Price calculation method for dynamically selecting price calculation rule on basis of dynamic scenario information, and computer

Abstract

The present invention relates to a price calculation method for dynamically selecting a price calculation rule on the basis of dynamic scenario information, and a computer. The method comprises the following steps: acquiring real-time scenario information before a price needs to be calculated each time, wherein the real-time scenario information refers to real-time data in a field related to the price of a product to be priced; using the real-time scenario information to train price calculation rules in a price calculation rule base, wherein the price calculation rule base includes at least two price calculation rules; and receiving product parameter information of the product to be priced, and using the price calculation rules in the price calculation rule base to calculate a real-time price corresponding to the product parameter information. In the present invention, the real-time scenario information is dynamically acquired, and the price calculation rules are trained on the basis of the real-time scenario information to realize dynamic changes in the price calculation rules, so that pricing is more timely and rational.

Description

Pricing method and computer for dynamically selecting pricing rules based on dynamic scene information Technical Field

[0001] The present invention relates to the field of commodity price calculation, and more specifically, to a pricing method and a computer for dynamically selecting pricing rules based on dynamic scene information. Background Art

[0002] Product prices are crucial to market competitiveness and profitability. Small businesses often rely on their own experience to set prices, but manual pricing is impossible for businesses selling large quantities of goods. For example, online platforms, which sell large quantities of goods, face the challenge of dynamically pricing products.

[0003] The pricing method in the existing technology is usually determined based on the analysis of historical data and market trends. The disadvantage of this method is that due to the uncertainty of market changes, historical data and market trends may not be able to accurately predict future market demand and price changes.

[0004] In addition, the current price determination usually requires the design of different pricing rules. For example, it is based on cost plus, and then adjusts the pricing considering factors such as market demand and competitors; or it is based on the price of competitors and then fine-tuned according to the characteristics of its own products or services.

[0005] However, these pricing rules are often fixed and difficult to adapt to the rapidly changing market and competitive landscape. In reality, due to numerous factors, including market demand, competitors, and the supply chain, an effective pricing rule may not remain effective forever. Therefore, a method that can dynamically capture pricing rules and calculate prices to address these changes is needed to ensure that companies remain competitive and maximize profits. Summary of the Invention

[0006] The technical problem to be solved by the present invention is to provide a pricing method and a computer for dynamically selecting pricing rules based on dynamic scene information.

[0007] The technical solution adopted by the present invention to solve the technical problem is to construct a pricing method for dynamically selecting pricing rules based on dynamic scene information, including the following steps:

[0008] Acquire real-time scenario information before calculating the price each time. The real-time scenario information refers to real-time data related to the price of the product to be priced;

[0009] Using the real-time scenario information to train a pricing rule in a pricing rule library, wherein the pricing rule library includes at least two of the pricing rules;

[0010] Receive commodity parameter information of the commodity to be priced, and use the pricing rules in the pricing rule library to calculate the real-time price corresponding to the commodity parameter information.

[0011] Furthermore, in the pricing method for dynamically selecting pricing rules based on dynamic scenario information according to the present invention, the real-time scenario information includes real-time market price information, and obtaining the real-time scenario information includes:

[0012] Obtain price information of the product to be priced on other product sales platforms through the Internet; and / or

[0013] Obtain price information of the commodity to be priced by accessing public databases or databases of professional market research institutions; and / or

[0014] The price information of the product to be priced is obtained through the API interface provided by a third-party data provider.

[0015] Furthermore, in the pricing method for dynamically selecting pricing rules based on dynamic scene information according to the present invention, the real-time scene information includes user behavior information, and obtaining the real-time scene information includes:

[0016] Recording user operation logs of users visiting product sales webpages or product sales applications; and / or

[0017] Collecting user behavior data through online survey platforms; and / or

[0018] Use the smart devices carried by users to capture users' location information and activity tracks.

[0019] Furthermore, in the pricing method for dynamically selecting pricing rules based on dynamic scene information according to the present invention, the real-time scene information includes social platform data, and obtaining the real-time scene information includes:

[0020] Obtaining user behavior data on social platforms through the social platform's API; and / or

[0021] Using keyword search functionality on social platforms to obtain data related to specific topics or events; and / or

[0022] Use natural language processing technology to crawl and analyze user content information from social platforms.

[0023] Furthermore, in the pricing method for dynamically selecting pricing rules based on dynamic scenario information according to the present invention, the real-time scenario information includes market research data, and obtaining the real-time scenario information includes:

[0024] Gathering market research data related to the product through online web research; and / or

[0025] Analyze past market trends through public databases and / or research reports and / or internal market research data.

[0026] Furthermore, in the pricing method for dynamically selecting pricing rules based on dynamic scenario information described in the present invention, the pricing rule library includes:

[0027] Direct pricing rules: pricing is based on a preset linear pricing relationship;

[0028] Freight pricing rules: pricing is based on the distance between the place of shipment and the place of receipt;

[0029] The promotion pricing rules are to determine the price based on the weight factors after the promotion conditions are met. The weight factors include at least one of the user identity, product batch, production date and equipment reputation.

[0030] Furthermore, in the pricing method for dynamically selecting pricing rules based on dynamic scene information according to the present invention, the using of the real-time scene information to train the pricing rules in the pricing rule library includes:

[0031] Selecting features that are relevant to price from the collected real-time scene information;

[0032] Dividing the real-time scene information into a training set and a test set;

[0033] Selecting a machine learning algorithm that is suitable for the data type of the real-time scene information;

[0034] Training the pricing rules using the training set and the machine learning algorithm;

[0035] The trained pricing rules are evaluated using the test set.

[0036] Furthermore, in the pricing method for dynamically selecting pricing rules based on dynamic scene information described in the present invention, the method further includes the steps of:

[0037] Setting a regular update mechanism to update the pricing rules in the pricing rule library according to the update time set by the regular update mechanism; or

[0038] An event update mechanism is set up to automatically update the pricing rules in the pricing rule library after detecting a preset event of the event update mechanism.

[0039] Furthermore, in the pricing method for dynamically selecting pricing rules based on dynamic scene information of the present invention, after calculating the real-time price corresponding to the commodity parameter information using the pricing rules in the pricing rule library, the method further includes the following steps:

[0040] Monitor and collect price feedback information, and evaluate and adjust the pricing rules in the pricing rule library based on the price feedback information.

[0041] In addition, the present invention also provides a computer, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the steps of the pricing method of dynamically selecting pricing rules based on dynamic scene information as described above by calling the computer program stored in the memory.

[0042] The pricing method and computer that implement the present invention for dynamically selecting pricing rules based on dynamic scene information have the following beneficial effects: the present invention dynamically obtains real-time scene information, and trains pricing rules based on the real-time scene information to achieve dynamic changes in pricing rules, thereby making pricing more timely and reasonable. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The present invention will be further described below with reference to the accompanying drawings and embodiments, in which:

[0044] FIG1 is a flowchart of a pricing method for dynamically selecting pricing rules based on dynamic scene information provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0045] In order to have a clearer understanding of the technical features, purposes and effects of the present invention, specific embodiments of the present invention are now described in detail with reference to the accompanying drawings.

[0046] In a preferred embodiment, referring to FIG1 , this embodiment provides a pricing method for dynamically selecting pricing rules based on dynamic scenario information. This method is applied to automatically pricing products, particularly products on online product sales platforms. Online product sales platforms (e-commerce platforms) sell a large number of products simultaneously, and this method is used to automatically price each product on the product sales platform. Specifically, this pricing method for dynamically selecting pricing rules based on dynamic scenario information includes the following steps:

[0047] Step S1: Before calculating the price each time, real-time scenario information is obtained. Real-time scenario information refers to real-time data in areas related to the price of the product to be priced.

[0048] Specifically, the real-time scenario information in this embodiment refers to real-time data related to the price of the commodity to be priced. Price-related fields refer to various factors that can affect the price of the commodity, and these factors are classified according to their attributes to obtain various fields. Real-time scenario information includes multiple scenarios, each of which affects different aspects of the commodity price. Real-time scenario information includes not only scenario information that has a direct impact on the price, but also scenario information that has an indirect impact on the price.

[0049] Understandably, because the various factors influencing commodity prices are constantly changing, it's important to obtain real-time scenario information before each price calculation to ensure that current pricing aligns with the current market environment. After acquiring real-time scenario data, it must be organized, cleaned, and converted according to pre-set algorithms to facilitate subsequent use.

[0050] Step S2: Use real-time scenario information to train pricing rules in a pricing rule library, where the pricing rule library contains at least two pricing rules.

[0051] Specifically, the pricing rule library of this embodiment includes a variety of pricing rules, and each pricing rule is used to price goods from one aspect. Pricing rules are a method of determining the price of model training. Unlike the unchanging pricing rules in the prior art, the pricing rules of this embodiment are in dynamic change. In order to maintain the real-time nature of the pricing rules, so that the pricing rules can reflect the current market status in real time, it is necessary to use real-time scene information to train the pricing rules in the pricing rule library before each price update. It can be understood that because the real-time scene information is the latest data, the pricing rules obtained through real-time scene information training are also the latest and can reflect the current market situation in a timely manner, so that the pricing of goods is more reasonable.

[0052] Furthermore, the process of using real-time scenario information to train the pricing rules in the pricing rule library is as follows:

[0053] 1) Select price-related features from the collected real-time scene information. Based on the problem requirements, select appropriate features from the collected dataset that are relevant to price.

[0054] 2) Divide the real-time scene information into training and test sets. Divide the collected data set into training and test sets, usually with a ratio of 70% to 80% being the training set and the rest being the test set.

[0055] 3) Select a machine learning algorithm that is appropriate for the data type of the real-time scenario information. Based on the problem type and data characteristics, choose an appropriate algorithm for model training. Common machine learning algorithms include linear regression, decision trees, random forests, and neural networks.

[0056] 4) Use the training set and machine learning algorithm to train the pricing rules. Use the training set to train the selected machine learning algorithm and optimize model performance by adjusting model parameters.

[0057] 5) Use the test set to evaluate the trained pricing rules. Use the test set to evaluate the trained model. Common evaluation metrics include mean squared error (MSE), mean absolute error (MAE), and coefficient of determination (R²).

[0058] Step S3: receiving the commodity parameter information of the commodity to be priced, and using the pricing rules in the pricing rule library to calculate the real-time price corresponding to the commodity parameter information.

[0059] Specifically, after using real-time scenario information to train the pricing rules in the pricing rule library, the product parameter information of the product to be priced is received. The product parameter information includes product attributes, transaction conditions and other information, and the pricing rules in the pricing rule library are used to calculate the real-time price corresponding to the product parameter information.

[0060] This embodiment dynamically obtains real-time scene information and trains pricing rules based on the real-time scene information to achieve dynamic changes in pricing rules, thereby making pricing more timely and reasonable.

[0061] In some embodiments of the pricing method for dynamically selecting pricing rules based on dynamic scenario information, the real-time scenario information includes real-time market price information. The real-time scenario information can be obtained in the following ways:

[0062] Method 1: Obtain price information for the product to be priced from other product sales platforms through the Internet. For example, use a web crawler to scrape price information from websites of other similar products on the market.

[0063] Method 2: Obtain price information of the product to be priced by accessing public databases or databases of professional market research institutions.

[0064] Method 3: Obtain price information of the product to be priced through the API interface provided by a third-party data provider.

[0065] The above three methods can be implemented individually or in combination. The above three methods for obtaining real-time market price information are merely illustrative of the methods of obtaining the information and are not intended to be exclusive. Using other technical means to obtain real-time market price information based on the design concept of this embodiment also falls within the scope of protection of this embodiment.

[0066] This embodiment provides a method for obtaining market price information, through which real-time market price information can be obtained in real time, providing a data basis for commodity price evaluation.

[0067] In some embodiments of the pricing method for dynamically selecting pricing rules based on dynamic scenario information, the real-time scenario information includes user behavior information, and obtaining the real-time scenario information includes:

[0068] Method 1: Record user operation logs when users visit product sales webpages or product sales applications. Record user behavior data in server logs, such as website access logs and application user operation logs.

[0069] Method 2: Collect user behavior data through online survey platforms.

[0070] Method 3: Using smart devices carried by users to capture their location information and activity traces. For example, using a smartphone or smartwatch to capture the user's location information and activity traces. It is understood that user behavior data must be authorized before use.

[0071] The above three methods can be implemented individually or in combination. The above three methods for obtaining user behavior information are only exemplary descriptions of the methods of obtaining the information and are not exclusive limitations. If other technical means are used to obtain user behavior information based on the design concept of this embodiment, it also falls within the scope of protection of this embodiment.

[0072] This embodiment provides a method for obtaining user behavior information, through which real-time user behavior information can be obtained in real time, providing a data basis for commodity price evaluation.

[0073] In some embodiments of the pricing method for dynamically selecting pricing rules based on dynamic scenario information, the real-time scenario information includes social platform data, and obtaining the real-time scenario information includes:

[0074] Method 1: Obtain user behavior data on the social platform through the social platform's API interface, such as obtaining user posting, likes, forwarding and other behavior data through the social media platform's API interface.

[0075] Method 2: Call the keyword search function of the social platform to obtain relevant data on a specific topic or event, for example, use the keyword search function provided by the social media platform to obtain data related to a specific topic or event.

[0076] Method 3: Use natural language processing technology to capture and analyze user content information from social platforms, such as using natural language processing technology to capture and analyze user comments, opinions, and other information from social media.

[0077] The above three methods can be implemented individually or in combination. The above three methods of obtaining social platform data are only exemplary descriptions of the methods of obtaining data, and are not exclusive limitations. Based on the design concept of this embodiment, other technical means of obtaining social platform data are also within the scope of protection of this embodiment.

[0078] This embodiment provides a method for obtaining social platform data, through which real-time social platform data can be obtained in real time, providing a data basis for commodity price evaluation.

[0079] In some embodiments of the pricing method for dynamically selecting pricing rules based on dynamic scenario information, the real-time scenario information includes market research data, and obtaining the real-time scenario information includes:

[0080] Method 1: Collect product-related market research data through online network surveys.

[0081] Method 2: Analyze past market trends through public databases and / or research reports and / or internal market research data.

[0082] Method 3: Conduct face-to-face interviews with potential customers, experts, or industry insiders to gain market insights and feedback, and input the resulting information into the platform.

[0083] The above three methods can be implemented individually or in combination. The above three methods for obtaining market research data are only exemplary descriptions of the methods of obtaining data and are not exclusive limitations. Using other technical means to obtain market research data based on the design concept of this embodiment also falls within the scope of protection of this embodiment.

[0084] This embodiment provides a method for obtaining market research data, through which real-time market research data can be obtained in real time, providing a data basis for commodity price evaluation.

[0085] In some embodiments of the pricing method for dynamically selecting pricing rules based on dynamic scenario information, the pricing rule library includes:

[0086] Direct pricing rules: pricing is based on a preset linear pricing relationship. The simplest direct pricing method for a product is a linear relationship.

[0087] Freight pricing rules are based on the distance between the shipping and receiving locations. For simple freight calculations, factors influencing the shipping and receiving locations, as well as the time of delivery, generally exhibit a linear relationship. For example, if a sudden weather event occurs in a certain area, the automated process will modify the freight pricing rules to appropriately increase the weight of the weather impact, retrain the model, and increase the freight rate.

[0088] Promotional pricing rules determine pricing based on weighting factors after promotional conditions are met. Weighting factors include at least one of user identity, product batch, production date, and device reputation. Generating promotional pricing rules is complex. First, it's necessary to determine whether the promotion conditions are met. Then, different weighting ratios are set based on factors such as user identity, product information (such as batch and production date), and other influences (such as social reputation). For example, if a product is currently using a promotional pricing rule, but the warehouse detects that the product is about to expire, the automated process will modify the promotional pricing rule to appropriately increase the expiration weight for the product, retrain the model, and reduce the promotional price accordingly.

[0089] It can be understood that the pricing rule library of this embodiment includes multiple pricing rules. The direct pricing rules, freight pricing rules and promotional pricing rules provided here are examples of pricing rules. According to this principle, corresponding pricing rules for other areas that affect pricing can be specified.

[0090] In some embodiments of the pricing method for dynamically selecting pricing rules based on dynamic scene information, the method further includes the steps of: setting a periodic update mechanism, and updating the pricing rules in the pricing rule library according to the update time set by the periodic update mechanism. The updating of the pricing rules in the pricing rule library can refer to the above embodiments. Periodicity can be a series of fixed time points, such as midnight every day. Periodicity can be a fixed time interval, such as one day, one week, etc. Data is obtained from the data source regularly and regularly, and the data is used to update and adjust the pricing rules. According to the different factor weights and changes in real-time data, the cost of each factor is recalculated, and the overall model training price is re-evaluated. This embodiment realizes the automatic update of prices by setting a periodic update mechanism, so that prices can change at any time with market conditions to make prices more reasonable.

[0091] In some embodiments of the pricing method for dynamically selecting pricing rules based on dynamic scene information, the method further includes the steps of: setting an event update mechanism to automatically update the pricing rules in the pricing rule library after detecting a preset event of the event update mechanism. The updating of the pricing rules in the pricing rule library can refer to the above embodiments. The preset events can be flexibly set according to the specific commodities and are not limited here. In this embodiment, the update mechanism is triggered by preset events to achieve automatic price updates, so that the price can change at any time according to market conditions to make the price more reasonable.

[0092] In some embodiments of the pricing method for dynamically selecting pricing rules based on dynamic scene information, after using the pricing rules in the pricing rule library to calculate the real-time price corresponding to the commodity parameter information, the steps are also included: monitoring and collecting price feedback information, and evaluating and adjusting the pricing rules in the pricing rule library based on the price feedback information. Ensure that a monitoring mechanism is established to track the cost and price changes of model training. Through the monitoring system, the accuracy of the data source can be monitored in real time, and corrections and adjustments can be made as needed. The pricing strategy may be adjusted according to the performance quality and performance of the trained model. If the performance of a model exceeds expectations, the pricing strategy may need to be re-evaluated to ensure that the price of model training matches the value it brings. Conversely, if the performance of the model does not meet the requirements, the pricing strategy may need to be reconsidered to improve the efficiency and quality of model training. This embodiment understands the effect of automatic pricing through price feedback information, adjusts the pricing rules in a timely manner, and continuously optimizes the pricing process.

[0093] Alternatively, based on the above embodiment, for the price calculation method: when applying the pricing rules and calculating the corresponding price, the price calculation method for the special formula can be performed using scripting languages ​​such as JavaScript, Groovy, and Python.

[0094] Alternatively, based on the above embodiment, for pricing rules: for a large number of pricing rules, the drools rule engine can be used to filter out the correct pricing rules to be used.

[0095] Alternatively, based on the above embodiments, for data mining and machine learning: for a large amount of scenario information and pricing parameter data, technical methods such as data mining and machine learning can be used to generate more accurate and personalized pricing rules through model training and optimization, automatically discover new rules and update the pricing rule library, and continuously improve and optimize the system in combination with user feedback and evaluation.

[0096] Alternatively, based on the above embodiment, parallel computing and distributed architecture can be used to improve the system's sorting efficiency and response speed. These technologies can be used to break down large computational tasks into multiple smaller tasks for parallel processing, shortening sorting time and increasing system throughput. Furthermore, clustering and load balancing technologies can be used to improve system availability and stability.

[0097] In a preferred embodiment, the computer of this embodiment includes a memory and a processor, the memory stores a computer program, and the processor executes the steps of the pricing method of dynamically selecting pricing rules based on dynamic scene information as in the above embodiment by calling the computer program stored in the memory.

[0098] In a preferred embodiment, this embodiment provides a system for dynamically acquiring pricing strategies and calculating prices based on pricing scenarios. The system includes the following components:

[0099] 1. Data collection module: used to collect relevant information such as environment, market and competitors, and convert it into a usable data format.

[0100] 2. Pricing rule library: It stores various pricing rules and can realize automatic pricing rule management and update.

[0101] 3. Pricing module: Apply pricing rules and calculate the corresponding price.

[0102] 4. API interface: An API interface provided to the client to facilitate the client to call the dynamic pricing service.

[0103] In summary, the present invention dynamically determines the pricing strategy and calculates the real-time price based on the attributes, inventory, sales status and other information of the goods, as well as the user's behavior, geographic location, time and other information. The technical feature of the present invention is that it performs calculations based on real-time data, has higher real-time and dynamic properties, and can dynamically adjust the pricing strategy and price according to different scenarios and needs to achieve more accurate pricing and sales. In short, this solution has broad application prospects and practical value, and has potential application space in the Internet industry and many other fields.

[0104] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Reference can be made to the common and similar parts between the various embodiments. For the devices disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method description.

[0105] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the components and steps of each example according to their functions. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0106] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0107] The above embodiments are intended only to illustrate the technical concepts and features of the present invention. Their purpose is to enable those skilled in the art to understand the present invention and implement it accordingly. They are not intended to limit the scope of protection of the present invention. All equivalent variations and modifications within the scope of the claims of the present invention are intended to be covered by the claims of the present invention.

Claims

1. A pricing method for dynamically selecting a pricing rule based on dynamic scene information, characterized in that Including the following steps: Obtain real-time scenario information before calculating the price each time. The real-time scenario information refers to real-time data in the field related to the price of the commodity to be priced; Use the real-time scenario information to train the pricing rules in the pricing rule library. The pricing rule library contains at least two of the pricing rules; Receive the commodity parameter information of the commodity to be priced, and use the pricing rules in the pricing rule library to calculate the real-time price corresponding to the commodity parameter information.

2. The pricing method for dynamically selecting a pricing rule based on dynamic scenario information according to claim 1, wherein If the real-time scenario information includes real-time market price information, then obtaining the real-time scenario information includes: Obtain the price information of the commodity to be priced on other commodity selling platforms through the network; and / or Obtain the price information of the commodity to be priced by accessing a public database or the database of a professional market research institution; and / or Obtain the price information of the commodity to be priced through the API interface provided by a third-party data provider.

3. The pricing method for dynamically selecting a pricing rule based on dynamic scenario information according to claim 1, wherein If the real-time scenario information includes user behavior information, then obtaining the real-time scenario information includes: Record the user operation logs of users accessing the commodity selling web page or commodity selling application; and / or Collect user behavior data through an online survey platform; and / or Use the smart device carried by the user to capture the location information and activity track of the user.

4. The pricing method for dynamically selecting a pricing rule based on dynamic scenario information according to claim 1, wherein If the real-time scenario information includes social platform data, then obtaining the real-time scenario information includes: Obtain the behavior data of users on the social platform through the API interface of the social platform; and / or Invoke the keyword search function of the social platform to obtain relevant data on specific topics or events; and / or Use natural language processing technology to scrape and analyze the content information published by users on the social platform.

5. The pricing method for dynamically selecting a pricing rule based on dynamic scene information according to claim 1, characterized in that, If the real-time scenario information includes market research data, then obtaining the real-time scenario information includes: Collect market research data related to the commodity through online network research; and / or Analyze the past market trends through a public database and / or research reports and / or the market research data within the enterprise.

6. The pricing method for dynamically selecting a pricing rule based on dynamic scenario information according to claim 1, wherein The pricing rule library includes: Direct pricing rules, pricing according to a preset linear pricing relationship; Freight pricing rules, pricing according to the distance between the place of dispatch and the place of receipt; Promotion pricing rules, pricing in combination with a weight factor after meeting the promotion conditions. The weight factor includes at least one of user identity, product batch, production date, and device reputation.

7. The pricing method for dynamically selecting a pricing rule based on dynamic scenario information according to claim 1, characterized in that, The using the real-time scenario information to train the pricing rules in the pricing rule library includes: Select the features relevant to the price from the collected real-time scenario information; Divide the real-time scenario information into a training set and a test set; Select a machine learning algorithm suitable for the data type of the real-time scenario information; Use the training set and the machine learning algorithm to train the pricing rules; 8. The pricing method for dynamically selecting a pricing rule based on dynamic scenario information according to claim 1, characterized in that, Use the test set to evaluate the trained pricing rules. The method further includes the steps of: Setting a regular update mechanism to update the pricing rules in the pricing rule library according to the update time set by the regular update mechanism; or Setting an event update mechanism to automatically update the pricing rules in the pricing rule library after monitoring the preset event of the event update mechanism.

9. The pricing method for dynamically selecting a pricing rule based on dynamic scenario information according to claim 1, characterized in that, After calculating the real-time price corresponding to the commodity parameter information using the pricing rules in the pricing rule library, the following steps are further included: Monitoring and collecting price feedback information, and evaluating and adjusting the pricing rules in the pricing rule library according to the price feedback information.

10. A computer, characterized in that, It includes a memory and a processor. A computer program is stored in the memory. The processor executes the steps of the pricing method for dynamically selecting pricing rules based on dynamic scenario information according to any one of claims 1 to 9 by calling the computer program stored in the memory.

Images