E-commerce platform promotion task allocation method, device, terminal and storage medium

Abstract

The present application discloses a method, device, terminal and storage medium for allocating promotion tasks on an e-commerce platform, wherein the method includes: obtaining an original promotion content data set and an original marketing account data set in the first promotion cycle, and judging the matching between each promotion content and each marketing account; initializing the expected promotion distribution income between each promotion content and each marketing account based on the matching; based on the expected promotion distribution income, solving and randomly rounding through a linear programming model to obtain an allocation scheme that meets preset constraints and has the maximum expected total income; entering the next promotion cycle and updating the original marketing account data set according to the allocation scheme, if the maximum promotion cycle has not been reached, obtaining the allocation scheme for the next promotion cycle based on the updated marketing account data set, otherwise terminating the task. The method of the present application realizes promotion task allocation through efficient matching between promotion content and marketing accounts, thereby improving promotion effect and resource utilization efficiency.

Description

Technical Field

[0001] The present application relates to the field of Internet e-commerce promotion, and in particular to a method, device, terminal and storage medium for allocating promotion tasks on an e-commerce platform. Background Art

[0002] In recent years, with the rapid development of internet technology and changing consumer shopping habits, e-commerce platforms have gradually become a vital channel for product sales. To enhance product promotion effectiveness, e-commerce platforms need to precisely match the characteristics of different marketing accounts with product features, thereby efficiently utilizing marketing resources. This process not only involves analyzing data from massive amounts of products and marketing accounts but also considers a variety of complex factors, such as promotion costs, effectiveness, and competitive relationships between products, to maximize returns within budget constraints.

[0003] In existing technologies, common approaches to solving task allocation problems on e-commerce platforms include manual judgment based on experience, simple matching based on historical data, and fixed-rule single-metric allocation strategies. Manual judgment relies primarily on the experience of operational personnel, manually selecting appropriate marketing accounts for promotion. Simple matching based on historical data statistically analyzes past promotion results and selects accounts with good performance for further promotion. Fixed-rule allocation strategies allocate tasks based on preset rules, such as the number of account followers or historical transaction volume.

[0004] However, these approaches often suffer from the following drawbacks: They fail to update marketing accounts' promotion capabilities in real time, leading to discrepancies between task allocation and actual promotion effectiveness; they fail to fully consider the trade-off between promotion costs and desired results, making it difficult to maximize revenue within budget constraints; and they ignore competition for promotional resources across products, potentially leading to poor promotional results or wasted resources for some products. Therefore, a task allocation technology for e-commerce platforms that can incorporate real-time, fine-grained traffic analysis is urgently needed to overcome the shortcomings of existing technologies. Summary of the Invention

[0005] In order to achieve efficient matching between multi-product marketing content and marketing accounts and realize task allocation with maximized benefits, this application provides an e-commerce platform promotion task allocation method, device, terminal and storage medium.

[0006] In the first aspect, this application provides a method for allocating promotion tasks on an e-commerce platform, which uses the following technical means:

[0007] A method for allocating promotion tasks on an e-commerce platform comprises the following steps:

[0008] During the first promotion cycle of a promotion assignment, the original promotion content dataset and the original marketing account dataset are obtained, and the matching status between each promotion content and each marketing account is determined.

[0009] Initialize the expected promotional distribution income between each promotional content and each marketing account in the current promotion period based on the matching situation;

[0010] During the current promotion cycle, based on the expected promotion distribution income, a linear programming model is solved and randomly rounded to obtain an allocation plan for the current promotion cycle that satisfies preset constraints and maximizes the total expected income;

[0011] Enter the next promotion cycle and update the original marketing account data set according to the allocation plan. If the maximum promotion cycle has not been reached, obtain the allocation plan for the next promotion cycle based on the updated marketing account data set; otherwise, terminate the promotion allocation task.

[0012] By adopting the above technical solution, each promotional content is matched with each marketing account, thereby initializing the expected promotion distribution benefits between the two during the current promotion cycle, thereby providing basic data support for subsequent distribution; the optimal distribution plan is obtained through the linear programming model, so that the final expected benefit meets the constraints while reaching the maximum, effectively combining historical data with real-time feedback, achieving efficient matching between promotional content and marketing accounts and dynamically updated promotion task allocation, thereby improving promotion effectiveness and resource utilization efficiency.

[0013] Preferably, the obtaining of the original promotion content dataset and the original marketing account dataset, and determining the matching between each piece of promotion content and each marketing account, specifically includes the following steps:

[0014] Obtain the original promotion content dataset and original marketing account dataset for the promotion assignment task, and generate promotion content based on the product categories and promotion content styles in the original promotion content dataset. Category label ,Right now, ;

[0015] in, For the promotional content Product categories, For the promotional content Promotional content style;

[0016] Combine the historical content style and historical promotion product category of each marketing account extracted from the original marketing account data set to determine the current promotion cycle Any of the marketing accounts mentioned above The promotional content described The matching situation;

[0017] Specifically, each marketing account is matched with the product category and the promotion content style of the current promotion content to obtain the following four matching results:

[0018] For the first matching result, the marketing account has never promoted the product with the current promotion content, and the style of the historical content of the marketing account is inconsistent with the style of the promotion content;

[0019] The second matching result is that the marketing account has promoted the product with the current promotion content, and the style of the historical content of the marketing account is consistent with the style of the promotion content;

[0020] The third matching result is that the marketing account has promoted the product with the current promotion content, but the style of the historical content of the marketing account is inconsistent with the style of the promotion content;

[0021] The fourth matching result is that the marketing account has never promoted the product with the current promotion content, but the historical content style of the marketing account is consistent with the promotion content style.

[0022] By adopting the above technical solution, category labels for promotional content are generated based on the promotional content dataset and the marketing account dataset. Combined with the historical content style and historical promotional product categories of the marketing account, the matching situation between each marketing account and the current promotional content is accurately judged. Through the matching analysis of product categories and promotional content styles, the adaptability of marketing accounts to current promotional content is divided into four matching results, thereby achieving comprehensive coverage of matching situations and improving the accuracy of task allocation.

[0023] Preferably, the initialization of the expected promotion distribution income between each promotion content and each marketing account in the current promotion period according to the matching situation specifically includes the following steps:

[0024] Initialize the first The promotional content and The matching degree of the marketing account and the promotion content In the marketing account historical returns;

[0025] The promotion content is matched according to the matching degree and the historical revenue. Assigned to the marketing account The expected promotion distribution benefit is initialized, that is, ;

[0026] in, For the current promotion period The expected promotion distribution income, For matching, For historical returns.

[0027] By adopting the above technical solution, by initializing the matching degree between promotion content and marketing accounts and historical revenue, we can ultimately achieve accurate initialization of the expected promotion allocation revenue, laying the foundation for the subsequent formulation of an efficient promotion allocation strategy, helping to maximize revenue under budget constraints, and improving the promotion efficiency of the e-commerce platform, thereby improving the overall promotion effect and resource utilization efficiency.

[0028] The original marketing account data set also includes transaction revenue of each category of goods;

[0029] For the first matching result, the matching degree and the historical benefit are:

[0030] ;

[0031] ;

[0032] For the second matching result, the matching degree and the historical benefit are:

[0033] ;

[0034] ;

[0035] For the third matching result, the matching degree and the historical benefits are:

[0036] ;

[0037] ;

[0038] For the fourth matching result, the matching degree and the historical benefits are:

[0039] ;

[0040] ;

[0041] in, For the marketing accounts within the promotion period The historical content style, For the marketing accounts within the promotion period Historical promotion product categories, For category The transaction revenue of the goods.

[0042] By adopting the above technical solution, different matching degrees and historical benefits are set for each matching result, making task allocation more scientific and reasonable, better reflecting the actual promotion effect, avoiding resource waste caused by blind allocation, and providing a reliable basis for initializing the subsequent expected promotion allocation benefits by quantifying the matching degree and historical benefits, which helps to maximize benefits under budget constraints and improve the promotion efficiency of e-commerce platforms.

[0043] Preferably, the expected promotion income distribution is solved by a linear programming model and randomly rounded to obtain an allocation plan for the current promotion cycle that satisfies preset constraints and maximizes the total expected income, specifically comprising the following steps:

[0044] Defining slack variables ,in , obtaining a constraint formula based on the slack variables and the preset constraint conditions;

[0045] A linear programming model is established based on the slack variables and the expected promotion distribution benefits. The linear programming model is:

[0046] ,

[0047] in, is the expected total return;

[0048] Solve the linear programming model according to the constraint formula to obtain the slack variable that maximizes the total expected benefit ;

[0049] The slack variables obtained by the solution are randomly rounded to obtain a rounding result, and an allocation plan for the current promotion cycle is generated according to the rounding result.

[0050] By adopting the above technical solution, a linear programming model is established based on the expected promotion allocation benefits and constraints. The slack variables that maximize the expected total benefits are obtained by solving the model. The slack variables are then randomly rounded to generate the optimal allocation plan. Random rounding effectively balances the gap between the theoretical optimal solution and practical operability, so that the final generated allocation plan can maximize the promotion benefits within the constraints, thereby improving the task allocation efficiency and promotion effect of the e-commerce platform.

[0051] Preferably, the constraints include:

[0052] A first constraint condition is that the total cost of the allocation plan does not exceed a preset period budget, and the total cost includes the cost of allocating each piece of promotional content to the corresponding marketing account;

[0053] The second constraint is that each piece of promotional content must be promoted at least once within each promotion cycle;

[0054] The third constraint condition is that the total number of promotional contents allocated to each marketing account does not exceed the carrying limit of the corresponding marketing account.

[0055] By adopting the above technical solution, detailed constraints were set, and refined management of the distribution of promotion tasks on the e-commerce platform was achieved. The first constraint ensured the financial controllability of the promotion activities and avoided the risk of overspending. The second constraint prevented the situation where some promotional content was ignored. The third constraint prevented the decline in effect caused by excessive promotion, while protecting the long-term promotion capabilities of the marketing account.

[0056] Preferably, the step of entering the next promotion cycle and updating the original marketing account dataset according to the allocation plan, and obtaining the allocation plan for the next promotion cycle based on the updated marketing account dataset if the maximum promotion cycle has not been reached, and otherwise terminating the promotion allocation task, specifically includes the following steps:

[0057] Incrementing the current promotion cycle by one, so that the promotion assignment task enters the next promotion cycle, and determining whether the current promotion cycle reaches a preset maximum promotion cycle;

[0058] If the maximum promotion period is reached, the promotion allocation task is terminated and the allocation plan for each promotion period is output;

[0059] If the maximum promotion cycle has not been reached, the promotion allocation task is executed according to the allocation plan and the promotion results of the previous promotion cycle are obtained. Based on the promotion results, the updated historical content style, historical promotion product categories and transaction revenue of each category of products are obtained, and the matching status of the current promotion cycle is obtained. Based on the matching status, each promotion content of the current promotion cycle and the expected promotion allocation revenue of each marketing account are initialized to obtain the allocation plan until the maximum promotion cycle is reached.

[0060] By adopting the above technical solution, when the maximum promotion cycle is not reached, the updated marketing account data set is used to reinitialize the expected promotion allocation income of each promotion content and each marketing account, which can timely reflect the changes in the actual promotion effect of the marketing account, thereby improving the accuracy of the allocation plan; a new allocation plan is generated based on the updated expected promotion allocation income, ensuring that the task allocation in each promotion cycle is optimized based on the latest data, maximizing the promotion income, and terminating the promotion task when the preset maximum promotion cycle is reached, thereby improving the controllability of the system and ultimately realizing the dynamic update and optimization of the promotion task allocation of the e-commerce platform.

[0061] In a second aspect, the present application provides an e-commerce platform promotion task allocation device, which adopts the following technical means:

[0062] An e-commerce platform promotion task allocation device includes the following modules:

[0063] a content matching module configured to obtain an original promotion content dataset and an original marketing account dataset during the first promotion cycle of a promotion assignment task, and determine a match between each piece of promotion content and each marketing account;

[0064] An expected revenue initialization module is configured to initialize the expected promotion distribution revenue between each promotional content and each marketing account in the current promotion cycle based on the matching situation;

[0065] an allocation plan generating module configured to, within the current promotion cycle, solve the expected promotion distribution income through a linear programming model and perform random rounding to obtain an allocation plan for the current promotion cycle that satisfies preset constraints and maximizes the expected total income;

[0066] The promotion update module is configured to enter the next promotion cycle and update the original marketing account data set according to the allocation plan. If the maximum promotion cycle has not been reached, the allocation plan for the next promotion cycle is obtained based on the updated marketing account data set; otherwise, the promotion allocation task is terminated.

[0067] By adopting the above technical solutions, a complete promotion task allocation system has been built, which provides the necessary software and technical support for the efficient promotion of e-commerce platform content, significantly improves the efficiency and intelligence of the matching between promotion content and marketing accounts, achieves the optimal allocation of resources, and meets the requirements of technological progress.

[0068] In a third aspect, the present application provides a smart terminal that adopts the following technical solution:

[0069] A smart terminal includes a memory and a processor, wherein the memory stores at least one instruction, at least one program, code set or instruction set, and the at least one instruction, at least one program, code set or instruction set is loaded and executed by the processor to implement the e-commerce platform promotion task allocation method as described above.

[0070] In a fourth aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution:

[0071] A computer-readable storage medium stores at least one instruction, at least one program, code set or instruction set, and the at least one instruction, at least one program, code set or instruction set is loaded and executed by a processor to implement the e-commerce platform promotion task allocation method as described above.

[0072] In summary, this application includes at least one of the following beneficial technical effects:

[0073] (1) This application can accurately reflect the actual promotion capabilities of marketing accounts by initializing and periodically updating each promotion content and the expected promotion distribution income of each marketing account, thereby improving the accuracy and real-time performance of task allocation.

[0074] (2) This application solves the linear programming model and generates an allocation plan by random rounding. While satisfying the constraints, it maximizes the total expected benefit and achieves the maximum benefit under the budget constraint.

[0075] (3) This application takes into account the competitive relationship between promotion resources among commodities. By dynamically adjusting the allocation strategy, it avoids the problems of resource waste and poor promotion effect, and improves the overall promotion efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 This is a flow chart of the method for allocating promotion tasks on an e-commerce platform according to an embodiment of the present application;

[0077] Figure 2 This is a flowchart of steps S1 and S2 of the method for allocating promotion tasks on an e-commerce platform according to an embodiment of the present application;

[0078] Figure 3 This is a flowchart of step S3 of the method for allocating promotion tasks on an e-commerce platform according to an embodiment of the present application;

[0079] Figure 4 This is a flowchart of step S4 of the method for allocating promotion tasks on an e-commerce platform according to an embodiment of the present application;

[0080] Figure 5 It is a structural diagram of the e-commerce platform promotion task allocation device in an embodiment of the present application. DETAILED DESCRIPTION

[0081] The following will be combined with the accompanying drawings to clearly and completely describe the technical solutions in the embodiments of the present invention. The described embodiments are only possible technical implementations of the present invention and are not all possible implementations. Those skilled in the art can fully combine the embodiments of the present invention to derive other embodiments without creative work, and these embodiments are also within the scope of protection of the present invention.

[0082] The inventors of this application discovered that the task allocation of an e-commerce platform requires selecting the most appropriate marketing account for the promotional content of each product on the basis of a given budget in order to maximize the expected promotional effect of the product. However, the matching between different marketing accounts and promotional content is complex and changeable, and is also affected by the cost, expected effect and the number of promotions. There is an urgent need for an efficient task allocation technology based on real-time fine-grained traffic analysis. To this end, this application mainly adopts the following method, which initializes and periodically updates the degree of adaptation of the promotional content and the marketing account, the promotion cost, and the historical promotion data, and combines the linear programming model to generate the optimal allocation plan, thereby achieving the effect of maximizing revenue under budget constraints.

[0083] The e-commerce platform promotion task allocation method provided in the embodiment of the present application includes obtaining and matching promotion content datasets and marketing account datasets, initializing the expected promotion distribution revenue, solving the linear programming model and randomly rounding to obtain an allocation plan that meets the preset constraints, and updating the marketing account dataset based on the execution results until the maximum promotion cycle is reached. The entire process can maximize the revenue under budget constraints and improve the promotion efficiency. The following is a further detailed description of this application.

[0084] The e-commerce platform promotion task allocation method provided in the embodiment of the present application is as follows: Figure 1 As shown, the following steps are included:

[0085] S1. When in the first promotion cycle of the promotion assignment task, obtain the original promotion content dataset and the original marketing account dataset, and determine the matching between each promotion content and each marketing account, such as Figure 2 As shown, the specific steps include:

[0086] S11. Get the current promotion period , to determine whether the current promotion is in the first promotion cycle of the promotion assignment task, that is, to determine whether the current promotion is in the first promotion cycle of the promotion assignment task Is it 1? , the current promotion cycle is the first promotion cycle of the promotion assignment task.

[0087] If the current promotion period is not the first promotion period, step S42 is executed.

[0088] S12. When in the first promotion cycle of the promotion assignment task, obtain an original promotion content dataset.

[0089] In this embodiment, the original promotion content dataset is defined The original promotion content data set includes product name, promotion content serial number , product category, promotional content copy, promotional content style, promotional content video.

[0090] Among them, the product category is a positive integer of no more than 9 digits, and each fixed digit represents a first-level product classification; the promotion content style is generated according to the style library and is a positive integer of no more than 4 digits.

[0091] The categories of promotional content are composed of product categories and promotional content styles.

[0092] For example, in one specific feasible method, a tweet promoting infant bath products could be: "Giving our little human babies a bath is so healing!!" The product category is "mother and baby products," and the promotional content style is "daily life."

[0093] S13. Generate promotional content based on the product categories and promotional content styles in the original promotional content dataset The category label of .

[0094] In this embodiment, Extract product categories separately and promotional content style , and generate promotional content Category tags, promotional content includes tweets and video ads.

[0095] Specifically, promotional content Category label , taking the category label as the first Category code of the promotional content.

[0096] S14. Obtain the original marketing account data set and define the original marketing account data set The original marketing account dataset includes the account ID, the account's historical content style, the account's historical promotion traffic conversion rate for different categories of products, the account's historical promotion product categories, the transaction revenue of different categories of products, and the promotion content traffic of different categories of products.

[0097] Specifically, the operation of obtaining the original promotion content dataset and the original marketing account dataset can include various methods, such as automatically extracting relevant data from the e-commerce platform database through a data crawler tool or API interface.

[0098] S15, combine the historical content style and historical promotion product categories of each marketing account extracted from the original marketing account data set to determine the current promotion cycle Any marketing account With current promotional content The matching situation.

[0099] Specifically, each marketing account is matched with the product category and promotional content style of the current promotional content.

[0100] In this embodiment, The first The historical content style of the marketing account is , the historical promotion product categories are , the transaction revenue of different categories of goods is , among which Transaction revenue of similar products .

[0101] According to Category tags for promoted content , No. Cycle marketing accounts Historical content style and Cycle marketing accounts Historical promotional product categories , to judge the matching situation, and get the following four matching results,

[0102] S151. First matching result: the marketing account has never promoted the product with the current promotion content, and the historical content style of the marketing account is inconsistent with the promotion content style.

[0103] S152. Second matching result: the marketing account has promoted the product of the current promotion content, and the historical content style of the marketing account is consistent with the promotion content style.

[0104] S153. The third matching result: the marketing account has promoted the product of the current promotion content, but the historical content style of the marketing account is inconsistent with the promotion content style.

[0105] S154. The fourth matching result is that the marketing account has never promoted the product of the current promotion content, but the historical content style of the marketing account is consistent with the promotion content style.

[0106] Therefore, the above marketing accounts With current promotional content The four matching results can be summarized into the following four cases in Table 1:

[0107] Table 1 Marketing Accounts With current promotional content Matching result table

[0108] ;

[0109] Among them, the first matching result corresponds to case 1, the second matching result corresponds to case 2, the third matching result corresponds to case 3, and the fourth matching result corresponds to case 4.

[0110] S2. Initialize the expected promotion distribution income of each promotion content and each marketing account in the current promotion cycle based on the matching situation, such as Figure 2 As shown, the specific steps include:

[0111] S21, initialize the first Promotional content and The matching degree of each marketing account and the promotional content In marketing account historical returns.

[0112] In this embodiment, the original marketing account data set also includes transaction revenue of each category of goods.

[0113] S211: For the first matching result, the marketing account Never promoted content Products whose style is consistent with the promotional content Inconsistency indicates a marketing account For promotional content Completely unfamiliar, that is , due to the category label ,therefore Can be used to indicate promotional content Product categories, Can be used to indicate promotional content Promotional content style.

[0114] To make marketing accounts Try to promote the content without paying too high a trial and error cost, and set the matching degree to a smaller value initially. Therefore, set the matching degree to 0.01, that is, , For matching degree.

[0115] At this time, the marketing account No promotional content Historical data on revenue, so the ability to promote other products to promote content We reason about the possible benefits and believe that the possible benefits are the sum of the benefits of all types of goods, that is, , For historical returns.

[0116] The basis for taking the sum of all types of commodity revenue as the possible revenue of the account is that the sum of all types of commodity revenue of this account can reflect the overall promotion and sales ability of this account. In order to balance the cost of trial and error, a smaller weight (matching degree) is assigned to it. If it still has a large estimated revenue, it means that it can be tried for promotion.

[0117] Therefore, the matching degree and historical returns are,

[0118] ;

[0119] .

[0120] S212: For the second matching result, the marketing account Previously promoted content Products, and the account style is also consistent with the content Consistent, that is , so the matching degree is set to 1.

[0121] At this time, due to the marketing account Promotional content exists If the historical data of income is ,have .

[0122] Therefore, the matching degree and historical returns are,

[0123] ;

[0124] .

[0125] S213: For the third matching result, it is known that the marketing account Previously promoted content Products, but the account style and content Inconsistency, that is , so the matching degree is set to 0.5.

[0126] At this time, due to the marketing account Promotional content exists If the historical data of income is ,have .

[0127] Therefore, the matching degree and historical returns are,

[0128] ;

[0129] .

[0130] S214: For the fourth matching result, it is known that the marketing account No content has been promoted Products, but the account style and content Consistent, that is , so the matching degree is set to 0.5.

[0131] At this time, the marketing account No promotional content Historical data on revenue, the ability to promote other products through the promotion of content We reason about the possible benefits and believe that the possible benefits are the sum of the benefits of all types of goods, that is, .

[0132] Therefore, the matching degree and historical returns are,

[0133] ;

[0134] .

[0135] in, For the Marketing accounts within the promotion period The historical content style, For the Marketing accounts within the promotion period Historical promotion product categories, For category The transaction revenue of the goods.

[0136] Therefore, the initialization process of the above matching degree and historical returns can be expressed as:

[0137] First judge Whether it belongs to ,if, ,make ,and ; Then judge Whether it belongs to ,if, , then let Otherwise, let ;

[0138] like ,make , then judge Whether it belongs to ,if, , then let , otherwise let .

[0139] This results in an initialization scheme for matching degree and historical returns.

[0140] S22, in Promotion cycle will promote content Assigned to marketing account The expected return is determined by the matching degree between the two and historical returns Jointly decide on the content to be promoted based on matching and historical revenue Initialize with the expected promotion distribution income allocated to the marketing account, that is,

[0141] ;

[0142] in, For the current promotion period The expected promotion distribution benefits.

[0143] The above steps initialize the expected promotion distribution income for each promotional content and each marketing account, which can accurately evaluate the adaptability of different marketing accounts to different promotional content and provide a scientific basis for subsequent allocation.

[0144] S3. In the current promotion cycle, based on the expected promotion distribution income, the linear programming model is solved and randomly rounded to obtain the current promotion cycle distribution plan that meets the preset constraints and has the largest expected total income, such as Figure 3 As shown, the specific steps include:

[0145] S31. Define slack variables ,in , the constraint formula is obtained based on the slack variables and the preset constraints.

[0146] Constraints include:

[0147] S311. First constraint: the total cost of the allocation plan does not exceed the preset period budget. The total cost includes the cost of allocating each piece of promotional content to the corresponding marketing account.

[0148] According to the first constraint condition, the first constraint formula is obtained, that is,

[0149] ;

[0150] in, To promote content Assigned to marketing account the cost, For cycle budget, and Both can be adjusted according to the needs of the promotion task.

[0151] S312. Second constraint: Each promotional content must be promoted at least once in each promotion cycle.

[0152] According to the second constraint condition, the second constraint formula is obtained, that is,

[0153] .

[0154] S313. The third constraint condition is that the total amount of promotional content allocated to each marketing account does not exceed the carrying limit of the corresponding marketing account.

[0155] According to the third constraint condition, the third constraint formula is obtained, that is,

[0156] ;

[0157] in, For marketing accounts The upper limit of the load.

[0158] S32. A linear programming model is established based on the slack variables and the expected promotion distribution benefits. The linear programming model is:

[0159] ,

[0160] in, is the expected total profit.

[0161] S33. Solve the linear programming model based on the first constraint formula, the second constraint formula, and the third constraint formula to obtain the slack variable that maximizes the total expected return. .

[0162] S34. Randomly round the solved slack variables to obtain a rounding result, and generate an allocation plan for the current promotion cycle based on the rounding result.

[0163] In this embodiment, the slack variables are randomly rounded to obtain , Indicates that during the promotion cycle Will the content be promoted? Assigned to marketing account , use 0 to indicate no allocation and 1 to indicate allocation;

[0164] By probability Round to 1 to get , output The plan with a value of 1 is the allocation plan for the promotion cycle.

[0165] In the above steps, defining slack variables and establishing a linear programming model in combination with constraint formulas helps to accurately quantify the feasibility and profit potential of the promotion task. Solving and randomly rounding based on the linear programming model ensures that the generated allocation plan meets the preset constraints while maximizing the total expected profit, thereby achieving optimal resource allocation.

[0166] S4. Enter the next promotion cycle and update the original marketing account dataset according to the allocation plan. If the maximum promotion cycle has not been reached, the allocation plan for the next promotion cycle is obtained based on the updated marketing account dataset. Otherwise, the promotion allocation task is terminated. Figure 4 As shown, the specific steps include:

[0167] S41, increment the current promotion cycle by one, so that the promotion allocation task enters the next promotion cycle, that is, set t=t+1;

[0168] S42: Determine whether the current promotion period reaches the preset maximum promotion period T, 1≤t≤T.

[0169] S43. If the maximum promotion cycle is reached, that is, t>T, the promotion allocation task is terminated and the allocation plan for each promotion cycle is output.

[0170] S44. If the maximum promotion period has not been reached, the promotion allocation task is executed according to the allocation plan and the promotion result of the previous promotion period is obtained, that is, the promotion result of the promotion period of step S3.

[0171] S45. Based on the promotion results, updated historical content styles, historical promotion product categories, and transaction revenue of each category of products are obtained.

[0172] S46. Based on the content of the updated marketing account, the matching status of the current promotion cycle is obtained. The steps are the same as those in S14 above and will not be repeated here.

[0173] Initialize the expected promotional income distribution of each promotional content and each marketing account in the current promotion cycle based on the matching situation.

[0174] In this embodiment, after obtaining the updated matching situation, the matching degree and historical revenue in step S21 are updated to obtain the matching degree and historical revenue initialized for the current promotion period.

[0175] Then recalculate the expected promotion distribution income for the current promotion cycle based on the updated matching degree and historical income.

[0176] S47. Obtain a distribution plan for the current promotion cycle based on the updated expected promotion distribution income.

[0177] The promotion allocation task will be terminated until the maximum promotion cycle is reached.

[0178] The above steps continuously optimize the allocation plan before the maximum promotion cycle is reached. Tasks are executed and the marketing account dataset is updated based on the allocation plan of the previous promotion cycle. This dynamically reflects the promotion effect, supports the accuracy of the next round of allocation decisions, ensures that the entire promotion process is always in the best state, and ultimately achieves maximum benefits within the budget constraint.

[0179] Based on the same inventive concept above, the embodiment of the present application also discloses an e-commerce platform promotion task allocation device, such as Figure 5 As shown, it includes the following modules:

[0180] a content matching module configured to obtain an original promotion content dataset and an original marketing account dataset during the first promotion cycle of a promotion assignment task, and determine a match between each piece of promotion content and each marketing account;

[0181] An expected revenue initialization module is configured to initialize the expected promotional revenue distribution between each promotional content and each marketing account within the current promotion cycle based on the matching situation;

[0182] The allocation plan generation module is configured to obtain an allocation plan for the current promotion period that satisfies preset constraints and maximizes the total expected revenue based on the expected promotion revenue during the current promotion period by solving a linear programming model and performing random rounding.

[0183] The promotion update module is configured to enter the next promotion cycle and update the original marketing account dataset according to the allocation plan. If the maximum promotion cycle has not been reached, the allocation plan for the next promotion cycle is obtained based on the updated marketing account dataset. Otherwise, the promotion allocation task is terminated.

[0184] In a specific implementation scheme, the content matching module includes the following units:

[0185] The category determination unit is configured to obtain the original promotion content dataset and the original marketing account dataset of the promotion assignment task, and generate promotion content based on the product category and promotion content style in the original promotion content dataset. Category label ,Right now,

[0186] ,

[0187] in, To promote content Product categories, To promote content Promotional content style;

[0188] The content matching unit is configured to combine the historical content style and historical promotion product categories of each marketing account extracted from the original marketing account data set to determine the content style of the marketing account in the current promotion cycle. Any marketing account With current promotional content The matching situation;

[0189] Specifically, each marketing account is matched with the product category and promotion content style of the current promotion content, and the following four matching results are obtained:

[0190] The first matching result is that the marketing account has never promoted the product with the current promotion content, and the style of the marketing account's historical content is inconsistent with the promotion content style;

[0191] The second matching result is that the marketing account has promoted the product of the current promotion content, and the style of the marketing account's historical content is consistent with the promotion content style;

[0192] The third matching result is that the marketing account has promoted the product of the current promotion content, but the style of the marketing account's historical content is inconsistent with the promotion content style;

[0193] The fourth matching result is that the marketing account has never promoted the product of the current promotion content, but the historical content style of the marketing account is consistent with the promotion content style.

[0194] In a specific implementation scheme, the expected return initialization module includes the following units:

[0195] The matching degree and historical income initialization unit is configured to initialize the first Promotional content and The matching degree of each marketing account and the promotional content In marketing account historical returns;

[0196] The expected revenue initialization unit is configured to promote content based on matching degree and historical revenue. Assigned to marketing accounts The expected generalization distribution benefit is initialized, that is,

[0197] ;

[0198] in, For the current promotion period The expected promotion distribution income, For matching, For historical returns.

[0199] In a specific implementation scheme, the matching degree and historical income initialization unit includes the following subunits:

[0200] The first initialization subunit is configured as the original marketing account data set including the transaction revenue of each category of goods.

[0201] For the first matching result, the matching degree and the historical benefit are:

[0202] ;

[0203] ;

[0204] The second initialization subunit is configured to, for the second matching result, the matching degree and the historical benefit are,

[0205] ;

[0206] ;

[0207] The third initialization subunit is configured to, for the third matching result, the matching degree and the historical benefit are,

[0208] ;

[0209] ;

[0210] The fourth initialization subunit is configured to: for the fourth matching result, the matching degree and the historical benefit are:

[0211] ;

[0212] ;

[0213] in, For the Marketing accounts within the promotion period The historical content style, For the Marketing accounts within the promotion period Historical promotion product categories, For category The transaction revenue of the goods.

[0214] In a specific implementation scheme, the allocation scheme generation module includes the following units:

[0215] Conditional constraint units, configured to define slack variables ,in , obtain the constraint formula based on the slack variables and the preset constraint conditions;

[0216] The linear solver is configured to establish a linear programming model based on the slack variables and the expected generalization distribution benefits. The linear programming model is ,in, is the total expected benefit. According to the constraint formula, the linear programming model is solved to obtain the slack variable that maximizes the total expected benefit. ;

[0217] The random rounding unit is configured to randomly round the solved slack variables to obtain a rounding result, and generate an allocation plan for the current promotion cycle according to the rounding result.

[0218] In a specific embodiment, the conditional constraint unit includes the following subunits:

[0219] The conditional constraint subunit is configured as a constraint including:

[0220] The first constraint is that the total cost of the allocation plan must not exceed the preset period budget. The total cost includes the cost of assigning each piece of promotional content to the corresponding marketing account.

[0221] The second constraint is that each piece of promotional content must be promoted at least once during each promotion cycle;

[0222] The third constraint is that the total amount of promotional content allocated to each marketing account does not exceed the carrying limit of the corresponding marketing account.

[0223] In a specific implementation scheme, the promotion and update module includes the following units:

[0224] The maximum period determination unit is configured to increment the current promotion period by one, so that the promotion assignment task enters the next promotion period, and determine whether the current promotion period reaches a preset maximum promotion period;

[0225] If the maximum promotion cycle is reached, the promotion allocation task is terminated and the allocation plan for each promotion cycle is output;

[0226] The periodic update unit is configured to execute the promotion allocation task according to the allocation plan and obtain the promotion results of the previous promotion cycle if the maximum promotion cycle has not been reached, obtain the updated historical content style, historical promotion product category and transaction revenue of each category of products based on the promotion results, and obtain the matching status of the current promotion cycle, initialize each promotion content of the current promotion cycle and the expected promotion allocation revenue of each marketing account based on the matching status, and obtain the allocation plan until the maximum promotion cycle is reached.

[0227] Based on the same inventive concept mentioned above, an embodiment of the present application also discloses a computer-readable storage medium, which stores at least one instruction, at least one program, code set or instruction set. The at least one instruction, at least one program, code set or instruction set can be loaded and executed by a processor to implement the e-commerce platform promotion task allocation method provided in the above method embodiment.

[0228] Also based on the same inventive concept mentioned above, an embodiment of the present application also discloses a computer-readable storage medium, in which at least one instruction, at least one program, code set or instruction set is stored. The at least one instruction, at least one program, code set or instruction set is loaded and executed by a processor to implement the e-commerce platform promotion task allocation method as described above.

[0229] Those skilled in the art will appreciate that all or part of the steps of implementing the above embodiments may be accomplished by hardware, or by a program instructing the relevant hardware to accomplish the steps. The program may be stored in the computer-readable storage medium, and the computer-readable storage medium may include, for example, various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0230] The above description is merely an optional embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A method for allocating promotion tasks on an e-commerce platform, characterized in that: The steps include: During the first promotion cycle of a promotion assignment, the original promotion content dataset and the original marketing account dataset are obtained, and the matching status between each promotion content and each marketing account is determined. Initialize the expected promotional revenue distribution between each promotional content and each marketing account within the current promotion period based on the matching situation; In the current promotion cycle, slack variables are defined, and a constraint formula is obtained based on the slack variables and preset constraint conditions; a linear programming model is established based on the slack variables and the expected promotion distribution benefits, and the linear programming model is solved based on the constraint formula to obtain the slack variables that maximize the expected total benefits; the solved slack variables are randomly rounded to obtain a rounding result, and an allocation plan for the current promotion cycle is generated based on the rounding result; The constraints include: a first constraint that the total cost of the allocation plan does not exceed a preset cycle budget; a second constraint that each piece of promotional content is promoted at least once during each promotion cycle; and a third constraint that the total number of promotional content allocated to each marketing account does not exceed the corresponding marketing account's carrying limit. Enter the next promotion cycle and update the original marketing account data set according to the allocation plan. If the maximum promotion cycle has not been reached, obtain the allocation plan for the next promotion cycle based on the updated marketing account data set; otherwise, terminate the promotion allocation task.

2. The e-commerce platform promotion task allocation method according to claim 1, characterized in that: The steps of obtaining the original promotion content dataset and the original marketing account dataset, and determining the matching between each promotion content and each marketing account, specifically include the following steps: Obtain the original promotion content dataset and original marketing account dataset for the promotion assignment task, and generate promotion content based on the product categories and promotion content styles in the original promotion content dataset. Category label ,Right now, ； in, For the promotional content Product categories, For the promotional content Promotional content style; Combine the historical content style and historical promotion product category of each marketing account extracted from the original marketing account data set to determine the current promotion cycle Any of the marketing accounts mentioned above The promotional content described The matching situation; Specifically, each marketing account is matched with the product category and the promotion content style of the current promotion content to obtain the following four matching results: For the first matching result, the marketing account has never promoted the product with the current promotion content, and the style of the historical content of the marketing account is inconsistent with the style of the promotion content; The second matching result is that the marketing account has promoted the product with the current promotion content, and the style of the historical content of the marketing account is consistent with the style of the promotion content; The third matching result is that the marketing account has promoted the product with the current promotion content, but the style of the historical content of the marketing account is inconsistent with the style of the promotion content; The fourth matching result is that the marketing account has never promoted the product with the current promotion content, but the historical content style of the marketing account is consistent with the promotion content style.

3. The e-commerce platform promotion task allocation method according to claim 2, characterized in that: Initializing the expected promotion distribution income between each promotional content and each marketing account in the current promotion cycle based on the matching situation specifically includes the following steps: Initialize the first The promotional content and The matching degree of the marketing account and the promotion content In the marketing account historical returns; The promotion content is matched according to the matching degree and the historical revenue. Assigned to the marketing account The expected generalization distribution benefit is initialized, that is, ； in, For the current promotion period The expected promotion distribution income, For matching, For historical returns.

4. The e-commerce platform promotion task allocation method according to claim 3, characterized in that: The original marketing account data set also includes the transaction volume of each category of goods; For the first matching result, the matching degree and the historical benefit are: ； ； For the second matching result, the matching degree and the historical benefit are: ； ； For the third matching result, the matching degree and the historical benefits are: ； ； For the fourth matching result, the matching degree and the historical benefits are: ； ； in, For the marketing accounts within the promotion period The historical content style, For the marketing accounts within the promotion period Historical promotion product categories, For category The transaction revenue of the goods.

5. The e-commerce platform promotion task allocation method according to claim 1, characterized in that: The slack variable is ,in ; The linear programming model is: , in, is the total expected return; The total cost in the first constraint condition includes the cost of allocating each piece of the promotional content to the corresponding marketing account.

6. The e-commerce platform promotion task allocation method according to claim 2, characterized in that: Entering the next promotion cycle and updating the original marketing account dataset according to the allocation plan, if the maximum promotion cycle has not been reached, obtaining the allocation plan for the next promotion cycle based on the updated marketing account dataset; otherwise, terminating the promotion allocation task, specifically includes the following steps: Incrementing the current promotion cycle by one, so that the promotion assignment task enters the next promotion cycle, and determining whether the current promotion cycle reaches a preset maximum promotion cycle; If the maximum promotion period is reached, the promotion allocation task is terminated and the allocation plan for each promotion period is output; If the maximum promotion cycle has not been reached, the promotion allocation task is executed according to the allocation plan and the promotion results of the previous promotion cycle are obtained. Based on the promotion results, the updated historical content style, historical promotion product categories and transaction volume of products in each category are obtained, and the matching status of the current promotion cycle is obtained. Based on the matching status, each promotion content of the current promotion cycle and the expected promotion allocation income of each marketing account are initialized to obtain the allocation plan until the maximum promotion cycle is reached.

7. An e-commerce platform promotion task allocation device, characterized in that: Includes the following modules: a content matching module configured to obtain an original promotion content dataset and an original marketing account dataset during the first promotion cycle of a promotion assignment task, and determine a match between each piece of promotion content and each marketing account; An expected revenue initialization module is configured to initialize the expected promotion distribution revenue between each promotional content and each marketing account in the current promotion cycle based on the matching situation; The allocation plan generation module is configured to define slack variables within the current promotion cycle, obtain a constraint formula based on the slack variables and preset constraint conditions; establish a linear programming model based on the slack variables and the expected promotion distribution benefits, solve the linear programming model based on the constraint formula to obtain the slack variables that maximize the expected total benefits; randomly round the solved slack variables to obtain a rounding result, and generate an allocation plan for the current promotion cycle based on the rounding result; The constraints include: a first constraint that the total cost of the allocation plan does not exceed a preset cycle budget; a second constraint that each piece of promotional content is promoted at least once during each promotion cycle; and a third constraint that the total number of promotional content allocated to each marketing account does not exceed the corresponding marketing account's carrying limit. The promotion update module is configured to enter the next promotion cycle and update the original marketing account data set according to the allocation plan. If the maximum promotion cycle has not been reached, the allocation plan for the next promotion cycle is obtained based on the updated marketing account data set; otherwise, the promotion allocation task is terminated.

8. An intelligent terminal, characterized in that: It includes a memory and a processor, wherein the memory stores at least one instruction, at least one program, code set or instruction set, and the at least one instruction, at least one program, code set or instruction set is loaded and executed by the processor to implement the e-commerce platform promotion task allocation method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that The readable storage medium stores at least one instruction, at least one program, code set or instruction set, and the at least one instruction, at least one program, code set or instruction set is loaded and executed by the processor to implement the e-commerce platform promotion task allocation method as described in any one of claims 1 to 6.

Images