Redeemable article depreciation drainage system based on digital points

Abstract

The invention, which relates to the technical field of e-commerce and marketing, discloses a digital-integral-based depreciation drainage system for exchangeable articles, comprising an integral and depreciation association module, a user drainage and retention module, and a data processing and system optimization module. Bidirectional driving of integral and depreciation promotion is realized by utilizing the integral and depreciation association module, and the integral is given when an article is depreciated, so that a consumer can obtain extra integral rights and interests while enjoying low price, the consumption enthusiasm is stimulated, and the effect of improving the article sales conversion rate and the user re-purchase rate is realized; new users, personalized schemes and loyalty level privileges are attracted by means of social drainage to enhance user stickiness, the effects of quickly expanding user groups and effectively retaining old users are achieved, and by means of a data processing and system optimization module, accurate prediction is performed and merchants are assisted in scientific decision making and stable processing of an efficient system architecture, so that the user experience is improved. The effects of optimizing the operation strategy and improving the response performance of the system are achieved.

Description

Technical Field

[0001] The present invention relates to the field of e-commerce and marketing technology, and in particular to a price reduction and traffic diversion system for redeemable items based on digital points. Background Art

[0002] In the field of business operations, the points system is a common customer relationship management and marketing tool. Points are a virtual value accumulated by enterprises after consumers consume, participate in activities or meet specific conditions. It represents the rights and interests or contribution of consumers in the business system. Consumers can use these points to redeem corresponding goods, services or enjoy specific discounts in the future. The price reduction and drainage system is a strategic system that attracts consumer traffic by lowering the price of goods. Its core purpose is to quickly increase the attention and sales of goods in a short period of time, attract new customers to try to buy products or services, and stimulate old customers to increase the frequency or purchase volume, so as to achieve various business goals such as improving brand awareness, clearing inventory, and expanding market share. It is of great significance to combine points with price reduction and drainage. For consumers, points increase the added value of consumption, making them have more expectations and sense of gain in the shopping process; while price reduction and drainage directly provide consumers with benefits. The combination of the two can further enhance consumers' shopping experience and satisfaction, and enhance their favorability and loyalty to merchants. For merchants, this combination can not only effectively promote commodity sales, but also make full use of the points mechanism to activate user resources, improve user activity and stickiness, enable merchants to stand out in the fierce market competition, and achieve long-term stable business operations and sustainable development.

[0003] However, the traditional points redemption system is mostly a one-way operation mode. Consumers only use accumulated points to redeem items or services of fixed value. There is a lack of effective interaction and coordination between points and price reduction promotion activities. This one-way nature leads to some shortcomings. The incentive effect of points is limited, and consumers are not motivated to obtain points because points lack more value extensions besides redeeming items. When items are discounted, the points mechanism is not fully utilized to further expand the promotion effect, and a virtuous cycle between promotion and points cannot be formed. It is difficult to deeply tap the purchasing potential of consumers. Traditional systems have shortcomings in data processing and coping with high concurrent operations. With the increase in the number of users and business volume, the system is prone to freezes and delays, which affects user experience and business efficiency. Therefore, it is necessary to propose a redeemable item price reduction drainage system based on digital points to solve the problems in the existing technology. Summary of the invention

[0004] The purpose of the present invention is to make up for the shortcomings of the prior art and to provide a redeemable item price reduction traffic diversion system based on digital points, which can realize the two-way drive of points and price reduction promotions, thereby improving the item sales conversion rate and user repurchase rate, attracting new users with the help of social diversion, and achieving the effect of quickly expanding the user base and effectively retaining old users, thereby optimizing the operation strategy and improving the system response performance.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: a system for attracting users to redeemable items at a price reduction based on digital points, the system comprising: a points and price reduction association module, a user traffic attraction and retention module, and a data processing and system optimization module;

[0006] The points and price reduction association module calculates the price reduction range of redeemable items based on the user's points acquisition speed, total amount, and usage frequency. The formula is: Where P d is the price reduction, I t is the current total points of the user, S f is the integral speed influence factor, I m is the intermediate amount of the integral threshold, k is the adjustment coefficient, and I is pressed when the item is discounted. a =P d ×C r ×I r Give extra points to users who participate in the redemption. a C is the number of gift points. r is the consumption amount correlation factor, I r The basic points redemption ratio, points speed influence factor S f The slope change rate is calculated by fitting the curve of the user's points acquisition change data and determined according to the corresponding table. The intermediate value of the integral threshold I m The median of the basic points required for redemption of items in the system multiplied by the market competition coefficient C c and seasonal sales fluctuation coefficient S s It turns out that C c Determined based on evaluation of competitor's points promotion activities, S s Determined by seasonal analysis based on historical sales data;

[0007] The user traffic diversion and retention module includes a social traffic diversion submodule based on point price reduction, a personalized point and price reduction scheme submodule, and a point loyalty level and differentiated price reduction privilege submodule;

[0008] The social traffic submodule based on point price reduction is used to attract new users to the system with the help of the communication power of social networks. Users can share the price reduction information of point redemption on social platforms. Invited friends can increase points for the sharer and reduce the points required for redemption through assistance operations. At the same time, the assistance friends can also get corresponding rewards, so as to achieve rapid growth in the number of users and increase the popularity of the system. The personalized points and price reduction plan submodule is used to deeply analyze the multi-dimensional data of users' consumption behavior, browsing history and point usage preferences by using big data analysis technology, and tailor a unique point acquisition plan and redeemable item price reduction recommendation plan for each user. The point loyalty level and differentiated price reduction privilege submodule is used to comprehensively evaluate the user loyalty level based on multiple indicators such as the total amount of user points, the stability of point acquisition, the frequency and amount of points redeemed for items, and the activeness of participating in system activities, and set differentiated redeemable item price reduction ranges, special redeemable item lists, and point redemption priority privileges for users of different levels;

[0009] The data processing and system optimization module includes a points and price reduction data prediction model submodule, a system architecture and points transaction processing submodule, and a real-time dynamic conversion system for points and monetary values. The points and price reduction data prediction model submodule uses a hybrid neural network algorithm to construct a prediction model, integrating the functions of the recurrent neural network RNN ​​and the convolutional neural network CNN, and pre-processes user points data, consumption data, item sales data, and market dynamics data according to time series and data feature dimensions and then inputs them into the model. An adaptive learning rate adjustment strategy is used during the model training process to dynamically adjust the learning rate according to changes in the prediction error.

[0010] Furthermore, the integral speed influencing factor S in the integral and price reduction association module f The method for determining S is as follows: collect the change data of the points obtained by the user within the set time period, fit it into a points acquisition curve, calculate the slope change rate of the curve, and find the corresponding impact factor value in the preset slope change rate-impact factor correspondence table according to the slope change rate as S f , and the corresponding table is generated based on the analysis of historical user data and machine learning algorithm training to adapt to the dynamic changes of different user groups and consumption scenarios.

[0011] Furthermore, the score threshold intermediate quantity I in the score and price reduction association module m The calculation method is: the median of the basic points redemption requirements of all redeemable items in the statistical system, multiplied by the current market competition coefficient C c and seasonal sales fluctuation coefficient S s , that is I m =M i ×C c ×S s , where Mi The median of the basic points redemption requirements, the market competition coefficient C c The seasonal sales fluctuation coefficient S is determined based on the frequency and intensity of similar point promotion activities of competitors in the same industry. s Based on the seasonal cycle analysis of historical sales data, the intermediate volume of the integral threshold can dynamically reflect changes in the market environment and sales cycle, and optimize the calculation of price reduction.

[0012] Furthermore, the social traffic submodule based on point price reduction in the user traffic and retention module, after the user shares the point redemption price reduction information on the social platform, counts the conversion rate of the traffic from different social platforms through tracking and analysis of the shared links, and dynamically adjusts the sharing reward mechanism according to the conversion rate. If the conversion rate of the traffic on the social platform is higher than the average level, the point reward range of the sharer and the booster on the platform is increased. The reward range adjustment formula is: adj =C tr ×A base , where A adj is the adjusted reward range, C tr is the conversion rate coefficient, A base The conversion rate ratio is calculated based on the current and historical conversion rate data through a specific weighted algorithm.

[0013] Furthermore, the personalized points and price reduction plan sub-module in the user traffic and retention module adopts a multi-dimensional user portrait construction algorithm when customizing personalized plans for users, comprehensively considers the user's consumption preferences, purchase frequency, consumption time period, and regional cultural characteristics, and constructs a user personalization model, wherein the consumption preference is determined by clustering analysis of the categories, brands, and price ranges of the user's historical purchases of goods, the purchase frequency is weighted calculation based on the number of purchases made by the user in different time periods and combined with the time decay function, the consumption time period analyzes the user's active purchasing time pattern within a day and a week, and is determined by counting the order time distribution of the user's historical purchase orders, and the regional cultural characteristics are matched with the consumption culture data of the corresponding region based on the user's registration address and common delivery address, and relevant information is obtained from a pre-constructed regional consumption culture database, and points acquisition tasks and redeemable item price reduction information are pushed to the user according to the personalized model.

[0014] Furthermore, the points loyalty level and differentiated price reduction privilege submodule in the user traffic and retention module are divided according to the user loyalty level, based on the user's total points, the stability of point acquisition, the frequency and amount of points redeemed for items, and the activeness of participating in system activities. The specific evaluation algorithm is L s =I t ×S i +Fi ×A i +E a ×V a , where L s is the loyalty score, S i is the integral stability coefficient, F i is the points redemption frequency, A i is the points redemption amount, E a is the activity participation, V a For the activity participation value coefficient, different levels are divided according to the loyalty score, and differentiated price reduction ranges and special lists of redeemable items are set for users of different levels to attract users to improve their loyalty level and increase consumption and participation depth in the system.

[0015] Furthermore, the system architecture and points transaction processing sub-module in the data processing and system optimization module adopt a distributed microservice architecture, which splits the points transaction processing function into multiple independent microservices, including points acquisition service, points redemption service, and price reduction calculation service. Each microservice is deployed in an independent container and managed and scheduled through container orchestration tools. In high-concurrency scenarios, message queues are used to achieve asynchronous communication and decoupling between microservices. When a large number of users perform points redemption operations at the same time, the points redemption requests first enter the message queue and are processed in sequence by the points redemption service. At the same time, cache technology is used to store hot data to reduce database query pressure.

[0016] Furthermore, the real-time dynamic conversion system of points and currency values ​​in the data processing and system optimization module establishes a real-time collection module for multi-currency exchange rate data, obtains the latest exchange rate information from the authoritative financial data platform, and combines the market price index data and the merchant's customized profit adjustment coefficient P adj Calculate the conversion ratio of points to currency. The conversion ratio calculation formula is: E r =E f ×P i ×P adj , where E r is the conversion ratio of points to currency, E f is the basic exchange rate factor, P i It is the price index influencing factor, and the basic exchange rate factor is calculated by weighted average based on the international exchange rate fluctuation range and historical exchange rate data of different currencies. The price index influencing factor is determined based on the year-on-year and month-on-month changes of the local market price index.

[0017] Furthermore, when the system is running as a whole, the data interaction between modules follows unified data standards and encrypted transmission protocols. The data standards specify the format, encoding and data structure of points data, user data, item data and transaction data. The encrypted transmission protocol uses an asymmetric encryption algorithm to encrypt and transmit sensitive data including user points balances and transaction amounts.

[0018] Compared with the prior art, the redeemable item price reduction and traffic diversion system based on digital points has the following beneficial effects:

[0019] The present invention utilizes the points and price reduction association module to realize the two-way drive of points and price reduction promotion. Points are given when the price of an item is reduced, so that consumers can obtain additional points rights while enjoying low prices, stimulate consumption enthusiasm, and achieve the effect of improving the sales conversion rate of items and the user repurchase rate. The present invention utilizes the user drainage and retention module to attract new users with the help of social drainage, and enhances user stickiness with personalized plans and loyalty level privileges, so as to achieve the effect of rapidly expanding the user group and effectively retaining old users. With the help of data processing and system optimization modules, accurate predictions are made and merchants are assisted in making scientific decisions. The stable processing of efficient system architecture achieves the effect of optimizing operation strategies and improving system response performance.

[0020] Other advantages, objectives and features of the present invention will be set forth in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0022] Figure 1 The figure is a structural diagram of a price reduction and traffic diversion system for redeemable items based on digital points. DETAILED DESCRIPTION

[0023] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. 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.

[0024] Embodiment 1

[0025] A large e-commerce platform has a vast amount of product resources and a huge user base. In order to increase user activity, promote product sales and enhance user stickiness, the platform has introduced a redeemable item price reduction system based on digital points. Merchants on the platform provide a rich variety of redeemable items, covering digital products, household items, beauty and skin care, food and beverages and other categories. Users can earn points by shopping on the platform and participating in platform activities (such as signing in, evaluating products, sharing product links, etc.). These points can be used to redeem their favorite items, and during specific promotional activities, the linkage between points and price reduction mechanisms can provide users with more benefits and shopping fun.

[0026] When users shop on e-commerce platforms, each purchase amount will be converted into points according to a certain ratio and accumulated into the user's points account. At the same time, the platform records in detail the time and method of obtaining points (such as consumption, event participation, etc.) and the use of points (exchanging items, participating in lucky draws, etc.). These data will serve as an important basis for the module linking points with price reductions. For example, when a user purchases a product worth M yuan, according to the points acquisition rules set by the platform, the user can obtain I t1 Points: The platform records the relevant information of the points obtained this time.

[0027] The points and price reduction association module regularly (such as weekly or monthly) analyzes and calculates based on the user's points data. Suppose a user's total points in a period of time reaches I t , the factor S affecting the speed of integral acquisition f According to the recent change data of the user's points acquisition, the points acquisition curve is fitted and the slope change rate is calculated, and then determined through the slope change rate-influence factor correspondence table (this correspondence table is generated by the platform based on the analysis of a large amount of historical user data and machine learning algorithm training). At the same time, the platform statistics the median M of the basic points redemption requirements of all redeemable items in the system i , and combined with the current market competition coefficient C c (determined based on the frequency and intensity of similar points promotion activities of competitors in the same industry) and seasonal sales fluctuation coefficient S s (Based on the seasonal analysis of historical sales data) Calculate the intermediate value of the integral threshold I m =M i ×C c ×S s , then, according to the formula Calculate the price reduction of the exchangeable item P d For example, for a redeemable digital product originally priced at P0 yuan, the price reduction is calculated to be P d Yuan, then the price of the product when redeeming points becomes P0-P d Yuan.

[0028] When the platform launches a price reduction promotion for a certain item, if a user participates in the price reduction redemption of the item, for example, the user uses the points to redeem the above digital products, the platform will calculate the price of the item according to formula I a =P d ×C r ×I r Give users extra points I a , where C r is the consumption amount correlation factor (related to the user's historical consumption amount on the platform and other data), I r The basic points redemption ratio.

[0029] After seeing the discounts for redeeming their points, users can share the price reduction information of the digital product points redeemed on social platforms (such as WeChat, Weibo, etc.). The platform tracks and analyzes the shared links, records the time when the shared links are clicked, the source social platform ID, the unique ID of the clicking user, and other information, and then classifies and counts the click data according to the source social platform ID, and calculates the conversion rate of the boosted traffic of each social platform (that is, the ratio of the number of successful boosts through the social platform to the total number of clicks on the platform). Assuming that the conversion rate of the boosted traffic of a certain social platform is C tr1 , the platform according to formula A adj =C tr ×A base Adjust the points reward range for the platform's sharers and contributors A adj (A base The amount is used as the basic reward range) to encourage more users to share and support on the social platform and expand the social communication influence of the platform.

[0030] The platform adopts a multi-dimensional user portrait construction algorithm, and builds a user personalized model by comprehensively considering factors such as user consumption preferences, purchase frequency, consumption time period, and regional cultural characteristics. Consumption preferences are determined by clustering analysis of the categories, brands, and price ranges of users' historical purchases (using a density-based clustering algorithm); purchase frequency is calculated by weighting the number of purchases made by users in different time periods and combined with a time decay function (exponential decay form); consumption time periods analyze the active purchase time patterns of users within a day or a week (determined by counting the order time distribution of users' historical purchase orders); regional cultural characteristics are matched with consumption culture data of the corresponding region based on the user's registered address or common delivery address (obtained from a pre-built regional consumption culture database). Based on this personalized model, the platform pushes points acquisition tasks that meet users' needs (such as extra points for purchasing specific categories of goods) and price reduction information of redeemable items (such as more substantial price reductions for beauty products preferred by users when redeeming points).

[0031] The platform comprehensively evaluates the user loyalty level based on indicators such as the total amount of user points, the stability of point acquisition, the frequency and amount of points redeemed for items, and the activeness of participating in system activities. The specific evaluation algorithm is L s =I t ×S i +F i ×A i +E a ×V a , where L s is the loyalty score, S i is the integral stability coefficient, F i is the points redemption frequency, A i is the points redemption amount, E a is the activity participation, V a For the activity participation value coefficient, different levels are divided according to the loyalty scores, and differentiated price reduction ranges and special lists of redeemable items are set for users of different levels. For example, users with high loyalty levels can enjoy greater price reductions when redeeming limited edition products, and have priority redemption rights, which helps to attract users to improve their loyalty levels and increase consumption and participation depth within the platform.

[0032] The precise points and price reduction data prediction model submodule uses a hybrid neural network algorithm to build a prediction model, integrating the advantages of recurrent neural networks (RNN) in processing time series data and the ability of convolutional neural networks (CNN) in extracting data features. User points data, consumption data, item sales data, market dynamics data, etc. are pre-processed according to time series and data feature dimensions and then input into the model. The model learns and trains on a large amount of historical data to predict changes in user points demand under different points strategies in the future and the impact of different price reduction ranges on the sales of redeemable items and overall business profits. During the model operation, an adaptive learning rate adjustment strategy is adopted to dynamically adjust the learning rate according to changes in prediction errors, thereby improving the prediction accuracy and stability of the model and providing data support for the platform's operational decisions. For example, the platform can adjust the points issuance rules and the pricing and price reduction strategies of redeemable items in advance based on the prediction results.

[0033] The efficient system architecture and points transaction processing submodule adopt a distributed microservice architecture, splitting the points transaction processing function into independent microservices such as points acquisition service, points redemption service, and price reduction calculation service. Each microservice is deployed in an independent container and managed and scheduled through container orchestration tools. In high-concurrency scenarios, such as during large-scale promotions, when a large number of users perform points redemption operations at the same time, the points redemption request first enters the message queue and is processed in sequence by the points redemption service to avoid system crashes due to high-concurrency requests. At the same time, caching technology is used to store the points information and price information of popular redeemable items, reducing database query pressure, improving system response speed, and ensuring users a smooth experience during shopping and points redemption.

[0034] The real-time dynamic conversion system of points and currency value establishes a real-time collection module for multi-currency exchange rate data, obtains the latest exchange rate information from the authoritative financial data platform, and combines the market price index data and the merchant's customized profit adjustment coefficient P adj Calculate the conversion ratio of points to currency, the calculation formula is E r =E f ×P i ×P adj , where E r is the conversion ratio of points to currency, E f is the basic exchange rate factor (calculated based on the weighted average of international exchange rate fluctuations of different currencies and historical exchange rate data), P i It is the price index influencing factor (determined according to the year-on-year and month-on-month changes in the local market price index). This ensures the accuracy and fairness of point redemption in multi-currency settlement scenarios and meets the needs of different regions and international businesses. For example, overseas users can get a reasonable redemption experience when shopping and redeeming points on the platform.

[0035] Effects brought by this embodiment: Through this embodiment, the e-commerce platform has achieved multi-faceted optimization and improvement. On the user side, the user's shopping experience has been greatly improved, and the value of points has been fully reflected. Not only can points be exchanged for more favorable items, but also additional points can be rewarded through social sharing, which increases the fun and interactivity of shopping. The personalized points and price reduction schemes have increased the user's attention and participation in the platform. Users can more conveniently obtain preferential information on the products they are interested in, thereby improving user satisfaction and loyalty. On the platform side, the points and price reduction association mechanism has effectively promoted the sales of goods, especially the overstocked goods or key promotion goods. The precise price reduction range is set to quickly attract users to redeem, thereby improving inventory turnover. The user drainage and retention module uses the power of social networks to achieve rapid growth in the number of users, and effectively retains old users through the loyalty level system, stabilizing the user group. The data processing and system optimization module ensures the stable operation of the system in high-concurrency scenarios, improves the system's response speed and processing efficiency, and provides solid technical support for the platform's large-scale operations. The accurate prediction model provides a scientific basis for the platform's operational decisions, enabling the platform to flexibly adjust its strategies to adapt to market changes, enhancing the platform's competitiveness in the e-commerce market, and achieving a win-win situation for users and the platform.

[0036] Embodiment 2

[0037] A redeemable item price reduction system based on digital points has been applied in the membership system of a large supermarket chain. The supermarket has many physical stores that sell various fresh foods, daily necessities, home appliances and other commodities. Supermarket members can accumulate points when shopping in the store. These points can not only be redeemed for goods in the supermarket, but also for goods or services in the supermarket's online mall. The supermarket hopes to increase the activity of members through this system, promote offline and online product sales, enhance member loyalty, and stand out in the fierce competition in the retail market.

[0038] When members shop at offline supermarkets, they will earn points for every amount they spend according to the points rules set by the supermarket. For example, if they spend X yuan, they can earn I points. t2 Points, at the same time, the supermarket cash register system records the members' consumption time, consumption product categories, consumption amount and corresponding points acquisition information. These data are transmitted to the supermarket's member management system as the data basis for the points and price reduction association module. Members' consumption in the supermarket's online mall and participation in online activities (such as online questionnaires, sharing online product links, etc.) will also accumulate points and be recorded according to the corresponding rules.

[0039] The points and price reduction association module conducts a comprehensive analysis of the members' points data every month. Assuming that the total points of a member reaches I t , integral speed influence factor Sf According to the member's recent changes in the number of points earned in stores and online, a curve is fitted and the slope change rate is calculated. Then, according to the slope change rate-influence factor correspondence table (the correspondence table is generated by the supermarket based on a large amount of historical member data and machine learning algorithms), the median M of the basic points redemption requirements for redeemable items in the supermarket statistical system is determined. i , combined with the current market competition coefficient C c (Considering similar points activities in other surrounding supermarkets) and seasonal sales fluctuation coefficient S s (Based on the sales data of various commodities in different seasons) calculate the intermediate amount of the integral threshold I m =M i ×C c ×S s , and then by the formula Calculate the price reduction of the exchangeable item P d For example, a household smart appliance with an original price of P'0 yuan is used as an exchangeable item. After calculation, its price reduction is P d Yuan, then members only need to pay P′0-P d Yuan (or deduct corresponding points and enjoy the price difference discount).

[0040] When a supermarket launches a price reduction promotion for a certain type of product and supports point redemption, if a member participates in the redemption, such as redeeming the above-mentioned smart appliances, the supermarket will calculate the points according to Formula I. a =P d ×C r ×I r Extra points for members a , where C r is the consumption amount correlation factor (related to the member’s historical consumption amount in the supermarket and the value of the redeemed goods this time, etc.), I r The basic points redemption ratio.

[0041] After seeing discounted products that can be redeemed for points in offline supermarket stores or finding attractive points redemption activities in the supermarket online mall, you can share the relevant information on social platforms (such as WeChat Moments, family groups, etc.). The supermarket member management system tracks and analyzes the shared links and counts the conversion rates of traffic from different social platforms. For example, the conversion rate of traffic from a social platform is C tr2 , according to formula A adj =C tr ×A base Adjust the points reward range for the platform's sharers and contributors A adj (A baseThe promotion is based on the basic reward range) to encourage more members to share on social platforms, attract more potential customers to pay attention to the supermarket's points activities, and expand the supermarket's brand influence and customer base.

[0042] The supermarket uses a multi-dimensional member portrait construction algorithm, which comprehensively considers members' consumption preferences (determined by analyzing members' historical purchases of commodity types, brand preferences and other data, and using cluster analysis), purchase frequency (based on statistics of members' offline and online purchases in different time periods and weighted calculations combined with time decay functions), consumption time periods (analyzing members' peak shopping hours within a day or week), and regional cultural characteristics (matching local consumer culture characteristics based on members' registered addresses or common delivery addresses, and obtaining information from the regional consumer culture database constructed by the supermarket) to build a member personalization model. Based on this model, the supermarket pushes personalized points acquisition tasks to members (such as obtaining extra points by purchasing branded goods that members frequently purchase) and price reduction information on redeemable items (such as launching larger point redemption price reductions for fresh foods preferred by members), thereby increasing members' attention and participation in the supermarket.

[0043] The supermarket comprehensively evaluates the member loyalty level based on the total number of member points, the stability of point acquisition (such as whether the points are acquired regularly), the frequency and amount of points redeemed for items, and the degree of participation in supermarket activities (including offline promotions, online interactive activities, etc.). The specific evaluation algorithm is L s =I t ×S i +F i ×A i +E a ×V a , where L s is the loyalty score, S i is the integral stability coefficient, F i is the points redemption frequency, A i is the points redemption amount, E a is the activity participation, V a For the activity participation value coefficient, different levels are divided according to the loyalty scores, and differentiated price reduction ranges and special lists of redeemable items are set for members of different levels. For example, high-loyalty level members can enjoy greater price reductions when redeeming limited edition high-end daily necessities, and have priority redemption rights, which helps to attract members to improve their loyalty levels and increase consumption and participation in the supermarket.

[0044] The accurate points and price reduction data prediction model submodule adopts a hybrid neural network algorithm to build a prediction model, integrating the advantages of recurrent neural network (RNN) in processing time series data and the ability of convolutional neural network (CNN) in extracting data features. Member points data, consumption data, commodity sales data, and market dynamics data (such as consumption trends in surrounding business districts, competitor promotions, etc.) are pre-processed according to time series and data feature dimensions and then input into the model. The model predicts the changes in members' points demand under different points strategies in the future and the impact of different price reduction ranges on the sales of redeemable items and the overall business profit of the supermarket through learning and training on a large amount of historical data. During the operation of the model, an adaptive learning rate adjustment strategy is adopted to dynamically adjust the learning rate according to the changes in the prediction error, so as to improve the prediction accuracy and stability of the model and provide data support for the supermarket's operational decisions. For example, the supermarket can adjust the points issuance rules, the types of redeemable items, and the pricing and price reduction strategies in advance according to the prediction results.

[0045] The efficient system architecture and points transaction processing submodule adopt a distributed microservice architecture, splitting the points transaction processing function into independent microservices such as points acquisition service, points redemption service, and price reduction calculation service. Each microservice is deployed in an independent container and managed and scheduled through container orchestration tools. In high-concurrency scenarios, such as when supermarkets carry out large-scale promotional activities or a large number of members redeem points at the same time during holidays, the points redemption request first enters the message queue and is processed in sequence by the points redemption service to avoid system crashes due to high-concurrency requests. At the same time, caching technology is used to store the points information and price information of popular redeemable items, reducing database query pressure, improving system response speed, and ensuring members a smooth experience during shopping and points redemption.

[0046] The real-time dynamic conversion system of points and currency value establishes a real-time collection module for multi-currency exchange rate data (taking into account that the supermarket may have international tourists or foreign members to consume), obtains the latest exchange rate information from the authoritative financial data platform, and combines the market price index data and the supermarket's customized profit adjustment coefficient P adj Calculate the conversion ratio of points to currency, the calculation formula is E r =E f ×P i ×P adj , where E r is the conversion ratio of points to currency, E f is the basic exchange rate factor (calculated based on the weighted average of international exchange rate fluctuations of different currencies and historical exchange rate data), P iIt is the price index influencing factor (determined according to the year-on-year and month-on-month changes in the local market price index). This ensures the accuracy and fairness of point redemption in multi-currency settlement scenarios and meets the needs of different regions and international businesses. For example, foreign members can get a reasonable redemption experience when shopping and redeeming points in supermarkets.

[0047] Effects of this embodiment: In this offline retail scenario embodiment, the supermarket has achieved remarkable results through the points system. For members, the shopping experience is richer and more personalized, the value of points is fully explored, members can redeem their favorite products at more favorable prices, and can also get extra points rewards through social sharing, which increases the interaction and stickiness between members and supermarkets. From the perspective of the supermarket, the points and price reduction association mechanism effectively stimulates commodity sales, the commodity turnover speed of both offline stores and online shopping malls is improved, inventory management is more efficient, the user flow and retention module expands the brand awareness and customer base of the supermarket with the help of members' social networks, the loyalty level system effectively improves the retention rate and consumption frequency of members, and stabilizes the core customer base of the supermarket. The data processing and system optimization module ensures the stability and response speed of the supermarket system during high-traffic operations. The accurate prediction model provides a scientific basis for the supermarket's commodity procurement, pricing, promotion and other operational decisions, enabling the supermarket to flexibly respond to market changes in the fierce retail market competition, improve the overall competitiveness and profitability of the supermarket, and achieve mutual benefit and win-win results between members and supermarkets.

[0048] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the same elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A price reduction and traffic diversion system for redeemable items based on digital points, characterized in that: The system includes: a points and price reduction association module, a user flow and retention module, and a data processing and system optimization module; The points and price reduction association module calculates the price reduction range of redeemable items based on the user's points acquisition speed, total amount, and usage frequency. The formula is: Where P d is the price reduction, I t is the current total points of the user, S f is the integral speed influence factor, I m is the intermediate amount of the integral threshold, k is the adjustment coefficient, and I is pressed when the item is discounted. a =P d ×C r ×I r Give extra points to users who participate in the redemption. a is the number of gift points, C r is the consumption amount correlation factor, I r The basic points redemption ratio, points speed influence factor S f The slope change rate is calculated by fitting the curve of the user's points acquisition change data and determined according to the corresponding table. The intermediate value of the integral threshold I m The median of the basic points required for redemption of items in the system multiplied by the market competition coefficient C c and seasonal sales fluctuation coefficient S s It turns out that C c Determined based on evaluation of competitor's points promotion activities, S s Determined by seasonal analysis based on historical sales data; The user traffic diversion and retention module includes a social traffic diversion submodule based on point price reduction, a personalized point and price reduction scheme submodule, and a point loyalty level and differentiated price reduction privilege submodule; The social traffic submodule based on point price reduction is used to attract new users to the system with the help of the communication power of social networks. Users can share the price reduction information of point redemption on social platforms. Invited friends can increase points for the sharer and reduce the points required for redemption through assistance. At the same time, the assistance friends can also get corresponding rewards, so as to achieve rapid growth in the number of users and increase the popularity of the system. The personalized points and price reduction plan submodule is used to deeply analyze the multi-dimensional data of users' consumption behavior, browsing history and point usage preference by using big data analysis technology, and tailor a unique points acquisition plan and redeemable item price reduction recommendation plan for each user. The points loyalty level and differentiated price reduction privilege submodule is used to comprehensively evaluate the user loyalty level based on multiple indicators such as the total amount of user points, the stability of point acquisition, the frequency and amount of points redeemed for items, and the activeness of participating in system activities, and set differentiated redeemable item price reduction ranges, special redeemable item lists and points redemption priority privileges for users of different levels; The data processing and system optimization module includes a points and price reduction data prediction model submodule, a system architecture and points transaction processing submodule, and a real-time dynamic conversion system for points and monetary values. The points and price reduction data prediction model submodule uses a hybrid neural network algorithm to construct a prediction model, integrating the functions of the recurrent neural network RNN ​​and the convolutional neural network CNN, and pre-processes user points data, consumption data, item sales data, and market dynamics data according to time series and data feature dimensions and then inputs them into the model. An adaptive learning rate adjustment strategy is used during model training to dynamically adjust the learning rate according to changes in prediction errors.

2. A price reduction and traffic diversion system for redeemable items based on digital points according to claim 1, characterized in that: The integral speed influencing factor S in the integral and price reduction association module f The method for determining S is as follows: collect the change data of the points obtained by the user within the set time period, fit it into a points acquisition curve, calculate the slope change rate of the curve, and find the corresponding impact factor value in the preset slope change rate-impact factor correspondence table according to the slope change rate as S f , and the corresponding table is generated based on the analysis of historical user data and machine learning algorithm training to adapt to the dynamic changes of different user groups and consumption scenarios.

3. A price reduction and traffic diversion system for redeemable items based on digital points according to claim 1, characterized in that: The score threshold intermediate amount I in the score and price reduction association module m The calculation method is: the median of the basic points redemption requirements of all redeemable items in the statistical system, multiplied by the current market competition coefficient C c and seasonal sales fluctuation coefficient S s , that is I m =M i ×C c ×S s , where M i The median of the basic points redemption requirements, the market competition coefficient C c The seasonal sales fluctuation coefficient S is determined based on the frequency and intensity of similar point promotion activities of competitors in the same industry. s Based on the seasonal cycle analysis of historical sales data, the intermediate volume of the integral threshold can dynamically reflect changes in the market environment and sales cycle, and optimize the calculation of price reduction.

4. A price reduction and traffic diversion system for redeemable items based on digital points according to claim 1, characterized in that: The social traffic submodule based on point price reduction in the user traffic and retention module, after the user shares the point redemption price reduction information on the social platform, tracks and analyzes the shared links, counts the conversion rate of the traffic from different social platforms, and dynamically adjusts the sharing reward mechanism according to the conversion rate. If the conversion rate of the traffic on the social platform is higher than the average level, the point reward range of the sharer and the booster on the platform will be increased. The reward range adjustment formula is: adj =C tr ×A base , where A adj is the adjusted reward range, C tr is the conversion rate coefficient, A base The conversion rate ratio is calculated based on the current and historical conversion rate data through a specific weighted algorithm.

5. The system for reducing the price of redeemable items based on digital points according to claim 1, characterized in that: The personalized points and price reduction scheme submodule in the user traffic diversion and retention module adopts a multi-dimensional user portrait construction algorithm when customizing personalized schemes for users, comprehensively considers the user's consumption preferences, purchase frequency, consumption time period, and regional cultural characteristics, and constructs a user personalized model, wherein the consumption preference is determined by clustering analysis of the categories, brands, and price ranges of the user's historical purchases of goods; the purchase frequency is weighted calculation based on the number of purchases made by the user in different time periods and combined with the time decay function; the consumption time period analyzes the user's active purchasing time pattern within a day and a week, and is determined by counting the order time distribution of the user's historical purchase orders; the regional cultural characteristics are matched with the consumption culture data of the corresponding region based on the user's registration address and common delivery address, and relevant information is obtained from a pre-constructed regional consumption culture database, and points acquisition tasks and redeemable item price reduction information are pushed to the user according to the personalized model.

6. The system for reducing the price of redeemable items based on digital points according to claim 1, characterized in that: The points loyalty level and differentiated price reduction privilege sub-module in the user flow and retention module are divided according to the user loyalty level. The total amount of user points, the stability of point acquisition, the frequency and amount of points exchanged for items, and the activity index of participating in system activities are comprehensively evaluated. The specific evaluation algorithm is L s =I t ×S i +F i ×A i +E a ×V a , where L s is the loyalty score, S i is the integral stability coefficient, F i is the points redemption frequency, A i is the points redemption amount, E a is the activity participation, V a For the activity participation value coefficient, different levels are divided according to the loyalty score, and differentiated price reduction ranges and special lists of redeemable items are set for users of different levels to attract users to improve their loyalty level and increase consumption and participation depth in the system.

7. The system for reducing the price of redeemable items based on digital points according to claim 1, characterized in that: The system architecture and point transaction processing sub-module in the data processing and system optimization module adopt a distributed microservice architecture, splitting the point transaction processing function into multiple independent microservices, including point acquisition service, point redemption service, and price reduction calculation service. Each microservice is deployed in an independent container and managed and scheduled through container orchestration tools. In high-concurrency scenarios, message queues are used to achieve asynchronous communication and decoupling between microservices. When a large number of users perform point redemption operations at the same time, the point redemption request first enters the message queue and is processed in sequence by the point redemption service. At the same time, cache technology is used to store hot data to reduce database query pressure.

8. The system for reducing the price of redeemable items based on digital points according to claim 1, characterized in that: The real-time dynamic conversion system of points and currency values ​​in the data processing and system optimization module establishes a real-time acquisition module for multi-currency exchange rate data, obtains the latest exchange rate information from the authoritative financial data platform, and combines the market price index data and the merchant's customized profit adjustment coefficient P adj Calculate the conversion ratio of points to currency. The conversion ratio calculation formula is: E r =E f ×P i ×P adj , where E r is the conversion ratio of points to currency, E f is the basic exchange rate factor, P i It is the price index influencing factor, and the basic exchange rate factor is calculated by weighted average based on the international exchange rate fluctuation range and historical exchange rate data of different currencies. The price index influencing factor is determined based on the year-on-year and month-on-month changes of the local market price index.

9. The system for reducing the price of redeemable items based on digital points according to claim 1, characterized in that: When the system is running as a whole, the data interaction between modules follows unified data standards and encrypted transmission protocols. The data standards specify the format, encoding and data structure of points data, user data, item data and transaction data. The encrypted transmission protocol uses an asymmetric encryption algorithm to encrypt and transmit sensitive data including user points balances and transaction amounts.

Images