketing area associated with the product, and clean and organize it into a user data set, wherein the basic user information includes transaction records, behavior logs, and evaluation data; Based on the user data set, stratify the users, and respectively identify static tags for each layer of users after stratification; According to the user data set, through preset tag indicators, respectively identify dynamic tags for each layer of users after stratification, wherein the preset tag indicator sections include a purchasing power threshold section, a user favorable comment threshold section, and a repeated purchase frequency section; Set corresponding update cycles for each marketing area respectively, and dynamically adjust the update cycles by monitoring the marketing indicators of each marketing area; Based on the dynamically adjusted update cycles, re-identify the dynamic tags of each layer of users after stratification to screen users.

[0006] Preferably, through the network big data platform, obtain the basic user information of each marketing area associated with the product, and clean and organize it into a user data set. The specific steps are as follows: Connect to the network big data platform, collect the basic user information of each marketing area associated with the product from the network big data platform, including transaction records, behavior logs, and evaluation data; Perform data cleaning on the collected basic user information, specifically including removing duplicate data, handling missing values, and unifying the data format; Integrate the cleaned basic user information according to the user ID to form a user data set.

[0007] Preferably, based on the user data set, stratify the users and respectively identify the static tags of each layer of users after stratification. The specific steps are as follows: According to the transaction records in the basic user information, set the user stratification strategy based on the transaction activity. Among them, the user stratification strategy specifically includes: those with a registration time within 30 days and a purchase behavior are classified as new customer users, those with a purchase behavior within 90 days are classified as active users, and those without a purchase behavior within 90 - 180 days are classified as dormant users; Based on the set user stratification strategy, divide the users into different levels to obtain the user stratification result. Among them, the different levels specifically include the new customer user layer, the active user layer, and the dormant user layer; According to the user stratification result and the user data set, respectively identify the preliminary static tags of each layer of users. Among them, the static tags specifically include the user's age group, gender, hobbies, or region.

[0008] Preferably, according to the user data set, through preset tag indicators, respectively identify the dynamic tags of each layer of users after stratification. The specific steps are as follows: Respectively set the specific numerical ranges of three tag indicators: the purchasing power threshold section, the user good review threshold section, and the repeat purchase frequency section; According to the user data set, calculate the purchasing power, good review rate, and repeat purchase frequency of each user, and determine the section to which it belongs to obtain the section attribution result; Based on the section attribution result, identify the dynamic tags corresponding to the section to which each layer of users after stratification belongs.

[0009] Preferably, corresponding update cycles are set for each marketing area, and the update cycles are dynamically adjusted by monitoring the marketing metrics of each marketing area. The specific steps are as follows: Set an initial update cycle for each marketing area according to business requirements; Monitor the marketing metrics of each marketing area at a preset cycle and calculate the changes in marketing metrics. Among them, the marketing metrics specifically include sales volume, user growth rate, and conversion rate; Based on the calculated results of the changes in marketing metrics, dynamically adjust the update cycles of each marketing area. Specifically, if the marketing metrics of the current marketing area increase, shorten the update cycle of the current marketing area.

[0010] Preferably, based on the dynamically adjusted update cycle, re-identify the dynamic labels of each layer of users after stratification to screen users. The specific steps are as follows: According to the dynamically adjusted update cycle, regularly recalculate and identify the dynamic labels of each layer of users after stratification; According to the re-identified dynamic labels, screen users to identify user groups with different marketing values.

[0011] An advertising marketing user screening system, the screening system includes: A data collection module, used to obtain the basic user information of each marketing area associated with the product through the network big data platform, and clean and organize it into a user data set. Among them, the basic user information includes transaction records, behavior logs, and evaluation data; An information identification module, used to stratify users based on the user data set, and respectively identify the static labels of each layer of users after stratification; A dynamic identification module, used to respectively identify the dynamic labels of each layer of users after stratification according to the user data set through preset label metrics. Among them, the preset label metric sections include purchasing power threshold sections, user favorable comment threshold sections, and repeat purchase frequency sections; A cycle adjustment module, used to set corresponding update cycles for each marketing area, and dynamically adjust the update cycles by monitoring the marketing metrics of each marketing area; A user screening module, used to re-identify the dynamic labels of each layer of users after stratification based on the dynamically adjusted update cycle to screen users.

[0012] Compared with the related technology, an advertising marketing user screening method and system provided by the present invention have the following beneficial effects: The present invention constructs a comprehensive and accurate user data foundation by collecting and cleaning user basic information; further refines user characteristics through user stratification and the identification of static and dynamic tags. By introducing dynamic tags, enterprises can more accurately target the target user group and formulate more effective marketing strategies; and by introducing dynamic adjustment of the update cycle, enterprises can flexibly arrange the update frequency of user data according to the actual situation, so that user screening is more flexible and efficient, can timely reflect changes in user behavior, ensure the timeliness and accuracy of screening, and the present invention provides strong support for formulating targeted marketing strategies and helps enterprises allocate marketing resources more reasonably. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a schematic flowchart of a method for screening users in advertising marketing in Embodiment 1 of the present invention; Figure 2 It is a schematic flowchart of step S4 in Embodiment 1 of the present invention; Figure 3 It is a system block diagram of a system for screening users in advertising marketing in Embodiment 2 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0014] The present invention will be further described below in conjunction with the drawings and embodiments.

[0015] Embodiment 1 As Figure 1 shown, a method for screening users in advertising marketing includes the following steps: S1. Through the network big data platform, obtain the user basic information of each marketing area associated with the product, and clean and organize it into a user data set. Among them, the user basic information includes transaction records, behavior logs, and evaluation data; S2. Based on the user data set, stratify the users, and respectively identify static tags for each layer of users after stratification; S3. According to the user data set, through preset tag indicators, respectively identify dynamic tags for each layer of users after stratification. Among them, the preset tag indicator sections include a purchasing power threshold section, a user favorable comment threshold section, and a repeated purchase frequency section; S4. Set corresponding update cycles for each marketing area respectively, and dynamically adjust the update cycle by monitoring the marketing indicators of each marketing area; S5. Based on the dynamically adjusted update cycle, re-identify the dynamic tags of each layer of users after stratification to screen users.

[0016] Among them, the specific implementation process of step S1 is: S101. Connect to the network big data platform and collect the user basic information of each marketing area associated with the product from the network big data platform, including transaction records, behavior logs, and evaluation data.

[0017] Specifically, establish a connection with the network big data platform through the docking of API interfaces and the establishment of data transmission protocols, and retrieve the user basic information associated with a specific product from the network big data platform after the connection, specifically including the user basic information of different marketing areas.

[0018] In this embodiment, an e-commerce platform screens users for a certain smart watch it sells. First, retrieve the user transaction records related to the smart watch, such as the user information who purchased the watch in the past year, and their purchase frequency, purchase amount, etc.; at the same time, collect the behavior logs of these users, such as the number of times the user browses the relevant pages of the smart watch, or whether they participated in the promotional activities of the smart watch, etc. In addition, the evaluation data of users is also an important object to collect, including their satisfaction scores for the smart watch, descriptions of the usage experience, etc.

[0019] S102. Perform data cleaning on the collected user basic information, specifically including removing duplicate data, handling missing values, and unifying the data format.

[0020] Specifically, perform data cleaning on the collected user basic information. The cleaning process includes removing duplicate data (such as the same transaction recorded multiple times), handling missing values (such as through interpolation, mean substitution, or deleting missing records, etc.), and unifying the data format (such as unifying the date format to YYYY-MM-DD format).

[0021] S103. Integrate the cleaned user basic information according to the user ID to form a user data set.

[0022] Specifically, merge all relevant information of the same user (such as transaction records, behavior logs, evaluation data) into one record, which helps to comprehensively understand the behavior characteristics and preferences of each user, and the integrated user data set will be used as the basis for subsequent user screening.

[0023] In this embodiment, integrate the cleaned user transaction records, behavior logs, and evaluation data according to the user ID. For example, for a certain user, the platform will merge all his records of purchasing the smart watch, the behavior logs of browsing the relevant pages of the smart watch, and the evaluation data of the watch into one record.

[0024] Among them, the specific implementation process of step S2 is as follows: S201. Set the user stratification strategy based on the transaction records in the user's basic information according to the transaction activity. Specifically, the user stratification strategy includes: users who registered within 30 days and have purchase behavior are classified as new customers; users who have purchase behavior within 90 days are classified as active users; users who have no purchase behavior within 90 - 180 days are classified as dormant users.

[0025] In this embodiment, when an e - commerce platform analyzes its user transaction records, it is found that user A registered an account and purchased a commodity one month ago, while user B registered three months ago and made multiple purchases, but has no purchase record in the recent two months. Then, according to the user stratification strategy, user A will be classified as a new customer, and user B will be regarded as a dormant user.

[0026] S202. Divide users into different levels based on the set user stratification strategy to obtain the user stratification result. Specifically, the different levels include the new customer layer, the active user layer, and the dormant user layer.

[0027] Specifically, based on the set user stratification strategy, use database query, script writing, or data analysis tools to divide the users in the user's basic information into different levels.

[0028] S203. Identify the preliminary static labels for users in each layer according to the user stratification result and the user data set. Specifically, the static labels include user age group, gender, hobbies, or region.

[0029] Specifically, use data analysis tools to identify the preliminary static labels for users in each layer.

[0030] In this embodiment, for the new customer layer, through data identification, it is found that they are mainly concentrated in the young population, so labels such as "young users" or "new users" can be added; for the active user layer, labels such as "fashion lovers" and "tech lovers" can be added according to their purchase history and preferences; for the dormant user layer, labels such as "potential churn users" or "users in need of activation" can be added by analyzing their registration information and purchase records.

[0031] Among them, the specific implementation process of step S3 is as follows: S301. Set the specific numerical ranges of three label indicators, namely the purchasing power threshold range, the user favorable comment threshold range, and the repeat purchase frequency range.

[0032] Specifically, clarify the target user group for advertising and marketing, and based on business requirements and user characteristics, determine three key label indicators: purchasing power, user favorable comments, and repeat purchase frequency. Then, according to historical data or industry standards, set the specific numerical range for each label indicator.

[0033] In this embodiment, the purchasing power threshold range: According to the consumption records and payment capabilities of users, the purchasing power is divided into three ranges: "low", "medium", and "high". Specifically, the low purchasing power range may be set as an annual consumption amount between 0 and 1000 yuan, the medium purchasing power range is 1000 - 5000 yuan, and the high purchasing power range is above 5000 yuan. The user good review threshold range: According to the evaluations and feedback of users, the good review rate is divided into four ranges: "poor", "average", "good", and "excellent". Specifically, a good review rate below 30% is "poor", 30% - 70% is "average", 70% - 90% is "good", and above 90% is "excellent". The repeat purchase frequency range: According to the purchase history of users, the repeat purchase frequency is divided into three ranges: "low frequency", "medium frequency", and "high frequency". Specifically, the low frequency is set as a purchase interval exceeding 6 months, the medium frequency is 3 - 6 months, and the high frequency is within 3 months.

[0034] S302. According to the user dataset, calculate the purchasing power, good review rate, and repeat purchase frequency of each user, and determine the range to which they belong to obtain the range attribution result.

[0035] Specifically, calculate the purchasing power, good review rate, and repeat purchase frequency of each user according to the user dataset, and set the specific numerical ranges of the three label indicators of the purchasing power threshold range, user good review threshold range, and repeat purchase frequency range to compare with the calculated results to obtain the range attribution results of the purchasing power, good review rate, and repeat purchase frequency of each user.

[0036] In this embodiment, for a user A, whose annual consumption amount is 3000 yuan, the good review rate is 85%, and the most recent purchase was 3 months ago. Then, according to the set numerical ranges, user A's purchasing power belongs to the "medium" range, the good review rate belongs to the "good" range, and the repeat purchase frequency belongs to the "medium frequency" range.

[0037] S303. Based on the range attribution result, label each layer of the stratified users with the dynamic label corresponding to the range to which they belong.

[0038] Specifically, based on the range attribution result determined in step S302, label each user. In this embodiment, for user A, it can be labeled with the label of "medium purchasing power - good review - medium frequency purchaser".

[0039] Among them, as Figure 2 shown, the specific implementation process of step S4 is as follows: S401. Set an initial update cycle for each marketing area according to business requirements.

[0040] Specifically, set an initial update cycle for each marketing area, where the initial update cycle is specifically weekly, monthly, or quarterly, and is used to regularly evaluate and adjust marketing strategies.

[0041] Exemplarily, for instance, there are two main marketing regions: first-tier cities and second-tier cities. Due to the fierce market competition and rapid changes in user demands in first-tier cities, it is decided to set a shorter update cycle for first-tier cities, such as once a week; while for second-tier cities, since the market competition is relatively mild and user demands change slowly, a longer update cycle is set, such as once a month.

[0042] S402. Monitor the marketing metrics of each marketing region according to the preset monitoring cycle, and calculate the changes in marketing metrics. Among them, the marketing metrics specifically include sales volume, user growth rate, and conversion rate.

[0043] Specifically, collect the marketing metric data of each marketing region through data analysis tools or market research, compare the data at different time points, and calculate the changes in marketing metrics, specifically the growth rate or decline rate.

[0044] S403. Dynamically adjust the update cycle of each marketing region based on the calculated results of the changes in marketing metrics. Specifically, if the marketing metrics of the current marketing region increase, shorten the update cycle of the current marketing region.

[0045] Specifically, the enterprise needs to dynamically adjust the update cycle of each marketing region according to the results of the changes in marketing metrics. It should be noted that if the marketing metrics of a certain marketing region increase rapidly, it indicates that the current marketing strategy has good effects and the market demand is strong. Therefore, the update cycle can be shortened to adjust and optimize the marketing strategy more timely; on the contrary, if the marketing metrics decline or remain stable, it indicates that the current marketing strategy may need to be adjusted or the market demand changes little. Therefore, the update cycle can be extended to reduce unnecessary resource waste.

[0046] In this embodiment, it is calculated that in the first-tier city region, it is found that the sales volume in a certain month has increased by 10%, the user growth rate has increased by 5%, and the conversion rate has remained stable; while in the second-tier city region, it is found that the sales volume has slightly decreased, the user growth rate has remained stable, but the conversion rate has increased. Then, shorten the update cycle of the first-tier city from once a week to once every three days to capture market changes and user demands more timely; while for the second-tier city, since the sales volume has slightly decreased but the conversion rate has increased, keep the current update cycle unchanged.

[0047] Among them, the specific implementation process of step S5 is as follows: S501. Regularly recalculate and identify the dynamic labels of each layer of users after stratification according to the dynamically adjusted update cycle.

[0048] Specifically, according to the dynamically adjusted update cycle, repeat step S3 to identify the dynamic labels of each layer of users after stratification.

[0049] S502. Screen users according to the re-identified dynamic labels to identify user groups with different marketing values.

[0050] Specifically, after obtaining the stratified user dynamic labels, users can be screened according to these labels. Exemplarily, users with strong purchasing power and high interest in specific categories among the "high-frequency purchasing users" are screened out as high-value user groups for key marketing.

[0051] In this embodiment, after obtaining the stratified user dynamic labels, users can be screened according to these labels. For example, users with strong purchasing power and high interest in specific categories among the "high-frequency purchasing users" are screened out as high-value user groups for key marketing. At the same time, for the "low-activity users" group, analyze the reasons for their inactivity (such as lack of interest, price sensitivity, etc.).

[0052] An advertising marketing user screening method provided in this embodiment constructs a comprehensive and accurate user data foundation by collecting and cleaning user basic information; further refines user characteristics through user stratification and the identification of static and dynamic labels. By introducing dynamic labels, enterprises can more accurately locate target user groups and formulate more effective marketing strategies; and the introduction of dynamic adjustment of the update cycle enables enterprises to flexibly arrange the update frequency of user data according to the actual situation, making user screening more flexible and efficient, being able to timely reflect changes in user behavior, ensuring the timeliness and accuracy of screening, providing strong support for formulating targeted marketing strategies, and helping enterprises more reasonably allocate marketing resources.

[0053] Embodiment 2 As Figure 3 shown, an advertising marketing user screening system applied to an advertising marketing user screening method specifically includes: A data collection module, configured to obtain user basic information of each marketing area associated with the product through a network big data platform and clean and organize it into a user data set, where the user basic information includes transaction records, behavior logs, and evaluation data; An information identification module, configured to stratify users based on the user data set and respectively identify static labels for each layer of the stratified users; A dynamic identification module, configured to respectively identify dynamic labels for each layer of the stratified users according to the user data set through preset label metrics, where the preset label metric sections include a purchasing power threshold section, a user favorable comment threshold section, and a repeat purchase frequency section; A cycle adjustment module, configured to set corresponding update cycles for each marketing area respectively, and dynamically adjust the update cycles by monitoring the marketing metrics of each marketing area; A user screening module, configured to re-identify the dynamic labels of each layer of users after stratification based on the dynamically adjusted update cycles to screen users.

[0054] This application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, and the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in one Figure 1 one process or multiple processes and / or blocks Figure 1 or multiple blocks.

[0055] Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing relevant hardware through a program. This program can be stored in a computer-readable storage medium, and the storage medium includes read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM), or other optical disc memories, magnetic disk memories, tape memories, or any other computer-readable medium capable of carrying or storing data.

[0056] It should also be noted that the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, commodity or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, commodity or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the existence of additional identical elements in the process, method, commodity or device comprising the element.

Claims

1. A method for screening advertising marketing users, characterized in that: The screening method comprises the following steps: Through the network big data platform, we obtain the basic user information of each marketing area associated with the product, and clean and organize it into a user data set, where the basic user information includes transaction records, behavior logs, and evaluation data; Based on the user data set, users are stratified and static labels are assigned to users in each stratification layer. According to the user data set, the users in each layer are dynamically labeled by using preset label indicators, wherein the preset label indicator segments include the purchasing power threshold segment, the user praise threshold segment and the repeated purchase frequency segment; Set corresponding update cycles for each marketing area, and dynamically adjust the update cycle by monitoring the marketing indicators of each marketing area; Based on the dynamically adjusted update cycle, the dynamic labels of the users in each layer after stratification are re-identified to screen the users.

2. The method for screening advertising and marketing users according to claim 1, characterized in that: The user basic information of each marketing area associated with the product is obtained through the network big data platform, and the basic information is cleaned and sorted into a user data set. The specific steps are as follows: Connect to the network big data platform and collect basic user information in various marketing areas associated with the product from the network big data platform, including transaction records, behavior logs and evaluation data; Clean the collected basic user information, including removing duplicate data, processing missing values, and unifying data formats; The cleaned basic user information is integrated according to the user ID to form a user data set.

3. The method for screening advertising and marketing users according to claim 1, characterized in that: The method of stratifying users based on the user data set and marking each stratified user with a static label is as follows: According to the transaction records in the user basic information, the user stratification strategy is set according to the transaction activity. Specifically, the user stratification strategy includes: those who registered within 30 days and made purchases are classified as new customers, those who made purchases within 90 days are classified as active users, and those who did not make purchases within 90-180 days are classified as dormant users; Based on the set user stratification strategy, users are divided into different levels to obtain user stratification results, where the different levels specifically include a new customer user level, an active user level, and a dormant user level; According to the user stratification results and the user data set, preliminary static labels are identified for each layer of users, where the static labels specifically include user age group, gender, interests or regions.

4. The method for screening advertising and marketing users according to claim 1, characterized in that: According to the user data set, the user of each layer after stratification is marked with dynamic labels by using preset label indicators, and the specific steps are as follows: Set specific numerical ranges for the three label indicators: purchasing power threshold segment, user praise threshold segment, and repeated purchase frequency segment; Based on the user data set, the purchasing power, favorable comment rate and repeat purchase frequency of each user are calculated, and the segment to which the user belongs is determined to obtain the segment attribution result; Based on the segment attribution result, each layer of users after stratification is identified as a dynamic label corresponding to the segment to which it belongs.

5. The method for screening advertising and marketing users according to claim 1, characterized in that: The specific steps of setting corresponding update cycles for the marketing areas respectively and dynamically adjusting the update cycles by monitoring the marketing indicators of each marketing area are as follows: Set up an initial update cycle for each marketing region based on business needs; Monitor the marketing indicators of each marketing area according to the preset cycle and calculate the changes in marketing indicators, where the marketing indicators specifically include sales, user growth rate and conversion rate; Based on the calculated marketing indicator change results, the update cycle of each marketing area is dynamically adjusted. Specifically, if the marketing indicator of the current marketing area increases, the update cycle of the current marketing area is shortened.

6. The method for screening advertising and marketing users according to claim 1, characterized in that: The specific steps of re-identifying the dynamic labels of users in each layer after stratification based on the dynamically adjusted update cycle to screen users are as follows: According to the dynamically adjusted update cycle, the dynamic labels of users in each layer after stratification are regularly recalculated and marked; Based on the re-identified dynamic tags, users are screened to identify user groups with different marketing values.

7. An advertising and marketing user screening system, applied to an advertising and marketing user screening method according to any one of claims 1 to 6, characterized in that: The screening system comprises: The data collection module is used to obtain the basic user information of each marketing area associated with the product through the network big data platform, and clean and organize it into a user data set, where the basic user information includes transaction records, behavior logs and evaluation data; The information identification module is used to stratify users based on the user data set and identify the users in each stratification layer with static labels; A dynamic labeling module is used to label each layer of users with dynamic labels according to the user data set and preset label indicators, wherein the preset label indicator segments include a purchasing power threshold segment, a user praise threshold segment, and a repeat purchase frequency segment; The cycle adjustment module is used to set corresponding update cycles for each marketing area and dynamically adjust the update cycle by monitoring the marketing indicators of each marketing area; The user screening module is used to re-identify the dynamic labels of the users in each layer after stratification based on the dynamically adjusted update cycle to screen the users.

Images