Pricing method, device and system of transaction object and electronic equipment
By updating the pricing function using the association relationship of predicted returns and historical transaction records in transaction pricing, the problem that existing transaction pricing plans cannot adapt to market changes and balance the returns of both parties in the transaction are solved, and a more reasonable transaction price and profit distribution is achieved.
Patent Information
- Application Number
- CN202410869241.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-28
- Publication Date
- 2024-11-26
AI Technical Summary
The existing transaction pricing plan cannot adapt to market changes, has poor flexibility, and cannot effectively balance the returns of both parties to the transaction.
The first pricing function determines the predicted returns of multiple trading entities of the transaction object, updates the direction, and adjusts the weight according to the association relationship and return in the historical transaction record, updates the pricing function to determine the transaction price, and ensures the rationality of the transaction price and balance.
It has achieved comprehensive consideration of the returns of multiple trading entities, adapted to market changes, and improved the rationality of transaction prices and the balance of returns of all parties to the transaction.
Smart Images

Figure CN119027193A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of data processing, and in particular to a pricing method, device, system and electronic equipment for a transaction object. Background Art
[0002] Transaction pricing is an important means to achieve optimal resource allocation. In order for buyers and sellers to successfully complete transactions, it is necessary to price the transaction objects (such as goods or services).
[0003] In related technologies, a fixed price pricing scheme is usually adopted, in which the price is fixed within a certain period of time. The fixed price scheme has the advantages of being simple and easy to use, and can reduce the communication cost of the bargaining link and quickly complete the transaction. However, the fixed price scheme cannot adapt to market changes, nor can it balance the benefits of both parties to the transaction well, and has poor flexibility. Summary of the invention
[0004] The present invention provides a pricing method, device and system for transaction objects, electronic equipment and storage medium.
[0005] In a first aspect, the present disclosure provides a pricing method for a transaction object, comprising:
[0006] Determining a plurality of subject predicted benefits of a plurality of transaction subjects corresponding to the transaction object through a first pricing function, and determining a plurality of subject update directions of the first pricing function corresponding to the plurality of transaction subjects according to the plurality of subject predicted benefits;
[0007] According to the historical transaction records of the transaction object, determine the association relationship between the historical income of multiple transaction subjects of the transaction object, and determine the income adjustment weight according to the association relationship; wherein the income adjustment weight is used to make the historical income of the multiple transaction subjects meet the preset income balance condition;
[0008] Processing the multiple subject update directions according to the revenue adjustment weights to obtain a function update direction of the first pricing function;
[0009] The first pricing function is updated according to the function update direction to obtain a second pricing function, and the transaction price of the transaction object is determined by the second pricing function.
[0010] In a second aspect, the present disclosure provides a pricing device for a transaction object, comprising:
[0011] a revenue determination module, adapted to determine a plurality of subject predicted revenues of a plurality of transaction subjects corresponding to the transaction object through a first pricing function, and determine a plurality of subject update directions of the first pricing function corresponding to the plurality of transaction subjects according to the plurality of subject predicted revenues;
[0012] A weight determination module, adapted to determine the association relationship between the historical earnings of multiple transaction subjects of the transaction object according to the historical transaction records of the transaction object, and determine the earnings adjustment weight according to the association relationship; wherein the earnings adjustment weight is used to make the historical earnings of the multiple transaction subjects meet a preset earnings balance condition;
[0013] a processing module, adapted to process the update directions of the plurality of entities according to the revenue adjustment weights, to obtain a function update direction of the first pricing function;
[0014] The updating pricing module is adapted to update the first pricing function according to the function updating direction to obtain a second pricing function, and determine the transaction price of the transaction object through the second pricing function.
[0015] In addition, the present disclosure also provides a pricing system for transaction objects, including: the pricing device for transaction objects mentioned above, a first network device corresponding to a first transaction subject, and a second network device corresponding to a second transaction subject.
[0016] In a third aspect, the present disclosure provides an electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores one or more computer programs executable by the at least one processor, and one or more of the computer programs are executed by the at least one processor so that the at least one processor can execute the above-mentioned pricing method for transaction objects.
[0017] In a fourth aspect, the present disclosure provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the above-mentioned pricing method for transaction objects when executed by a processor / processing core.
[0018] The pricing method of the transaction object provided by the embodiment of the present disclosure first determines the predicted earnings of multiple transaction subjects corresponding to the transaction object through the first pricing function, and then determines the update direction of the multiple transaction subjects corresponding to the first pricing function; then, according to the historical transaction records of the transaction object, the correlation relationship between the historical earnings of the multiple transaction subjects and the earnings adjustment weight are determined, and then the update directions of the multiple subjects are processed through the earnings adjustment weight to obtain the function update direction of the first pricing function, so as to update the first pricing function and then determine the transaction price of the transaction object. It can be seen that this method comprehensively considers the earnings of multiple transaction subjects, and through the earnings adjustment weight, it can adapt to market changes and better balance the earnings of all parties to the transaction.
[0019] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] The accompanying drawings are used to provide a further understanding of the present disclosure and constitute a part of the specification. Together with the embodiments of the present disclosure, they are used to explain the present disclosure and do not constitute a limitation of the present disclosure. The above and other features and advantages will become more apparent to those skilled in the art by describing detailed example embodiments with reference to the accompanying drawings, in which:
[0021] Figure 1 A schematic diagram showing a process of a pricing method for a transaction object provided by an embodiment of the present disclosure;
[0022] Figure 2 A flowchart of a pricing method for a transaction object provided by an example of the present disclosure;
[0023] Figure 3 A schematic diagram showing the structure of a pricing device for a transaction object provided by another embodiment of the present application is shown;
[0024] Figure 4 A block diagram of an electronic device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION
[0025] In order to enable those skilled in the art to better understand the technical solutions of the present disclosure, the exemplary embodiments of the present disclosure are described below in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, the description of well-known functions and structures is omitted in the following description.
[0026] In the absence of conflict, the various embodiments of the present disclosure and the various features therein may be combined with each other.
[0027] As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.
[0028] The terms used herein are only used to describe specific embodiments and are not intended to limit the present disclosure. As used herein, the singular forms "a" and "the" are also intended to include plural forms, unless the context clearly indicates otherwise. It will also be understood that when the terms "including" and / or "made of" are used in this specification, the presence of the features, wholes, steps, operations, elements and / or components is specified, but the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or groups thereof is not excluded. "Connected" or "connected" and similar words are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect.
[0029] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art. It will also be understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant art and the present disclosure, and will not be interpreted as having an idealized or overly formal meaning unless explicitly defined as such herein.
[0030] Figure 1 A flowchart of a pricing method for a transaction object provided by an embodiment of the present disclosure is shown. Figure 1 As shown, the pricing method of the transaction object includes:
[0031] Step S110: Determine multiple subject predicted benefits of multiple transaction subjects corresponding to the transaction object through the first pricing function, and determine multiple subject update directions of the first pricing function corresponding to the multiple transaction subjects based on the multiple subject predicted benefits.
[0032] Among them, the transaction objects include various objects available for trading, such as various physical objects or virtual objects such as goods, services, data, and electronic vouchers. The transaction subject of the transaction object refers to: the subject participating in the transaction process of the transaction object. The transaction subject of the transaction object includes at least: a first transaction subject corresponding to the buyer user and a second transaction subject corresponding to the seller user. In addition, in the scenario where the two parties to the transaction conduct a transaction through a third-party trading platform, the transaction subject of the transaction object also includes a third transaction subject corresponding to the trading platform. In addition, taking into account the special circumstances that there may be multiple parties purchasing together or multiple parties selling together, the number of the first transaction subject corresponding to the buyer user and the second transaction subject corresponding to the seller user can be multiple. Therefore, this application does not limit the specific number of transaction subjects of the transaction object.
[0033] Among them, the subject predicted income of each transaction subject corresponding to the transaction object is determined by the first pricing function, thereby obtaining multiple subject predicted income corresponding to multiple transaction subjects one by one. Among them, the subject predicted income is used to characterize the predicted income of the corresponding transaction subject in the current round of transactions. For example, the subject predicted income of the first transaction subject corresponding to the buyer user is used to characterize the predicted income of the first transaction subject in the current round of transactions, the subject predicted income of the second transaction subject corresponding to the seller user is used to characterize the predicted income of the second transaction subject in the current round of transactions, and the subject predicted income of the third transaction subject corresponding to the transaction platform is used to characterize the predicted income of the agent corresponding to the transaction platform in the current round of transactions.
[0034] According to the predicted revenue of multiple entities, it is possible to determine the multiple subject update directions of the first pricing function corresponding to the multiple transaction entities. Among them, the multiple subject update directions correspond to the multiple transaction entities one by one. Each subject update direction refers to: a function update direction that can increase the predicted revenue of the corresponding transaction entity. For example, according to the three subject predicted revenues corresponding to the three transaction entities, the three subject update directions of the first pricing function can be determined. Among them, when the first pricing function is updated in the subject update direction corresponding to the first transaction entity, the predicted revenue of the first transaction entity will increase; when the first pricing function is updated in the subject update direction corresponding to the second transaction entity, the predicted revenue of the second transaction entity will increase; when the first pricing function is updated in the subject update direction corresponding to the third transaction entity, the predicted revenue of the third transaction entity will increase. Among them, each subject update direction can be calculated by performing gradient derivation on the first pricing function.
[0035] Step S120: Determine the correlation relationship between the historical income of multiple transaction subjects of the transaction object based on the historical transaction records of the transaction object, and determine the income adjustment weight based on the correlation relationship; wherein the income adjustment weight is used to make the historical income of the multiple transaction subjects meet the preset income balance condition.
[0036] Among them, the historical transaction records of the transaction object include: the historical transaction prices of the transaction object in multiple historical transaction rounds in the past. Accordingly, based on the historical transaction records of the transaction object, the historical returns of multiple transaction subjects of the transaction object can be determined. The historical returns of a transaction object are used to characterize the returns obtained by the transaction object in historical transactions. The correlation between the historical returns of multiple transaction subjects is used to characterize the correlation between the historical returns of multiple transaction subjects. For example, the correlation between the historical returns of multiple transaction subjects can be characterized by various calculation results such as variance, covariance, and difference between the historical returns of multiple transaction subjects.
[0037] Accordingly, the income adjustment weight can be determined according to the correlation between the historical income of multiple transaction subjects of the transaction object; wherein the income adjustment weight is used to make the historical income of multiple transaction subjects meet the preset income balance condition. Specifically, through the correlation between the historical income of multiple transaction subjects, the distribution of the historical income of each transaction subject can be determined, and then the income distribution between multiple transaction subjects can be adjusted through the income adjustment weight, so that the income distribution between multiple subjects is more balanced.
[0038] Step S130: Processing multiple subject update directions according to the revenue adjustment weights to obtain the function update direction of the first pricing function.
[0039] Among them, each subject update direction is calculated based on the interests of the corresponding transaction subject. Therefore, in order to make the final function update direction take into account the interests of each transaction subject and make the profit distribution more reasonable, in this step, the weight adjustment processing (such as weighted processing) of multiple subject update directions can be performed through the profit adjustment weight to obtain the final function update direction.
[0040] Step S140: Update the first pricing function according to the function update direction to obtain a second pricing function, and determine the transaction price of the transaction object through the second pricing function.
[0041] The first pricing function is updated according to the function update direction to obtain an updated second pricing function. The second pricing function is obtained after updating the first pricing function according to the function update direction. Since the above update process comprehensively considers the profit balance of multiple transaction entities, determining the transaction price of the transaction object according to the updated second pricing function can improve the rationality of the transaction price.
[0042] It can be seen that in the pricing method of the transaction object provided in the embodiment of the present disclosure, first, the predicted earnings of multiple subjects corresponding to the multiple transaction subjects of the transaction object are determined through the first pricing function, and then the update direction of the multiple subjects corresponding to the multiple transaction subjects of the first pricing function is determined; then, according to the historical transaction records of the transaction object, the correlation relationship between the historical earnings of the subjects of the multiple transaction subjects and the earnings adjustment weight are determined, and then the update directions of the multiple subjects are processed through the earnings adjustment weight to obtain the function update direction of the first pricing function, so as to update the first pricing function and then determine the transaction price of the transaction object. It can be seen that this method comprehensively considers the earnings of multiple transaction subjects, and through the earnings adjustment weight, it can adapt to market changes and better balance the earnings of all parties to the transaction.
[0043] In addition, those skilled in the art may also make various modifications and variations to the above embodiments:
[0044] In an optional implementation, the pricing function is updated in each transaction round, and accordingly, the first pricing function is the pricing function in the previous transaction round, and the second pricing function is the pricing function in the current transaction round. In addition, the transaction subject may include: a first transaction subject corresponding to the buyer user, a second transaction subject corresponding to the seller user, and a third transaction subject corresponding to the transaction platform. Generally, the execution subject of the pricing method in this application corresponds to the third transaction subject.
[0045] In an optional implementation, before determining the predicted returns of multiple transaction subjects corresponding to the transaction object through the first pricing function, the following operations are further performed: receiving the object description data of the transaction object in the current transaction round sent by the first transaction subject; querying the candidate transaction object matching the object description data according to the object attribute data of the candidate transaction object published by the second transaction subject. It can be seen that in this implementation, the buyer user (i.e., the first transaction subject) can send a transaction request containing object description data to the transaction platform according to actual needs. In addition, the seller user (i.e., the second transaction subject) can send a transaction object containing object attribute information (i.e., a transaction object available for sale) to the transaction platform in advance. Accordingly, the transaction platform matches the object description information in the transaction request with the object attribute data of the transaction object published by each seller user, and determines the candidate transaction object that meets the buyer's needs according to the matching results. Accordingly, the predicted returns of multiple transaction subjects corresponding to the transaction object can be determined in the following manner: for the candidate transaction object matching the object description data, determining the predicted price of the candidate transaction object through the first pricing function; and calculating the predicted returns of multiple transaction subjects of the transaction object according to the predicted price.
[0046] In an optional implementation, considering that price is proportional to quality, when determining the predicted price of a candidate transaction object through the first pricing function, first, a quality score corresponding to the object attribute data of the candidate transaction object is determined through a preset quality evaluation function. The object attribute data is used to characterize the characteristics of the transaction object, and the quality evaluation function is used to determine the quality score of the transaction object based on the characteristics of the transaction object, and the quality score is used to evaluate the quality of the transaction object. Then, the predicted price corresponding to the quality score is determined through the first pricing function; wherein the predicted price increases monotonically with respect to the quality score. The quality evaluation function can improve the accuracy of the predicted price.
[0047] In an optional implementation, when calculating the predicted returns of multiple trading entities of a trading object based on the predicted price, it can be implemented in the following way: obtaining the expected price of each trading entity, and calculating the predicted returns of multiple trading entities of the trading object based on the expected price and the predicted price. The expected price is the psychological price of the trading entity, and the predicted return can be determined more accurately in combination with the user's psychological price.
[0048] In an optional implementation, the function update direction of the first pricing function may be determined in the following manner:
[0049] First, according to the income adjustment weight, multiple subject update directions are weighted to obtain the average update direction of the first pricing function. The average update direction is the update direction determined according to the principle of income balance in this application.
[0050] Then, the target update direction of the first pricing function is determined. The target update direction of the first pricing function can be determined by various known methods. The present application aims to adjust the target update direction determined by the existing method by using the average update direction, in order to obtain a more accurate update direction.
[0051] Next, a first objective optimization problem is constructed based on the average update direction of the first pricing function, the target update direction of the first pricing function, and the reweighting parameter to be solved. The first objective optimization problem includes: under the constraint that the target update direction is the same as the average update direction and the difference between the target update direction and the average update direction is less than a preset threshold, the preset operation result between the reweighting parameter and the target update direction satisfies the preset optimization goal. It can be seen that the core of the first objective optimization problem is: (1) ensuring that the target update direction is the same as the average update direction: if the directions of the two update directions are different, it often means that the currently determined update direction is incorrect; (2) the difference between the target update direction and the average update direction is less than a preset threshold: if the difference between the two update directions is too large, it usually means that at least one update direction is not accurate enough.
[0052] Finally, the first objective optimization problem is solved to obtain the parameter value of the reweighted parameter to be solved; and according to the parameter value of the reweighted parameter to be solved, the target update direction of the first pricing function is adjusted to obtain the function update direction of the first pricing function. Under the condition that the constraint conditions of the first objective optimization problem are met, the first objective optimization problem is solved to obtain the parameter value of the reweighted parameter to be solved. Among them, the parameter value of the reweighted parameter is used to adjust the target update direction, so as to obtain a more accurate function update direction as the final update direction with the help of reweighting processing.
[0053] In another optional implementation, the number of transaction entities is N, the correlation between the historical returns of multiple transaction entities of the transaction object can be represented by a covariance matrix, and the return adjustment weight can be represented by an N-dimensional return adjustment weight vector; wherein the covariance matrix can be determined in the following manner: for any two transaction entities, the covariance between the historical returns of the two transaction entities is calculated to obtain N*N covariance values; based on the N*N covariance values, a covariance matrix with N rows and N columns is generated.
[0054] In another optional implementation, the income adjustment weight can be determined by the following method: constructing a second objective optimization problem, the second objective optimization problem includes: under the condition of satisfying the income balance condition, minimizing the preset operation result between the income adjustment weight vector and the covariance matrix; wherein the income balance condition includes: the preset operation result between the mean of the subject historical income of multiple transaction subjects of the transaction object and the income adjustment weight matches the preset expected income value. It can be seen that the core of the second objective optimization problem is to ensure the income balance between multiple transaction subjects.
[0055] For ease of understanding, the following takes a specific example as an example to describe in detail the specific implementation details of the pricing system of the transaction object in this application:
[0056] This example enables data buyers to add query requests (also called transaction requests) with data quality restrictions in the data market (i.e., trading platform), and also enables data buyers to control and restrict the quality of data sellers. Secondly, this example can also achieve overall utility balance in the data market and promote the overall smooth operation of the data market.
[0057] Among them, data pricing is an important part of realizing the data market, so obtaining high-quality training data is crucial to developing high-precision and robust machine learning models. However, obtaining such data may incur huge costs and take a lot of time. In related technologies, data markets mostly conduct data pricing through face-to-face negotiations. Quantitative pricing is based on simple linear relationship pricing or complex game pricing, and these pricing methods cannot meet the quality requirements of data buyers for purchased data. Although there are also pricing methods based on data quality in related technologies, it is impossible to achieve the goal of balanced market benefit distribution while controlling quality.
[0058] It can be seen that the pricing methods in the relevant technologies have at least the following defects: First, most data markets cannot enable buyers to control the quality of the data they want to purchase, that is, they lack the ability to control the quality of the purchased data. Second, the basic optimization goal of existing trading platforms is to maximize the income of seller users, but this may lead to a reduction in the purchase volume of buyers, which in turn affects the data demand of data buyers and ultimately affects the sellers' future profitability and the overall transaction volume of the market.
[0059] This example aims to solve the above problems and proposes a pricing method for transaction objects. Among them, the transaction objects in this example can be virtual data or physical goods, and this application does not limit this. Among them, this example aims to balance the benefits (also called benefits) of all parties to the transaction. To this end, this example defines the method of calculating the benefits of each transaction subject.
[0060] Among them, the profit calculation method of the agent (i.e. trading platform) is as follows:
[0061]
[0062] Among them, U α represents the revenue of the trading platform, ρ represents the ratio of each transaction taken by the trading platform, and p ij Indicates the data buyer b in the transaction i Query data sellers j Provided transaction object Q ij The data price (i.e. the predicted price mentioned above), i represents the buyer's ID, and j represents the seller's ID. ij Data Q representing the data buyer’s demand for it ij The expected value of (i.e., the expected price of the transaction subject), H(·) is used to represent the current query Q ij The probability of a deal is defined as:
[0063]
[0064] Among them, κ is a continuous random variable defined in [0,1], which means that when Q ij The price of ij Greater than the data buyer's expected psychological price v ij When querying the probability of successful transaction, κ can be obtained through various mathematical distributions. In this example, Beta distribution (α, β) is used in the experimental stage. Of course, κ can also be obtained based on normal distribution, Poisson distribution, etc., which is not limited in this application.
[0065] In addition, the seller's profit can be calculated as follows:
[0066]
[0067] In this example, the seller user's profit is: the total price of all data transactions minus a portion of the agency fee (i.e., the agency benefit). s represents the revenue of the seller user, and λ is the adjustment parameter.
[0068] In addition, the buyer's profit can be calculated as follows:
[0069]
[0070] In this example, the revenue of the data buyer (i.e., the buyer user) is the total expected return of all data transactions minus the price, minus the remaining agency fee (agent benefit). b Profits for buyer users.
[0071] Additionally, in this example, a price reference compensation is further set:
[0072] This example assumes that the data seller can provide a price reference set Among them, q i Indicates the quality score corresponding to the data, p i Indicates the reference price of the data on the seller's side. Considering that the price cannot deviate too far from the reference price, this example sets a reference price compensation function to control the query price:
[0073]
[0074] Among them, the mass fraction q i It can be obtained by weighted average of multiple data quality assessment functions φ(·), such as the cleanliness of the dataset label, the separability of the dataset features, etc. i It is a data quality score composed of quality parameters of multiple dimensions. According to the buyer's needs, the data quality evaluation function φ(·) can have multiple methods, which can be determined according to the dimensions of the buyer's quality requirements. This application does not limit the specific implementation details.
[0075] Among them, π represents the agent's data pricing function for the current market (i.e., the first pricing function). The specific definition of the pricing function is:
[0076] p ij =π(q ij )
[0077] The query Q ij The mass fraction of
[0078]
[0079] The demand put forward by the data buyer is defined as q ij ,qij It can be calculated according to the object description information of the transaction object to be purchased.
[0080] k is generated by the data buyer in the query Q ij Determine that each data quality assessment function φ k (Q ij ) can determine the score of the specific quality requirement put forward by a data buyer in the query, and the final query Q can be obtained through weighted aggregation ij The fraction q ij .
[0081] The overall optimization goal of this example is to update the pricing strategy so that U α ,U s ,U b ,T can be maximized, that is, the seller's benefit, the buyer's benefit, the trading platform benefit, and the price reference compensation value are all maximized. It should be noted that the price reference compensation value is determined by the reference price p i and the predicted price π(q i ), so its maximum value should be 0. In other words, in addition to maximizing the benefits of the three parties to the transaction, the optimization goal of this application also needs to make the value of the price reference compensation approach zero.
[0082] The above overall optimization goal can be expressed as follows:
[0083] arg max π (U α ,U s ,U b ,T)
[0084] stπ(q i )>π(q j ),i <j
[0085] q i >q j ,i <j.
[0086] The above expression is intended to illustrate that the price of the pricing function increases as the quality score increases.
[0087] In addition, in this example, in order to balance the benefits of all parties, a long-term risk balance factor B is further set. The long-term risk balance factor can also be called a weight factor, which is a specific implementation of the benefit adjustment weight mentioned above. Specifically, in order to control the long-term stability of the market, this example proposes a long-term risk balance factor B = [β1, β2, …, β m ] m×1, where m represents the number of different types of entities in the market. In the present invention, m=3 by default, that is, the transaction entities in the market include: agents, data sellers and data buyers, a total of three.
[0088] This example is mainly divided into three parts: one is the data seller, one is the data buyer (i.e. the first transaction subject), and the last part is the data market composed of agents (i.e. the trading platform). The general process of this example is that the data buyer makes a data query request with quality requirements to the agent. The agent searches for data that can meet the buyer's quality requirements based on the data query request provided by the data buyer, and generates a preliminary price for the current round based on the existing pricing function. In addition, the market initialization method in this example is as follows: initialize the utility value (i.e. the return value) of each party in the market, initialize the market's long-term risk balance factor Among them, B is a column vector consisting of three real numbers, β s ,β b ,β α Represent the balance factors of the seller, buyer and agent respectively.
[0089] Figure 2 A flowchart of the pricing method in this example is shown in FIG. Figure 2 As shown, taking the transaction process of round t as an example, the method includes the following steps:
[0090] Step S201: receiving object description data of a transaction object in a current transaction round sent by a first transaction subject.
[0091] The trading platform obtains the object description data contained in the data query request, and the object description data is used to determine the quality level of the transaction object. Specifically, taking the transaction process of round t as an example, the data buyer submits a data query with quality requirements to the agent.
[0092] Step S202: According to the object attribute data of the candidate transaction object released by the second transaction subject, query the candidate transaction object that matches the object description data.
[0093] Specifically, this step aims to determine candidate transaction objects in the transaction platform that match the object description data.
[0094] Step S203: Determine multiple subject predicted benefits of multiple transaction subjects corresponding to the transaction object through the first pricing function, and determine multiple subject update directions of the first pricing function corresponding to the multiple transaction subjects based on the multiple subject predicted benefits.
[0095] Among them, in round t, the market collects buyer demand Q ij , according to the pricing function π of the previous round t-1 t-1(i.e. the first pricing function) calculates the initial benefit value of each subject in the data market (i.e. the subject's predicted income): the seller's subject's predicted income is The buyer's principal forecasts a return of The agent's principal forecasted return is And, the current pricing function π is determined by the calculated benefit value t-1 Gradient Represents π t-1 Find the partial derivative to calculate the update direction g of the pricing function n ,n=s,b andα.
[0096] Among them, n in the above text represents any one of the three parties: the buyer, the seller, and the agent. It can be seen that by predicting the returns of multiple entities and performing gradient derivation operations on the first pricing function, multiple entity update directions corresponding to multiple transaction entities (i.e., g n ).
[0097] Step S204: Determine the covariance matrix between the historical returns of multiple transaction entities of the transaction object based on the historical transaction records of the transaction object, and determine the long-term risk balance factor based on the covariance matrix.
[0098] Among them, the long-term risk balance factor can be updated according to the previous transaction history. The update process is as follows:
[0099] First, the benefit mean, variance and covariance of each transaction participant (i.e., transaction subject) are calculated.
[0100] The average values of each trading entity are as follows:
[0101]
[0102] The variances of each trading entity are as follows:
[0103]
[0104] The covariance of each trading entity is as follows:
[0105]
[0106] Where i represents the transaction round, cov(U m ,U n ) t represents the covariance between any two trading entities in the tth trading round.
[0107] Through the above formula, the covariance matrix C∈R can be calculated 3×3, this example prevents excessive fluctuations in the benefits of one side of the market by optimizing the following minimum variance:
[0108]
[0109] Among them, B is the long-term risk balance factor, C is the covariance matrix, That is, E(U) represents the mean return of each trading entity in the tth round of transactions. U0 represents the average return value expected by the market, which can be pre-set by the agent based on historical records.
[0110] Specifically, the above problem can be optimized by the Lagrange multiplier method. The Lagrange equation is:
[0111]
[0112] Among them, L represents the Lagrange equation, λ1 and λ2 represent the Lagrange multipliers, which are parameters used to deal with constraints. By taking partial derivatives of B, λ1 and λ2 respectively, we can get
[0113]
[0114]
[0115]
[0116] By setting the above three derivatives equal to 0, we can find the long-term risk balance factor B that minimizes the market variance. That is, we can solve the following linear equations to obtain the long-term risk balance factor B for this round:
[0117]
[0118] Step S205: Process the update directions of multiple entities according to the long-term risk balance factor to obtain the average update direction of the first pricing function.
[0119] Among them, the update direction of multiple subjects is g n ,n=s,b andα. Specifically, the update direction g of the average expected benefit of the current round is calculated by the long-term risk balance factor B calculated previously. a :
[0120] g α =∑ n β n g n ,n=s,b andα
[0121] where β n for The equilibrium factor of a certain entity in the market, g nIt is the update direction of the market pricing function expected by a certain trading entity.
[0122] It can be seen that the long-term risk balance factor updates the direction of multiple entities g n After weighted processing, the average update direction g of the first pricing function can be obtained a .
[0123] Step S206: Determine the target update direction of the first pricing function, construct and solve the first target optimization problem according to the average update direction of the first pricing function, the target update direction of the first pricing function, and the reweighted parameter to be solved, and obtain the parameter value of the reweighted parameter to be solved.
[0124] Among them, the target update direction of the first pricing function can be determined through various known implementation methods, and this application does not limit this.
[0125] Because this example requires solving the following multi-objective optimization problem:
[0126] arg max π (U α ,U s ,U b ,T)
[0127] stπ(q i )>π(q j ),i <j
[0128] q i >q j ,i <j.
[0129] Therefore, it can be achieved through The formula is used to obtain the first step direction weight distribution [γ1,γ2,...,γ n ], thereby obtaining the preliminary weighted directions (i.e., the target update direction of the first pricing function).
[0130] Wherein, the target update direction of the first pricing function d = ∑ n γ n g n is the weighted average of the update direction, It is guaranteed to find the update direction that satisfies Pareto optimality and has the smallest norm. g n ,n=s,b andαThe convex hull of all combinations, Represents a standard simplex in 3D space. This example will use Instead of γ n g nIn order to optimize the market to a balanced market faster, this example will pass First find an update gradient with the largest difference from the update direction d (i.e. the least effective gradient). In order to improve the least effective gradient, all gradients close to the average (i.e. the average update direction g above) are balanced. a ) to ensure that when improving a certain gradient, other gradients will not be harmed. This example obtains the updated gradient by optimizing the following minimum and maximum problems (i.e., the first objective optimization problem) The reweighting parameter w of:
[0131]
[0132]
[0133] Among them, d represents the update direction, N g represents the dimension of the gradient, N g dimensional real number space, w=[w s ,w b ,w α ] indicates the update gradient The reweighting parameter, represents the optimal update direction, g a represents the average update direction obtained by the long-term risk balance factor, and ξ represents a hyperparameter that controls the stability of the algorithm.
[0134] According to Sion's minimum-maximum problem, this example can transform the above optimization problem into a maximum-minimum dual problem:
[0135]
[0136] The maximum-minimum dual problem can ensure that the target update direction is consistent with the average update direction, and the difference between the target update direction and the average update direction is small.
[0137] The Lagrange multiplier method is also used to transform the solution equation:
[0138]
[0139] Among them, the λ Lagrange multiplier is a parameter used to deal with constraints, and the solution to the above optimization problem is obtained as
[0140]
[0141] The optimal update direction d is obtained by * Substituting into the original Lagrangian equation, the reweighted parameter w can be obtained through the following optimization problem:
[0142]
[0143] It can be seen that the first objective optimization problem can be specifically the minimum-maximum problem mentioned above (such as the maximum-minimum dual problem). Among them, the reweighting parameter w is the parameter value of the reweighting parameter to be solved. * That is the function update direction of the first pricing function.
[0144] Step S207: According to the parameter value of the re-weighted parameter to be solved, the target update direction of the first pricing function is adjusted to obtain the function update direction of the first pricing function, and the first pricing function is updated according to the function update direction to obtain the second pricing function.
[0145] After determining the optimal update direction of the pricing function, To get the next pricing function (i.e., the second pricing function), where e represents the e-th update in the t-th round, η represents the update step of gradient descent, and the update process will iterate e times, and finally obtain a pricing function about the utility equilibrium of all parties in the market.
[0146] Step S208: Determine the transaction price of the transaction object through the second pricing function.
[0147] Specifically, according to the obtained pricing function, complete this round of transactions, distribute equity, and save the pricing function and corresponding transaction records.
[0148] In addition, those skilled in the art may make various changes and modifications to the above examples: for example, regarding the selection of hyperparameters, the parameter selection range in the present invention may be replaced in various ways, such as ξ, the number of updates e, and the expected average benefit of the market U0. For another example, the random distribution followed by the transaction probability may be diverse. For another example, the measurement functions of data quality may be diverse, and all of them convert the information in the data itself into a real number.
[0149] In summary, the present invention proposes a data pricing method that fairly distributes benefits. The multi-objective optimization problem can be solved by re-weighting the update direction of each optimization benefit, and a relatively balanced Pareto optimal solution can be obtained. The present invention proposes a method for determining the long-term development goals of the data market based on mean variance. The present invention proposes a data market based on data quality control, which can respond to and price queries with data quality requirements raised by users.
[0150] Figure 3 A schematic diagram of the structure of a pricing device for a transaction object provided by another embodiment of the present application is shown. Figure 3 As shown, the system includes:
[0151] The revenue determination module 31 is adapted to determine a plurality of subject predicted revenues of a plurality of transaction subjects corresponding to the transaction object through a first pricing function, and determine a plurality of subject update directions of the first pricing function corresponding to the plurality of transaction subjects according to the plurality of subject predicted revenues;
[0152] The weight determination module 32 is adapted to determine the association relationship between the subject historical earnings of multiple transaction subjects of the transaction subject according to the historical transaction records of the transaction subject, and determine the earnings adjustment weight according to the association relationship; wherein the earnings adjustment weight is used to make the subject historical earnings of the multiple transaction subjects meet the preset earnings balance condition;
[0153] A processing module 33, adapted to process the multiple subject update directions according to the revenue adjustment weights to obtain a function update direction of the first pricing function;
[0154] The updating pricing module 34 is adapted to update the first pricing function according to the function updating direction to obtain a second pricing function, and determine the transaction price of the transaction object through the second pricing function.
[0155] In an optional implementation, the first pricing function is a pricing function in a previous transaction round, and the second pricing function is a pricing function in a current transaction round;
[0156] Furthermore, the transaction subject comprises: a first transaction subject corresponding to the buyer user, a second transaction subject corresponding to the seller user, and a third transaction subject corresponding to the transaction platform, and the execution subject of the method corresponds to the third transaction subject.
[0157] In an optional implementation, the profit determination module is also used to: receive object description data of the transaction object in the current transaction round sent by the first transaction entity; query the candidate transaction object that matches the object description data based on the object attribute data of the candidate transaction object released by the second transaction entity; determine the predicted price of the candidate transaction object through a first pricing function for the candidate transaction object that matches the object description data; and calculate multiple entity predicted profits of multiple transaction entities of the transaction object based on the predicted price.
[0158] In an optional implementation, the revenue determination module is specifically used to:
[0159] Determining a quality score corresponding to the object attribute data of the candidate transaction object through a preset quality evaluation function;
[0160] A predicted price corresponding to the quality score is determined through the first pricing function; wherein the predicted price increases monotonically relative to the quality score.
[0161] In an optional implementation, the revenue determination module is specifically used to:
[0162] The expected price of each transaction subject is obtained, and multiple subject predicted returns of multiple transaction subjects of the transaction object are calculated according to the expected price and the predicted price.
[0163] In an optional implementation, the processing module is specifically used to:
[0164] According to the revenue adjustment weight, weighted processing is performed on the update directions of the multiple entities to obtain an average update direction of the first pricing function;
[0165] determining a target update direction of the first pricing function;
[0166] A first objective optimization problem is constructed according to the average update direction of the first pricing function, the target update direction of the first pricing function, and the reweighting parameter to be solved; wherein the first objective optimization problem includes: under the condition that the target update direction is the same as the average update direction and the difference between the target update direction and the average update direction is less than a preset threshold, making the preset operation result between the reweighting parameter and the target update direction meet the preset optimization target;
[0167] Solving the first objective optimization problem to obtain the parameter value of the reweighted parameter to be solved;
[0168] According to the parameter value of the re-weighted parameter to be solved, the target update direction of the first pricing function is adjusted to obtain the function update direction of the first pricing function.
[0169] In an optional implementation, the number of the transaction subjects is N, the correlation relationship between the subject historical earnings of the multiple transaction subjects of the transaction object is represented by a covariance matrix, and the earnings adjustment weight is represented by an N-dimensional earnings adjustment weight vector;
[0170] The covariance matrix is determined in the following manner: for any two transaction entities, the covariance between the historical returns of the two transaction entities is calculated to obtain N*N covariance values; and a covariance matrix of N rows and N columns is generated based on the N*N covariance values.
[0171] In an optional implementation, the revenue adjustment weight is determined by:
[0172] Constructing a second objective optimization problem, the second objective optimization problem comprising: minimizing a preset operation result between the revenue adjustment weight vector and the covariance matrix while satisfying the revenue balance condition;
[0173] The profit balance condition includes: a preset calculation result between the average of the historical profits of multiple transaction subjects of the transaction object and the profit adjustment weight matches a preset expected profit value.
[0174] In addition, the present application also provides a pricing system for transaction objects, including: the pricing device for transaction objects mentioned above, a first network device corresponding to a first transaction subject, and a second network device corresponding to a second transaction subject.
[0175] It can be understood that the above-mentioned various method embodiments mentioned in the present disclosure can be combined with each other to form a combined embodiment without violating the principle logic. Due to space limitations, the present disclosure will not repeat them. It can be understood by those skilled in the art that in the above-mentioned method of the specific implementation method, the specific execution order of each step should be determined according to its function and possible internal logic.
[0176] In addition, the present disclosure also provides a pricing device, an electronic device, and a computer-readable storage medium for a transaction object, all of which can be used to implement any pricing method for a transaction object provided by the present disclosure. The corresponding technical solutions and descriptions are referred to in the corresponding records of the method part and will not be repeated here.
[0177] Figure 4 A block diagram of an electronic device provided in an embodiment of the present disclosure.
[0178] Reference Figure 4 An embodiment of the present disclosure provides an electronic device, which includes: at least one processor 701; at least one memory 702, and one or more I / O interfaces 703, connected between the processor 701 and the memory 702; wherein the memory 702 stores one or more computer programs that can be executed by the at least one processor 701, and the one or more computer programs are executed by the at least one processor 701 so that the at least one processor 701 can execute the above-mentioned pricing method for transaction objects.
[0179] Each module in the above electronic device can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.
[0180] The present disclosure also provides a computer-readable storage medium on which a computer program is stored, wherein the computer program implements the above-mentioned pricing method for transaction objects when executed by a processor / processing core. The computer-readable storage medium may be a volatile or non-volatile computer-readable storage medium.
[0181] The embodiments of the present disclosure also provide a computer program product, including a computer-readable code, or a non-volatile computer-readable storage medium carrying the computer-readable code. When the computer-readable code runs in a processor of an electronic device, the processor in the electronic device executes the pricing method for the transaction object.
[0182] It will be appreciated by those skilled in the art that all or some of the steps, systems, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In a hardware implementation, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or may be implemented as hardware, or may be implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable storage medium, which may include a computer storage medium (or a non-transitory medium) and a communication medium (or a temporary medium).
[0183] As is known to those of ordinary skill in the art, the term computer storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable program instructions, data structures, program modules or other data). Computer storage media include, but are not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), static random access memory (SRAM), flash memory or other memory technology, portable compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and can be accessed by a computer. In addition, it is known to those of ordinary skill in the art that communication media typically contain computer-readable program instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transmission mechanism, and may include any information delivery medium.
[0184] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.
[0185] The computer program instructions for performing the operation of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages, such as Smalltalk, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. Computer-readable program instructions may be executed completely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be customized by utilizing the state information of the computer-readable program instructions, and the electronic circuit may execute the computer-readable program instructions, thereby realizing various aspects of the present disclosure.
[0186] The computer program product described herein may be implemented in hardware, software, or a combination thereof. In one optional embodiment, the computer program product is embodied as a computer storage medium, and in another optional embodiment, the computer program product is embodied as a software product, such as a software development kit (SDK), etc.
[0187] Various aspects of the present disclosure are described herein with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer-readable program instructions.
[0188] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processor of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.
[0189] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operating steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.
[0190] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to multiple embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of the module, program segment or instruction includes one or more executable instructions for realizing the specified logical function. In some alternative implementations, the function marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or action, or can be implemented with a combination of special hardware and computer instructions.
[0191] Example embodiments have been disclosed herein, and although specific terms are employed, they are used and should be interpreted only in a general illustrative sense and not for limiting purposes. In some instances, it will be apparent to those skilled in the art that, unless otherwise expressly noted, features, characteristics, and / or elements described in conjunction with a particular embodiment may be used alone or in combination with features, characteristics, and / or elements described in conjunction with other embodiments. Therefore, those skilled in the art will appreciate that various changes in form and detail may be made without departing from the scope of the present disclosure as set forth in the appended claims.
Claims
1. A pricing method for a transaction object, characterized in that: include: Determining a plurality of subject predicted benefits of a plurality of transaction subjects corresponding to the transaction object through a first pricing function, and determining a plurality of subject update directions of the first pricing function corresponding to the plurality of transaction subjects according to the plurality of subject predicted benefits; According to the historical transaction records of the transaction object, determine the association relationship between the historical income of multiple transaction subjects of the transaction object, and determine the income adjustment weight according to the association relationship; wherein the income adjustment weight is used to make the historical income of the multiple transaction subjects meet the preset income balance condition; Processing the multiple subject update directions according to the revenue adjustment weights to obtain a function update direction of the first pricing function; The first pricing function is updated according to the function update direction to obtain a second pricing function, and the transaction price of the transaction object is determined by the second pricing function.
2. The method according to claim 1, characterized in that The first pricing function is a pricing function in a previous transaction round, and the second pricing function is a pricing function in a current transaction round; Furthermore, the transaction subject comprises: a first transaction subject corresponding to the buyer user, a second transaction subject corresponding to the seller user, and a third transaction subject corresponding to the transaction platform, and the execution subject of the method corresponds to the third transaction subject.
3. The method according to claim 2, characterized in that Before determining the predicted returns of multiple transaction entities corresponding to the transaction object through the first pricing function, the method further includes: receiving object description data of the transaction object in the current transaction round sent by the first transaction entity; querying the candidate transaction object matching the object description data according to the object attribute data of the candidate transaction object released by the second transaction entity; The method of determining the predicted returns of multiple transaction entities corresponding to the transaction object through the first pricing function includes: determining the predicted price of the candidate transaction object that matches the object description data through the first pricing function; and calculating the predicted returns of multiple transaction entities of the transaction object based on the predicted price.
4. The method according to claim 3, characterized in that Determining the predicted price of the candidate transaction object by using the first pricing function includes: Determining a quality score corresponding to the object attribute data of the candidate transaction object through a preset quality evaluation function; A predicted price corresponding to the quality score is determined through the first pricing function; wherein the predicted price increases monotonically relative to the quality score.
5. The method according to claim 3 or 4, characterized in that: The calculating of the predicted returns of multiple transaction entities of the transaction object according to the predicted price includes: The expected price of each transaction subject is obtained, and multiple subject predicted returns of multiple transaction subjects of the transaction object are calculated according to the expected price and the predicted price.
6. The method according to any one of claims 1 to 4, characterized in that: The processing of the multiple subject update directions according to the revenue adjustment weight to obtain the function update direction of the first pricing function includes: According to the revenue adjustment weight, weighted processing is performed on the update directions of the multiple entities to obtain an average update direction of the first pricing function; determining a target update direction of the first pricing function; A first objective optimization problem is constructed according to the average update direction of the first pricing function, the target update direction of the first pricing function, and the reweighting parameter to be solved; wherein the first objective optimization problem includes: under the condition that the target update direction is the same as the average update direction and the difference between the target update direction and the average update direction is less than a preset threshold, making the preset operation result between the reweighting parameter and the target update direction meet the preset optimization target; Solving the first objective optimization problem to obtain the parameter value of the reweighted parameter to be solved; According to the parameter value of the re-weighted parameter to be solved, the target update direction of the first pricing function is adjusted to obtain the function update direction of the first pricing function.
7. The method according to any one of claims 1 to 4, characterized in that: The number of the transaction subjects is N, the correlation relationship between the subject historical returns of the multiple transaction subjects of the transaction object is represented by a covariance matrix, and the return adjustment weight is represented by an N-dimensional return adjustment weight vector; The covariance matrix is determined in the following manner: for any two transaction entities, the covariance between the historical returns of the two transaction entities is calculated to obtain N*N covariance values; and a covariance matrix of N rows and N columns is generated based on the N*N covariance values.
8. The method according to claim 7, characterized in that The income adjustment weight is determined in the following way: Constructing a second objective optimization problem, wherein the second objective optimization problem includes: minimizing a preset operation result between the revenue adjustment weight vector and the covariance matrix while satisfying the revenue balance condition; The profit balance condition includes: a preset calculation result between the average of the historical profits of multiple transaction subjects of the transaction object and the profit adjustment weight matches a preset expected profit value.
9. A pricing device for a transaction object, characterized in that: include: a revenue determination module, adapted to determine a plurality of subject predicted revenues of a plurality of transaction subjects corresponding to the transaction object through a first pricing function, and determine a plurality of subject update directions of the first pricing function corresponding to the plurality of transaction subjects according to the plurality of subject predicted revenues; A weight determination module, adapted to determine the association relationship between the historical earnings of multiple transaction subjects of the transaction object according to the historical transaction records of the transaction object, and determine the earnings adjustment weight according to the association relationship; wherein the earnings adjustment weight is used to make the historical earnings of the multiple transaction subjects meet a preset earnings balance condition; a processing module, adapted to process the update directions of the plurality of entities according to the revenue adjustment weights to obtain a function update direction of the first pricing function; The updating pricing module is adapted to update the first pricing function according to the function updating direction to obtain a second pricing function, and determine the transaction price of the transaction object through the second pricing function.
10. A pricing system for transaction objects, characterized in that: include: The pricing device for the transaction object as described in claim 9, the first network device corresponding to the first transaction subject, and the second network device corresponding to the second transaction subject.
11. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores one or more computer programs executable by the at least one processor, and the one or more computer programs are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 8.
12. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the computer program implements the method according to any one of claims 1 to 8.
13. A computer program product, characterized in that The invention comprises a computer-readable code, or a non-volatile computer-readable storage medium carrying the computer-readable code. When the computer-readable code runs in a processor of an electronic device, the processor in the electronic device executes the method as claimed in any one of claims 1 to 8.
