Target resource value generation and display method and system, computer device and storage medium

Abstract

Embodiments of the present application relate to the field of artificial intelligence, and provide a target resource value generation and display method and system, a computer device and a storage medium, the method comprising: obtaining resource value update rule data, determining multi-dimensional resource value update constraints according to the resource value update rule data; taking the multi-dimensional resource value update constraints as constraint conditions of a target object, and taking maximum resource gain data as a target function, to construct a resource value dynamic optimization model; solving based on reinforcement learning technology to obtain an optimized resource value; adjusting the optimized resource value according to a resource value adjustment operation instruction to generate a target resource value corresponding to the target object; and matching according to object feature information of the target object to obtain a hardware type of a hardware display device. The implementation of the method improves the data configuration efficiency and data display efficiency of the target object resource value.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to a method, system, computer device, and storage medium for generating and displaying target resource values. Background Technology

[0002] During resource configuration, it is necessary to configure and display the resource values ​​of certain target objects. These target objects can be server objects, product objects, etc., and the resource values ​​can be the configured values ​​of computing or storage resources for server objects, or the configured values ​​of price for product objects.

[0003] For example, in the traditional retail industry, traditional pricing and inventory management models are often used to manually configure resource values ​​such as prices for goods. This manual resource configuration method is time-consuming and labor-intensive, resulting in low efficiency in price configuration. In addition, the display of price configuration values ​​also relies on manual methods for formulation and operation, resulting in low efficiency in data display.

[0004] In summary, the efficiency of data configuration and data display for target object resource values ​​is relatively low. Summary of the Invention

[0005] This application provides a method, system, computer device, and storage medium for generating and displaying target resource values. More specifically, this application provides a method, system, computer device, computer storage medium, and computer program product for generating and displaying target resource values, thereby improving the data configuration efficiency and data display efficiency of target resource values.

[0006] In a first aspect, embodiments of this application provide a method for generating and displaying target resource values, including: Obtain resource value update rule data, and determine multi-dimensional resource value update constraints based on the resource value update rule data; Using the multi-dimensional resource value update constraints as the target constraints and maximizing resource gain data as the objective function, a dynamic optimization model for resource values ​​is constructed. The resource numerical dynamic optimization model is solved using reinforcement learning techniques to obtain the optimized resource values ​​associated with the target object. In response to a resource value adjustment operation instruction input by the resource decision-maker based on the instruction input terminal, the optimized resource value is adjusted according to the resource value adjustment operation instruction to generate the target resource value corresponding to the target object; The hardware type of the hardware display device is obtained by performing matching processing based on the object feature information of the target object. The target resource value is displayed in the hardware display device corresponding to the hardware type.

[0007] Secondly, embodiments of this application provide a system for generating and displaying target resource values, which has the function of implementing the target resource value generation and display method provided in the first aspect above. The function can be implemented by hardware or by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the above functions, and the modules can be software and / or hardware.

[0008] In one possible design, the system includes: The constraint generation module is used to acquire resource value update rule data and determine multi-dimensional resource value update constraints based on the resource value update rule data. The model building module is used to construct a dynamic optimization model for resource values, with the multi-dimensional resource value update constraints as the target object and the maximization of resource gain data as the objective function. The model solving module is used to solve the resource numerical dynamic optimization model based on reinforcement learning technology to obtain the optimized resource values ​​associated with the target object. The target resource value generation module is used to respond to the resource value adjustment operation instruction input by the resource decision-maker based on the instruction input terminal, adjust the optimized resource value according to the resource value adjustment operation instruction, and generate the target resource value corresponding to the target object. The hardware type matching module is used to perform matching processing based on the object feature information of the target object to obtain the hardware type of the hardware display device; The target resource value display module is used to display the target resource value in the hardware display device corresponding to the hardware type.

[0009] In another aspect, this application provides a computer device including at least one connected processor and a memory, wherein the memory is used to store program code, and the processor is used to call the program code in the memory to execute the methods described in the above aspects.

[0010] In another aspect, embodiments of this application provide a computer storage medium including instructions that, when executed on a computer, cause the computer to perform the methods described in the above aspects.

[0011] In another aspect, this application provides a computer program product containing instructions that, when run on a computer, cause the computer to perform the methods described in the above aspects.

[0012] Compared to traditional technologies, the technical solution of this application sets multi-dimensional resource value update constraints, achieves the goal of maximizing resource gain data, reduces the human and time costs of manually configuring resource values, improves the rationality of resource value configuration, and integrates object characteristics to continue matching and selecting the hardware type of the hardware display device, thereby improving the data configuration efficiency and data display efficiency of the target object resource values. Attached Figure Description

[0013] Figure 1 This is an application environment diagram from one embodiment; Figure 2 This is a flowchart of one embodiment; Figure 3 This is a structural block diagram of the system in one embodiment; Figure 4 Here is a system architecture diagram from one embodiment; Figure 5 This is an internal structural diagram of a computer device in one embodiment; Figure 6 This is a diagram of the internal structure of a computer device in another embodiment. Detailed Implementation

[0014] The terms "first," "second," etc., used in the embodiments of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or modules is not necessarily limited to those steps or modules explicitly listed, but may include other steps or modules not explicitly listed or inherent to these processes, methods, products, or devices. The division of modules appearing in the embodiments of this application is only a logical division. In actual applications, there may be other division methods. For example, multiple modules may be combined or integrated into another system, or some features may be ignored or not performed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be through some interface, and the indirect coupling or communication connection between modules may be electrical or other similar forms. None of these are limited in the embodiments of this application. Furthermore, the modules or sub-modules described as separate components may or may not be physically separated, may or may not be physical modules, or may be distributed among multiple circuit modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the embodiments of this application.

[0015] Figure 1As an application environment diagram in one embodiment, this application provides a method for generating and displaying target resource values, which can be applied to, for example... Figure 1 In the application scenario shown, terminal 102 communicates with server 104 via a network.

[0016] The terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, etc. The server 104 can be implemented using a standalone server or a server cluster consisting of multiple servers.

[0017] It should be noted that the terminal 102 involved in the embodiments of this application can be a wired terminal or a wireless terminal, and can be a device that provides voice and / or data connectivity to a user, a handheld device with wireless connectivity, or other processing devices connected to a wireless modem. The wireless terminal can communicate with one or more core networks via a wireless access network, and the wireless terminal can be a mobile terminal, such as a mobile phone or a computer with a mobile terminal.

[0018] Figure 2 This is a flowchart illustrating one embodiment, such as... Figure 2 As shown in the embodiments of this application, the method for generating and displaying target resource values ​​includes: S2100: Obtain resource value update rule data, and determine multi-dimensional resource value update constraints based on the resource value update rule data.

[0019] Resource value update rule data refers to the preset rule-related data used to adjust or update the resource values ​​of a target object; multi-dimensional resource value update constraints refer to the conditions that restrict resource value adjustments from multiple aspects. Multi-dimensional resource value update constraints can be obtained by converting various types of resource value update rule data.

[0020] The target object can be a server object, a product object, etc., and the resource values ​​can be the configuration values ​​of the server object's computing resources, the configuration values ​​of the server object's storage resources, the configuration values ​​of the product object's price, etc.

[0021] Taking the target object as the product object and the resource value as the price configuration value as an example, the resource value update rule data can be dynamic price adjustment rule data, including strategies, mechanisms, or rules used for dynamic price adjustments. Correspondingly, multi-dimensional resource value update constraints can be multi-dimensional dynamic price adjustment constraints, referring to conditions that restrict price adjustments from multiple aspects.

[0022] Specifically, the dynamic price adjustment rule data includes data on automatic price reduction strategies for near-expiry products, dynamic pricing data during peak and off-peak seasons, competitor price monitoring and automatic price adjustment data, personalized pricing mechanism data, and intelligent clearance mechanism data for slow-moving goods. Corresponding to the content of the dynamic price adjustment rule data, the multi-dimensional dynamic price adjustment constraints include constraints on automatic price reduction strategies for near-expiry products, dynamic pricing constraints during peak and off-peak seasons, competitor price monitoring and automatic price adjustment constraints, personalized pricing mechanism constraints, and intelligent clearance mechanism constraints for slow-moving goods.

[0023] S2200 uses multi-dimensional resource numerical update constraints as the constraint conditions for the target object and maximizes resource gain data as the objective function to construct a dynamic optimization model for resource numerical values.

[0024] Among them, the constraint condition refers to the rule that limits the solution range of the resource numerical dynamic optimization model; maximizing resource gain data refers to the goal of achieving the highest resource benefit through optimization; the objective function refers to the function used for quantifying the resource gain optimization objective; and the resource numerical dynamic optimization model refers to the analytical model used to solve for the optimal resource value.

[0025] Taking the target object as the commodity object and the resource value as the price configuration value as an example, the resource gain data refers to the profit of the commodity object. Taking maximizing the resource gain data as the objective function means taking maximizing the profit of the commodity object as the objective function.

[0026] S2300 uses reinforcement learning technology to solve the dynamic optimization model of resource values, and obtains the optimized resource values ​​associated with the target object.

[0027] Among them, reinforcement learning technology refers to the technique in the field of artificial intelligence that optimizes the model solution through trial and error iteration; the optimized resource value refers to the optimal resource value reference data obtained by the model solution.

[0028] Taking the product as the target object as an example, this embodiment provides a dynamic optimization algorithm in steps S2200 and S2300. The dynamic optimization algorithm outputs specific price configuration values ​​down to the individual product, store, and time period. Resource decision-makers can directly view the price configuration values ​​provided by the system and use them as price recommendations. The system supports one-click adoption or manual modification of price recommendations.

[0029] This dynamic optimization algorithm integrates constraints such as near-expiry dates, competing products, and slow-moving goods, with profit maximization as the objective function. It uses reinforcement learning to search for the optimal price configuration value, transforming various constraints into executable and precise numbers, reducing manual trial calculations, and improving the efficiency of price configuration value allocation.

[0030] For example, the objective function is that profit equals (price minus cost) multiplied by sales volume minus inventory cost; constraints may include a price range from cost price to the upper limit of market price, inventory greater than 0, and price not significantly higher than competitors; the optimization algorithm adopts a deep reinforcement learning algorithm, with the current inventory, price configuration value, competitor price configuration value, and date as the state, and the actions as increasing the price configuration value by 5%, decreasing it by 5%, or keeping it unchanged, with the daily profit as the reward, allowing the model to autonomously learn the optimal configuration strategy for the price configuration value.

[0031] S2400, in response to the resource value adjustment operation command input by the resource decision-maker based on the command input terminal, adjusts the optimized resource value according to the resource value adjustment operation command, and generates the target resource value corresponding to the target object.

[0032] Among them, the resource decision-maker, also known as the manager or administrator, refers to the relevant entity responsible for making decisions on resource value adjustments.

[0033] Among them, the instruction input terminal refers to the operating device for the decision-maker to input adjustment instructions; the resource value adjustment operation instruction refers to the resource value adjustment instruction issued by the resource decision-maker; taking the target object as the commodity object as an example, the resource value adjustment operation instruction can be an instruction to confirm the price configuration value or an instruction to change the price configuration value.

[0034] The target resource value refers to the final resource value after adjustment.

[0035] S2500 performs matching processing based on the object feature information of the target object to obtain the hardware type of the hardware display device.

[0036] Among them, object feature information refers to the attributes and characteristics of the target object itself, such as resource exchange scenario information.

[0037] Among them, matching processing refers to the operation of adapting object characteristics to hardware type.

[0038] Among them, the hardware type of the hardware display device refers to the specific category of the device used to display numerical values. For example, the hardware type of electronic price tag refers to the specific category of electronic price tags used to display price configuration values.

[0039] S2600 displays the target resource value in the hardware display device corresponding to the hardware type.

[0040] Among them, hardware display devices refer to terminal devices that carry and display the target resource values, including electronic price tags.

[0041] Compared to traditional technologies, this application embodiment first acquires resource value update rule data, determines multi-dimensional resource value update constraints, then uses these multi-dimensional resource value update constraints as the constraint conditions for the target object, and uses maximizing resource gain data as the objective function to construct a dynamic optimization model for resource values. This model is then solved using reinforcement learning techniques to obtain optimized resource values. Next, adjustments are made according to resource value adjustment operation instructions to generate target resource values. Finally, matching processing is performed based on the object feature information of the target object, and the target resource values ​​are displayed on the hardware display device corresponding to the matched hardware type. The technical solution of this application embodiment sets multi-dimensional resource value update constraints, achieves the goal of maximizing resource gain data, reduces the human and time costs of manual resource value configuration, improves the rationality of resource value configuration, and integrates object features to further match and select the hardware type of the hardware display device, thereby improving the data configuration efficiency and data display efficiency of the target object's resource values.

[0042] In one embodiment, a resource value update rule data can be converted into a resource value update constraint. Specifically, taking a product object as the target object as an example, a dynamic price adjustment rule data can be converted into a dynamic price adjustment constraint. The rule content displayed in the dynamic price adjustment rule data, and the specific meaning of the dynamic price adjustment constraint, are as follows.

[0043] (1) The strategy data for automatic price reduction of near-expiry goods includes four parts: First, there is a near-expiration detection mechanism. When goods are put into storage, the product ID, product name, production date, shelf life, expiration date and other shelf life information must be entered. The system automatically scans the inventory every morning at midnight, calculates the number of days left until the product expires, and triggers a near-expiration warning when the threshold is reached.

[0044] Secondly, a tiered price reduction rule is implemented, dividing the product into six stages based on its remaining time before expiration and applying corresponding price adjustment strategies. When the product is more than 50% away from expiration, it is sold at the original price. When it is 30% to 50% away from expiration, a 10% discount is applied for early promotion. When it is 10% to 30% away from expiration, an 80% discount is applied for accelerated clearance. When it is 7-10 days away from expiration, a 50-70% discount is applied for a near-expiration special price. When it is 3-7 days away from expiration, a 30-50% discount is applied for emergency clearance. When it is 1-3 days away from expiration, the product is sold at cost price or even lower for extreme clearance. The price tags for different stages display corresponding promotional information.

[0045] Thirdly, a product category differentiation strategy is adopted, with different price reduction strategies implemented for different types of products with different shelf lives. Fresh produce with a shelf life of 1-7 days has more aggressive price reductions, with discounts of 20%, 50%, and 30% respectively at 8 pm, 9 pm, and 10 pm on the same day. Dairy products / bread with a shelf life of 7-30 days have standard tiered price reductions, while snacks / beverages with a shelf life of 6-12 months have more moderate price reductions, with discounts of 10%, 30%, and 50% respectively when there are 3 months, 1 month, and 1 week left before the expiration date.

[0046] Fourthly, intelligent inventory allocation: after the system detects products nearing their expiration date, it will determine the inventory status of the product in other stores. If only this store has the product, it will directly implement a price reduction and clearance sale. If other stores also have the product and it is about to expire, the entire chain will uniformly implement a price reduction and clearance sale. If other stores have fresh inventory, it is recommended to transfer the product. While clearing out products nearing their expiration date at a price reduction, fresh products will be transferred from other stores to replenish the inventory.

[0047] During the optimization phase, the above-mentioned automatic price reduction strategy data for near-expiry goods needs to be transformed into automatic price reduction strategy constraints for near-expiry goods. For example, the automatic price reduction strategy constraints for near-expiry goods can be: the price of near-expiry goods ∈ the first preset price range, where the first preset price range refers to the price range determined after the near-expiry goods are reduced by the automatic price reduction strategy data for near-expiry goods.

[0048] The mechanism of this constraint is as follows: if a product is identified as near-expiry, and its price does not meet the price constraint of the price reduction for near-expiry products, then the price of the near-expiry product will not be accepted, that is, the specific price will not be output as the recommended price.

[0049] (2) The peak-and-valley dynamic pricing rules in the peak-and-valley dynamic pricing data include three parts: First, peak passenger flow identification is performed by analyzing historical data based on time period, weekday, holiday, and weather to identify peak periods such as 7:00-9:00 am, 11:30-13:00 pm, 17:30-20:00 pm on weekdays, and 14:00-18:00 pm on weekends. At the same time, an LSTM time series passenger flow prediction model is used to output hourly passenger flow for the next 24 hours by taking historical passenger flow, date, weather, and promotional activities as input.

[0050] Second, a dynamic pricing strategy is implemented. During peak periods, when real-time customer flow exceeds the predicted flow by 120% and there are more than 5 people in the queue, the prices of non-essential items other than the traffic-driving items will be moderately increased by 5% to 10%, with higher-profit-margin items being prioritized for price increases. Price tags will only display the current price. During off-peak periods, when real-time customer flow is less than the predicted flow by 50% and there is sufficient inventory, the prices of high-margin items will be reduced by 5% to 15% and combined with a buy-A-get-B-free promotion. Price tags will indicate the corresponding special offer information and the limited-time period.

[0051] Thirdly, there is a time-based pricing calendar, which presets time-based prices that can be dynamically covered by artificial intelligence. From 06:00 to 09:00, breakfast items are discounted by 10%; from 11:00 to 13:00, lunch items are sold at the original price; from 15:00 to 17:00, coffee and beverages are discounted by 10%; and from 19:00 to 21:00, fresh produce is discounted by 20%. Prices are relatively stable throughout the day on weekends with few adjustments. On holidays, due to high demand, some items may see slight increases or remain at the original price.

[0052] During the optimization phase, the aforementioned peak-and-valley dynamic pricing data needs to be transformed into peak-and-valley dynamic pricing constraints. For example, the peak-and-valley dynamic pricing constraints can be: the price of a preset commodity during a preset time period ∈ a second preset price range, where the second preset price range is the price range determined after the commodity is reduced by the peak-and-valley dynamic pricing rules.

[0053] The mechanism of this constraint is as follows: if the price of a certain product during a certain period does not conform to the price constraint after dynamic pricing based on peak and trough, then the price of the product will not be accepted, that is, the specific price will not be output as the recommended price.

[0054] (3) The competitor price monitoring and automatic price adjustment rules in the competitor price monitoring and automatic price adjustment data include four parts: competitor price data collection, product matching algorithm, intelligent price tracking strategy, and price war early warning.

[0055] Competitor price data collection includes two methods: online web crawling and (Method A) offline manual collection (Method B).

[0056] Method A involves web scraping of online competitors (e-commerce / O2O platforms). This is achieved through a scheduled task every hour, which scrapes the product prices of competitors' websites / apps, parses the product names, specifications, and price information, and matches them with the corresponding products in our store. At the same time, we use proxy IP pools, request frequency control, User-Agent rotation, and CAPTCHA recognition (OCR / captcha solving platform) to deal with anti-scraping restrictions.

[0057] Method B involves manual data collection from offline competitors, which includes three specific options: Option 1 involves store inspectors taking photos and recording prices at competitor stores and uploading them to the system for manual entry; Option 2 involves encouraging customers to take photos and upload competitor prices and awarding points after verification; and Option 3 involves collaborating with a third-party price monitoring platform to purchase competitor price data services.

[0058] The two methods of collecting competitor pricing data are not simply allocated proportionally, but rather paired as complementary nodes in a "data chain": web crawling serves as the "high-frequency benchmark," while manual collection handles "low-frequency calibration and blind spot filling," ultimately transforming the two data streams into a single "credible price" for the pricing engine. This combination, through a mechanism of "web crawling for broad coverage, manual verification for in-depth validation, and discrepancy-triggered review," ensures that 99% of SKUs have real-time online prices while also achieving manual physical verification of all prices at least once every 7 days, thereby keeping the error rate below 0.3%.

[0059] Before collecting competitor pricing data, it's crucial to identify the competitors. The competitor identification process follows a core path of "defining the business scope and precise AI matching," ensuring high recall and accuracy through five verification stages. First, a professional team develops a benchmark list, clearly defining the platforms, stores, brands, and categories to monitor, mitigating the risk of unauthorized web scraping. Then, dedicated access parameters are used to stably capture competitor data from the list. Next, the captured information undergoes three layers of verification: text semantics, product specifications, and image feature similarity comparison. Products meeting these criteria are marked as highly reliable competitors. Then, the qualified results are manually sampled, and feedback on the results is provided. Finally, anomaly monitoring is conducted, with a secondary verification for significant price reductions by competitors to avoid misjudgments.

[0060] By following the steps above, relevant information about competing products corresponding to a certain product is determined, and competitor identification is completed. This facilitates the use of competitors to execute subsequent product matching algorithms, intelligent price tracking strategies, and price war warnings, so as to achieve subsequent price adjustments.

[0061] The product matching algorithm is designed to address the issue of discrepancies between competing product names and specifications. The algorithm sequentially performs brand identification, specification identification, and category identification, then uses NLP semantic similarity calculation. When the similarity exceeds 80%, the products are considered the same and their prices are compared.

[0062] The intelligent price-following strategy includes three types: mainstream products follow the rule of not exceeding the mainstream level of competitors, adjusting prices to be slightly lower than or equal to the median price to avoid blindly competing on the lowest price; differentiated products can be priced at a premium of 5% to 10% based on their advantages in quality, service, and brand, and their selling points can be displayed on the price tag; regional pricing is adjusted according to the store location, with prices slightly higher (5%) in high-end community stores and low-price competitiveness in ordinary community or suburban stores.

[0063] The price war warning system detects when competitors significantly reduce prices by more than 20%, determines whether it is a promotional activity or the start of a price war, and then sends a warning notification to the operations staff. The operations staff then decide whether to follow up with price adjustments, maintain the original price to emphasize differentiation, or temporarily refrain from taking action and monitor sales changes.

[0064] During the optimization phase, the aforementioned competitor price monitoring and automatic price adjustment data needs to be transformed into competitor price monitoring and automatic price adjustment constraints. For example, the competitor price monitoring and automatic price adjustment constraints can be: the price of the product associated with the competitor is ∈ a third preset price range, where the third preset price range refers to the price range after adjustment based on the competitor price monitoring and automatic price adjustment rules and the actual price of the competitor.

[0065] The mechanism of this constraint is as follows: if the price of a product associated with a competitor does not conform to the price constraint adjusted based on the actual price of the competitor, then the pricing of that product is not accepted, meaning that the specific pricing will not be output as the recommended price.

[0066] (4) The principle of the personalized pricing mechanism in the data is to feed back the fact of "the user identity of the currently browsing product" to the price engine in real time, and then present different prices to different users on the same product and the same shelf through electronic price tags, APP, and mini-program. The core is to embed "user identity" into the price query interface, send back "dynamic exclusive price" and ensure compliance and transparency.

[0067] The specific process can be as follows: the electronic price tag only displays the "basic price" visible to everyone, while a "member code" area is added in the lower right corner; after the user scans the code with their mobile phone, they are redirected to a mini program, which automatically carries the WeChat or Alipay user identifier. The backend price service generates and returns an "exclusive price" based on the member level and price sensitivity model, which is finally displayed on the user's mobile phone.

[0068] Personalized pricing mechanisms consist of three parts: First, user segmentation is performed based on user data across four dimensions: Dimension 1 is membership level, divided into new users, regular members, silver card members, gold card members, and diamond card members; Dimension 2 is price sensitivity, divided into high sensitivity, medium sensitivity, and low sensitivity; Dimension 3 is purchase frequency, divided into high frequency, medium frequency, and low frequency; and Dimension 4 is average order value, divided into high average order value, medium average order value, and low average order value.

[0069] Second, differentiated pricing rules are set up for three application scenarios. Scenario A is for attracting new users, offering exclusive prices for new users, discount coupons, and free delivery for the first order. New users can scan the code to view the exclusive price. Scenario B is for member-exclusive prices, with higher membership levels offering larger discounts. The electronic price tag displays the base price on the front, and members can scan the code to view the dynamic exclusive price. Scenario C is for targeted discounts for price-sensitive users, pushing exclusive discounts to win back lost users.

[0070] Thirdly, compliance assurance requires attention to three legal risk prevention principles. Principle 1 is compliance with price discrimination: pricing based on membership level and spending amount is legal, while pricing based on new and old users should be approached with caution. The solution is to clearly indicate the scope of discounts for new users and provide exclusive discounts for old users. Principle 2 is price transparency: electronic price tags should display the base price visible to everyone, and personalized prices should be viewed by scanning a QR code in the app with the reason for the discount clearly stated. Principle 3 is informed consent from users: the privacy policy should clearly state the use of data, and users should be able to opt out of personalized pricing.

[0071] During the optimization phase, the aforementioned personalized pricing mechanism data needs to be transformed into personalized pricing mechanism constraints. For example, a personalized pricing mechanism constraint could be: the price of goods to be purchased by users belonging to a preset user segment is ∈ the fourth preset price range, where the fourth preset price range refers to the price range of goods determined based on the personalized pricing mechanism and user segmentation results.

[0072] The mechanism of this constraint is as follows: if the price of a product that a user belonging to a certain user group wants to purchase does not meet the price constraint corresponding to the personalized pricing mechanism, then the price of the product will not be accepted, that is, the price will not be output as the recommended price.

[0073] (5) The intelligent clearance mechanism for slow-moving goods in the data includes four parts: The first step is to identify slow-moving goods. This is judged from four dimensions: Indicator 1 is inventory turnover rate, calculated as: Turnover rate = Sales volume / Average inventory. A turnover rate greater than 0.5 is considered normal, and a turnover rate less than 0.2 is considered slow-moving. Indicator 2 is sales trend. A sales decline of more than 30% in the past 30 days compared to the previous 30 days indicates a decline in demand. Indicator 3 is inventory backlog. Current inventory is more than 60 times the average daily sales in the past 30 days, indicating excessive inventory. Indicator 4 is seasonality. Off-season goods need to be cleared out in advance. If two or more indicators are met, the product is marked as slow-moving.

[0074] Secondly, a tiered clearance strategy is adopted, with different strategies depending on the severity of the slow-moving inventory. Mildly slow-moving goods (inventory turnover between 30 and 60 days) are subject to a mild 10% discount without special promotion. Moderately slow-moving goods (inventory turnover between 60 and 90 days) are subject to a 20-30% discount for accelerated clearance, combined with price tag labeling, end-of-shelf display, and promotional information pushed through the APP. Severely slow-moving goods (inventory turnover greater than 90 days) are subject to an extreme clearance sale with a 50% discount or lower, combined with price tag labeling, special display areas, discounts for purchases over a certain amount or buy-one-get-one-free offers, and employee purchase incentives.

[0075] Thirdly, there is the combined promotional strategy. When slow-moving products are difficult to clear out on their own, three strategies are adopted: selling them with best-selling products, bundling them together, and offering discounts for purchases over a certain amount.

[0076] Fourth, channel clearance: when offline clearance is not effective, expand to three channels: online special sales, wholesale disposal, and charitable donations.

[0077] During the optimization phase, the data of the above-mentioned intelligent clearance mechanism for slow-moving goods needs to be transformed into constraints for the intelligent clearance mechanism for slow-moving goods. For example, the constraints for the intelligent clearance mechanism for slow-moving goods can be: the price of slow-moving goods ∈ the fifth preset price range, where the fifth preset price range refers to the price range of the goods under the clearance strategy determined by the intelligent clearance mechanism for slow-moving goods.

[0078] The mechanism of this constraint is as follows: if the price of a certain slow-moving product does not meet the price constraint determined by the slow-moving product intelligent clearance mechanism, then the price of the product will not be accepted, that is, the price will not be output as the recommended price.

[0079] Optionally, in some embodiments of this application, before the step of adjusting the optimized resource value according to the resource value adjustment operation instruction input by the resource decision-maker based on the instruction input terminal to generate the target resource value corresponding to the target object, the method further includes: making a prediction based on the demand forecasting model to generate an object delivery scale prediction result associated with the target object; obtaining the resource value elasticity analysis result associated with the target object; and displaying the object delivery scale prediction result and the resource value elasticity analysis result so that the resource decision-maker can input the resource value adjustment operation instruction in the instruction input terminal according to the displayed results.

[0080] Among them, the demand forecasting model refers to the analytical model used to predict the delivery scale of the target object; the object delivery scale prediction result refers to the object delivery volume prediction data output by the model; specifically, the object delivery scale prediction result can be the prediction data of the future sales volume of the product object.

[0081] Among them, the resource numerical elasticity analysis results are used to reflect the impact of resource value changes on the delivery of the target object; specifically, the resource numerical elasticity analysis results can be the price elasticity analysis results.

[0082] For example, by Figure 2 The technical solution of the corresponding embodiment can realize the dynamic adjustment of resource values ​​(such as price configuration values) and output a preliminary recommended result (i.e., optimized resource values) to the resource decision-maker. Then, the object delivery scale prediction result and resource value elasticity analysis result obtained in this embodiment are also given to the resource decision-maker as data analysis suggestions to assist the resource decision-maker in making decisions. After the resource decision-maker decides whether to modify this recommended result, the final result (i.e., target resource value) is obtained and displayed.

[0083] Taking the target object as the product object as an example, the system first provides dynamically recommended price configuration values, then pushes the predicted data of future sales and the price elasticity analysis results to the manager, who decides whether to adopt or fine-tune them, and finally displays them.

[0084] In this embodiment, the prediction results of object delivery scale and the results of resource numerical elasticity analysis are used to assist decision-making, thereby improving the scientific nature of resource numerical adjustments and the rationality of decisions.

[0085] Optionally, in some embodiments of this application, prediction based on a demand forecasting model to generate a predicted object delivery scale for the target object includes: constructing a demand forecasting model by combining a time series model with a machine learning model; inputting the object features and environmental data associated with the target object as input features into the demand forecasting model, and outputting the predicted object delivery scale for the target object corresponding to the input features.

[0086] Among them, the time series model refers to a model that predicts trends based on the time-series patterns of data; the machine learning model refers to a model that achieves autonomous learning and prediction through data training; and the demand forecasting model refers to a model that combines the above two models to predict the delivery scale of a target object. The delivery scale of the target object can be the sales volume data of the product object.

[0087] Among them, object characteristics and environmental data refer to the attributes of the target object itself and the data of the external environment; specifically, object characteristics and environmental data include data related to the product object such as historical sales, price, promotional activities, date characteristics, weather, competitor prices, and inventory levels.

[0088] Here, model input features refer to the feature data input into the model to support the prediction calculation; object delivery scale prediction results refer to the object delivery volume prediction data output by the model.

[0089] For example, this embodiment provides a demand forecasting model, which uses artificial intelligence to predict the delivery scale of objects (such as future sales volume data) to assist in the configuration decision of price configuration values.

[0090] More specifically, the model input features include historical sales volume, price, promotional activities, date features, weather, competitor prices, and inventory levels. Historical sales volume refers to daily sales over the past 90 days, and date features include weekdays, holidays, and seasons. Weather factors affect the sales of fresh produce and beverages. The model algorithm combines time series models and machine learning models. The time series models include ARIMA and Prophet, while the machine learning models include XGBoost and LSTM. The model output is the expected daily sales volume for the next 7 days. Taking a certain beverage as an example, if the current price is ¥5, the predicted sales volume for tomorrow is 200 bottles. If the price is reduced to ¥4.5, the predicted sales volume for tomorrow is 280 bottles, representing a 40% increase in sales. The future sales prediction results will be displayed to managers to facilitate price adjustment decisions based on the forecast.

[0091] In this embodiment, a demand prediction model is constructed by fusing different models and inputting multi-dimensional features, which improves the accuracy of object delivery scale prediction, enhances the applicability of resource numerical allocation, facilitates decision support, and improves the efficiency of resource numerical allocation.

[0092] Optionally, in some embodiments of this application, obtaining the resource numerical elasticity analysis results associated with the target object includes: obtaining the percentage change data of the object delivery scale of the target object and the percentage change data of the resource value; determining the resource numerical elasticity coefficient of the target object based on the percentage change data of the object delivery scale and the percentage change data of the resource value; and determining the resource numerical elasticity analysis results based on the resource numerical elasticity coefficient.

[0093] Among them, the percentage change data of object delivery scale refers to the quantitative data of the change in object delivery volume; for example, the percentage change in sales volume; the percentage change data of resource value refers to the quantitative data of the change in resource value; for example, the percentage change in price.

[0094] Among them, the resource numerical elasticity coefficient is used to characterize the sensitivity of the delivery scale of an object to changes in resource value. The resource numerical elasticity coefficient can be the price elasticity coefficient.

[0095] Among them, the results of the resource numerical elasticity analysis are used to reflect the conclusions on the correlation between resource values ​​and object delivery scale.

[0096] For example, this embodiment provides a price elasticity analysis method to analyze the impact of price changes on sales volume. The core calculation method is that the price elasticity coefficient equals the percentage change in sales volume divided by the percentage change in price. Taking a certain product as an example, if the original price is ¥10 and the sales volume is 100 units, and the price is reduced to ¥9 (a 10% price decrease), the sales volume increases to 130 units (a 30% increase). The calculated price elasticity of this product is 3, classifying it as a high-elasticity product, and the system recognizes the price reduction as effective. Based on this, the resource value elasticity analysis results (i.e., configuration suggestions regarding price allocation values) are given: "For high-elasticity products with an elasticity greater than 2, price reductions have a good promotional effect; for low-elasticity products with an elasticity less than 1, price reductions have an insignificant effect, and it is recommended to maintain the price."

[0097] In this embodiment, the elasticity coefficient is determined by quantifying the relationship between delivery scale and resource value changes, which improves the scientific nature of resource allocation recommendations and enhances the adaptability of resource allocation values.

[0098] Optionally, in some embodiments of this application, the object feature information includes resource exchange scenario information, total cost of ownership value, and object feature score value. Matching processing is performed based on the object feature information of the target object to obtain the hardware type of the hardware display device, including: obtaining resource exchange scenario information corresponding to the target object; generating a total cost of ownership value based on the total cost of ownership model for each target object's inventory unit; performing a weighted score based on the object features of the target object to obtain an object feature score value; and determining the hardware type of the hardware display device based on the resource exchange scenario information, total cost of ownership value, and object feature score value.

[0099] Among them, resource exchange scenario information refers to information related to the scenario in which the target object exchanges resources; resource exchange scenario information includes business scenario information such as gas stations, tea stalls, bakery counters, fast food cashier islands, and member day flash sale walls.

[0100] Among them, the Stock Keeping Unit (SKU) refers to the basic unit for inventory management of the target object; the Total Cost of Ownership model refers to the model used to calculate the comprehensive cost of the target object; and the Total Cost of Ownership value refers to the comprehensive cost data calculated through the cost model, which can be denoted as TCO.

[0101] Among them, object features refer to the attribute characteristics of the target object itself; weighted scoring refers to the evaluation method of calculating the comprehensive score according to the feature weights; and object feature score value refers to the quantitative score obtained after weighted scoring. Among them, hardware display devices refer to terminal devices used to display target resource values, such as electronic price tags; hardware type refers to the category of hardware display devices according to their attributes, such as the type of electronic price tags.

[0102] Specifically, the types of electronic shelf labels include two categories: mainstream e-ink screen shelf labels and high-end backlit electronic shelf labels (including LED / LCD shelf labels). This application selects the hardware type in the following manner.

[0103] For example, the specific decision-making process for each product's inventory unit is as follows: First, for any scenario requiring "instant price changes, color posters, video animations, and instant QR code displays" (such as gas stations, tea stalls, bakery counters, fast food checkout areas, and member day flash sale walls), LED / LCD price tags will be mandatory; other scenarios will proceed to the next step for scoring.

[0104] Then, each SKU is input into the configured Total Cost of Ownership (TCO) model to obtain the model's output. The model's output principle is: Total Cost of Ownership (TCO) = Hardware Cost + Electricity Cost + Battery / Wiring Installation + Labor Maintenance - Gross Profit Margin Improvement Revenue. Based on the calculated TCO, a judgment is made. Specifically, if the difference between the TCO of LED / LCD and e-ink screen is ≤12 months of payback period, and the store's average daily price change frequency is ≥8 times, then LED / LCD is allowed; otherwise, e-ink screen is the default.

[0105] Finally, among the SKUs that have passed the economic threshold designed in the above steps, a weighted score is then applied based on product characteristics. For example, short-shelf-life fresh food SKUs (price changes > 5 times / day) receive 2 points; high-margin impulse-buy beverage and snack SKUs receive 1 point; and low-margin standard products SKUs (such as mineral water) receive 1 point less. This yields a score for product characteristics. Based on the score, if the total score is ≥ 1 and the store's LED power supply wiring can cover it, then LED / LCD is selected; otherwise, e-ink screens are still used.

[0106] Once the hardware type is determined, the system will tag each SKU. Subsequently, the pricing strategy engine will only enable features such as "flash sales" and "animated posters" for SKUs that are eligible for LCD displays, thereby linking hardware selection with dynamic pricing strategies.

[0107] In this embodiment, by matching the hardware type of the hardware display device with multi-dimensional information, the accuracy of hardware selection is improved, the overall cost of device use is reduced, the compatibility between the target resource value and the hardware is enhanced, and the technical effect of efficiently displaying the target resource value is achieved.

[0108] Optionally, in some embodiments of this application, the method further includes: upon detecting that a target object has been added to the database, recording the quality period information of the target object; calculating the remaining time information of the quality period of the target object based on the quality period information; determining the near-expiration judgment result of the target object based on the remaining time information of the quality period; and updating the target resource value using a tiered update strategy based on the near-expiration judgment result.

[0109] Among them, the quality period information refers to the effective period for the target object to maintain the specified quality, such as the shelf life; the remaining quality period information refers to the time from the current time of the target object to the end of the quality period, such as the number of days until the product expires; the near-expiration judgment result is used to reflect the judgment conclusion on whether the target object is in a near-expiration state.

[0110] Among them, the tiered update strategy refers to the strategy of adjusting resource values ​​according to their degree of impending expiration, such as the tiered price reduction strategy.

[0111] For example, this embodiment provides a price reduction process for near-expiry goods. When goods are received into the warehouse, their expiration dates are recorded. The system scans the inventory daily and calculates the number of days until each item expires, then determines whether the item is nearing its expiration date: if the item is not nearing its expiration date, the original price is maintained; if the item is nearing its expiration date, a tiered price reduction strategy is implemented: 20% off for items 7-10 days from expiration, 50% off for items 3-7 days from expiration, and a clearance sale for items 1-3 days from expiration. The system automatically sends the new prices to electronic price tags, which are updated within 3 seconds.

[0112] In this embodiment, by tracking the quality deadline and performing tiered resource value updates, the turnover efficiency of near-expiration target objects is improved, loss costs are reduced, and the flexibility of resource allocation is enhanced.

[0113] Optionally, in some embodiments of this application, the method further includes: obtaining the competitive resource values ​​of the competitor objects associated with the target object based on web crawling technology; determining the pairing result between the target object and the competitor objects based on a product matching algorithm based on natural language processing; and updating the target resource value according to the competitive resource values ​​of the competitor objects and the target resource value of the target object if the pairing result is successful.

[0114] Among them, web crawling technology refers to the technical means of automatically crawling the relevant resource values ​​of competitors; competitors refer to similar objects that have a market competition relationship with the target object; and competitor resource values ​​refer to the quantifiable resource indicators corresponding to competitors.

[0115] Among them, the product matching algorithm of natural language processing refers to the algorithm that achieves object matching based on natural language processing; the matching result is used to reflect the judgment of whether the target object and the competitor object are successfully matched.

[0116] For example, this embodiment provides a competitor monitoring and automatic price adjustment process. The competitor monitoring and automatic price adjustment process is started by a scheduled task once per hour. First, competitor prices are crawled, then a product matching algorithm based on natural language processing is used to complete product matching, followed by price comparison analysis; then it is determined whether the store's price needs to be adjusted. If the store's price is reasonable, it remains unchanged; if the competitor's price is lower, the corresponding price-following strategy is executed.

[0117] In this embodiment, by crawling and capturing competitor resource values ​​and completing object matching, the accuracy of competitor monitoring is improved, the timeliness and rationality of resource value adjustments are enhanced, thereby improving the efficiency of updating and displaying target resource values.

[0118] The technical research process and other technical details of this application are described below with reference to a specific embodiment.

[0119] In traditional technology, the retail industry mostly adopts traditional pricing and inventory management models. Goods are subject to fixed pricing strategies, and prices remain stable in the long term, relying solely on manual periodic adjustments. Price tags are still in paper form, and replacing them requires a lot of manpower and time. Promotional activities are also relatively simple and crude, with slow-moving goods being judged manually and discounted uniformly. Inventory checks are mainly conducted periodically, which not only fails to achieve real-time updates of inventory status but also makes it easy for near-expiry goods to be difficult to identify in a timely manner, leading to a series of operational problems.

[0120] Problems with traditional technologies include: (1) Rigid pricing, specifically including: fixed prices cannot cope with market changes (competitor price reductions / supply and demand fluctuations); missing the best sales opportunities (holidays / peak periods); long price adjustment cycles (usually monthly / quarterly adjustments); low efficiency and easy error in manual price adjustments.

[0121] (2) Serious inventory losses, including: failure to discover near-expiry goods in time, resulting in expired and scrapped goods; stockpiling of slow-moving goods tying up capital and storage space; loss rate of 5% to 15% in the fresh / food industry; annual losses of hundreds of thousands to millions of yuan.

[0122] (3) The pricing lacks a basis, including: pricing based on experience, which is not scientific enough; pricing blindly without understanding the prices of competitors; not considering inventory status and not clearing inventory when it should be cleared; not considering demand elasticity, and price reduction does not necessarily increase sales.

[0123] (4) Poor consumer experience, including: opaque pricing and delayed discount information; no differentiated membership benefits; missing out on discounted items near their expiration date (not knowing which items are on sale); and limited price tag information (only price, no promotional details).

[0124] (5) High operating costs, including: time-consuming manual replacement of price tags (1-2 days for more than 1,000 items); price tag printing costs (paper / manpower); high price error rate (manual errors are easy to make); difficulty in implementing complex promotional strategies (discounts for purchases over a certain amount / tiered pricing).

[0125] Overall, the traditional pricing mechanisms used in the existing retail industry are rigid, slow to respond to pricing, resulting in high inventory losses, poor consumer experience, and pricing methods that cannot adapt to the dynamic market environment.

[0126] Based on this, this application provides a method for generating and displaying target resource values, as detailed below.

[0127] The method for generating and displaying target resource values ​​in this application involves the fields of smart retail, artificial intelligence, and big data analysis. Specifically, it involves the field of intelligent pricing of goods based on demand forecasting, inventory status, and competitive analysis. By monitoring competitors, forecasting demand, and segmenting users to build multi-dimensional price adjustment constraints, and combining reinforcement learning to construct a profit maximization optimization model, it outputs accurate prices, which can effectively reduce product loss and labor costs, accelerate inventory turnover, and achieve refined operation and efficiency improvement.

[0128] The advantage of the method for generating and displaying the target resource values ​​in this application is that: (1) Losses are significantly reduced. The loss rate of near-expiry or expired goods in the traditional scheme is 5% to 15%. This scheme reduces the loss rate to 2% to 5% by automatically lowering prices and clearing inventory in advance. Medium-sized supermarkets can save 500,000 to 1 million yuan in loss costs per year.

[0129] (2) Inventory turnover is accelerated. The traditional solution requires an average inventory turnover of 30-45 days. This solution relies on the intelligent clearance of slow-moving products to shorten the turnover time to 20-30 days, reduce capital occupation by 30%, and improve cash flow.

[0130] (3) Improved price competitiveness: Traditional solutions are prone to missing competitive opportunities due to lagging price adjustments. This solution can monitor competitors in real time and achieve hourly response, helping to increase customer traffic by 15% to 25%.

[0131] (4) Gross profit margin optimization: Traditional schemes use fixed pricing, which may lead to missed opportunities to raise prices during peak periods. This scheme can intelligently raise prices slightly during peak periods and carry out promotions during off-peak periods, thereby increasing the overall gross profit margin by 2-5 percentage points.

[0132] (5) Improved operational efficiency: Traditional solutions require 1-2 days to manually replace price tags on more than a thousand items, while this solution can complete batch updates in 3 minutes with the help of electronic price tags, saving 60% to 80% of labor costs and with an error rate close to zero.

[0133] (6) Improved user experience: Personalized pricing allows members to receive exclusive discounts, increasing member loyalty by 40%. At the same time, transparent pricing of near-expiry products allows users to buy discounted near-expiry products, increasing satisfaction by 35%.

[0134] It should be noted that any technical feature in any of the above embodiments provided in this application is also applicable to any of the following embodiments provided in this application, and similar details will not be repeated hereafter.

[0135] Figure 3 Here is a structural block diagram of the system in one embodiment, with reference to Figure 3 The system for generating and displaying target resource values ​​includes: The constraint generation module 301 is used to acquire resource value update rule data and determine multi-dimensional resource value update constraints based on the resource value update rule data. The model building module 302 is used to construct a dynamic optimization model of resource values ​​with constraints that take multi-dimensional resource value update constraints as the target object and the objective function that maximizes resource gain data. The model solving module 303 is used to solve the resource numerical dynamic optimization model based on reinforcement learning technology to obtain the optimized resource values ​​associated with the target object; The target resource value generation module 304 is used to respond to the resource value adjustment operation command input by the resource decision-maker based on the command input terminal, adjust the optimized resource value according to the resource value adjustment operation command, and generate the target resource value corresponding to the target object. The hardware type matching module 305 is used to perform matching processing based on the object feature information of the target object to obtain the hardware type of the hardware display device; The target resource value display module 306 is used to display the target resource value in a hardware display device corresponding to the hardware type.

[0136] In this embodiment of the application, based on, as follows Figure 3 The connections between the various modules or units shown in the document improve the efficiency of data configuration and data display of the target object's resource values ​​through the cooperation between these modules or units.

[0137] Figure 4 The system architecture diagram in one embodiment shows that the target resource value generation and display system provided in this application has a four-layer architecture from top to bottom. The core is the cloud pricing system, which contains five major functional modules: demand forecasting module, dynamic pricing engine, competitor monitoring module, membership management system, and inventory management system. The cloud pricing system is connected to the store server (or gateway), electronic price tag base station (also known as base station AP) in sequence, and finally connected to the electronic price tag terminal to form a complete instruction transmission link.

[0138] Each base station (AP) can cover 50-100 price tags, and each product corresponds to one electronic price tag terminal. Data transmission between different levels is achieved through wireless communication. The cloud-based pricing system can update prices in batches, and the electronic price tag terminal is used to display the latest prices in real time.

[0139] A product price data generation system based on dynamic pricing requires the deployment of an electronic price tag system, as detailed below: In terms of hardware type selection, there are two categories: mainstream e-ink screen price tags and high-end backlit electronic price tags (including LED / LCD price tags).

[0140] E-ink price tags utilize e-paper display technology, ranging in size from 2.13 inches to 7.5 inches, with resolutions from 250×122 to 800×480 pixels. Full refresh speeds are 1-3 seconds, and partial refresh speeds are less than 1 second. They feature extremely low power consumption and a 3-5 year battery life with a button cell battery. Colors available include black and white, black and white with red, and black and white with yellow. Advantages include a wide viewing angle (170°), readability in sunlight, power saving, and a paper-like visual texture. Disadvantages include a slow refresh rate, making them unsuitable for instant price changes, and they do not support video or animation.

[0141] LED / LCD price tags use LED or LCD display technology, with sizes ranging from 3 to 10 inches. They offer millisecond-level real-time refresh rates and support color, video, and animation. Their advantages include rich display content, real-time price changes, and vibrant colors. However, their disadvantages include higher power consumption, the need for continuous power supply, and costs that are 3 to 5 times higher than e-ink price tags.

[0142] Regarding the communication method, the wireless communication protocol of the electronic price tag system to be deployed in this application includes three options: Option A is 2.4G wireless, with a frequency band of 2.4GHz and a communication distance of 100-200 meters. It has the advantages of being mature, stable, and low-cost, but its anti-interference capability is average. Option B is Sub-1G (433MHz / 868MHz / 915MHz), with a communication distance of 300-500 meters. Its advantages are strong penetration, anti-interference, and low power consumption, but its disadvantage is relatively small bandwidth. Option C is Bluetooth Mesh, with a single-hop communication distance of 10-30 meters and a multi-hop mode that can cover the entire store. Its advantages are standardization and easy expansion, but its disadvantage is that it requires dense deployment of nodes.

[0143] In summary, the system recommends using the Sub-1G protocol in large supermarkets and the 2.4G protocol in convenience stores.

[0144] In another embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 5As shown, it includes a processor, memory, input / output interfaces, and a communication interface. The processor, memory, and input / output interfaces are connected via a system bus, and the communication interface is connected to the system bus via the input / output interfaces. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database stores relevant data. The input / output interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. The computer program can be executed by the processor to implement the various methods described in the above embodiments.

[0145] In yet another embodiment, a computer device is provided, such as a terminal, whose internal structure diagram may be as follows: Figure 6 As shown, it includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. The computer program can be executed by the processor to implement the various methods described in the above embodiments.

[0146] Those skilled in the art will understand that Figure 5 and Figure 6 The structure shown is only a block diagram of a part of the structure related to the solution of this application, and does not constitute a limitation on the computer device on which the solution of this application is applied. It may also include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements, in order to realize the function of the terminal or server.

[0147] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0148] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the systems, devices, equipment, modules or units described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0149] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, devices, or methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0150] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0151] Furthermore, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium.

[0152] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product.

[0153] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, optical fiber) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium, or a semiconductor medium (e.g., a solid-state drive), etc.

[0154] The technical solutions provided by the embodiments of this application have been described in detail above. Specific examples have been used in the embodiments of this application to illustrate the principles and implementation methods of the embodiments of this application. The description of the above embodiments is only for the purpose of helping to understand the methods and core ideas of the embodiments of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the embodiments of this application. Therefore, the content of this specification should not be construed as a limitation on the embodiments of this application.

Claims

1. A method for generating and displaying target resource values, characterized in that, The method includes: Obtain resource value update rule data, and determine multi-dimensional resource value update constraints based on the resource value update rule data; Using the multi-dimensional resource value update constraints as the target constraints and maximizing resource gain data as the objective function, a dynamic optimization model for resource values ​​is constructed. The resource numerical dynamic optimization model is solved using reinforcement learning techniques to obtain the optimized resource values ​​associated with the target object. In response to a resource value adjustment operation instruction input by the resource decision-maker based on the instruction input terminal, the optimized resource value is adjusted according to the resource value adjustment operation instruction to generate the target resource value corresponding to the target object; The hardware type of the hardware display device is obtained by performing matching processing based on the object feature information of the target object. The target resource value is displayed in the hardware display device corresponding to the hardware type.

2. The method according to claim 1, characterized in that, Before the step of responding to a resource value adjustment operation instruction input by the resource decision-maker based on an instruction input terminal, adjusting the optimized resource value according to the resource value adjustment operation instruction, and generating the target resource value corresponding to the target object, the method further includes: Based on the demand forecasting model, a prediction result of the delivery scale of the target object is generated. Obtain the numerical elasticity analysis results of the resources associated with the target object; The system displays the predicted delivery scale of the object and the resource numerical elasticity analysis results, allowing the resource decision-maker to input resource numerical adjustment commands in the command input terminal based on the displayed results.

3. The method according to claim 2, characterized in that, The process of generating a predicted delivery scale result for the target object based on a demand forecasting model includes: The demand forecasting model is constructed by combining time series models and machine learning models. The object features and environmental data associated with the target object are used as model input features and input into the demand prediction model to output the object delivery scale prediction result corresponding to the model input features.

4. The method according to claim 2, characterized in that, The step of obtaining the numerical resilience analysis results of the resources associated with the target object includes: Obtain the percentage change data of the object delivery scale and the percentage change data of the resource value of the target object; Based on the percentage change data of the object delivery scale and the percentage change data of the resource value, determine the resource value elasticity coefficient of the target object; The resource numerical elasticity analysis results are determined based on the resource numerical elasticity coefficients.

5. The method according to claim 1, characterized in that, The object feature information includes resource exchange scenario information, total cost of ownership, and object feature score. The matching process based on the object feature information of the target object to obtain the hardware type of the hardware display device includes: Obtain the resource exchange scenario information corresponding to the target object; For each inventory unit of the target object, a total cost of ownership value is generated based on the total cost of ownership model; A weighted score is calculated based on the object features of the target object to obtain the object feature score value; The hardware type of the hardware display device is determined based on the resource exchange scenario information, the total cost of ownership, and the object feature score.

6. The method according to claim 1, characterized in that, The method further includes: Upon detecting that the target object has been added to the database, the quality period information of the target object is recorded; The remaining time information of the target object's quality period is calculated based on the quality period information. The near-expiration judgment result of the target object is determined based on the remaining time information of the quality period; Based on the near-expiration judgment result, the target resource value is updated using a tiered update strategy.

7. The method according to claim 1, characterized in that, The method further includes: The competitor resource values ​​of the target object are obtained based on web crawling technology. A product matching algorithm based on natural language processing is used to determine the pairing result between the target object and the competitor object; If the pairing result is a successful pairing, the target resource value is updated based on the competitor resource value of the competitor object and the target resource value of the target object.

8. A system for generating and displaying target resource values, characterized in that, The system includes: The constraint generation module is used to acquire resource value update rule data and determine multi-dimensional resource value update constraints based on the resource value update rule data. The model building module is used to construct a dynamic optimization model for resource values, with the multi-dimensional resource value update constraints as the target object and the maximization of resource gain data as the objective function. The model solving module is used to solve the resource numerical dynamic optimization model based on reinforcement learning technology to obtain the optimized resource values ​​associated with the target object. The target resource value generation module is used to respond to the resource value adjustment operation instruction input by the resource decision-maker based on the instruction input terminal, adjust the optimized resource value according to the resource value adjustment operation instruction, and generate the target resource value corresponding to the target object. The hardware type matching module is used to perform matching processing based on the object feature information of the target object to obtain the hardware type of the hardware display device; The target resource value display module is used to display the target resource value in the hardware display device corresponding to the hardware type.

9. A computer device, characterized in that, The computer device includes: At least one processor and memory; The memory is used to store program code, and the processor is used to call the program code stored in the memory to execute the method as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that, It includes instructions that, when executed on a computer, cause the computer to perform the method as described in any one of claims 1 to 7.

Images