Auction item value dynamic evaluation and precise pricing model construction system of shared auction platform

Abstract

The invention discloses an auction item value dynamic evaluation and precise pricing model construction system of a shared auction platform, and relates to the technical field of auction. Comprising a data acquisition module, an index analysis module, a dynamic evaluation module, a pricing model construction module, a strategy execution module, an application monitoring module, a service management module, an auction management module and an interaction feedback module. According to the auction item value dynamic evaluation and precise pricing model construction system of the shared auction platform, the auction item value can be dynamically evaluated based on real-time data, the timeliness and the accuracy of an evaluation result are ensured, the auction platform can capture market changes in time, and a reliable basis is provided for auction item pricing; and a precise pricing model is constructed based on the dynamic evaluation result, so that real-time precise pricing is performed on the auction items according to the auction item value dynamic evaluation result, subjectivity and uncertainty of manual pricing are reduced, and better auction experience and service are provided for sellers and buyers.

Description

Technical Field

[0001] The present invention relates to the field of auction technology, and in particular to a system for dynamically evaluating the value of auction items and building an accurate pricing model for a shared auction platform. Background Art

[0002] A shared auction platform is an online platform that allows sellers to upload auction items to the platform and buyers to bid for them, allowing buyers and sellers to trade easily. Shared auction platforms are used in many fields such as second-hand goods trading, art auctions, and cloud resource allocation. The auction item value assessment refers to the assessment of the value of the auction items or property to determine its market price, and the accuracy of the results is crucial to the success of the auction. In the prior art, there are many methods for auction item value assessment, such as the market method, the income method, and the cost method. The market method assesses the value of the auction item by comparing the transaction prices of similar assets in the market; the income method assesses the value of the auction item by predicting its future income; the cost method assesses the value of the auction item by estimating its replacement cost and deducting the depreciation factor.

[0003] On shared auction platforms, the pricing of auction items is usually based on their value assessment results, but it is also affected by many factors such as market supply and demand, auction rules, and sellers' expected profits. The formulation of pricing strategies requires weighing these factors to ensure that the auction items can be sold at a reasonable price and maximize the seller's profit. However, in the actual pricing process, the existing auction item value assessment methods have their own limitations and scope of application, and are usually based on historical data or static assumptions. It is difficult to accurately reflect the changes in market supply and demand and the true value of the auction items. However, the market environment is constantly changing. New competing products, buyer demand, policy regulations, etc. may have an impact on the value of the auction items. This uncertainty makes it very difficult to dynamically assess the value of the auction items. Even if the shared auction platform can obtain and process data in real time, it is impossible to accurately predict the impact of market changes on the value of the auction items.

[0004] Therefore, it is urgent to improve this shortcoming. The present invention studies and improves the existing technology and its shortcomings, and provides a system for dynamically evaluating the value of auction items and building an accurate pricing model for a shared auction platform. Summary of the invention

[0005] The purpose of the present invention is to provide a system for dynamically evaluating the value of auction items and building an accurate pricing model for a shared auction platform, so as to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above-mentioned purpose, the present invention provides the following technical solution: a system for dynamically evaluating the value of auction items and constructing an accurate pricing model on a shared auction platform, comprising: A data collection module, which is responsible for collecting various data related to the evaluation of the value of auction items, including market data (competing item prices, market trends, auction history records, etc.), user data (user behavior, purchase history, payment ability, etc.), and auction item data (attributes of auction items, appraisal reports, historical transaction records, etc.); An indicator analysis module, which is responsible for determining the evaluation indicators that affect the value of the auction items according to the attributes of the auction items and the market environment, and quantifying the indicators, converting non-quantitative indicators into quantitative indicators for subsequent mathematical calculations and model building; A dynamic evaluation module, which is responsible for dynamically evaluating the value of the auction item based on quantified evaluation indicators using machine learning or statistical methods; A pricing model building module, which is responsible for building an accurate pricing model based on the dynamic evaluation results of the auction item value and formulating a reasonable pricing strategy for the auction item; A strategy execution module, which is responsible for applying the formulated price strategy to the auction platform to achieve automated pricing; Application monitoring module, which is responsible for monitoring the execution of price strategies, including key indicators such as transaction rate and profit margin, and timely adjusting price strategies based on monitoring results to cope with market changes and changes in user needs; A service management module, which is responsible for coordinating and managing various service resources on the auction platform, such as customer service and technical support, to ensure that these resources can respond to user needs efficiently and accurately, and to continuously monitor the quality of services provided by the platform, including key indicators such as response time and problem resolution rate, to ensure that users receive a high-quality service experience; The auction management module is responsible for the entire process management of the auction, including the creation, publication, bidding, settlement and other links of the auction, ensuring the smooth progress of the auction, formulating the rules and processes of the auction, and supervising the implementation of these rules to prevent fraud and violations. At the same time, the auction is analyzed for data, including the value of the auction items, bidding conditions, user participation, etc., to provide data support for platform operation and pricing model optimization; The interactive feedback module is responsible for designing a user-friendly interface to realize interactive functions between users and the platform, such as searching, bidding, purchasing, etc., to ensure that users can conveniently participate in auction activities, collect user opinions and suggestions, respond to user feedback in a timely manner, and solve problems encountered by users during use.

[0007] Furthermore, the data collection module captures relevant data from data sources such as e-commerce platforms and auction websites through web crawlers, cleans the captured data, removes duplicate, invalid or erroneous data, and performs necessary preprocessing, such as data format conversion and missing value filling.

[0008] Furthermore, the evaluation indicators include quality indicators (such as material, craftsmanship, preservation status, etc.), market indicators (such as supply and demand relationship, market popularity, competitive product prices, etc.), user preference indicators (such as user purchase history, willingness to pay, etc.), and the quantitative processing of indicators includes direct quantification, indirect quantification and conversion quantification; The direct quantization specifically includes: Numerical quantification: directly use specific numerical values ​​to express the size or degree of indicators. For example, use numerical values ​​such as sales (10,000 yuan), number of users (people), click-through rate (%) to quantify performance indicators. Proportional quantification: The size of an indicator is expressed by calculating a ratio or percentage. For example, market share (%), customer satisfaction (%) and other ratios are used to quantify market performance. The indirect quantification specifically includes: Grading: Divide the indicator into different grades or score ranges and assign corresponding values ​​to each grade or score. For example, a five-grade rating system (1-5 points) is used to quantify product quality or service quality. Frequency statistics: The number of times or frequency of occurrence of statistical indicators within a certain time or space range. For example, the number of user visits, purchase frequency, etc. are used to quantify user activity; Ranking and assignment: sort the indicators in a certain order and assign corresponding values ​​to each ranking position. For example, use ranking (1, 2, 3...) to quantify competitiveness or influence; The conversion quantization specifically includes: Questionnaire scoring: Through questionnaire surveys, respondents are asked to score indicators based on their perceptions or experiences; Expert evaluation: Invite experts or industry authorities to evaluate the indicators and give corresponding values ​​or grades; Data analysis: Use data mining, machine learning and other technologies to analyze large amounts of data to extract quantitative information related to indicators.

[0009] Furthermore, the dynamic evaluation module includes the following steps: S1. Model selection: According to the data characteristics and evaluation requirements, select the appropriate evaluation model, such as linear regression, decision tree, random forest, neural network, etc.; S2. Training model: Use historical data to train the model to learn the mapping relationship between features and auction item values, and use methods such as cross-validation to evaluate the performance of the model and adjust and optimize parameters; S3, real-time evaluation: when new data arrives, the dynamic evaluation module uses the trained model to evaluate the new data and calculate the real-time value of the auction item; S4. Result output: The evaluation results are immediately output to the user and pricing model building module in the form of numerical values, charts, etc., and the evaluation results include the real-time value of the auction item, the trend of value changes, the importance of relevant features, etc.

[0010] Furthermore, the evaluation model sets a strategy for regular updates, such as daily, weekly or monthly updates, and during updates, the model is incrementally updated or retrained using new data to reflect the latest trends in market changes and user preferences.

[0011] Furthermore, the pricing model building module includes the following submodules: Data integration submodule: receives the output data from the auction item value dynamic evaluation module, including various evaluation indicators and quantitative results, and cleans, organizes and formats the data to ensure data quality and consistency; Feature screening submodule: Use correlation analysis to screen features, and perform feature engineering based on business needs and data characteristics to extract more valuable features; Model training submodule: Select appropriate machine learning algorithms (such as linear regression, support vector machine, neural network, etc.) as the basis of the pricing model, and use the preprocessed data and filtered feature set to train the model; Strategy formulation submodule: Based on the prediction results of the pricing model, combined with market demand, competitive situation and other factors, formulate pricing strategies, including price range, pricing method, etc.; Verification and optimization sub-module: Use new data sets to verify the pricing model, evaluate the accuracy and stability of the model, and adjust and optimize the model based on the verification results to improve the model's prediction accuracy and practicality. Continuously collect user feedback and market data, and iteratively update the model to adapt to market changes.

[0012] Furthermore, the feature screening submodule uses statistical methods (such as correlation coefficient, chi-square test, etc.) to evaluate the correlation between each feature and the auction item price, and selects features based on the correlation score.

[0013] Furthermore, the feature engineering of the feature screening submodule specifically includes: Feature combination: combine two or more features into a new feature to capture the relationship between them. The combination methods include product features, ratio features, difference features and logical combination; Feature transformation: Convert the original features into new features through mathematical transformation to improve the performance of the model. This is achieved by using logarithmic transformation, square root transformation, polynomial transformation, and inverse function transformation to handle the nonlinear relationship between features, and by using feature scaling (including standardization and normalization) to eliminate the impact of the dimensions and value ranges between features. Feature encoding: converting categorical features into numerical features, for example, converting categorical features into binary vectors through One-Hot Encoding; Feature binning: Divide continuous features into multiple discrete intervals; Time feature extraction: extract features such as year, month, day, and hour from the timestamp.

[0014] Furthermore, the pricing model training submodule specifically adjusts model parameters, optimizes model performance, and evaluates the model's prediction accuracy and generalization ability through methods such as cross-validation and grid search, to ensure that the model can accurately reflect the relationship between the value and price of the auction item.

[0015] Furthermore, the strategy execution module includes an automated pricing system, and the operating logic of the strategy execution module is: by inputting the formulated price strategy into the automated pricing system, so that it can perform pricing according to preset rules and logic, adjust the auction price in real time, and ensure that the price strategy is applied to all auction activities in real time and accurately.

[0016] The present invention provides a system for dynamically evaluating the value of auction items and building an accurate pricing model for a shared auction platform, which has the following beneficial effects: 1. The present invention can dynamically evaluate the value of auction items based on real-time data to ensure the timeliness and accuracy of the evaluation results, which helps the auction platform to capture market changes in a timely manner and provide a reliable basis for auction item pricing; and construct an accurate pricing model based on the dynamic evaluation results, so as to accurately price the auction items in real time according to the dynamic evaluation results of the auction item value, and further formulate a pricing strategy based on the prediction results of the pricing model, which helps the auction platform to improve pricing efficiency, reduce the subjectivity and uncertainty of artificial pricing, and provide sellers and buyers with a better auction experience and service.

[0017] 2. Through the system's real-time and accurate pricing, users can clearly understand the pricing basis and price changes of the auction items, which helps to enhance users' trust in the platform and improve user loyalty.

[0018] 3. This system can realize the automated management of the auction process, including the creation, publication, bidding, settlement and other aspects of the auction, which will help the auction platform improve operational efficiency, reduce manual intervention and error rates, and at the same time, it can intelligently monitor auction activities, promptly detect and handle abnormal situations, which will help ensure the fairness and impartiality of auction activities and maintain the good reputation of the platform.

[0019] 4. This system can conduct in-depth mining and analysis of large amounts of auction data, providing valuable market insights and decision-making support for auction platforms, which will help auction platforms better understand market demand and trends and formulate more accurate market strategies. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A schematic diagram of the system architecture of a system for dynamically evaluating the value of auction items and building a precise pricing model for a shared auction platform according to the present invention; Figure 2 This is a schematic flow chart of the steps of a dynamic evaluation module of a system for dynamically evaluating the value of auction items and building an accurate pricing model for a shared auction platform of the present invention. DETAILED DESCRIPTION

[0021] The following embodiments of the present invention are described in further detail in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0022] like Figure 1-Figure 2 As shown, a system for dynamically evaluating the value of auction items and building a precise pricing model for a shared auction platform includes a data collection module, an indicator analysis module, a dynamic evaluation module, a pricing model building module, a strategy execution module, an application monitoring module, a service management module, an auction management module, and an interactive feedback module; Data collection module: collects various data related to the evaluation of the value of auction items, including market data (competing product prices, market trends, auction history, etc.), user data (user behavior, purchase history, payment ability, etc.), auction item data (attributes of auction items, appraisal reports, historical transaction records, etc.); this module uses web crawlers to capture relevant data from data sources such as e-commerce platforms and auction websites, cleans the captured data, removes duplicate, invalid or erroneous data, and performs necessary preprocessing, such as data format conversion and missing value filling.

[0023] Index analysis module: Determine the evaluation indicators that affect the value of the auction items based on the properties of the auction items and the market environment, quantify the indicators, and convert non-quantitative indicators into quantitative indicators for subsequent mathematical calculations and model construction; among them, the evaluation indicators include quality indicators (such as material, craftsmanship, preservation status, etc.), market indicators (such as supply and demand relationship, market popularity, competitor prices, etc.), user preference indicators (such as user purchase history, willingness to pay, etc.), and the quantitative processing of indicators includes direct quantification, indirect quantification and conversion quantification.

[0024] Direct quantification specifically includes: Numerical quantification: directly use specific numerical values ​​to express the size or degree of indicators. For example, use numerical values ​​such as sales (10,000 yuan), number of users (people), click-through rate (%) to quantify performance indicators. Proportional quantification: The size of an indicator is expressed by calculating the ratio or percentage. For example, market share (%), customer satisfaction (%) and other ratios are used to quantify market performance.

[0025] Indirect quantification specifically includes: Grading: Divide the indicator into different grades or score ranges and assign corresponding values ​​to each grade or score. For example, a five-grade rating system (1-5 points) is used to quantify product quality or service quality. Frequency statistics: The number of times or frequency of occurrence of statistical indicators within a certain time or space range. For example, the number of user visits, purchase frequency, etc. are used to quantify user activity; Ranking and assignment: Sort the indicators in a certain order and assign a corresponding value to each ranking position. For example, use rankings (1, 2, 3...) to quantify competitiveness or influence.

[0026] Conversion quantization specifically includes: Questionnaire scoring: Through questionnaire surveys, respondents are asked to score indicators based on their perceptions or experiences; Expert evaluation: Invite experts or industry authorities to evaluate the indicators and give corresponding values ​​or grades; Data analysis: Use data mining, machine learning and other technologies to analyze large amounts of data to extract quantitative information related to indicators.

[0027] In this embodiment, the quality indicator quantification process is specifically as follows: Material: Grading and scoring are based on the type, rarity, and durability of the material. For example, for artworks, the rarity and market value of materials such as gold, silver, diamonds, and wood can be scored.

[0028] Craftsmanship: The level of craftsmanship is graded through expert evaluation or industry standards. For example, handicrafts can be evaluated based on the exquisiteness of the craftsmanship, the perfection of the details, etc.

[0029] Condition: A quantitative assessment of the item’s condition, such as condition, wear, signs of restoration, etc. Condition can be quantified using a percentage or a graded system.

[0030] The specific quantitative processing of market indicators is as follows: Supply and demand relationship: Through market research and data analysis, determine the supply and demand of similar items in the market, and calculate the supply-demand ratio. The higher the supply-demand ratio, the stronger the market demand, and the higher the value of the item may be.

[0031] Market heat: Market heat is measured using data such as search engine index, social media attention, and auction website visits. These data can be converted into a heat index to reflect the market's attention to the auction item.

[0032] Competitive item prices: Collect the transaction prices of similar items in the market and calculate the average price, median price or price range. These price data can be used as a reference for evaluating the value of the items.

[0033] The specific quantitative processing of user preference indicators is as follows: User purchase history: Analyze the user's purchase history and count the purchase frequency and purchase amount of different types and price ranges of auction items. These data can reflect the user's purchase preferences and spending power.

[0034] Willingness to pay: Understand users’ willingness to pay for auction items through questionnaire surveys, bidding behavior analysis, etc. Willingness to pay can be converted into specific price ranges or scores to reflect users’ perception of the value of the auction items.

[0035] Dynamic evaluation module: Use machine learning or statistical methods to dynamically evaluate the value of the lot based on quantified evaluation indicators; this module includes the following steps: S1. Model selection: According to the data characteristics and evaluation requirements, select the appropriate evaluation model, such as linear regression, decision tree, random forest, neural network, etc.; S2. Training model: Use historical data to train the model to learn the mapping relationship between features and auction item values, and use methods such as cross-validation to evaluate the performance of the model and adjust and optimize parameters; S3, real-time evaluation: when new data arrives, the dynamic evaluation module uses the trained model to evaluate the new data and calculate the real-time value of the auction item; S4. Result output: The evaluation results are immediately output to the user and the pricing model building module in the form of numerical values, charts, etc., and the evaluation results include the real-time value of the auction item, the trend of value change, the importance of relevant features, etc.; In addition, the evaluation model sets a strategy for regular updates, such as daily, weekly, or monthly updates, and when updating, the model is incrementally updated or retrained using new data to reflect the latest trends in market changes and user preferences.

[0036] Pricing model building module: Based on the dynamic evaluation results of the auction item value, an accurate pricing model is built to formulate a reasonable pricing strategy for the auction item; this module includes the following sub-modules: Data integration submodule: receives the output data from the auction item value dynamic evaluation module, including various evaluation indicators and quantitative results, and cleans, organizes and formats the data to ensure data quality and consistency; Feature screening submodule: Use correlation analysis to screen features, and perform feature engineering based on business needs and data characteristics to extract more valuable features; In this embodiment, the feature screening submodule uses statistical methods (such as correlation coefficient, chi-square test, etc.) to evaluate the correlation between each feature and the auction item price, and selects features based on the correlation score; at the same time, the feature engineering of the feature screening submodule specifically includes: Feature combination: combine two or more features into a new feature to capture the relationship between them. The combination methods include product features, ratio features, difference features and logical combination; Feature transformation: Convert the original features into new features through mathematical transformation to improve the performance of the model. This is achieved by using logarithmic transformation, square root transformation, polynomial transformation, and inverse function transformation to handle the nonlinear relationship between features, and by using feature scaling (including standardization and normalization) to eliminate the impact of the dimensions and value ranges between features. Feature encoding: converting categorical features into numerical features, for example, converting categorical features into binary vectors through One-Hot Encoding; Feature binning: Divide continuous features into multiple discrete intervals; Time feature extraction: extract features such as year, month, day, and hour from the timestamp; Model training submodule: Select a suitable machine learning algorithm (such as linear regression, support vector machine, neural network, etc.) as the basis of the pricing model, and use the preprocessed data and the screened feature set to train the model; in this embodiment, the pricing model training submodule specifically adjusts the model parameters and optimizes the model performance through methods such as cross-validation and grid search, and evaluates the prediction accuracy and generalization ability of the model to ensure that the model can accurately reflect the relationship between the value of the auction item and the price; Strategy formulation submodule: Based on the prediction results of the pricing model, combined with market demand, competitive situation and other factors, formulate pricing strategies, including price range, pricing method, etc.; Verification and optimization submodule: Use new data sets to verify the pricing model, evaluate the accuracy and stability of the model, and adjust and optimize the model based on the verification results to improve the model's prediction accuracy and practicality. Continue to collect user feedback and market data, and iterate and update the model to adapt to market changes. Strategy execution module: applies the formulated price strategy to the auction platform to realize automatic pricing; in this embodiment, the strategy execution module includes an automatic pricing system, and the strategy execution module inputs the formulated price strategy into the automatic pricing system so that it performs pricing according to preset rules and logic, thereby ensuring that the price strategy is applied to all auction activities in real time and accurately; Application monitoring module: monitors the execution of price strategies, including key indicators such as transaction rate and profit margin, and adjusts price strategies in a timely manner based on monitoring results to cope with market changes and changes in user needs; Service management module: coordinates and manages various service resources on the auction platform, such as customer service and technical support, to ensure that these resources can respond to user needs efficiently and accurately, and continuously monitors the quality of services provided by the platform, including key indicators such as response time and problem resolution rate, to ensure that users receive a high-quality service experience; Auction management module: manage the entire process of auction activities, including the creation, release, bidding, settlement and other links of the auction, ensure the smooth progress of the auction activities, formulate the rules and processes of the auction activities, and supervise the implementation of these rules to prevent fraud and violations. At the same time, conduct data analysis on auction activities, including the value of the auction items, bidding conditions, user participation, etc., to provide data support for platform operation and pricing model optimization; Interactive feedback module: Design a user-friendly interface to realize interactive functions between users and the platform, such as search, bidding, purchase, etc., to ensure that users can easily participate in auction activities, collect user opinions and suggestions, respond to user feedback in a timely manner, and solve problems encountered by users during use; in this embodiment, user feedback is collected through multiple channels such as online customer service, email, and telephone, and a special team is set up to handle user feedback and complaints.

[0037] The embodiments of the present invention are given for the purpose of illustration and description, and are not intended to be exhaustive or to limit the invention to the disclosed forms. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present invention and to enable those of ordinary skill in the art to understand the present invention and thereby design various embodiments with various modifications suitable for specific uses.

Claims

1. A system for building a dynamic evaluation and accurate pricing model for auction items on a shared auction platform, comprising a data collection module, which is responsible for collecting various data related to the evaluation of auction item value, including market data, user data, and auction item data, and is characterized in that: It also includes an indicator analysis module, a dynamic evaluation module, a pricing model building module, and a strategy execution module; The index analysis module is responsible for determining the evaluation indexes that affect the value of the auction items according to the attributes of the auction items and the market environment, and quantifying the indexes to convert non-quantitative indexes into quantitative indexes; The dynamic evaluation module is responsible for dynamically evaluating the value of the auction item based on the quantified evaluation indicators using machine learning or statistical methods; The pricing model building module is responsible for building an accurate pricing model based on the dynamic evaluation results of the auction item value and formulating a reasonable pricing strategy for the auction item; and the pricing model building module includes the following sub-modules: Data integration submodule: receives the output data from the auction item value dynamic evaluation module, including various evaluation indicators and quantitative results, and cleans, organizes and formats the data; Feature screening submodule: Use correlation analysis to screen features, and perform feature engineering based on business needs and data characteristics to extract more valuable features; Model training submodule: selects a suitable machine learning algorithm as the basis of the pricing model and trains the model using the preprocessed data and the filtered feature set; Strategy formulation submodule: formulates pricing strategies based on the forecast results of the pricing model, combined with factors such as market demand and competition situation; Verification and optimization submodule: Use new data sets to verify the pricing model, and adjust and optimize the model based on the verification results. Continue to collect user feedback and market data, and iterate and update the model to adapt to market changes. The strategy execution module is responsible for applying the formulated price strategy to the auction platform to achieve automated pricing.

2. According to claim 1, the system for building a dynamic evaluation and accurate pricing model for auction items on a shared auction platform is characterized in that: The system further comprises: Application monitoring module: responsible for monitoring the implementation of price strategies and adjusting price strategies in a timely manner based on monitoring results to cope with market changes and changes in user needs; Service management module: responsible for coordinating and managing various service resources on the auction platform, and continuously monitoring the service quality provided by the platform to ensure that users receive a high-quality service experience; Auction management module: responsible for the entire process management of auction activities, as well as formulating the rules and processes of auction activities and supervising the implementation of these rules. At the same time, it conducts data analysis on auction activities to provide data support for platform operation and pricing model optimization; Interactive feedback module: responsible for designing a user-friendly interface, realizing the interactive function between users and the platform, collecting users' opinions and suggestions, responding to user feedback in a timely manner, and solving problems encountered by users during use.

3. According to claim 1, the system for building a dynamic evaluation and accurate pricing model for auction items on a shared auction platform is characterized in that: The data acquisition module captures relevant data from e-commerce platforms and auction websites through web crawlers, cleans the captured data, removes duplicate, invalid or erroneous data, and performs necessary preprocessing.

4. The system for building a dynamic evaluation and accurate pricing model for auction items on a shared auction platform according to claim 1, characterized in that: The evaluation indicators include quality indicators, market indicators, and user preference indicators, and the quantitative processing of the indicators includes direct quantification, indirect quantification, and conversion quantification; The direct quantization specifically includes: Numerical quantification: directly using specific numerical values ​​to express the size or degree of an indicator; Proportional quantification: expressing the size of an indicator by calculating the ratio or percentage; The indirect quantification specifically includes: Grading: Divide the indicators into different grades or score ranges and assign corresponding values ​​to each grade or score; Frequency statistics: the number of times or frequency at which a statistical indicator appears within a certain time or space range; Sorting and assignment: sort the indicators in a certain order and assign corresponding values ​​to each sorting position; The conversion quantization specifically includes: Questionnaire scoring: Through questionnaire surveys, respondents are asked to score indicators based on their perceptions or experiences; Expert evaluation: Invite experts or industry authorities to evaluate the indicators and give corresponding values ​​or grades; Data analysis: Use data mining and machine learning techniques to analyze large amounts of data to extract quantitative information related to indicators.

5. The system for building a dynamic evaluation and accurate pricing model for auction items on a shared auction platform according to claim 1, characterized in that: The dynamic assessment module comprises the following steps: S1. Model selection: Select an appropriate evaluation model based on data characteristics and evaluation requirements; S2. Training model: Use historical data to train the model to learn the mapping relationship between features and auction item values, and use cross-validation methods to evaluate the performance of the model and adjust and optimize parameters; S3, real-time evaluation: when new data arrives, the dynamic evaluation module uses the trained model to evaluate the new data and calculate the real-time value of the auction item; S4. Result output: The evaluation results are immediately output to the user and pricing model building module in the form of numerical values ​​and charts, and the evaluation results include the real-time value of the auction item, the trend of value changes, and the importance of related features.

6. The system for building a dynamic evaluation and accurate pricing model for auction items on a shared auction platform according to claim 5, characterized in that: The evaluation model sets a strategy for regular updates, and when updating, the model is incrementally updated or retrained using new data to reflect the latest trends in market changes and user preferences.

7. The system for building a dynamic evaluation and accurate pricing model for auction items on a shared auction platform according to claim 1, characterized in that: The feature screening submodule uses statistical methods to evaluate the correlation between each feature and the auction item price, and selects features based on the correlation score.

8. The system for building a dynamic evaluation and accurate pricing model for auction items on a shared auction platform according to claim 1, characterized in that: The feature engineering of the feature screening submodule specifically includes: Feature combination: combine two or more features into a new feature to capture the relationship between them. The combination methods include product features, ratio features, difference features and logical combination. Feature transformation: Convert the original features into new features through mathematical transformation to improve the performance of the model. This is achieved by processing the nonlinear relationship between features through logarithmic transformation, square root transformation, polynomial transformation, and inverse function transformation, and by using feature scaling to eliminate the impact of the dimensions and value ranges between features. Feature encoding: converting categorical features into numerical features; Feature binning: Divide continuous features into multiple discrete intervals; Time feature extraction: Extract year, month, day, and hour features from the timestamp.

9. The system for building a dynamic evaluation and accurate pricing model for auction items on a shared auction platform according to claim 1, characterized in that: The pricing model training submodule specifically adjusts model parameters, optimizes model performance, and evaluates the model's prediction accuracy and generalization ability through cross-validation and grid search methods to ensure that the model can accurately reflect the relationship between the value and price of the auction item.

10. The system for building a dynamic evaluation and accurate pricing model for auction items on a shared auction platform according to claim 1, characterized in that: The strategy execution module includes an automated pricing system, and the operation logic of the strategy execution module is: by inputting the formulated price strategy into the automated pricing system, it is priced according to preset rules and logic, and the auction price is adjusted in real time.

Images