A storage method and system based on e-commerce tag data

Abstract

The application relates to an e-commerce label data-based storage method and system. The method combines the business characteristics of the e-commerce industry, sets multiple label dimensions to comprehensively divide label data, comprehensively calculates and analyzes label data according to the weights of customer active behavior, purchase behavior, marketing behavior and repurchase behavior, divides the label data into low active degree and high active degree label groups, then establishes a label horizontal table to store low active degree label group data, and establishes a label vertical table to store high active degree label group data, combines the performance of label horizontal table query labels and the flexibility of label vertical table increase and delete label data, effectively avoids business collapse caused by multiple changes of table structures, and reduces system performance loss. In addition, the weight values of various customer behaviors and the customer active threshold in the method are all dynamic configuration parameters, can be flexibly configured according to the actual application requirements of e-commerce business, and the flexibility and expansibility of label data storage are improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of data storage methods, and particularly relates to a storage method and system based on e-commerce tag data. BACKGROUND

[0002] In today's rapid development of mobile Internet, the e-commerce mode platforms based on B2B (Business To Business), B2C (Business To Consumer), C2C (Consumer To Consumer), O2O (Online To Offline) and C2M (Customer to Manufacturer) are everywhere in our life. On these e-commerce platforms such as the above, a large amount of user data, business data and merchant data is generated based on user purchase behavior. The more the original data accumulates, the more it becomes the key data asset of an enterprise. However, the disordered original data cannot directly generate business value. Only through the construction of a tag data layer and the tagging operation of data objects in different business dimensions can the data tagging process maximize the data value. At present, it has become a common marketing strategy for most large and medium-sized e-commerce platforms to extract key data under target tags through data tagging, output user portrait maps, and then achieve different precise marketing purposes based on user portraits. In view of the particularly huge amount of tag data of large and medium-sized e-commerce platforms, it is particularly important to find a high-efficiency, high-performance and high-reliability tag data storage solution.

[0003] In the existing tag data storage technology, when the tag data management platform stores the tag data, the tag data has an association relationship between the tag data and the tag data. Therefore, the tag data is usually completely stored in a vertical table or a horizontal table. However, with the development of business, the tag data is constantly changing, and a large amount of query and application operations need to be performed in the running process of the business. At this time, when the tag data changes, the table structure of the horizontal table needs to be changed if the horizontal table is used to store the tag data, and the vertical table needs to be converted into a horizontal table if the vertical table is used to store the tag data. This makes the workload of the tag data management platform in the running process very huge, which may cause the business being executed on the tag data management platform to collapse and cause a large amount of performance loss. In addition, the low-activity tag group and high-activity tag group division form in the current tag data storage method is too single, the division dimension is relatively narrow, and lacks flexibility and scalability, which cannot meet the actual application requirements of the e-commerce industry.

[0004] In view of the above problems, an ideal solution has not been proposed at present. SUMMARY

[0005] In order to solve the problems of inconvenient operation, large workload, easy to cause label data management platform business collapse, cause a large amount of performance loss, poor flexibility and expansibility, and not suitable for e-commerce platform label data management of the existing label data storage technology, we provide a storage scheme based on e-commerce label data, the purpose of the present application aims to improve the storage efficiency, storage performance and storage reliability of e-commerce label data.

[0006] The present application combines the business characteristics of the e-commerce industry, sets up multiple label dimensions to comprehensively weight divide various label data, comprehensively calculates and analyzes the weight of the to-be-stored label data according to the customer active behavior, purchase behavior, marketing behavior and repurchase behavior, divides the label data into a low active degree label group and a high active degree label group, then establishes a label horizontal table to store the low active degree label group data, and establishes a label vertical table to store the high active degree label group data, combines the performance of the label horizontal table query label and the flexibility of the label vertical table to add and delete label data, effectively avoids the business collapse caused by the multiple changes of the table structure, reduces the system performance loss, and greatly improves the flexibility and expansibility of the label data storage system.

[0007] Specifically, in a first aspect, the present application provides a storage method based on e-commerce label data, which comprises:

[0008] S1: constructing a customer active degree value calculation system, the calculation system includes the following four types of customer behavior statistical characteristics: active behavior, purchase behavior, marketing behavior, and repurchase behavior, and the statistical data of the above four types of customer behavior is weighted calculated according to a preset weight value to obtain a customer active degree value;

[0009] S2: inputting the to-be-stored label data into the customer active degree value calculation system to obtain the customer active degree value of the label data;

[0010] S3: setting a customer active threshold, and comparing the customer active degree value of the label data obtained in the above step with the customer active threshold; if the customer active degree value of the label data is less than the customer active threshold, the label data is divided into a low active degree label group; if the customer active degree value of the label data is greater than or equal to the customer active threshold, the label data is divided into a high active degree label group;

[0011] S4: establishing a label horizontal table, storing the label data of the low active degree label group in the label horizontal table; establishing a label vertical table, storing the label data of the high active degree label group in the label vertical table;

[0012] S5: receiving a label output request, judging whether the label output request carries a specified label name and a specified label value;

[0013] S6: If the specified tag name and the specified tag value are carried in the tag output request, the tag longitudinal table is converted into a temporary transverse table, and the specified tag name and the specified tag value carried in the tag output request are filtered based on the tag transverse table and the temporary transverse table; if the specified business subject is carried in the tag output request but the specified tag name and the specified tag value are not carried, the specified business subject is directly filtered based on the tag transverse table and the tag longitudinal table.

[0014] Further, in the storage method of the e-commerce tag data, the weight values of the four types of customer behaviors in S1 and the customer activity threshold in S3 are dynamic configuration parameters, which can be flexibly configured according to the actual application scene and specific application requirements of the e-commerce business.

[0015] Further, according to some embodiments of the present application, the statistical features of the four types of customer behaviors in the storage method of the e-commerce tag data in step S1 include the following statistical data respectively:

[0016] The active behavior includes registration duration, number of monthly platform logins, number of monthly platform order placements, and sleep duration;

[0017] The purchase behavior includes the number of order placements and the single price of a customer;

[0018] The marketing behavior includes the single average price of using a coupon, the number of expired coupons, and the number of coupons received;

[0019] The repeat purchase behavior includes the number of times of repeatedly purchasing the same SKU product and the single average price of the same SKU product.

[0020] Further, according to some embodiments of the present application, the label group value calculation method of the statistical features of the four types of customer behaviors in the above storage method of the e-commerce tag data is as follows:

[0021] The active behavior label group value = registration duration + number of monthly platform logins + number of monthly platform order placements - sleep duration;

[0022] The purchase behavior label group value = single price of a customer * number of order placements;

[0023] The marketing behavior label group value = (number of received coupons - number of expired coupons) * single average price of using a coupon;

[0024] The repeat purchase behavior label group value = number of times of repeatedly purchasing the same SKU product * single average price of the same SKU product.

[0025] Further, according to some embodiments of the present application, the customer activity value calculation method in the above storage method of the e-commerce tag data is as follows:

[0026] Customer activity value = active behavior tag group value * active behavior weight value + purchase behavior tag group value * purchase behavior weight value + marketing behavior tag group value * marketing behavior weight value + repeat purchase behavior tag group value * repeat purchase behavior weight value.

[0027] Further, according to some embodiments of the present application, the present application stores the low-activity tag group tag data in the tag horizontal table and stores the high-activity tag group tag data in the tag vertical table based on the establishment of the tag horizontal table and the tag vertical table in the step S4 of the storage method of e-commerce tag data.

[0028] The establishment method of the tag horizontal table is as follows:

[0029] The description object of at least one first to-be-stored tag data is counted as a first business subject in the low-activity tag group; a first table structure is established according to a first association relationship between the at least one first to-be-stored tag data and the first business subject; and the at least one first to-be-stored tag data is filled into the first table structure to generate the tag horizontal table.

[0030] The establishment method of the tag vertical table is as follows:

[0031] The description object of at least one second to-be-stored tag data is counted as a second business subject in the high-activity tag group; a second table structure is established with the subject name of the second business subject, the tag name of the at least one second to-be-stored tag data, and the tag value as columns; and the tag name and the tag value of the at least one second to-be-stored tag data are filled into the second table structure to generate the tag vertical table.

[0032] Further, according to some embodiments of the present application, the format conversion of the tag vertical table to generate the temporary horizontal table in the step S6 of the storage method of e-commerce tag data includes: establishing a third table structure according to a second association relationship between the at least one second to-be-stored tag data and the second business subject of the tag vertical table; and filling the at least one second to-be-stored tag data into the third table structure to generate the temporary horizontal table.

[0033] In a second aspect, the present application further provides a storage system of e-commerce tag data, which comprises:

[0034] A customer activity value calculation module is configured to calculate a customer activity value according to the statistical characteristics of multiple types of customer behaviors and according to preset weight values, wherein the customer behaviors include active behaviors, purchase behaviors, marketing behaviors, and repeat purchase behaviors.

[0035] The customer activity comparison module is configured to compare the customer activity value with a preset customer activity threshold, and divide the label data into a low activity label group or a high activity label group according to the comparison result.

[0036] The label cross table management module is configured to establish a label cross table, and store the label data of the low activity label group in the label cross table.

[0037] The label vertical table management module is configured to establish a label vertical table, and store the label data of the high activity label group in the label vertical table.

[0038] The task processing module is configured to receive a label output request, and determine whether the label output request carries a specified label name and a specified label value; if the label output request carries the specified label name and the specified label value, the task processing module is configured to perform format conversion on the label vertical table, generate a temporary cross table, and filter the specified label name and the specified label value carried in the label output request based on the label cross table and the temporary cross table; if the label output request carries a specified business subject but does not carry the specified label name and the specified label value, the task processing module is configured to directly filter the specified business subject based on the label cross table and the label vertical table.

[0039] Further, according to some embodiments of the present application, the customer activity value calculation module in the storage system of the e-commerce label data calculates the customer activity value according to the statistical characteristics of multiple types of customer behaviors in a manner of weighted calculation according to preset weight values, and the customer behaviors include active behaviors, purchase behaviors, marketing behaviors, and repeat purchase behaviors.

[0040] The statistical characteristics of the customer behaviors respectively include the following statistical data.

[0041] The active behaviors include registration duration, monthly login platform times, monthly platform order placing times, and sleep duration.

[0042] The purchase behaviors include order placing times and single customer price.

[0043] The marketing behaviors include single average order value of using coupons, number of expired coupons, and number of obtained coupons.

[0044] The repeat purchase behaviors include number of repeated purchase of the same SKU product and single average order value of the same SKU product.

[0045] The label group value calculation method of the statistical characteristics of the customer behaviors is as follows.

[0046] The active behavior label group value = registration duration + monthly login platform times + monthly platform order placing times - sleep duration.

[0047] The purchase behavior label group value = single customer price * order placing times.

[0048] Marketing behavior tag group value = (number of coupons taken - number of expired coupons) * single price of using coupons;

[0049] Repeat purchase behavior tag group value = number of times of repeatedly purchasing the same SKU * single price of the same SKU;

[0050] The customer activity value calculation method is as follows:

[0051] Customer activity value = active behavior tag group value * active behavior weight value + purchase behavior tag group value * purchase behavior weight value + marketing behavior tag group value * marketing behavior weight value + repeat purchase behavior tag group value * repeat purchase behavior weight value.

[0052] In a third aspect, the present application also provides a computer readable storage medium, wherein the storage medium stores a computer program, and the program is executed by a processor to realize the steps of the above-mentioned e-commerce tag data storage method.

[0053] In summary, the e-commerce tag data storage method and system of the present application has the following characteristics:

[0054] (1) The present application method combines the business characteristics of the e-commerce industry, sets multiple tag dimensions to comprehensively weight various types of tag data, and comprehensively calculates the weight of the to-be-stored tag data according to the customer activity, purchase behavior, marketing behavior, and repeat purchase behavior, divides the to-be-stored tag data into a low activity tag group and a high activity tag group, and the division standard is more comprehensive and objective, and can better reflect the accurate distinction of customer activity.

[0055] (2) The weight values of the customer activity, purchase behavior, marketing behavior, and repeat purchase behavior in the present application method and the customer activity threshold are all dynamic configuration parameters, which can be flexibly configured according to the actual application scene and specific application requirements of the e-commerce business, thereby increasing the flexibility of dividing the low activity tag group and the high activity tag group.

[0056] (3) The low activity tag group data is stored in a tag horizontal table, and the high activity tag group data is stored in a tag vertical table, which combines the performance of querying tags in the tag horizontal table and the flexibility of adding and deleting tag data in the tag vertical table, effectively avoids business collapse caused by multiple changes of table structure, reduces performance loss, and significantly improves the flexibility and scalability of tag data storage.

[0057] (4) The present application method is simple to operate, and can greatly improve the storage efficiency, storage performance, and storage reliability of tag data, and can well meet the actual application requirements of e-commerce platform tag data management. BRIEF DESCRIPTION OF DRAWINGS

[0058] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings used in the embodiment description are briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, not all embodiments, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0059] Figure 1 The architecture diagram of the e-commerce user label system designed by the present application.

[0060] Figure 2 The implementation flowchart of the storage method based on e-commerce label data of the present application.

[0061] Figure 3 The implementation flowchart of the storage method based on e-commerce label data of the present application.

[0062] Figure 4 The composition structure diagram of the storage system based on e-commerce label data of the present application. DETAILED DESCRIPTION

[0063] In order to make the purpose, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be described clearly and completely in combination with specific embodiments. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments, and the present application can also be implemented or applied by other different specific embodiments, and each detail in the specification can be modified or changed based on different views and applications without departing from the spirit of the present application.

[0064] It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict; and all other embodiments obtained by those skilled in the art based on the embodiments in the present disclosure without creative labor are within the scope of protection of the present disclosure.

[0065] It should be noted that the aspects described herein can be embodied in a wide variety of forms and any specific structure and / or function described herein is merely illustrative. Based on the present disclosure, those skilled in the art should understand that one aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways, for example, a device can be implemented and / or a method can be practiced using any number of the aspects described herein.

[0066] The present application is described in detail below using the embodiments shown in the drawings.

[0067] According to Figures 1-3 , the storage method based on e-commerce label data of the present application comprises:

[0068] According to Figure 1 The user label system shown in the figure, through the customer active behavior, purchase behavior, marketing behavior, repeat purchase behavior four aspects of statistical characteristics weighted calculation customer activity value, and then the customer activity value and customer activity threshold comparison. When the customer activity value is less than the customer activity threshold, then the storage label data is divided in the low activity label group; when the customer activity value is greater than or equal to the customer activity threshold, then the storage label data is divided in the high activity label group.

[0069] Table 1: User label system weight setting

[0070] Statistical feature tag group Weight (dynamically configurable) Active behavior 10 Purchase behavior 5 Marketing behavior 3 Repeat purchase behavior 4

[0071] Ua: customer activity value (User activity);

[0072] X: active behavior label group value, X = registration duration + login platform times per month + platform order times per month - sleep duration;

[0073] Y: purchase behavior label group value, Y = single price * order times;

[0074] Z: marketing behavior label group value, Z = (the number of received coupons - the number of expired coupons) * the single price of using coupons;

[0075] W: repeat purchase behavior label group value, W = the number of times of repeated purchase of the same SKU product * the single price of the same SKU product;

[0076] Then the customer activity value Ua = X * 10 + Y * 5 + Z * 3 + W * 4.

[0077] Set the customer activity threshold Uat (User Active Threshold) as: Uat = 120, this threshold can be flexibly configured according to the actual application, it is a dynamic configuration parameter.

[0078] When Ua < Uat, then the storage label data is divided in the low activity label group;

[0079] When Ua ≥ Uat, then the storage label data is divided in the high activity label group.

[0080] The division of low activity label group and high activity label group combines the weights of active behavior, purchase behavior, marketing behavior and repeat purchase behavior, and the division standard is more comprehensive and objective, which can better reflect the accurate distinction of customer activity.

[0081] Then, a label horizontal table is established to store low-activity label group data, and a label vertical table is established to store high-activity label group data, the performance of querying labels based on the label horizontal table is combined with the flexibility of adding and deleting label data of the label vertical table, business collapse caused by multiple changes of table structures is avoided, performance loss is reduced, and application flexibility and expansibility are improved.

[0082] A label horizontal table is established, and low-activity label group data is stored based on the label horizontal table. In the low-activity label group, a description object of at least one first to-be-stored label data is counted as a first business subject. A first table structure is established according to a first association relationship between the at least one first to-be-stored label data and the first business subject. The at least one first to-be-stored label data is filled into the first table structure to generate the label horizontal table.

[0083] A label vertical table is established, and high-activity label group data is stored based on the label vertical table. In the high-activity label group, a description object of at least one second to-be-stored label data is counted as a second business subject. A second table structure is established with a subject name of the second business subject, a label name of the at least one second to-be-stored label data, and a label value as columns. The label name and the label value of the at least one second to-be-stored label data are filled into the second table structure to generate the label vertical table.

[0084] A label output request is received. If the label output request carries a specified business subject but does not carry a specified label name and a specified label value, the specified business subject is filtered based on the label horizontal table and the label vertical table. If the label output request carries the specified label name and the specified label value, the label vertical table is format-converted to generate a temporary horizontal table, and the specified label name and the specified label value carried in the label output request are filtered based on the label horizontal table and the temporary horizontal table.

[0085] The format conversion of the label vertical table to generate the temporary horizontal table includes: establishing a third table structure according to a second association relationship between the at least one second to-be-stored label data and the second business subject of the label vertical table; and filling the at least one second to-be-stored label data into the third table structure to generate the temporary horizontal table.

[0086] According to Figure 4 As shown in FIG. 1, the storage system of the e-commerce label data according to the present application includes:

[0087] The customer activity value calculation module is configured to calculate the customer activity value by weighting according to the statistical characteristics of the multiple types of customer behaviors according to the preset weight values, wherein the customer behaviors include active behaviors, purchase behaviors, marketing behaviors, and repeat purchase behaviors.

[0088] The customer activity comparison module is configured to compare the customer activity value with a preset customer activity threshold, and divide the label data into a low activity label group or a high activity label group according to the comparison result.

[0089] The label cross table management module is configured to establish a label cross table, and store the label data of the low activity label group in the label cross table.

[0090] The label vertical table management module is configured to establish a label vertical table, and store the label data of the high activity label group in the label vertical table.

[0091] The task processing module is configured to receive a label output request, and determine whether the label output request carries a specified label name and a specified label value; if the label output request carries the specified label name and the specified label value, the label vertical table is converted into a temporary cross table, and the specified label name and the specified label value carried in the label output request are filtered based on the label cross table and the temporary cross table; if the label output request carries a specified business subject but does not carry the specified label name and the specified label value, the specified business subject is directly filtered based on the label cross table and the label vertical table.

[0092] The modules are implemented and operated according to the storage method of the e-commerce label data.

[0093] The embodiments in the present application are described in a progressive manner, and the same or similar parts of the embodiments can be referred to each other.

[0094] The above only describes the embodiments of the present application, and is not used to limit the present application. The present application can be variously changed and modified by those skilled in the art. Any modification, replacement and the like within the spirit and principle of the present application should be included in the protection scope of the claims of the present application.

Claims

1. A method for storing e-commerce tag data, characterized in that, The method includes: S1: Construct a customer activity value calculation system, which includes the following four types of customer behavior statistical characteristics: active behavior, purchase behavior, marketing behavior, and repeat purchase behavior. The statistical data of the above four types of customer behavior are weighted according to preset weight values ​​to obtain the customer activity value. The statistical characteristics of the four types of customer behavior include the following statistical data: Active behaviors include registration duration, number of times you log in to the platform per month, number of orders you place on the platform per month, and inactive duration; Purchasing behavior includes the number of orders placed and the average order value; Marketing activities include average price per unit using coupons, number of coupons expiring, and number of coupons claimed; Repeat purchase behavior includes the number of times the same SKU is purchased repeatedly and the average unit price of the same SKU. The calculation method for the label group values ​​of the statistical characteristics of the four types of customer behavior is as follows: Active behavior tag group value = Registration duration + Monthly platform login count + Monthly platform order count - Dormant duration; Purchase behavior tag group value = average order value * number of orders; Marketing behavior tag group value = (number of coupons claimed - number of coupons expired) * average unit price of coupons used; Repeat purchase behavior tag group value = number of times the same SKU is purchased repeatedly * average unit price of the same SKU; The customer activity score is calculated as follows: Customer activity score = Active behavior tag group score * Active behavior weight score + Purchase behavior tag group score * Purchase behavior weight score + Marketing behavior tag group score * Marketing behavior weight score + Repeat purchase behavior tag group score * Repeat purchase behavior weight score; S2: Input the tag data to be stored into the customer activity value calculation system to calculate the customer activity value of the tag data; S3: Set a customer activity threshold and compare the customer activity value of the tag data obtained in the previous step with the customer activity threshold; if the customer activity value of the tag data is less than the customer activity threshold, then the tag data is classified into the low activity tag group; if the customer activity value of the tag data is greater than or equal to the customer activity threshold, then the tag data is classified into the high activity tag group. S4: Create a horizontal tag table to store the tag data of low-activity tag groups; create a vertical tag table to store the tag data of high-activity tag groups. S5: Receive tag output request and determine whether the tag output request carries the specified tag name and the specified tag value; S6: If the tag output request carries a specified tag name and a specified tag value, then the tag vertical table is formatted to generate a temporary horizontal table, and the specified tag name and specified tag value carried in the tag output request are filtered based on the tag horizontal table and the temporary horizontal table; if the tag output request carries a specified business entity but does not carry a specified tag name and specified tag value, then the specified business entity is directly filtered based on the tag horizontal table and the tag vertical table.

2. The method for storing e-commerce tag data according to claim 1, characterized in that, The weight values ​​of the four types of customer behavior mentioned in S1 and the customer activity threshold mentioned in S3 are all dynamically configurable parameters. These parameters can be flexibly configured according to the actual application scenarios and specific application needs of e-commerce business.

3. The method for storing e-commerce tag data according to claim 1, characterized in that, As described in S4, a horizontal tag table is created to store the tag data of the low-activity tag group; a vertical tag table is created to store the tag data of the high-activity tag group. The method for creating the label table is as follows: In the low-activity tag group, at least one description object of the first tag data to be stored is selected as the first business entity; based on the first association relationship between the at least one first tag data to be stored and the first business entity, a first table structure is established; the at least one first tag data to be stored is filled into the first table structure to generate the tag horizontal table; The method for creating the label column table is as follows: In the high-activity tag group, at least one description object of the second tag data to be stored is counted as the second business subject; a second table structure is established with the subject name of the second business subject, the tag name of the at least one second tag data to be stored, and the tag value as columns; The tag name and tag value of the at least one second tag data to be stored are filled into the second table structure to generate the tag vertical table.

4. The method for storing e-commerce tag data according to claim 3, characterized in that, The process of converting the format of the vertical tag table in S6 to generate a temporary horizontal table includes: establishing a third table structure based on the second association relationship between the at least one second tag data to be stored and the second business entity of the vertical tag table; filling the at least one second tag data to be stored into the third table structure to generate the temporary horizontal table.

5. A storage system based on e-commerce tag data, comprising: Customer activity value calculation module: used to calculate the customer activity value by weighting multiple types of customer behavior according to preset weight values ​​based on statistical characteristics. The customer behavior includes active behavior, purchasing behavior, marketing behavior, and repeat purchase behavior. The statistical characteristics of the four types of customer behavior include the following statistical data: Active behaviors include registration duration, number of times you log in to the platform per month, number of orders you place on the platform per month, and inactive duration; Purchasing behavior includes the number of orders placed and the average order value; Marketing activities include average price per unit using coupons, number of coupons expiring, and number of coupons claimed; Repeat purchase behavior includes the number of times the same SKU is purchased repeatedly and the average unit price of the same SKU. The calculation method for the label group values ​​of the statistical characteristics of the four types of customer behavior is as follows: Active behavior tag group value = Registration duration + Monthly platform login count + Monthly platform order count - Dormant duration; Purchase behavior tag group value = average order value * number of orders; Marketing behavior tag group value = (number of coupons claimed - number of coupons expired) * average unit price of coupons used; Repeat purchase behavior tag group value = number of times the same SKU is purchased repeatedly * average unit price of the same SKU; The customer activity score is calculated as follows: Customer activity score = Active behavior tag group score * Active behavior weight score + Purchase behavior tag group score * Purchase behavior weight score + Marketing behavior tag group score * Marketing behavior weight score + Repeat purchase behavior tag group score * Repeat purchase behavior weight score; Customer activity comparison module: This module compares customer activity values ​​with preset customer activity thresholds and classifies the tag data into low activity tag groups or high activity tag groups based on the comparison results. Tag Horizontal Table Management Module: Used to create tag horizontal tables and store low-activity tag group tag data in the tag horizontal table; Tag column management module: used to create tag columns and store high-activity tag group data in the tag column. Task processing module: Receives tag output requests and determines whether the tag output request carries a specified tag name and a specified tag value. If the tag output request carries a specified tag name and a specified tag value, the tag vertical table is formatted to generate a temporary horizontal table, and the specified tag name and specified tag value carried in the tag output request are filtered based on the tag horizontal table and the temporary horizontal table. If the tag output request carries a specified business entity but does not carry a specified tag name and specified tag value, the specified business entity is directly filtered based on the tag horizontal table and the tag vertical table.

6. A computer-readable storage medium storing a computer program, wherein the program, when executed by a processor, implements the steps of the storage method based on e-commerce tag data as described in any one of claims 1-4.

Images