E-commerce data management method and system based on big data analysis

Abstract

The invention provides an e-commerce data management method and system based on big data analysis, and belongs to the technical field of data management, and the method specifically comprises the steps: taking the historical purchase data of a user in a user portrait interval in different long-period commodities and the remaining available sales deadline as the basis, and when determining that the correlation degree of the user and the long-period commodities meets the requirement, determining that the user is in the user portrait interval; according to a preset time period, user data of a user in an e-commerce platform is divided into different time intervals, historical purchase association conditions of commodities purchased by the user and long-period commodities in the different time intervals are determined, and composition conditions of the user data in the different time intervals are combined. And the management processing strategies of the user data in different time intervals are determined, so that the description processing efficiency of the user portrait is improved.

Description

Technical Field

[0001] The present invention belongs to the field of data management technology, and in particular relates to an e-commerce data management method and system based on big data analysis. Background Art

[0002] In order to segment e-commerce users and perform targeted product recommendations, the invention patent application CN202111655687.9, "E-commerce Cloud Data Analysis Method and System," generates a sentiment quantification matrix based on the original shopping data and original evaluation data in the historical shopping data of the analysis object. The sentiment quantification matrix is ​​then input into a data analysis model so that the data analysis model outputs a product that the analysis object is inclined to purchase. This improves the accuracy of the user's product recommendation process, but it suffers from the following technical drawbacks: In the process of analyzing and processing the purchase tendency of the analysis object, when the sales cycle of the product corresponding to the user portrait of the analysis object is long, the demand for the analysis accuracy of the user portrait of the analysis object is high. At the same time, the reference value of the user's historical purchase data will also change with the change of time. Therefore, how to manage the purchase data of the characterization processing of the user portrait of the analysis object according to the sales cycle of the product corresponding to the user portrait of the analysis object, and ensure the reliability of the characterization processing of the user portrait of some users on the basis of improving the efficiency of the update processing of the user portrait has become a technical problem that needs to be solved urgently.

[0003] In response to the above technical problems, the present invention provides an e-commerce data management method and system based on big data analysis. Summary of the Invention

[0004] To achieve the purpose of the present invention, the present invention adopts the following technical solutions: Specifically, in a first aspect, the present application provides an e-commerce data management method based on big data analysis, which specifically includes: S1 divides users of the e-commerce platform into different user profile intervals based on user profiles, selects products corresponding to the user profile intervals as matching products, and determines long-cycle products among the matching products based on sales cycle data of different matching products; S2 determines the demand coefficient of the portrait update processing accuracy rate in the user portrait interval based on the inventory data of the long-cycle product and the remaining available sales period. If it is within the preset range, proceed to the next step; S3 divides the user data on the e-commerce platform into different time intervals according to a preset time period when it is determined that the degree of association between the user and the long-cycle product meets the requirements based on the historical purchase data and the remaining available sales period of the user in the user profile interval for different long-cycle products; S4 determines the correlation between the user's purchased goods and the historical purchase of long-cycle goods in different time intervals, and determines the management and processing strategy of the user data in different time intervals based on the composition of the user data in different time intervals.

[0005] The beneficial effects of the present invention are: Based on the historical purchase data of users in the user portrait range for different long-cycle commodities and the remaining available sales period, it is determined whether the degree of association between the user and the long-cycle commodities meets the requirements. Not only the difference in the degree of association between the user portrait and different long-cycle commodities due to the difference in the historical purchase data of the long-cycle commodities is taken into account, but also the difference in the accuracy requirements of the user portrait for different long-cycle commodities due to the difference in the remaining available sales period is taken into account. The difference in the degree of association between the user portrait and the long-cycle commodity is realized, and a differentiated user portrait characterization and processing strategy is determined, thereby ensuring the accuracy of the characterization processing of user portraits with a higher degree of association.

[0006] Based on the correlation between users' purchased goods and historical purchases of long-cycle goods in different time intervals and the composition of user data, the management and processing strategies for user data in different time intervals are determined. The differences in the amount of user data in different time intervals and the degree of correlation with long-cycle goods are fully taken into consideration, and the time intervals with a high degree of correlation with long-cycle goods and a large amount of data are screened, thereby laying the foundation for the accuracy of constructing user portraits for long-cycle goods, and also ensuring the reliability of user recommendation processing for long-cycle goods.

[0007] A further technical solution is to divide the users of the e-commerce platform into different user profile intervals, including: Group users with the same user profile into the same user profile range.

[0008] A further technical solution is that the commodities corresponding to the user portrait interval are determined based on a preset correspondence between the user portraits corresponding to the user portrait interval and the commodities.

[0009] A further technical solution is that the method for determining the long-cycle products among the matching products is: Based on the sales cycle of the matching product, sales batches with purchase quantities within a preset quantity range are used as reference sales batches, and the sales duration of the matching product in different reference sales batches is determined; By matching the sales duration of the products in different reference sales batches, the reference sales batches whose sales duration exceeds the preset sales duration threshold are determined and regarded as the over-limit batches; According to the number of the over-limit batches, it is determined whether the matching product is a long-cycle product.

[0010] A further technical solution is that when the number of over-limit batches of the matching goods is greater than a preset over-limit batch threshold, the matching goods are determined to be long-cycle goods.

[0011] A further technical solution is that the value range of the preset time period is 1 month or half a month.

[0012] A further technical solution is that the historical purchase association between the purchased goods and the long-cycle goods is determined based on purchase data of the purchased goods and the long-cycle goods by the same user.

[0013] A further technical solution is that the method for determining the management and processing strategy of user data in the time interval is: Based on the historical purchase association between the user's purchased goods and the long-cycle goods within the time period, determine whether the purchased goods and the long-cycle goods were purchased simultaneously by different users, and identify users on the e-commerce platform who have purchased both the purchased goods and the long-cycle goods as associated users; Determine the associated purchased products of different long-cycle products based on the number of users associated with different purchased products and different long-cycle products, and determine the data correlation coefficient between the long-cycle products and the time period based on the proportion of the associated purchased products to the number of users purchased in the time period; Based on the average value of the data correlation coefficients between different long-cycle commodities and the time period, and the proportion of the amount of user data in the time period to the amount of user data of the user, the portrait correlation coefficient of the time period is determined, and the management and processing strategy of the user data in the time period is determined based on the portrait correlation coefficient.

[0014] A further technical solution is that the associated purchased goods are purchased goods whose number of associated users is greater than a preset threshold value of the number of associated users.

[0015] A further technical solution is that the portrait correlation coefficient is determined based on the sum of the average value of the data correlation coefficients between different long-cycle commodities and the time period and the proportion of the amount of user data in the time period to the amount of user data of the user.

[0016] A further technical solution is to determine a management and processing strategy for user data in the time interval according to the portrait correlation coefficient, specifically including: When the portrait correlation coefficient is greater than a preset correlation coefficient threshold, when updating the user portrait of the user, the user data of the user's time interval is combined to update the user portrait; When the portrait correlation coefficient is not greater than the preset correlation coefficient threshold, determine whether the portrait correlation coefficient is less than the preset value of the portrait correlation coefficient. If so, the user data of the time interval will be deleted in the e-commerce platform, and when updating the user portrait of the user, the user data of the time interval of the user does not need to be considered. If not, determine whether to consider the user data of the time interval of the user based on the update processing type of the user portrait.

[0017] A further technical solution is to determine whether to consider the user data of the user's time interval according to the update processing type of the user portrait, specifically including: When the update processing type of the user portrait is a type of update processing type, determining the user data of the time interval of the user; When the update processing type of the user portrait is the second update processing type, it is not necessary to consider the user data of the user's time interval.

[0018] A further technical solution is that the update processing period of the first update processing type is greater than that of the second update processing type.

[0019] In a second aspect, the present invention provides a computer system comprising: a memory and a processor in communication connection, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the above-mentioned e-commerce data management method based on big data analysis when running the computer program.

[0020] Other features and advantages will be described in the following description. The objectives and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description and drawings.

[0021] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The above and other features and advantages of the present invention will become more apparent by describing in detail example embodiments thereof with reference to the accompanying drawings; Figure 1 It is a flow chart of a thermoforming process control method; Figure 2 It is a flow chart of a method for determining long-cycle products among matching products; Figure 3 This is a flow chart of a method for determining a required coefficient of accuracy of a profile update process in a user profile interval; Figure 4 This is a flowchart of a method for determining a management processing policy for user data in a time interval. DETAILED DESCRIPTION

[0023] To help those skilled in the art better understand the technical solutions in this specification, the following will provide a clear and complete description of the technical solutions in the embodiments of this specification, in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of this specification, not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this specification without creative work should fall within the scope of protection of this specification.

[0024] In this application, the correlation between user portraits and long-cycle commodities is utilized to achieve differentiated management of users' historical purchase data, thereby reducing the data storage pressure of the e-commerce platform and improving the efficiency of user portrait update processing.

[0025] Example 1 like Figure 1 As shown, this application provides an e-commerce data management method based on big data analysis, which specifically includes: S1 divides users of the e-commerce platform into different user profile intervals based on user profiles, selects products corresponding to the user profile intervals as matching products, and determines long-cycle products among the matching products based on sales cycle data of different matching products; S2 determines the demand coefficient of the portrait update processing accuracy rate in the user portrait interval based on the inventory data of the long-cycle product and the remaining available sales period. If it is within the preset range, proceed to the next step; S3 divides the user data on the e-commerce platform into different time intervals according to a preset time period when it is determined that the degree of association between the user and the long-cycle product meets the requirements based on the historical purchase data and the remaining available sales period of the user in the user profile interval for different long-cycle products; S4 determines the correlation between the user's purchased goods and the historical purchase of long-cycle goods in different time intervals, and determines the management and processing strategy of the user data in different time intervals based on the composition of the user data in different time intervals.

[0026] Furthermore, users of the e-commerce platform are divided into different user profiles, including: Group users with the same user profile into the same user profile range.

[0027] Specifically, the commodities corresponding to the user portrait interval are determined according to the preset correspondence between the user portraits corresponding to the user portrait interval and the commodities.

[0028] Specifically, such as Figure 2As shown, the method for determining the long-cycle products among the matching products is: Based on the sales cycle of the matching product, sales batches with purchase quantities within a preset quantity range are used as reference sales batches, and the sales duration of the matching product in different reference sales batches is determined; By matching the sales duration of the products in different reference sales batches, the reference sales batches whose sales duration exceeds the preset sales duration threshold are determined and regarded as the over-limit batches; According to the number of the over-limit batches, it is determined whether the matching product is a long-cycle product.

[0029] Furthermore, when the number of over-limit batches of the matching product is greater than a preset over-limit batch threshold, the matching product is determined to be a long-cycle product.

[0030] In another possible embodiment, a method for determining long-cycle products among the matching products is: Taking sales batches with purchase quantities within a preset quantity range as reference sales batches, and determining the sales duration of the matching products in different reference sales batches based on the sales cycles of the matching products; By matching the sales duration of the products in different reference sales batches, the average sales duration of the different reference sales batches is determined; Whether the matching product is a long-cycle product is determined based on the average sales time of different reference sales batches.

[0031] Furthermore, when the average sales time of different reference sales batches is greater than a preset sales time threshold, the matching product is determined to be a long-cycle product.

[0032] Optionally, a method for determining the long-cycle products among the matching products is: S11 takes the sales batches with purchase quantities within a preset quantity range as reference sales batches, and determines the sales duration of the matching products in different reference sales batches based on the sales cycles of the matching products; It should be noted that before entering step S12, it is necessary to consider that there are reference sales batches whose sales time does not meet the requirements, then the matching product is determined to be a long-cycle product. When there are no reference sales times whose sales time does not meet the requirements, the average sales time of different reference sales batches within the preset time period is obtained. When the average sales time within the preset time period is greater than the preset time threshold, the matching product is determined to be a long-cycle product.

[0033] When the average sales time within the preset time period is not greater than the preset time threshold, when the average sales time within the preset time period is less than the preset value of the sales time, it is determined that the matching product does not belong to a long-cycle product; when the average sales time within the preset time period is not less than the preset value of the sales time, proceed to step S12.

[0034] S12 determines the abnormal coefficient of sales duration variation within a preset time period based on the variation of sales durations of different adjacent reference sales batches; Exemplarily, before proceeding to step S13, it is also necessary to check whether the duration variation abnormality coefficient of the sales duration within the preset time period meets the requirements. If not, it is determined that the matching product is a long-cycle product. If so, proceed to step S13.

[0035] S13 determines the average sales time of different reference sales batches based on the sales time of the matching products in different reference sales batches, determines the sales cycle abnormality factor of the matching products based on the average sales time of different reference sales batches and the changes in the sales time of different adjacent reference sales times, and combines the abnormal coefficient of the sales time change within a preset time period, and determines whether the matching products are long-cycle products based on the sales cycle abnormality factor.

[0036] Furthermore, when the sales cycle abnormality factor of the matching product is greater than a preset abnormality factor threshold, the matching product is determined to be a long-cycle product.

[0037] Specifically, the inventory data includes the inventory quantity of the long-cycle commodity.

[0038] It should be noted that the remaining available sales period is determined based on the remaining shelf life of the long-cycle product.

[0039] Specifically, such as Figure 3 As shown, the method for determining the required coefficient of the portrait update processing accuracy rate in the user portrait interval is: Determining the inventory quantity of the long-cycle commodity based on the inventory data of the long-cycle commodity; Determining the remaining shelf life of the long-life product based on the remaining available sales period; Long-cycle products whose remaining shelf life is less than the preset shelf life threshold or whose inventory is greater than the preset inventory threshold are selected as screening products. The demand coefficient of the portrait update processing accuracy in the user portrait interval is determined based on the proportion of the number of matching products of the screened products in the user portrait interval.

[0040] Furthermore, the value range of the demand coefficient of the portrait update processing accuracy of the user portrait interval is between 0 and 1, wherein the larger the demand coefficient of the portrait update processing accuracy of the user portrait interval is, the higher the demand degree of the portrait update processing accuracy of the user portrait interval is.

[0041] It can be understood that when the demand coefficient of the portrait update processing accuracy in the user portrait interval is not within the preset range, when updating the user portraits of users in the user portrait interval, the user portraits are updated based on the user's full historical purchase data.

[0042] In another possible embodiment, the method for determining the required coefficient of the portrait update processing accuracy rate in the user portrait interval is: Determining the inventory quantity of the long-cycle commodity based on the inventory data of the long-cycle commodity; Determining the remaining shelf life of the long-life product based on the remaining available sales period; The sales demand coefficients of different long-cycle commodities are determined based on the inventory and remaining shelf life of the long-cycle commodities, and the demand coefficient of the portrait update processing accuracy of the user portrait interval is determined based on the sales demand coefficients of the different long-cycle commodities.

[0043] In another possible embodiment, the method for determining the required coefficient of the portrait update processing accuracy rate in the user portrait interval is: S21 obtains the number of long-cycle products in the user profile interval; It should be noted that before proceeding to the next step, it is also necessary to determine that when the number of long-cycle commodities in the user portrait interval does not meet the requirements, it is determined that the demand coefficient of the portrait update processing accuracy of the user portrait interval is not within the preset range. When the number of long-cycle commodities in the user portrait interval meets the requirements, the basic demand coefficient of the portrait update processing accuracy of the user portrait curve is determined based on the number of long-cycle commodities in the user portrait interval and the proportion of the number of matching commodities in the user portrait interval.

[0044] It can be understood that when the basic demand coefficient is less than the basic coefficient preset value, the demand coefficient for determining the portrait update processing accuracy of the user portrait interval is within the preset range. When the basic demand coefficient is not less than the basic coefficient preset value, the inventory data of the long-cycle commodity is used as the basis to determine the inventory quantity of the long-cycle commodity. When the sum of the inventory quantities of different long-cycle commodities does not meet the requirements, it is determined that the demand coefficient for determining the portrait update processing accuracy of the user portrait interval is not within the preset range.

[0045] If and only if the sum of the inventory quantities of different long-cycle commodities meets the requirements, proceed to the next step.

[0046] S22: Based on the inventory data of the long-lead commodity, determine the inventory quantity of the long-lead commodity, determine the remaining shelf life of the long-lead commodity according to the remaining available sales period, and select the long-lead commodity whose remaining shelf life is less than a preset shelf life threshold or whose inventory quantity is greater than a preset inventory quantity threshold as the screening commodity; It should be further explained that before proceeding to the next step, it is necessary to further determine that when the proportion of the number of matching products of the filtered products in the user portrait interval does not meet the requirements, it is determined that the demand coefficient of the portrait update processing accuracy of the user portrait interval is not within the preset range. When the proportion of the number of matching products of the filtered products in the user portrait interval meets the requirements, the sales demand coefficients of different long-cycle products are determined based on the inventory and remaining shelf life of the long-cycle products.

[0047] When the sum of the inventory quantities of long-cycle commodities whose sales demand coefficient is greater than the preset sales demand coefficient threshold does not meet the requirements, it is determined that the demand coefficient of the portrait update processing accuracy of the user portrait interval is not within the preset interval. When the sum of the inventory quantities of long-cycle commodities whose sales demand coefficient is greater than the preset sales demand coefficient threshold meets the requirements, proceed to the next step.

[0048] S23 determines the demand coefficient of the portrait update processing accuracy rate in the user portrait interval according to the sales demand coefficients of different long-cycle commodities.

[0049] Furthermore, determining whether the degree of association between the user and the long-cycle product meets the requirements specifically includes: Determining the number of historical purchases of the user for the different long-cycle products based on the user's historical purchase data for the different long-cycle products; Determining the remaining shelf life of the long-life product based on the remaining available sales period; Long-cycle products with historical purchase data are used as matching purchase products, and whether the degree of association between the user and the long-cycle products meets the requirements is determined based on the remaining shelf life of different matching purchase products.

[0050] Specifically, when the number of matching purchased products whose remaining shelf life is less than a preset shelf life threshold is greater than a preset shelf life deviation product number threshold, it is determined that the degree of association between the user and the long-cycle product does not meet the requirements.

[0051] It should be noted that when the degree of association between the user and the long-cycle product does not meet the requirements, the user's portrait is updated based on the user's full historical purchase data.

[0052] In another possible embodiment, determining whether the degree of association between the user and the long-life commodity meets the requirements specifically includes: S31 determines the number of historical purchases of the user for different long-cycle commodities based on the user's historical purchase data for different long-cycle commodities, and determines a profile matching coefficient between the user and different long-cycle commodities based on the time interval between different historical purchase times and the current moment and the time interval between different historical purchase times; S32: determining the remaining shelf life of the long-lead commodity based on the remaining available sales period, and determining saleable abnormality coefficients of different long-lead commodities based on the remaining shelf life; S33 determines the purchase correlation coefficient between the user and the long-cycle commodity based on the sum of the product of the portrait matching coefficient between the user and different long-cycle commodities and the saleable abnormality coefficient, and determines whether the degree of correlation between the user and the long-cycle commodity meets the requirements based on the purchase correlation coefficient.

[0053] A further technical solution is that the purchase correlation coefficient between the user and the long-cycle commodity ranges from 0 to 1. When the purchase correlation coefficient between the user and the long-cycle commodity is greater than a preset correlation coefficient threshold, it is determined that the degree of correlation between the user and the long-cycle commodity does not meet the requirements.

[0054] Optionally, the above step S31 includes the following contents: S311 determines the number of historical purchases of different long-cycle commodities by the user based on their historical purchase data for different long-cycle commodities. If the number of long-cycle commodities for which purchase data exists exceeds a preset purchase commodity quantity threshold, it is determined that the degree of association between the user and the long-cycle commodities does not meet the requirement. If the number of long-cycle commodities for which purchase data exists does not exceed the preset purchase commodity quantity threshold, the process proceeds to step S312. At step S312, when the number of long-term commodities with purchase data is within a preset quantity range, it is determined that the degree of association between the user and the long-term commodities meets the requirement; when the number of long-term commodities with purchase data is not within the preset quantity range, the process proceeds to step S313; S313 determines the profile matching coefficients between the user and different long-cycle products based on the number of historical purchases of the user for different long-cycle products, the time intervals between different historical purchases and the current moment, and the time intervals between different historical purchases. If there is a long-cycle product with a profile matching coefficient greater than a preset profile matching coefficient threshold, the process proceeds to step S314. If there is no long-cycle product with a profile matching coefficient greater than the preset profile matching coefficient threshold, the process proceeds to step S315. At step S314, when the number of long-cycle commodities whose portrait matching coefficients are greater than the preset portrait matching coefficient threshold does not meet the requirement, it is determined that the degree of association between the user and the long-cycle commodity does not meet the requirement. When the number of long-cycle commodities whose portrait matching coefficients are greater than the preset portrait matching coefficient threshold meets the requirement, the process proceeds to step S315. S315 When the sum of the portrait matching coefficients with different long-cycle commodities is greater than the preset portrait matching coefficient threshold, it is determined that the degree of association between the user and the long-cycle commodities does not meet the requirements. When the sum of the portrait matching coefficients with different long-cycle commodities is not greater than the preset portrait matching coefficient threshold, proceed to step S32.

[0055] Optionally, the above step S32 includes the following contents: S321 identifies the long-cycle commodity with purchase data as a purchased long-cycle commodity. If there is no purchased long-cycle commodity with a remaining shelf life less than the preset shelf life threshold, the process proceeds to step S323. If there is a purchased long-cycle commodity with a remaining shelf life less than the preset shelf life threshold, the process proceeds to step S322. At step S322, if the number of long-life commodities purchased with a remaining shelf life less than the preset shelf life threshold does not meet the requirement, it is determined that the degree of association between the user and the long-life commodity does not meet the requirement. If the number of long-life commodities purchased with a remaining shelf life less than the preset shelf life threshold does meet the requirement, the process proceeds to step S323. S323 determines the remaining shelf life of the long-cycle commodity based on the remaining available sales period, determines the saleable anomaly coefficients of different long-cycle commodities based on the remaining shelf life, and determines the product matching coefficients of different long-cycle commodities by multiplying the portrait matching coefficients of different long-cycle commodities by the saleable anomaly coefficients. When the number of long-cycle commodities whose product matching coefficients are greater than the preset product matching coefficient threshold does not meet the requirements, it is determined that the degree of association between the user and the long-cycle commodity does not meet the requirements. When the number of long-cycle commodities whose product matching coefficients are greater than the preset product matching coefficient threshold meets the requirements, proceed to step S33.

[0056] Furthermore, the preset time period has a value range of 1 month or half a month.

[0057] Specifically, the historical purchase association between the purchased product and the long-cycle product is determined based on purchase data of the purchased product and the long-cycle product by the same user.

[0058] Specifically, such as Figure 4 As shown, the method for determining the management and processing strategy of user data in the time interval is: Based on the historical purchase association between the user's purchased goods and the long-cycle goods within the time period, determine whether the purchased goods and the long-cycle goods were purchased simultaneously by different users, and identify users on the e-commerce platform who have purchased both the purchased goods and the long-cycle goods as associated users; Determine the associated purchased products of different long-cycle products based on the number of users associated with different purchased products and different long-cycle products, and determine the data correlation coefficient between the long-cycle products and the time period based on the proportion of the associated purchased products to the number of users purchased in the time period; Based on the average value of the data correlation coefficients between different long-cycle commodities and the time period, and the proportion of the amount of user data in the time period to the amount of user data of the user, the portrait correlation coefficient of the time period is determined, and the management and processing strategy of the user data in the time period is determined based on the portrait correlation coefficient.

[0059] Furthermore, the associated purchased goods are purchased goods whose number of associated users is greater than a preset threshold of the number of associated users.

[0060] It can be understood that the portrait correlation coefficient is determined based on the sum of the average value of the data correlation coefficients between different long-cycle products and the time period and the proportion of the user data in the time period to the user data of the user.

[0061] It should be noted that the management and processing strategy for user data in the time interval is determined based on the portrait correlation coefficient, specifically including: When the portrait correlation coefficient is greater than a preset correlation coefficient threshold, when updating the user portrait of the user, the user data of the user's time interval is combined to update the user portrait; When the portrait correlation coefficient is not greater than the preset correlation coefficient threshold, determine whether the portrait correlation coefficient is less than the preset value of the portrait correlation coefficient. If so, the user data of the time interval will be deleted in the e-commerce platform, and when updating the user portrait of the user, the user data of the time interval of the user does not need to be considered. If not, determine whether to consider the user data of the time interval of the user based on the update processing type of the user portrait.

[0062] Furthermore, according to the update processing type of the user portrait, determining whether to consider the user data of the user's time interval specifically includes: When the update processing type of the user portrait is a type of update processing type, determining the user data of the time interval of the user; When the update processing type of the user portrait is the second update processing type, it is not necessary to consider the user data of the user's time interval.

[0063] It can be understood that the update processing period of the first update processing type is greater than that of the second update processing type.

[0064] Example 2 In a second aspect, the present invention provides a computer system comprising: a memory and a processor in communication connection, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the above-mentioned e-commerce data management method based on big data analysis when running the computer program.

[0065] The various embodiments in this specification are described in a progressive manner. Similar portions between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from the other embodiments. In particular, the device, apparatus, and non-volatile computer storage medium embodiments are generally similar to the method embodiments, so their descriptions are relatively simplified. For relevant details, refer to the descriptions of the method embodiments.

[0066] The foregoing description of this specification describes specific embodiments. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that described in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0067] The foregoing description is merely one or more embodiments of this specification and is not intended to limit this specification. It will be apparent to those skilled in the art that various modifications and variations may be made to one or more embodiments of this specification. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of one or more embodiments of this specification are intended to be within the scope of the claims of this specification.

Claims

1. An e-commerce data management method based on big data analysis, characterized in that: Specifically include: Divide the users of the e-commerce platform into different user profile intervals based on user profiles, use the products corresponding to the user profile intervals as matching products, and determine the long-cycle products among the matching products based on the sales cycle data of different matching products; Based on the inventory data and the remaining available sales period of long-cycle products, when the demand coefficient of the portrait update processing accuracy of the user portrait interval is determined to be within the preset range, proceed to the next step; Based on the historical purchase data and remaining available sales period of users in the user profile interval for different long-cycle products, when it is determined that the degree of association between the user and the long-cycle product meets the requirements, the user data on the e-commerce platform is divided into different time intervals according to the preset time period; Determine the correlation between the user's purchased goods and the historical purchase of long-cycle goods in different time intervals, and determine the management and processing strategy of user data in different time intervals based on the composition of user data in different time intervals.

2. The e-commerce data management method based on big data analysis according to claim 1, characterized in that: Divide the users of the e-commerce platform into different user profiles, including: Group users with the same user profile into the same user profile range.

3. The e-commerce data management method based on big data analysis according to claim 1, characterized in that: The commodities corresponding to the user portrait interval are determined based on the preset correspondence between the user portraits corresponding to the user portrait interval and the commodities.

4. The e-commerce data management method based on big data analysis according to claim 1, characterized in that: The method for determining the long-cycle products among the matching products is as follows: Based on the sales cycle of the matching product, sales batches with purchase quantities within a preset quantity range are used as reference sales batches, and the sales duration of the matching product in different reference sales batches is determined; By matching the sales duration of the products in different reference sales batches, the reference sales batches whose sales duration exceeds the preset sales duration threshold are determined and regarded as the over-limit batches; According to the number of the over-limit batches, it is determined whether the matching product is a long-cycle product.

5. The e-commerce data management method based on big data analysis according to claim 4, characterized in that: When the number of over-limit batches of the matching product is greater than a preset over-limit batch threshold, the matching product is determined to be a long-cycle product.

6. The e-commerce data management method based on big data analysis according to claim 1, characterized in that: The remaining available sales period is determined based on the remaining shelf life of the long-cycle product.

7. The e-commerce data management method based on big data analysis according to claim 1, characterized in that: The method for determining the required coefficient of the portrait update processing accuracy rate in the user portrait interval is: Determining the inventory quantity of the long-cycle commodity based on the inventory data of the long-cycle commodity; Determining the remaining shelf life of the long-life product based on the remaining available sales period; Long-cycle products whose remaining shelf life is less than the preset shelf life threshold or whose inventory is greater than the preset inventory threshold are selected as screening products. The demand coefficient of the portrait update processing accuracy in the user portrait interval is determined based on the proportion of the number of matching products of the screened products in the user portrait interval.

8. The e-commerce data management method based on big data analysis according to claim 1, characterized in that: The method for determining the management and processing strategy of user data in the time interval is: Based on the historical purchase association between the user's purchased goods and the long-cycle goods within the time period, determine whether the purchased goods and the long-cycle goods were purchased simultaneously by different users, and identify users on the e-commerce platform who have purchased both the purchased goods and the long-cycle goods as associated users; Determine the associated purchased products of different long-cycle products based on the number of users associated with different purchased products and different long-cycle products, and determine the data correlation coefficient between the long-cycle products and the time period based on the proportion of the associated purchased products to the number of users purchased in the time period; Based on the average value of the data correlation coefficients between different long-cycle commodities and the time period, and the proportion of the amount of user data in the time period to the amount of user data of the user, the portrait correlation coefficient of the time period is determined, and the management and processing strategy of the user data in the time period is determined based on the portrait correlation coefficient.

9. The e-commerce data management method based on big data analysis according to claim 8, characterized in that: The portrait correlation coefficient is determined based on the sum of the average value of the data correlation coefficients between different long-cycle commodities and the time period and the proportion of the amount of user data in the time period to the amount of user data of the user.

10. A computer system comprising: A memory and a processor in communication connection, and a computer program stored in the memory and capable of running on the processor, characterized in that when the processor runs the computer program, it executes an e-commerce data management method based on big data analysis as described in any one of claims 1-9.

Images