Special commodity commission discount configuration method based on big data analysis

Abstract

The invention relates to the technical field of electronic commerce, and discloses a special commodity commission discount configuration method based on big data analysis. The method comprises the following steps: S1, acquiring historical promotion data of a commodity, wherein the historical promotion data comprises commodity sales data and marketing activity data; s2, analyzing the historical data by using a big data analysis technology, and screening out special commodities which need to be emphatically popularized; s3, determining matched activity categories and commission discount setting parameters for the screened special commodities; and S4, automatically generating a commission discount configuration template, and filling the commission discount parameters into the template. By adopting the technical scheme based on big data analysis, the commodities are intelligently screened through the multi-dimensional data model and the clustering algorithm, and the technical effect of accurately screening the key promotion commodities is achieved. Compared with the technical scheme of selecting commodities depending on manual rule configuration or single data index in the prior art, the problems of poor flexibility, low efficiency and indefinite promotion goal are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of e-commerce, and in particular to a method for configuring special commodity rebates based on big data analysis. Background Art

[0002] With the rapid development of Internet technology, e-commerce and digital marketing have become the core model of commodity sales. Through the Internet platform, merchants can quickly cover a large number of user groups and promote commodity sales through diversified marketing activities. In this process, rebate configuration, as an effective incentive, has gradually become an important strategy for merchants to improve the effectiveness of commodity promotion. Reasonable rebate rules can not only motivate platform dealers and promoters to actively participate in promotion, but also optimize the market performance of commodities, thereby increasing overall sales profits.

[0003] The existing commodity rebate configuration system usually relies on manual setting of rules. For all categories of goods, promoters need to manually set parameters such as rebate ratio, activity conditions and effective time. This method is inefficient and prone to configuration errors in scenarios with a wide variety of products and frequent marketing activities.

[0004] The existing rebate configuration methods mostly adopt the "fixed rules" model, that is, setting a unified rebate ratio or conditions for all categories of goods, or setting rebate rules through simple product classification. This fixed model cannot flexibly adapt to market changes, nor can it formulate accurate rebate strategies based on the characteristics of different products.

[0005] The existing rebate configuration system usually monitors promotional activities only through simple sales data statistics, such as total sales and number of participants. This shallow monitoring cannot fully reflect the actual effect of the rebate configuration. Some products may waste budgets due to overly high rebate ratios, but the existing system cannot detect and provide feedback in real time. Summary of the invention

[0006] In view of the shortcomings of the prior art, the present invention provides a special commodity rebate configuration method based on big data analysis, which solves the problems of low efficiency, insufficient flexibility, low data utilization and weak promotion effect monitoring ability in the commodity rebate configuration method.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: A special commodity rebate configuration method based on big data analysis includes the following steps: S1. Obtain historical promotion data of products, including product sales data and marketing activity data; S2. Analyze the historical data using big data analysis technology to select special products that need to be promoted; S3. Determine the matching activity categories and rebate setting parameters for the selected special products; S4. Automatically generate a rebate configuration template and fill the rebate parameters into the template; S5. Send the rebate configuration template to the product promotion system and execute the promotion activities to achieve the optimal configuration of product promotion.

[0008] Preferably, the historical promotion data in step S1 includes the following multi-dimensional data: commodity profit margin, sales volume, inventory level and conversion rate.

[0009] Preferably, the method for screening special commodities in step S2 comprises the following steps: S2.1. Build a multi-dimensional data model based on commodity profit margin, sales volume, and inventory level; S2.2. Analyze the multi-dimensional data model using a data clustering algorithm to determine special products that need to be promoted.

[0010] Preferably, the rebate setting parameters in step S3 include: rebate ratio, rebate method and activity time range.

[0011] Preferably, the rebate setting parameters in step S3 are calculated according to the following formula: R=(P×C) / T Among them; R is the rebate ratio; P is the product profit margin; C is the sales conversion rate; T is the target activity condition.

[0012] Preferably, the method for generating a rebate configuration template in step S4 comprises the following steps: S4.1. Extract product characteristic data from the analysis results; S4.2. Automatically match product feature data with corresponding rebate parameters based on preset template rules; S4.3. Generate a rebate configuration template suitable for promotion activities.

[0013] Preferably, the rebate configuration template in step S4 includes the following fields: product category, promotion activity name, rebate ratio and activity conditions.

[0014] Preferably, when the rebate configuration template is sent to the product promotion system in step S5, the promotion system supports real-time monitoring of promotion effects and feedback to adjust rebate parameters, and supports batch configuration and real-time update.

[0015] Preferably, in step S5, the product promotion system dynamically adjusts the rebate configuration parameters by acquiring the returned execution feedback data to optimize the promotion effect.

[0016] Preferably, the execution feedback data includes the click rate, conversion rate and sales volume of the special product.

[0017] The present invention provides a method for configuring special commodity rebates based on big data analysis. It has the following beneficial effects: 1. The present invention adopts a technical solution based on big data analysis, uses a multi-dimensional data model and clustering algorithm to intelligently screen commodities, and achieves the technical effect of accurately screening key promotion commodities. Compared with the technical solution in the prior art that relies on manual rule configuration or a single data indicator to select commodities, it solves the problems of poor flexibility, low efficiency, and unclear promotion goals.

[0018] 2. The present invention adopts the technical solution of automatically generating a rebate configuration template, automatically filling the selected special products and their rebate parameters into the preset template, and achieves the technical effect of quickly and efficiently generating a promotion configuration file. Compared with the technical solution of manually configuring templates or relying on fixed rules to generate configurations in the prior art, the problems of low configuration efficiency, easy errors and insufficient adaptability are solved.

[0019] 3. The present invention adopts the technical solution of real-time monitoring and dynamic optimization, monitors key indicators such as sales volume, click-through rate and conversion rate during the promotion activities through the promotion system, and adjusts the rebate parameters and activity conditions in real time, thereby achieving the technical effect of improving the promotion effect and resource utilization. Compared with the technical solution in the prior art that lacks real-time monitoring and optimization capabilities, it solves the problems of uncontrollable activity execution effects and uneven resource allocation. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in the specification of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] Please see attached Figure 1 , an embodiment of the present invention provides a special commodity rebate configuration method based on big data analysis, including the following steps; S1. Obtain historical promotion data of products, including product sales data and marketing activity data; Specifically, step S1 obtains historical promotion data of the product, including product sales data and marketing activity data. This step is the basis for subsequent data analysis, special product screening, and rebate parameter setting. The accuracy and completeness of data acquisition directly affect the implementation effect of the entire method. This step focuses on combining different data sources, data dimensions, and processing methods to ensure that it can provide complete, accurate, and efficient data support for subsequent steps; In this embodiment, firstly, the data collection module connects with the product promotion system through an interface to obtain the sales data of the product in real time. The sales data includes the daily sales, weekly sales, monthly sales and cumulative sales of the product, etc., which are used to reflect the sales performance of the product. The sales data can be classified according to the SKU of the product so that the sales of different products can be accurately distinguished in the subsequent analysis. Secondly, the marketing activity data is extracted from the enterprise marketing management system through another interface. These data include the type of promotional activities that the product participates in, the duration of the activity, the activity budget, and the consumer participation. To ensure the accuracy and consistency of the data, the acquired historical data was preprocessed. Specifically, data preprocessing includes the following steps: Data cleaning: removing redundant or invalid data, such as duplicate records, incomplete sales records, or outliers; Data formatting: Integrate data from different sources according to a unified time dimension and product classification; Data verification: Ensure the consistency between sales data and inventory data through verification rules; Specifically, in this embodiment, the data consistency check is performed using the following formula: Q total ≤I init +P added -I remaining in: Q total Indicates the cumulative sales quantity of the product; I init Indicates the initial inventory of the product; P added Indicates the quantity of inventory added during the sales period; I remaining Indicates the current remaining inventory of the product; Adjustments can be made through manual intervention prompted by the system, or automatic completion can be performed based on historical data; The amount of historical data is large, and the processing efficiency of a single node is low. Therefore, by adopting big data technology, data storage and computing can be distributed to multiple nodes, thereby improving the efficiency of data processing; The distributed computing framework used in the present invention can support the following operations: Data sharding storage, dividing different data blocks according to product classification or time dimension; Batch processing of data based on MapReduce algorithm; Specifically, among the acquired data, the sales conversion rate of goods is an important indicator for screening special goods; In this embodiment, the sales conversion rate is calculated by the following formula: in: T represents the sales conversion rate of the product; O represents the number of times the product is purchased; C represents the number of clicks on the product; To improve the comprehensiveness of data acquisition, historical promotion data also includes customer evaluation data and return rates of products; Specifically: Customer evaluation data can reflect the quality of goods and consumer satisfaction, which can be further quantified into positive evaluation rate through sentiment analysis algorithms; The return rate can provide a reference for the after-sales performance of products to avoid promoting inefficient or high-return products. To improve the credibility of conversion rate data, the system performs denoising on click data in conversion rate calculation to eliminate invalid clicks; the data collection module connects with a third-party big data service platform through an API to supplement a wider range of market data, which provides more comparison basis for subsequent screening of special products; Step S1 provides a comprehensive data foundation for subsequent big data analysis and special product screening through multi-dimensional and multi-source data collection and preprocessing technology. After the historical promotion data is acquired, the system can enter step S2 to further analyze and screen the data.

[0023] S2. Analyze the historical data using big data analysis technology to select special commodities that need to be promoted. Specifically, after completing the acquisition and preprocessing of historical promotion data in step S1, step S2 aims to use these data to select special commodities that need to be promoted. The screening results provide a direct basis for the generation of subsequent rebate setting parameters. The screening of special commodities requires comprehensive consideration of multi-dimensional data, such as commodity profit margins, sales volumes, inventory levels, and conversion rates. These data are analyzed through a multi-dimensional model to quantify the importance score of each commodity, and finally select key commodities. Combine historical data with real-time dynamic data to ensure that the screening results can reflect current market demand and promotion goals. In the embodiment, a multi-dimensional data model is first constructed based on the historical promotion data of the product to represent the performance of each product in each dimension. The multi-dimensional data model is constructed in the following manner: Using merchandise profit margin, sales volume, and inventory levels as core dimensions; Assign weights to each dimension to highlight the importance of different dimensions to the promotion goals; Eliminate the influence of different dimensional data units and magnitudes through normalization; Specifically, in this embodiment, the following formula is used to calculate the comprehensive score of the product: S=w1×P+w2×V+w3×T in: S represents the comprehensive rating of the product; P represents the profit margin of the commodity; V represents the sales volume of the product; T represents the conversion rate of the product; w1, w2, and w3 represent the weight factors of the corresponding dimensions respectively; The selection of weight factors is based on the company's promotion goals. When the promotion goal is to increase overall profits, w1>w2 and w1>w3 can be set; when the goal is to increase sales, w2>w1 and w2>w3 can be set. The weight factors can also be dynamically adjusted according to specific market needs. Based on the calculation results of the above comprehensive scoring formula, the system sorts the products from high to low according to the score and sets the score threshold S min Filter out special products that need to be promoted. Specifically, the scores are higher than the threshold S. min Goods that are considered special goods; To further optimize the screening process, a data clustering algorithm was used to analyze the multi-dimensional data model. Specifically: First, the normalized multidimensional data is input into the K-means clustering algorithm, and the number of cluster centers K is determined according to the number of product categories or promotion needs; Secondly, the clustering results are used to identify the product group that is closest to the characteristics of the target group; Finally, select several products with the highest comprehensive scores from the clustering results as special products; In this embodiment, through the outlier detection mechanism, when the comprehensive score of a product is significantly higher or lower than similar products, the integrity and accuracy of its historical data are further checked to eliminate the deviation results that may be caused by abnormal data; In this embodiment, dynamic data is used to optimize the screening results. Dynamic data includes real-time inventory changes, market demand fluctuations, and consumer behavior data, etc. Specifically: By analyzing real-time inventory change data, avoid including products with insufficient inventory in the key promotion scope; Adjust the scoring weights according to market demand fluctuations to make the screening results more in line with the current market environment; Combined with real-time consumer behavior data, the timeliness of screening can be further improved; Improve screening efficiency and system performance. Use a distributed computing framework to perform parallel computing on multi-dimensional data models. Utilize Spark's distributed data processing capabilities to complete scoring calculations and cluster analysis of large-scale data in a short time. In this embodiment, additional constraints are added to the scoring formula to balance the weight effects between different dimensions. The optimized formula is as follows: Among them, the denominator Indicates the normalization of weight factors to ensure the comparability of scoring results; Step S2 can effectively screen out special products that need to be promoted through the construction of a multi-dimensional data model, the calculation of a comprehensive scoring formula, and the application of a data clustering algorithm. The implementation ensures the scientificity and accuracy of product screening and provides high-quality input data for the generation of subsequent rebate parameters.

[0024] S3. Determine the matching activity categories and rebate setting parameters for the selected special products; In this embodiment, step S3 determines the matching activity category and rebate setting parameters for the special products selected in step S2, and generates rebate configuration parameters suitable for product promotion by combining product characteristic data and historical promotion data, thereby providing accurate support for subsequent template generation and promotion execution; The rebate setting parameters include rebate ratio, rebate method and activity time range. In this step, the rebate parameters are dynamically adjusted according to different promotion goals and product characteristics to improve the accuracy and flexibility of product promotion; The determination of commission setting parameters mainly includes the following three aspects: matching of activity categories, calculation of rebate ratio and setting of activity time range. In terms of matching of activity categories, by analyzing the historical sales data of the product, the types of promotional activities involved and their effects, select activity categories that match the characteristics of the product. If a product has good historical sales data in a full-reduction activity, then the product will be matched to the full-reduction activity category first. Specifically, in this embodiment, the following rules are used to match activity categories: if the conversion rate of a product is significantly improved in a discount activity, it will be matched to the discount activity category; If the sales volume of a product performs well in a full-reduction activity, it will be matched to the full-reduction activity category; If the product has a high user rating and strong consumer participation in historical data, it will be matched to the points return activity category; the matching of activity categories can also be optimized in combination with market dynamic data; Secondly, in terms of the calculation of the rebate ratio, this embodiment uses the following formula to calculate the rebate ratio by analyzing data such as product profit margin, conversion rate and promotion target; in: C represents the rebate ratio; P represents the profit margin of the commodity; T represents the conversion rate of the product; K represents the activity adjustment factor, which is used to reflect the impact of activity type on the rebate ratio; G represents the promotion target conditions, such as expected sales or profits; The calculation of the rebate ratio combines historical data and real-time dynamic data. When market demand fluctuates greatly, the value of K can be dynamically adjusted to improve the flexibility of the rebate strategy; In this embodiment, the best activity time range is determined by analyzing the sales cycle and inventory changes of the product. If the historical sales data of a product shows that its sales peak is concentrated on weekends, the activity time range can be set to Friday to Sunday. Specifically, the activity time range can be set using the following formula: T a =max(T s ,T d ) in: T a Indicates the time range of the activity; T s Indicates the historical sales peak period; T d Indicates a reasonable time period for inventory consumption; In this embodiment, the adaptability of the rebate setting parameters is improved. By analyzing historical data and real-time feedback data, the rebate ratio, rebate method and activity time range are continuously adjusted. If the conversion rate of a certain product during the activity execution period does not meet expectations, the system can increase the rebate ratio or extend the activity time range in real time to improve the promotion effect; This step also uses a distributed computing framework to parallelize the calculation of rebate parameters. By dividing the products into different data blocks by category, the rebate setting parameters can be calculated simultaneously on multiple computing nodes, thereby improving the computing efficiency; Step S3 generates adaptive rebate setting parameters by matching activity categories, calculating rebate ratios, and setting activity time ranges, combining historical data and dynamic data, ensuring the accuracy and flexibility of product promotion strategies, and laying a solid foundation for subsequent rebate template generation and promotion activity execution. S4. Automatically generate a rebate configuration template and fill the rebate parameters into the template; Step S4 automatically generates a rebate configuration template based on the rebate setting parameters generated in step S3, and fills the relevant parameters into the template, directly connecting the results of the aforementioned activity category matching and rebate parameter setting, and providing standardized structural support for the issuance of the rebate configuration template in the subsequent step S5. Through automated template generation and parameter filling, the configuration efficiency can be significantly improved and the possibility of manual intervention can be reduced, thereby achieving accurate and efficient rebate promotion configuration; The rebate configuration template includes standardized field design, which is used to record the promotion category, rebate ratio, activity conditions and product feature data of the product. The field content and format of the template can be flexibly adjusted according to different promotion systems and business needs to ensure its compatibility and applicability; In this embodiment, the specific implementation of generating the rebate configuration template and filling in parameters includes the following steps: First, the template generation module creates a basic rebate configuration template by calling the rebate setting parameters and activity category information calculated and generated in step S3; The field design of the basic template includes the following: Product category field: records the classification information of the product; Activity category field: records the type of promotion activity to which the product belongs; Commission Ratio Field: records the commission ratio set for the product; Activity Condition Field: records the conditions for the promotion to take effect; Product feature field: records basic information about the product; During the template parameter filling process, the corresponding data is automatically filled into the corresponding fields of the template in combination with the product feature data and the rebate setting parameters; In this embodiment, the following rules are used to achieve automatic filling: Based on the product classification information and activity category, match the product category field and activity category field of the template; Fill the rebate ratio field with the rebate ratio parameter calculated in step S3; Populate the activity condition fields with the results of setting the activity time range; Extract the product's inventory information and sales summary, and fill in the product feature fields; In this step, you only need to call the stored calculation results to complete the automatic filling. In order to improve the accuracy of filling, a data verification mechanism is also set in this embodiment. When filling the rebate ratio field, the system will verify whether its value is within the preset upper and lower limits. If it exceeds the range, a warning will be triggered and the filling operation will be suspended; This embodiment adopts a parallel processing mechanism of template generation and filling, specifically; Through the distributed processing framework, the data of multiple commodities are distributed to different computing nodes; Each node independently generates a commodity rebate configuration template and completes parameter filling; All generated templates are aggregated on the master node to form a complete rebate configuration result; In this embodiment, the adaptability of the rebate configuration template is enhanced. Specifically, each time a template is generated, the field content of the template and its filled value are recorded to form a version record. By comparing the field changes of different versions of templates, the adjustment effect of the promotion strategy can be analyzed, and a basis for subsequent optimization can be provided; The template generation module also supports multi-language output function. A new language field is added to the template, and a rebate configuration template adapted to different language environments is generated through the translation engine. After the template generation is completed, this embodiment supports exporting the template in multiple formats. The formatted template can be directly used for interface docking with the promotion system to ensure the smooth implementation of subsequent steps; In this embodiment, through standardized template field design, automatic filling of parameters and diversified template output methods, efficient generation and flexible adaptation of rebate configuration templates are achieved. This step ensures the accuracy of product rebate configuration and system compatibility, and provides reliable support for the subsequent issuance of templates and the implementation of promotional activities.

[0025] S5. Sending the rebate configuration template to the product promotion system to execute the promotion activities to achieve the optimal configuration of the product promotion. Step S5 is to send the template to the product promotion system to support the execution of the actual promotion activities after the rebate configuration template is generated. Step S5 sends the rebate configuration template generated in step S4 to the product promotion system and executes the promotion activities. This step needs to ensure that the template can be correctly transmitted and efficiently applied to the promotion system, while supporting real-time monitoring and feedback adjustment of the promotion effect. Through this process, closed-loop management from parameter configuration to promotion execution is achieved, further optimizing the overall effect of product promotion; The issuance of rebate configuration templates involves multiple links, including template format conversion, interface transmission, promotion system reception and activity launch. Through the real-time feedback mechanism, the execution effect of the promotion activities is continuously monitored, and the rebate parameters are dynamically adjusted to ensure the achievement of the activity goals; In this embodiment, before the template is issued, the correctness and compatibility of the rebate configuration template are ensured. In this embodiment, the template content is format converted and data verified; Specifically, in this embodiment, the following formula is used to dynamically adjust the rebate parameters to ensure the optimization of the activity effect; in: C ′ is the adjusted rebate ratio; C is the initial rebate ratio; R real is the actual conversion rate of the current promotion; R target The preset target conversion rate; According to the above formula, when the actual conversion rate is lower than the target value, the system will automatically increase the rebate ratio to enhance the promotion efforts; otherwise, the rebate ratio will be reduced to optimize resource utilization; After the template is issued, in this embodiment, the execution status of the rebate configuration is tracked through the real-time monitoring module of the promotion system. The monitoring module will collect the following key data; Product click-through rate: reflects consumers’ attention to the promoted products; Conversion rate: measures the actual sales effect of promoted products; Campaign budget consumption rate: Tracks the use of funds for the campaign; The above monitoring data is counted and analyzed by the following formula; in: R c Indicates conversion rate; C o Indicates the actual number of purchases; C t Indicates the total number of clicks; The monitoring data is fed back to the template generation module in real time through the interface to update the rebate parameters and optimize the promotion strategy. If the conversion rate of a product is lower than the preset threshold, its rebate ratio can be dynamically adjusted or the activity time range can be extended; In this embodiment, the dynamic adjustment mechanism automatically optimizes the rebate parameters based on real-time feedback data; If you find that the inventory of a certain product is decreasing rapidly, you can automatically reduce the rebate ratio or shorten the activity time range to avoid insufficient inventory; if the product has a high click-through rate but a low conversion rate, you can increase the rebate ratio or adjust the activity conditions to attract more purchases.

[0026] This embodiment supports cross-platform promotion, generates corresponding rebate configuration templates, and issues them through the platform-specific interface. This design ensures the flexible adaptability of the template and reduces the need for manual intervention; Step S5 ensures the efficient execution of rebate configuration through mechanisms such as template verification, batch distribution, real-time monitoring and dynamic adjustment. The implementation of this step significantly improves the accuracy and flexibility of product promotion, and provides important support for enterprises to achieve precision marketing in a complex market environment.

[0027] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A special commodity rebate configuration method based on big data analysis, characterized in that: The steps include: S1. Obtain historical promotion data of products, including product sales data and marketing activity data; S2. Analyze the historical data using big data analysis technology to select special products that need to be promoted; S3. Determine the matching activity categories and rebate setting parameters for the selected special products; S4. Automatically generate a rebate configuration template and fill the rebate parameters into the template; S5. Send the rebate configuration template to the product promotion system and execute the promotion activities to achieve the optimal configuration of product promotion.

2. The method for configuring special commodity rebates based on big data analysis according to claim 1 is characterized in that: The historical promotion data in step S1 includes the following multi-dimensional data: commodity profit margin, sales volume, inventory level and conversion rate.

3. The method for configuring special commodity rebates based on big data analysis according to claim 1 is characterized in that: The method for screening special commodities in step S2 comprises the following steps: S2.

1. Build a multi-dimensional data model based on commodity profit margin, sales volume, and inventory level; S2.

2. Analyze the multi-dimensional data model using a data clustering algorithm to determine special products that need to be promoted.

4. The method for configuring special commodity rebates based on big data analysis according to claim 1 is characterized in that: The rebate setting parameters in step S3 include: rebate ratio, rebate method and activity time range.

5. The method for configuring special commodity rebates based on big data analysis according to claim 1 is characterized in that: The rebate setting parameters in step S3 are calculated according to the following formula: R=(P×C) / T Among them; R is the rebate ratio; P is the product profit margin; C is the sales conversion rate; T is the target activity condition.

6. The method for configuring special commodity rebates based on big data analysis according to claim 1 is characterized in that: The method for generating a rebate configuration template in step S4 comprises the following steps: S4.

1. Extract product characteristic data from the analysis results; S4.

2. Automatically match product feature data with corresponding rebate parameters based on preset template rules; S4.

3. Generate a rebate configuration template suitable for promotion activities.

7. The method for configuring special commodity rebates based on big data analysis according to claim 1 is characterized in that: The rebate configuration template in step S4 includes the following fields: Product category, promotional activity name, rebate ratio and activity conditions.

8. The method for configuring special commodity rebates based on big data analysis according to claim 1, characterized in that: When the rebate configuration template is sent to the product promotion system in step S5, the promotion system supports real-time monitoring of promotion effects and feedback to adjust rebate parameters, and supports batch configuration and real-time update.

9. The method for configuring special commodity rebates based on big data analysis according to claim 1, characterized in that: In step S5, the product promotion system dynamically adjusts the rebate configuration parameters by obtaining the returned execution feedback data to optimize the promotion effect.

10. The method for configuring special commodity rebates based on big data analysis according to claim 9, characterized in that: The execution feedback data includes the click rate, conversion rate and sales volume of special products.

Images