Auction risk prediction method, device and equipment for idle and waste material disposal and storage medium

Abstract

The invention discloses an auction risk prediction method, device and equipment for idle and waste material disposal and a storage medium, and relates to the technical field of auction, and the method comprises the steps: obtaining the basic attribute information and market environment information of a target auction subject matter; according to the basic attribute information and the market environment information, performing calculation to obtain a multi-dimensional risk factor; and inputting the multi-dimensional risk factor into a pre-trained machine learning model, and outputting a default probability of risk prediction of the target auction subject matter. According to the method, the default probability prediction accuracy can be improved.

Description

Technical Field

[0001] This application relates to the field of auction technology, and in particular to a method, apparatus, equipment and storage medium for predicting auction risks for the disposal of idle waste materials. Background Technology

[0002] To ensure the seriousness of the auction process and prevent the risk of malicious bidding or default after a successful bid, the auction deposit system is an indispensable core risk control element in such platforms. Essentially, the deposit is designed to cover potential default losses, and its reasonable amount should be determined primarily by two factors: first, the intrinsic value of the auctioned item (appraised value), which forms the basis for compensation; and second, the predicted probability of default by the bidder in this transaction (predicted default probability), which is a quantitative estimate of the probability of the risk occurring.

[0003] Currently, the two main methods for calculating auction deposits commonly used in the industry are: one is the fixed percentage method, which calculates the deposit based on a preset uniform percentage (such as 10%) according to the appraised value of the subject matter. This method is simple to implement but lacks flexibility; the other is the manual experience method, which involves operators manually setting a fixed amount based on subjective judgments such as the type of goods and market conditions. Although this method is somewhat targeted, it is inefficient and difficult to standardize and scale up.

[0004] However, existing technical solutions are inaccurate in predicting the risks of auction systems, specifically the probability of default. Summary of the Invention

[0005] This application provides a method, apparatus, equipment, and storage medium for predicting auction risks in the disposal of idle waste materials, which can improve the accuracy of predicting the probability of default.

[0006] To achieve the above objectives, this application adopts the following technical solution: Firstly, this application provides a method for predicting auction risks for the disposal of idle waste materials, including: Obtain basic attribute information and market environment information of the target auction item; Based on the aforementioned basic attribute information and market environment information, multi-dimensional risk factors are calculated; The multi-dimensional risk factors are input into a pre-trained machine learning model, which outputs the default probability of the target auctioned item.

[0007] Optionally, the multi-dimensional risk factors include category risk coefficient, new product rate, and market popularity index. The calculation of these multi-dimensional risk factors based on the basic attribute information and market environment information includes: Based on the aforementioned basic attribute information, determine the type of goods or services of the target auction item; Based on the historical average default rate and basic risk coefficient of the aforementioned material category, the category risk coefficient of the aforementioned material category is obtained. The condition rate of the target auction item is obtained based on the material's service life, maintenance status score, and appearance score included in the basic attribute information. The market heat index is obtained based on the historical performance rate of the current registered bidders in the market environment information.

[0008] Optionally, the method further includes: The deposit amount is calculated based on the probability of default and the appraised value of the auctioned item.

[0009] Optionally, calculating the deposit amount based on the probability of default and the appraised value of the auctioned item includes: The margin adjustment factor is determined based on the aforementioned default probability; The deposit amount is obtained by multiplying the appraised value of the auctioned item by the deposit adjustment factor.

[0010] Optionally, obtaining the category risk coefficient of the material category based on the historical average default rate and the basic risk coefficient of the material category includes:

[0011] in, Indicates time The Category risk coefficient of different types of goods. Indicates the first Basic risk coefficient of each type of material. This represents the risk sensitivity adjustment parameter. Indicates the first The historical average default rate for this type of goods category This represents the historical average default rate for all product categories across the entire platform.

[0012] Optionally, obtaining the condition rate of the target auction item based on the material's service life, maintenance status score, and appearance score included in the basic attribute information includes:

[0013] in, Indicates the rate of newness. Indicates the service life of the materials. This indicates the standard maximum service life of this type of material. Indicates the maintenance status score. Indicates appearance rating. Indicates the first weight. Indicates the second weight. This indicates the third weight.

[0014] Optionally, the step of obtaining the market popularity index based on the historical performance rate of currently registered bidders in the market environment information includes:

[0015] in, Indicates time Market popularity index This indicates the cumulative number of applicants for the current target item. Indicates the first The historical contract fulfillment rate of each registered bidder This represents the maximum historical fulfillment rate of all bidders in the system. Indicates the time decay coefficient. Indicates the first The registration period for each bidder This represents the average number of applicants for similar items in the past. This represents the market benchmark adjustment factor.

[0016] Secondly, this application provides a bidding risk prediction device for the disposal of idle waste materials, comprising: The acquisition module is used to acquire basic attribute information and market environment information of the target auction item; The processing module is used to calculate multi-dimensional risk factors based on the basic attribute information and market environment information. The prediction module is used to input the multi-dimensional risk factors into a pre-trained machine learning model and output the default probability of the target auctioned item.

[0017] Thirdly, this application provides a computing device, including a memory and a processor; The memory stores one or more computer programs, the one or more computer programs including instructions; when the instructions are executed by the processor, the computing device performs the method as described in any one of the first aspects.

[0018] Fourthly, this application provides a computer-readable storage medium for storing a computer program for performing the method as described in any one of the first aspects.

[0019] As can be seen from the above technical solution, this application has at least the following beneficial effects: This application utilizes a system to acquire basic attribute information and market environment information of the target auction item, extracting multi-dimensional risk factors such as category risk coefficient, newness rate, and market popularity index. This approach abandons the existing fixed-ratio method's singular and homogeneous risk assessment model and avoids the subjective bias of manual experience-based methods. By comprehensively characterizing the inherent attributes and dynamic market risks of the target item through multi-dimensional risk factors, and then inputting this information into a pre-trained machine learning model for intelligent analysis, the potential for bidder default can be more accurately quantified. This solves the main problem of inaccurate default probability prediction in existing technologies and provides a reliable risk basis for the reasonable setting of margin deposits.

[0020] Furthermore, based on the accurately predicted probability of default, the deposit amount is determined in conjunction with the appraised value of the auctioned item. This approach ensures basic coverage of loss compensation through the appraised value and achieves dynamic matching between the deposit and risk through the adjustment coefficient corresponding to the probability of default. In scenarios with high default risk, the deposit amount is increased accordingly to effectively cover potential default losses; in scenarios with low default risk, excessive deposits are avoided from tying up bidders' funds, thus lowering the participation threshold. This approach balances the platform's risk control needs with market participation vitality, overcoming the shortcomings of the fixed percentage method (lack of flexibility) and the inefficiency and difficulty in standardization and scaling of the manual experience method.

[0021] Furthermore, the calculation of multi-dimensional risk factors is based on objective data and quantified through clear mathematical formulas, ensuring the standardization and repeatability of the evaluation process. It can be adapted to auction scenarios in different categories and market environments, without relying on subjective human judgment, which greatly improves the efficiency and consistency of margin calculation and provides strong support for the large-scale operation of the auction platform.

[0022] It should be understood that the descriptions of technical features, technical solutions, beneficial effects, or similar language in this application do not imply that all features and advantages can be achieved in any single embodiment. Rather, it is understood that the description of a feature or beneficial effect means that a specific technical feature, technical solution, or beneficial effect is included in at least one embodiment. Therefore, the descriptions of technical features, technical solutions, or beneficial effects in this specification do not necessarily refer to the same embodiment. Furthermore, the technical features, technical solutions, and beneficial effects described in this embodiment can be combined in any suitable manner. Those skilled in the art will understand that embodiments can be implemented without one or more specific technical features, technical solutions, or beneficial effects of a particular embodiment. In other embodiments, additional technical features and beneficial effects may be identified in specific embodiments that do not embody all embodiments. Attached Figure Description

[0023] Figure 1 A flowchart illustrating a bidding risk prediction method for the disposal of idle waste resources, provided as an embodiment of this application; Figure 2 A schematic diagram of an auction risk prediction device for the disposal of idle waste materials provided in this application embodiment; Figure 3 This is a schematic diagram of a computing device provided in an embodiment of this application. Detailed Implementation

[0024] The terms "first," "second," and "third," etc., used in this application specification and accompanying drawings are used to distinguish different objects, not to limit a specific order.

[0025] In the embodiments of this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design that is described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design. Specifically, the use of the terms "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0026] To ensure clarity and conciseness in the description of the following embodiments, a brief introduction to the related technologies is given first: The subject of an auction refers to a specific asset or item that is the object of a transaction in an auction activity. It can be tangible assets such as idle or waste materials, equipment, and raw materials, or intangible assets such as property rights and usage rights. It is the object of competition among bidders, and its own attributes and market attention directly determine the risk level of the auction transaction.

[0027] The auction deposit is a certain amount of money that bidders must pay according to the platform rules before participating in the auction. Its purpose is to ensure the seriousness of the auction process and prevent malicious bidding, breach of contract after the transaction, etc. The amount must match the potential losses of breach of contract and is a risk control measure of the auction platform.

[0028] In the operation of auction platforms, the auction deposit system is a key risk control element. The reasonableness of its amount directly determines the balance between risk control effectiveness and market participation. The main problem with existing technologies is the inaccurate prediction of the probability of transaction default, which leads to unreasonable deposit amounts. This makes it impossible to achieve the goal of matching risk with deposits. In low-risk transaction scenarios, excessively high deposits will tie up bidders' capital costs, reduce their willingness to participate, and even cause auctions to fail. In high-risk transaction scenarios, insufficient deposits are difficult to cover potential losses such as price difference losses and re-auction operation costs caused by default. At the same time, manual experience-based methods also suffer from low calculation efficiency and poor standardization, making them unsuitable for the needs of large-scale auctions.

[0029] The main reason for the above problems lies in the obvious defects of the existing risk assessment system: the fixed-ratio method only uses the appraised value of the target asset as the basis for calculation, ignoring key risk factors such as differences in the type of target asset, the rate of newness, and market attention, and homogenizing transactions with different risk characteristics, resulting in an incomplete risk characterization; the manual experience method relies on the subjective judgment of operators and has not established a quantitative system based on objective data. It cannot accurately capture the complex risks of multiple factors intertwined with the attributes of the target asset and market dynamics, nor can it ensure the consistency of assessment standards, and it is inefficient; at the same time, the existing technology has not constructed a complete logical chain of data-factor-prediction, has not comprehensively characterized risks through multi-dimensional risk factors, and has not used machine learning models to explore historical risk patterns, resulting in a lack of scientific support for the prediction of default probability, ultimately causing the margin amount to be out of sync with the actual risk.

[0030] In view of this, embodiments of this application provide a method for predicting auction risks for the disposal of idle waste resources, which can be executed by a processing device. This processing device can be a terminal or a server. Terminals include, but are not limited to, smartphones, tablets, laptops, personal digital assistants, or smart wearable devices. Servers can be cloud servers, such as central servers in a central cloud computing cluster or edge servers in an edge cloud computing cluster. Alternatively, servers can be located in a local data center. A local data center refers to a data center directly controlled by the user.

[0031] To address the technical problems of inaccurate default probability prediction, limited risk assessment dimensions, and mismatch between margin and actual risk in existing auction margin calculation methods, this application proposes an auction risk prediction and margin calculation scheme based on multi-dimensional risk factors and machine learning models. First, by collecting data on the basic attributes of the auctioned item and the market environment, multi-dimensional quantifiable risk factors such as category risk coefficient, newness rate, and market popularity index are extracted to comprehensively characterize transaction risk. Then, a pre-trained machine learning model is used to intelligently analyze these risk factors and accurately output the default probability. Finally, a dynamic margin calculation rule is constructed based on the default probability and the appraised value of the item, achieving accurate matching between risk and margin amount. This approach solves the problems of low efficiency in manual experience-based methods and poor flexibility in fixed-ratio methods, balancing platform risk control needs with market participation vitality.

[0032] To make the technical solution of this application clearer and easier to understand, the following describes, in conjunction with the accompanying drawings, a method for predicting auction risks for the disposal of idle waste materials provided by an embodiment of this application. Figure 1 As shown, this figure is a flowchart of a bidding risk prediction method for the disposal of idle waste resources provided in an embodiment of this application. The method includes: S201. The processing equipment acquires basic attribute information and market environment information of the target auction item.

[0033] The target auction item refers to the specific transaction object to be auctioned. It can be tangible assets such as idle waste materials, equipment, and raw materials, or intangible assets such as property rights and usage rights. Its own attributes and market attention directly determine the risk level of the auction transaction, and it is the object of data collection and risk assessment in this application.

[0034] Basic attribute information is objective data that describes the inherent characteristics of the target auction item. It is the basis for characterizing the stability of the item's value and the risk of disposal. Specifically, it includes the type of material (such as general steel and special precision instruments), service life, maintenance condition score, appearance score, appraised value (the value assessment amount of the target item), and the standard maximum service life of this type of material.

[0035] Market environment information refers to external dynamic data related to the target auction item, reflecting the market's attention to the item, the competitive landscape, and potential transaction risks. Specifically, it includes the historical performance rate of current registered bidders, the cumulative number of registered bidders for the current item, and the average number of registered bidders for similar items in the past.

[0036] The auction platform's processing equipment serves as the execution vehicle. Through pre-set data acquisition channels, such as the transferor's information entry port, system backend database calls, and bidder information registration modules, two types of key data are acquired simultaneously. The first is basic attribute information that directly describes the inherent characteristics of the target auction item itself. The second is market environment information that reflects the dynamics of the auction market and the competitive environment of the target item. This provides complete and reliable raw data support for the subsequent comprehensive characterization of transaction risks and extraction of risk factors from both the target item itself and the market environment, avoiding one-sided risk assessment due to missing or singular data.

[0037] S202. The processing equipment calculates multi-dimensional risk factors based on basic attribute information and market environment information.

[0038] Multi-dimensional risk factors are quantifiable risk indicators derived through standardized calculations based on fundamental attribute information and market environment information. They serve as a bridge connecting raw data with risk prediction models, comprehensively depicting the potential risks of auction transactions from different dimensions. Multi-dimensional risk factors include category risk coefficient, newness rate, and market popularity index.

[0039] Specifically, the processing equipment determines the category of the target auction item based on basic attribute information; and obtains the category risk coefficient of the category of the item based on the historical average default rate and basic risk coefficient of the category of the item.

[0040] Material categories are classifications of target auction items based on their inherent attributes, uses, and characteristics. Examples include general steel, special precision instruments, hazardous chemical containers, and used computers. Different categories of target items differ significantly in terms of disposal difficulty, number of potential bidders, and disposal costs after default, making it one of the dimensions of risk assessment.

[0041] The historical average default rate refers to the proportion of transactions in the same category of goods as the target auctioned item within a preset statistical period (such as the past six months or one year) where the bidder failed to fulfill the contract after winning the bid (such as failure to pay on time or failure to complete delivery) out of the total number of transactions in that category. It is a historical data support reflecting the inherent default risk of that category.

[0042] The basic risk coefficient is initially set based on industry experience, expert judgment, or historical data. It corresponds to the initial risk quantification value (such as a value in the range of 0-1) for different categories of materials. It is the benchmark parameter for calculating the category risk coefficient and reflects the inherent risk differences of different categories of materials.

[0043] The category risk coefficient is a quantitative indicator calculated by combining the historical average default rate of a category of goods with the basic risk coefficient. It is a core factor that accurately depicts the inherent risk level of the target goods in that category and directly reflects the potential for default in auction transactions.

[0044] Specifically, the processing equipment first extracts key features related to the material category from the basic attribute information of the target auction item, such as keywords in the material name and description of its purpose, to clarify the specific material category to which the target item belongs, for example, determining that a certain item belongs to the category of specialized precision instruments. Then, the processing equipment calls the historical transaction database stored in the system's backend to extract the historical average default rate of the material category within a preset period, for example, to calculate the default transaction ratio of the specialized precision instrument category in the past six months, and at the same time retrieves the preset basic risk coefficient corresponding to the category. Finally, through preset standardized calculation rules, the historical average default rate of the category and the basic risk coefficient are integrated and calculated to obtain a category risk coefficient that can accurately quantify the inherent risk level of the category.

[0045] The formula for calculating the category risk coefficient is as follows:

[0046] in, Indicates time The Category risk coefficient of different types of goods. Indicates the first Basic risk coefficient of each type of material. This represents the risk sensitivity adjustment parameter. Indicates the first The historical average default rate for this type of goods category This represents the historical average default rate for all product categories across the entire platform.

[0047] This step addresses the problem of existing technologies neglecting the risk differences between different product categories and conducting homogeneous risk assessments. By first determining the product category and then quantifying the risk based on historical default data, it achieves an accurate characterization of the risk at the product category level, laying the foundation for subsequent multi-dimensional risk integration and default probability prediction.

[0048] The processing equipment obtains the depreciation rate of the target auction item based on the material's service life, maintenance status score, and appearance score included in the basic attribute information.

[0049] The service life of materials refers to the length of time (in years or months) that the target auctioned item has actually been used from the time it was put into use until the current auction point. It is an important indicator reflecting the degree of wear and tear and performance degradation of the target item. The longer the service life, the more serious the aging of the target item, and the lower its value and condition.

[0050] The maintenance status score is a quantitative score based on the daily maintenance records, inspection reports, and on-site inspection results of the subject property. For example, it is a score of 0-10 or a value in the range of 0-1. It is used to assess the maintenance level and functional integrity of the subject property. The higher the maintenance status score, the better the performance of the subject property is maintained and the lower the degree of wear and tear.

[0051] Appearance rating is a quantitative score given by visual observation (or image recognition analysis) of the appearance of the object, such as wear, corrosion, deformation, missing parts, etc. For example, a score of 0-10 or a value in the range of 0-1. It is an indicator that directly reflects the physical condition of the object. The higher the appearance rating, the better the physical integrity of the object and the less external damage.

[0052] The condition rate refers to the current condition and value retention ratio of the target auction item. It is a risk factor that quantifies the newness and value stability of the target item. It is expressed as a percentage or a value in the range of 0-1. The higher the condition rate, the more stable the value of the target item, the lower the difficulty of handling default after the auction, and the smaller the transaction risk.

[0053] Specifically, the processing equipment first extracts three main characteristic data directly related to the condition of the target item from the acquired basic attribute information: the item's service life, maintenance status score, and appearance score. Then, according to the preset standardized calculation rules, it performs weighted fusion calculation on these three types of characteristic data, and balances the impact of different dimensions on the depreciation rate by reasonably allocating the weight of each characteristic. Finally, it outputs a depreciation rate index that can accurately quantify the current condition and value retention ratio of the target auction item.

[0054] The formula for calculating the depreciation rate is:

[0055] in, Indicates the rate of newness. Indicates the service life of the materials. This indicates the standard maximum service life of this type of material. Indicates the maintenance status score. Indicates appearance rating. Indicates the first weight. Indicates the second weight. This indicates the third weight.

[0056] This step addresses the problem in existing technologies that rely solely on a single attribute to determine the value of an asset and ignore multi-dimensional state characteristics. By integrating three types of data—years of use, maintenance status, and appearance—it achieves a scientific quantification of the stability of the asset's intrinsic value, providing a basis for subsequent comprehensive assessment of transaction risks and matching reasonable margin amounts.

[0057] The processing equipment calculates the market heat index based on the historical performance rate of current registered bidders in the market environment information.

[0058] Currently registered bidders refer to all entities (individuals or enterprises) that have completed the registration process for the target auction item and are qualified to participate in the auction. Their ability and willingness to perform their obligations directly affect the risk of default in this transaction.

[0059] Historical performance rate refers to the proportion of a single bidder's successful bids within a preset statistical period (such as the past year) and the number of transactions in which they fulfill their obligations such as payment and delivery according to the platform rules, out of the total number of successful bids. It is expressed as a value or percentage in the range of 0-1. It is a key indicator for quantifying a bidder's individual willingness to perform and their credit level. The higher the historical performance rate, the lower the probability of default and the better the credit level of the bidder.

[0060] The market heat index is a quantitative indicator calculated based on data such as the quality of current auction participants (historical performance rate) and the scale of participation. It is a risk factor that reflects the market attention, competitive rationality, and potential transaction risks of the target auction item. The index value not only reflects the market's attention to the target item, but also reflects the credit level of the participating group through the bidders' historical performance rate, and is directly related to the level of default risk of this transaction.

[0061] Specifically, the processing equipment first extracts the individual historical performance rate data of all currently registered bidders from the acquired market environment information, while simultaneously retrieving auxiliary data stored in the system's backend (such as the maximum historical performance rate of bidders across the entire platform, the average number of bidders for similar historical targets, etc.). Subsequently, according to preset standardized calculation rules, the historical performance rates of the currently registered bidders are weighted, normalized, and dynamically adjusted, taking into account both the differences in individual bidder creditworthiness and the impact of registration time on transaction decisions, while also being calibrated in conjunction with market benchmark data for similar historical targets. Finally, it outputs a market heat index that comprehensively reflects the market activity level, the credit level of the participating groups, and the potential default risk of this auction.

[0062] The formula for calculating the market popularity index is:

[0063] in, Indicates time Market popularity index This indicates the cumulative number of applicants for the current target item. Indicates the first The historical contract fulfillment rate of each registered bidder This represents the maximum historical fulfillment rate of all bidders in the system. Indicates the time decay coefficient. Indicates the first The registration period for each bidder This represents the average number of applicants for similar items in the past. This represents the market benchmark adjustment factor.

[0064] This step addresses the problem of existing technologies that rely solely on the number of applicants to judge market activity and neglect the quality of participants. By combining bidders' historical performance rate with factors such as market participation scale and time, it achieves an accurate characterization of market environment risks, providing a basis for subsequent comprehensive assessment of transaction risks and dynamic matching of margin amounts.

[0065] S203 The processing equipment inputs multi-dimensional risk factors into a pre-trained machine learning model and outputs the default probability of the target auctioned item.

[0066] Pre-trained machine learning models refer to algorithmic models that have the ability to predict risks after being trained and optimized using historical auction transaction data (including historical multi-dimensional risk factors and corresponding transaction default results). Examples include classification models and regression models. Their function is to learn the mapping pattern between risk factors and default behavior. After training, they can be directly used for risk quantification prediction of new transactions and can quickly output results without retraining.

[0067] Risk prediction refers to the process of using a trained machine learning model, based on multi-dimensional risk factors, to estimate the probability of default that may occur in subsequent transactions of a target auction item. It is a transformation from risk characteristics to risk quantification.

[0068] The probability of default refers to the likelihood that, after a successful bid, the bidder will fail to fulfill its obligations, such as payment or physical delivery, within the stipulated period. It is an output indicator that quantifies the risk of this transaction and is expressed as a value or percentage in the range of 0-1. The higher the value, the greater the risk of default, and vice versa.

[0069] Specifically, after the processing equipment completes the calculation and integration of multi-dimensional risk factors, it organizes the standardized factor data (such as category risk coefficient 0.6, newness rate 0.7, market popularity index 1.2, etc.) into input data according to the format required by the model. Subsequently, the processing equipment calls the pre-trained machine learning model already deployed in the system and inputs the processed multi-dimensional risk factors into the model; based on the historical risk patterns learned during the training phase, the model performs intelligent reasoning analysis on the input new risk factors and comprehensively evaluates the synergistic effects of the target product category risk, its own integrity risk, and market environment risk. The final output is an accurate quantitative value for the probability of default in the target auction item transaction (e.g., 0.35, meaning a 35% probability of default). This step solves the problem of low prediction accuracy in existing technologies that rely on subjective human judgment or single rules to assess risk. By using machine learning models to uncover the complex correlation between multiple risk factors and default behavior, it achieves the scientific quantification of transaction risk and lays the foundation for the subsequent construction of dynamic calculation rules for risk and margin matching.

[0070] The processing equipment calculates the deposit amount based on the probability of default and the appraised value of the auctioned item.

[0071] Specifically, the processing equipment determines the deposit adjustment factor based on the probability of default; the deposit amount is obtained by multiplying the appraised value of the auctioned item by the deposit adjustment factor.

[0072] The margin adjustment factor is a risk adaptation parameter set based on the probability of default. It serves as a bridge connecting transaction risk and margin amount. Its function is to transform the abstract probability of default into a quantitative coefficient that can be directly used for calculation, thereby achieving dynamic adaptation of high risk with a high coefficient and low risk with a low coefficient.

[0073] The appraised value of the auctioned item refers to the objective value amount determined by a professional appraisal agency or platform based on factors such as the market value, condition, and category attributes of the item. It serves as the basic benchmark for calculating the deposit amount, ensuring that the deposit can cover the value loss of the item itself.

[0074] The deposit amount refers to the final amount of funds that bidders must pay according to the platform rules before participating in the bidding for the target auction item. It is a risk control measure to prevent the risk of transaction default. The amount must cover potential default losses without adding extra financial burden to bidders in low-risk transactions.

[0075] Specifically, the processing device first receives the default probability of the target auction item output from the previous steps. Based on preset mapping rules, such as the tiered rule: when the default probability is ≤0.2, the adjustment coefficient is 0.8; when the default probability is 0.2 < default probability ≤0.5, the adjustment coefficient is 1.0; when the default probability is >0.5, the adjustment coefficient is 1.5. It automatically matches and determines the corresponding margin adjustment coefficient. Essentially, it converts the risk level corresponding to the default probability into a calculable coefficient value. Subsequently, the processing equipment retrieves the appraised value of the target auction item, multiplies the appraised value by the determined deposit adjustment coefficient, and finally obtains the amount of deposit that the bidder needs to pay.

[0076] This step addresses the disconnect between margin requirements and actual risk in existing technologies. By using the logic of determining the adjustment coefficient based on the probability of default and adjusting the valuation based on the adjustment coefficient, it achieves dynamic adaptation between margin amount and transaction risk: in high default risk scenarios, the adjustment coefficient increases, and the margin amount increases accordingly, effectively covering potential default losses; in low default risk scenarios, the adjustment coefficient decreases, and the margin amount is reasonably reduced, lowering the cost of capital occupation for bidders and increasing their willingness to participate, ultimately achieving the dual goals of effective risk control and market activity.

[0077] Based on the above description, this application has the following beneficial effects: This application utilizes a system to acquire basic attribute information and market environment information of the target auction item, extracting multi-dimensional risk factors such as category risk coefficient, newness rate, and market popularity index. This approach abandons the existing fixed-ratio method's singular and homogeneous risk assessment model and avoids the subjective bias of manual experience-based methods. By comprehensively characterizing the inherent attributes and dynamic market risks of the target item through multi-dimensional risk factors, and then inputting this information into a pre-trained machine learning model for intelligent analysis, the potential for bidder default can be more accurately quantified. This solves the main problem of inaccurate default probability prediction in existing technologies and provides a reliable risk basis for the reasonable setting of margin deposits.

[0078] Furthermore, based on the accurately predicted probability of default, the deposit amount is determined in conjunction with the appraised value of the auctioned item. This approach ensures basic coverage of loss compensation through the appraised value and achieves dynamic matching between the deposit and risk through the adjustment coefficient corresponding to the probability of default. In scenarios with high default risk, the deposit amount is increased accordingly to effectively cover potential default losses; in scenarios with low default risk, excessive deposits are avoided from tying up bidders' funds, thus lowering the participation threshold. This approach balances the platform's risk control needs with market participation vitality, overcoming the shortcomings of the fixed percentage method (lack of flexibility) and the inefficiency and difficulty in standardization and scaling of the manual experience method.

[0079] Furthermore, the calculation of multi-dimensional risk factors is based on objective data and quantified through clear mathematical formulas, ensuring the standardization and repeatability of the evaluation process. It can be adapted to auction scenarios in different categories and market environments, without relying on subjective human judgment, which greatly improves the efficiency and consistency of margin calculation and provides strong support for the large-scale operation of the auction platform.

[0080] The above text combined Figure 1 The auction risk prediction method for the disposal of idle waste materials provided in this application embodiment has been described in detail. The apparatus and equipment provided in this application embodiment will be described below with reference to the accompanying drawings.

[0081] like Figure 2 As shown in the figure, this is a schematic diagram of an auction risk prediction device for the disposal of idle waste materials provided in an embodiment of this application. The device includes: The acquisition module 301 is used to acquire basic attribute information and market environment information of the target auction item; Processing module 302 is used to calculate multi-dimensional risk factors based on the basic attribute information and market environment information; The prediction module 303 is used to input the multi-dimensional risk factors into a pre-trained machine learning model and output the default probability of the target auctioned item.

[0082] Optionally, the processing module 302 is specifically used to determine the material category of the target auction item based on the basic attribute information; Based on the historical average default rate and basic risk coefficient of the aforementioned material category, the category risk coefficient of the aforementioned material category is obtained. The condition rate of the target auction item is obtained based on the material's service life, maintenance status score, and appearance score included in the basic attribute information. The market heat index is obtained based on the historical performance rate of the current registered bidders in the market environment information.

[0083] Optionally, the processing module 302 is also used to calculate the deposit amount based on the default probability and the appraised value of the auctioned item.

[0084] Optionally, the processing module 302 is specifically used to determine the margin adjustment coefficient based on the default probability; The deposit amount is obtained by multiplying the appraised value of the auctioned item by the deposit adjustment factor.

[0085] Optionally, processing module 302 is specifically used to obtain the category risk coefficient of the material category based on the historical average default rate and the basic risk coefficient of the material category, including:

[0086] in, Indicates time The Category risk coefficient of different types of goods. Indicates the first Basic risk coefficient of each type of material. This indicates the risk sensitivity adjustment parameter. Indicates the first The historical average default rate for this type of goods category This represents the historical average default rate for all product categories across the entire platform.

[0087] Optionally, the processing module 302 is specifically used to obtain the condition rate of the target auction item based on the material's service life, maintenance status score, and appearance score included in the basic attribute information, including:

[0088] in, Indicates the rate of newness. Indicates the service life of the materials. This indicates the standard maximum service life of this type of material. Indicates the maintenance status score. Indicates appearance rating. Indicates the first weight. Indicates the second weight. This indicates the third weight.

[0089] Optionally, processing module 302 is specifically used to obtain a market popularity index based on the historical performance rate of currently registered bidders in the market environment information, including:

[0090] in, Indicates time Market popularity index This indicates the cumulative number of applicants for the current target item. Indicates the first The historical contract fulfillment rate of each registered bidder This represents the maximum historical fulfillment rate of all bidders in the system. Indicates the time decay coefficient. Indicates the first The registration period for each bidder This represents the average number of applicants for similar items in the past. This represents the market benchmark adjustment factor.

[0091] The auction risk prediction device for the disposal of idle waste materials according to the embodiments of this application can correspond to the execution of the method described in the embodiments of this application, and the other operations and / or functions of each module / unit of the auction risk prediction device for the disposal of idle waste materials are respectively for the purpose of implementing Figure 1 For the sake of brevity, the corresponding processes of each method in the illustrated embodiments will not be described in detail here.

[0092] This application also provides a computing device. For example... Figure 3 As shown in the figure, this is a schematic diagram of a computing device provided in an embodiment of this application. The computing device 700 includes a bus 701, a processor 702, a communication interface 703, and a memory 704. The processor 702, the memory 704, and the communication interface 703 communicate with each other via the bus 701.

[0093] The 701 bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, Figure 3 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0094] The processor 702 can be any one or more of the following processors: central processing unit (CPU), graphics processing unit (GPU), microprocessor (MP), or digital signal processor (DSP).

[0095] The communication interface 703 is used for communication with external devices.

[0096] Memory 704 may include volatile memory, such as random access memory (RAM). Memory 704 may also include non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid state drive (SSD).

[0097] The memory 704 stores executable code, which the processor 702 executes to perform the aforementioned auction risk prediction method for the disposal of idle waste materials.

[0098] Specifically, in achieving Figure 2 In the case of the illustrated embodiment, and Figure 2 When the modules or units of the auction risk prediction device for the disposal of idle waste resources described in the embodiments are implemented by software, the execution... Figure 2 The software or program code required for the functions of each module / unit can be partially or wholly stored in the memory 704. The processor 702 executes the program code corresponding to each unit stored in the memory 704, and executes the aforementioned auction risk prediction method for the disposal of idle waste resources.

[0099] This application also provides a computer-readable storage medium. The computer-readable storage medium can be any available medium that a computing device can store, or a data storage device such as a data center containing one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive). The computer-readable storage medium includes instructions that instruct a computing device to execute the above-described auction risk prediction method for the disposal of idle waste resources.

[0100] This application also provides a computer program product comprising one or more computer instructions. When the computer instructions are loaded and executed on a computing device, all or part of the processes or functions described in this application are generated.

[0101] 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, or data center to another website, computer, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means.

[0102] When the computer program product is executed by a computer, the computer executes any of the aforementioned methods of the auction risk prediction method for the disposal of waste materials. The computer program product can be a software installation package; when any of the aforementioned methods of the auction risk prediction method for the disposal of waste materials needs to be used, the computer program product can be downloaded and executed on the computer.

[0103] The descriptions of the processes or structures corresponding to the above figures each have their own emphasis. For parts of a process or structure that are not described in detail, please refer to the relevant descriptions of other processes or structures.

[0104] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions within the technical scope disclosed in this application should be covered within the scope of protection of this application.

Claims

1. A method for predicting auction risks for the disposal of idle and waste materials, characterized in that, The method includes: Obtain basic attribute information and market environment information of the target auction item; Based on the aforementioned basic attribute information and market environment information, multi-dimensional risk factors are calculated; The multi-dimensional risk factors are input into a pre-trained machine learning model, which outputs the default probability of the target auctioned item.

2. The method according to claim 1, characterized in that, The multi-dimensional risk factors include category risk coefficient, new product rate, and market popularity index. These multi-dimensional risk factors are calculated based on the basic attribute information and market environment information, and include: Based on the aforementioned basic attribute information, determine the type of goods in the target auction item; Based on the historical average default rate and basic risk coefficient of the aforementioned material category, the category risk coefficient of the aforementioned material category is obtained. The condition rate of the target auction item is obtained based on the material's service life, maintenance status score, and appearance score included in the basic attribute information. The market heat index is obtained based on the historical performance rate of the current registered bidders in the market environment information.

3. The method according to claim 1, characterized in that, The method further includes: The deposit amount is calculated based on the probability of default and the appraised value of the auctioned item.

4. The method according to claim 3, characterized in that, The calculation of the deposit amount based on the probability of default and the appraised value of the auctioned item includes: The margin adjustment factor is determined based on the aforementioned default probability; The deposit amount is obtained by multiplying the appraised value of the auctioned item by the deposit adjustment factor.

5. The method according to claim 2, characterized in that, The category risk coefficient for the material category is obtained based on the historical average default rate and the basic risk coefficient of the material category, including: in, Indicates time The Category risk coefficient of different types of goods. Indicates the first Basic risk coefficient of each type of material. This indicates the risk sensitivity adjustment parameter. Indicates the first The historical average default rate for this type of goods category This represents the historical average default rate for all product categories across the entire platform.

6. The method according to claim 2, characterized in that, The process of obtaining the condition rate of the target auction item based on the material's service life, maintenance status score, and appearance score included in the basic attribute information includes: in, Indicates the rate of newness. Indicates the service life of the materials. This indicates the standard maximum service life of this type of material. Indicates the maintenance status score. Indicates appearance rating. Indicates the first weight. Indicates the second weight. This indicates the third weight.

7. The method according to claim 2, characterized in that, The market heat index is obtained based on the historical performance rate of currently registered bidders in the market environment information, including: in, Indicates time Market popularity index This indicates the cumulative number of applicants for the current target item. Indicates the first The historical contract fulfillment rate of each registered bidder This represents the maximum historical fulfillment rate of all bidders in the system. Indicates the time decay coefficient. Indicates the first The registration period for each bidder This represents the average number of applicants for similar items in the past. This represents the market benchmark adjustment factor.

8. A device for predicting auction risks for the disposal of idle waste resources, characterized in that, The device includes: The acquisition module is used to acquire basic attribute information and market environment information of the target auction item; The processing module is used to calculate multi-dimensional risk factors based on the basic attribute information and market environment information. The prediction module is used to input the multi-dimensional risk factors into a pre-trained machine learning model and output the default probability of the target auctioned item.

9. A computing device, characterized in that, Including memory and processor; The memory stores one or more computer programs, the one or more computer programs including instructions; when the instructions are executed by the processor, the computing device performs the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program for performing the method as described in any one of claims 1 to 7.

Images