A method and apparatus for allocating resource bits

Abstract

The application discloses a resource bit allocation method and device. The method comprises the following steps: acquiring an available resource bit set and a delivery subject set of a creative script object according to search keyword information; acquiring an effect score of each resource bit in the available resource bit set when each resource bit is allocated to each delivery subject in the delivery subject set based on a value parameter of the resource bit, an influence parameter of an audience of the creative script object and a characteristic parameter of the creative script object; and obtaining a resource bit allocation scheme which makes the sum of effect scores of all resource bits in the available resource bit set maximum based on the effect score of each resource bit allocated to each delivery subject and according to a set limited condition. The resource bit allocation method can balance indexes concerned by three parties, i.e. media, an audience of a creative script object and a delivery subject of the creative script object.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer, in particular to a resource position allocation method and device. BACKGROUND

[0002] The bidding way is used by the creative script object's delivery subject to participate in the resource position allocation of the media, and then the traffic is obtained by delivering the creative script object in the resource position obtained by bidding. The inventor finds that the success of the resource position allocation mechanism needs to consider the goals of three parties, i.e. the long-term revenue of the media, the audience satisfaction and the revenue of the creative script object's delivery subject, wherein the long-term revenue of the media is embodied by the social welfare (SW) index, the audience satisfaction is embodied by the Click-Through Rate (CTR) and Click Value Rate (CVR) indexes of the creative script object, and the revenue of the creative script object's delivery subject is embodied by the Gross Merchandise Volume (GMV) and Social Welfare (SW) indexes. However, due to the internal conflict of the indexes concerned by the above three parties, it is difficult to maximize all indexes of the three parties at the same time. Therefore, it is necessary to provide a resource position allocation scheme which can ensure the balance between the indexes concerned by the three parties. SUMMARY

[0003] In view of the above problems, the present application is proposed in order to provide a resource position allocation method and device which can overcome the above problems or at least partially solve the above problems.

[0004] In a first aspect, an embodiment of the present application provides a resource position allocation method, comprising the following steps:

[0005] According to the search keyword information, a set of available resource positions and a set of creative script object's delivery subjects are obtained;

[0006] Based on the value parameter of the resource position, the influence parameter of the audience of the creative script object and the characteristic parameter of the creative script object, an effect score is obtained when each resource position in the set of available resource positions is allocated to each delivery subject in the set of delivery subjects;

[0007] Based on the effect score of each resource position allocated to each delivery subject, a resource position allocation scheme is obtained according to a set of limiting conditions, which makes the sum of the effect scores of all resource positions in the set of available resource positions maximum.

[0008] In some optional embodiments, the value parameter of the resource position comprises: the bid of the delivery subject for the resource position;

[0009] The influence parameter of the audience of the creative script object comprises a click probability of the creative script object put by the putting subject when the resource position is allocated to the putting subject.

[0010] The characteristic parameter of the creative script object comprises a conversion probability of a promoted commodity of the creative script object and a commodity price of the promoted commodity.

[0011] The effect score of each resource position in the available resource position set allocated to each putting subject in the putting subject set is obtained based on the value parameter of the resource position, the influence parameter of the audience of the creative script object and the characteristic parameter of the creative script object, comprising:

[0012] The effect score of each resource position in the available resource position set allocated to each putting subject in the putting subject set is determined according to the preset parameter, the bid, the click probability, the conversion probability and the commodity price.

[0013] In some optional embodiments, the method further comprises:

[0014] The historical creative script object data in a preset time length of media is obtained, and the average click amount of the creative script object, the average commodity conversion amount of the creative script object promoted commodity and the average commodity sales amount in the preset time length are determined.

[0015] The preset parameter is determined according to the average click amount of the creative script object, the average commodity conversion amount and the average commodity sales amount, and preset lower limit values T1, T2 and T3 of the three; wherein the preset parameter comprises a first characteristic value, a second characteristic value and a third characteristic value, and each characteristic value is greater than or equal to 0.

[0016] In some optional embodiments, the preset parameter is obtained by the following way:

[0017] The Lagrange function is obtained by transforming the maximization social benefit function and its constraint condition by using the Lagrange multiplier method, and the preset parameter is obtained when the Lagrange function is maximized; the constraint condition comprises that the average click amount of the creative script object is greater than the lower limit value T1, the average commodity conversion amount is greater than the lower limit value T2 and the average commodity sales amount is greater than the lower limit value T3.

[0018] The maximization social benefit function represents a function of maximizing the social benefit representing the long-term revenue of the media.

[0019] In some optional embodiments, the effect score of each resource position in the set of available resource positions when allocated to each of the set of delivery subjects is determined according to the preset parameter, the bid, the click probability, the conversion probability, and the price of the commodity, including:

[0020] multiplying the bid by the click probability to obtain a parameter component of social benefit;

[0021] multiplying the first feature value in the preset parameter by the click probability to obtain a parameter component of the click rate of the creative script object;

[0022] multiplying the second feature value in the preset parameter, the click probability, and the conversion probability to obtain a parameter component of the conversion rate of the creative script object;

[0023] multiplying the third feature value in the preset parameter, the click probability, the conversion probability, and the price of the commodity to obtain a parameter component of the sales amount of the commodity in the benefit of the delivery subject;

[0024] adding the parameter component of social benefit, the parameter component of the click rate of the creative script object, the parameter component of the conversion rate of the creative script object, and the parameter component of the sales amount of the commodity in the benefit of the delivery subject to obtain the effect score of one resource position in the set of available resource positions when allocated to one delivery subject in the set of delivery subjects.

[0025] In some optional embodiments, the resource position allocation scheme that maximizes the sum of the effect scores of all resource positions in the set of available resource positions is obtained according to the set limit condition based on the effect score of each resource position allocated to each delivery subject, including:

[0026] establishing the limit condition: any resource position in the set of available resource positions cannot be allocated to multiple delivery subjects; any delivery subject in the set of delivery subjects participating in the delivery of the creative script object cannot occupy multiple resource positions; any resource position in the set of available resource positions can be allocated to any delivery subject in the set of delivery subjects participating in the delivery of the creative script object;

[0027] using the maximum weighted matching algorithm to determine the resource position allocation scheme that maximizes the sum of the effect scores of all resource positions in the set of available resource positions under the condition of satisfying the limit condition.

[0028] In some optional embodiments, the bid is equal to the estimated value of the resource position by the delivery subject when clicking the creative script object through the search keyword information.

[0029] In a second aspect, an embodiment of the present application provides a resource position allocation device, including:

[0030] The acquisition module is configured to acquire a set of available resource positions and a set of delivery subjects of creative script objects according to the search keyword information.

[0031] The score determination module is configured to acquire an effect score of each resource position in the set of available resource positions when each resource position is allocated to each delivery subject in the set of delivery subjects based on the value parameter of the resource position, the influence parameter of the audience of the creative script object, and the characteristic parameter of the creative script object.

[0032] The allocation module is configured to obtain a resource position allocation scheme that maximizes the sum of effect scores of all resource positions in the set of available resource positions based on the effect score of each resource position allocated to each delivery subject and the set limit condition.

[0033] In a third aspect, an embodiment of the present application provides a computer readable storage medium having computer instructions stored thereon, the instructions being executed by a processor to implement the resource position allocation method described above.

[0034] In a fourth aspect, an embodiment of the present application provides a computer device, comprising a processor and a memory for storing processor executable instructions; wherein the processor is configured to execute the resource position allocation method described above.

[0035] The beneficial effects of the above technical solutions provided by the embodiments of the present application at least include:

[0036] The resource position allocation method provided by the embodiments of the present application acquires an effect score of each resource position in the set of available resource positions when each resource position is allocated to each delivery subject in the set of delivery subjects based on the value parameter of the resource position, the influence parameter of the audience of the creative script object, and the characteristic parameter of the creative script object. Since the value parameter of the resource position and the influence parameter of the audience of the creative script object can reflect the social benefit SW indicator, the influence parameter of the audience of the creative script object can reflect the click-through rate CTR indicator of the creative script object, and the influence parameter of the audience of the creative script object and the characteristic parameter of the creative script object can reflect the conversion rate CVR indicator and the gross merchandise volume GMV indicator of the creative script object, the resource position allocation scheme that maximizes the sum of effect scores of all resource positions in the set of available resource positions can simultaneously optimize the following indicators: the social benefit SW, the click-through rate CTR of the creative script object, the conversion rate CVR of the creative script object, and the gross merchandise volume GMV, that is, the long-term revenue of the media, the audience satisfaction, and the benefit of the delivery subject of the creative script object are simultaneously optimized, the balance between the indicators concerned by the three parties is ensured, the indicators concerned by the three parties are simultaneously optimized, the maximization of the benefits of the three parties is achieved, and the optimality of the resource position allocation is ensured.

[0037] Additional features and advantages of the present application will be set forth in the description that follows, and in part will be apparent from the description, or can be learned by practice of the application. The objectives and other advantages of the present application will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

[0038] The technical solutions of the present application are described in further detail below with reference to the accompanying drawings and examples. BRIEF DESCRIPTION OF DRAWINGS

[0039] The accompanying drawings are included to provide a further understanding of the present application and are incorporated in and constitute a part of this specification, illustrate embodiments of the present application and together with the description serve to explain the present application. In the drawings:

[0040] Figure 1 A flow chart of a resource bit auction implementation method in an embodiment of the present application;

[0041] Figure 2 A flow chart of another resource bit auction implementation method in an embodiment of the present application;

[0042] Figure 3 A schematic diagram of a resource bit auction implementation device in an embodiment of the present application;

[0043] Figure 4 A schematic diagram of another resource bit auction implementation device in an embodiment of the present application. DETAILED DESCRIPTION

[0044] Exemplary embodiments of the present disclosure will be described more fully hereinafter with reference to the accompanying drawings. While example embodiments of the present disclosure are shown in the drawings, it is understood that the present disclosure can be embodied in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the present disclosure to those skilled in the art.

[0045] Embodiments of the present application aim at the problems existing in the prior art described above, and provide a resource bit allocation method, the flow of which is shown in Figure 1 For any search a in the media, the following steps are included:

[0046] S101: Obtain a set of available resource bits I according to search keyword information τ τa and a set of creative script objects J τa ;

[0047] In step S101 above, the search keyword information τ for searching 'a' contains various information about the search, including keywords, user location, and related historical search information. When a search 'a' is submitted to the media, information on multiple resource slots i that can be used for auction is obtained based on the search keyword information τ, resulting in the available resource slot set I. τa ; Obtain information about the target entities j of multiple creative script objects participating in the resource slot auction, and obtain the set J of target entities participating in the creative script object auction. τa .

[0048] S102: Based on the value parameters of the resource slots, the audience impact parameters of the creative script object, and the characteristic parameters of the creative script object, obtain the available resource slot set I. τa Each resource bit i in the set is assigned to the delivery entity set J. τa The performance score for each delivery subject j in the data;

[0049] In step S102 above, the value parameters of the resource slot include: the bid b of the offering entity j for resource slot i. ja The audience impact parameters of creative script objects include: the click probability of the creative script object placed by advertiser j when resource slot i is allocated to advertiser j. The following is referred to as the click probability of the creative script object. The characteristic parameters of the creative script object include: the conversion probability of the product promoted by the creative script object. That is, the probability that the audience of the creative script object will purchase the product promoted by the creative script object after the creative script object is clicked, and the price of the product promoted by the creative script object. Specifically, the available resource bit set I can be obtained through the following method. τa Each resource bit i in the set is assigned to the delivery entity set J. τa Performance score for each subject j in the distribution:

[0050] According to preset parameters The bid b of the advertising entity j for resource slot i ja The probability of a creative script object being clicked. Creative script object to promote product conversion probability Product prices for products promoted using creative script objects. Determine the set of available resource bits I τa Each resource bit i in the set is assigned to the delivery entity set J. τa Performance score for each delivery subject j in the data.

[0051] Among them, preset parameters The first eigenvalue θ1 in* The second eigenvalue θ2 * and the third eigenvalue θ3 * All are greater than or equal to 0.

[0052] It should be noted that, in this embodiment of the invention, although the available resource bit set I τa Each resource slot i in the creative script object has a different value, but the target audience J is the same. τa Each bidder j is only allowed to submit one bid b for resource position i. ja Furthermore, a single advertising entity j can only be allocated one resource slot i, therefore, the click probability of the creative script object is... Conversion probability of products promoted by creative script objects That's also certain.

[0053] S103: Based on the effect score of each resource position i allocated to each delivery subject j, obtain a resource position allocation scheme that maximizes the sum of the effect scores of all resource positions in the available resource position set according to the set constraints.

[0054] In this embodiment of the invention, the effectiveness score of allocating each resource position i to each delivery entity j is determined through the above step S102. Subsequently, in step S103, a selected matching algorithm may be used to obtain the set of available resource bits I according to the set constraints. τa Performance score for all resource bits i The resource allocation scheme with the largest sum is used to obtain the resource allocation vector. The vector This includes the delivery subject j of the creative script object to which each resource position i is assigned.

[0055] The resource bit allocation method provided by the embodiment of the present application is based on the value parameter of the resource bit, the influence parameter of the audience of the creative script object and the characteristic parameter of the creative script object, and the effect score of each resource bit in the available resource bit set when allocated to each delivery principal in the delivery principal set is obtained. Since the value parameter of the resource bit and the influence parameter of the audience of the creative script object can reflect the social benefit SW index, the influence parameter of the audience of the creative script object can reflect the click rate CTR index of the creative script object, and the influence parameter of the audience of the creative script object and the characteristic parameter of the creative script object can reflect the conversion rate CVR index and the gross merchandise volume GMV index of the creative script object, therefore, the resource bit allocation scheme obtained by maximizing the sum of the effect scores of all resource bits in the available resource bit set can simultaneously optimize the following indexes: the social benefit SW, the click rate CTR of the creative script object, the conversion rate CVR of the creative script object and the gross merchandise volume GMV, that is, the long-term revenue of the media, the audience satisfaction and the benefit of the delivery principal of the creative script object are simultaneously optimized, the balance between the indexes concerned by the three parties is ensured, the indexes concerned by the three parties are simultaneously optimized, the maximization of the benefits of the three parties is realized, and the optimality of the resource bit allocation is ensured.

[0056] In an optional embodiment, since the media cannot completely know the estimated value of the delivery principal j to the resource bit i when performing resource bit allocation, but can only obtain the bid b ja of the delivery principal j to the resource bit i, in order to realize the best resource bit allocation optimization effect, the bid b ja of the delivery principal j to the resource bit i is set to be equal to the estimated value of the delivery principal j to the resource bit i when clicking the creative script object by searching the keyword information τ. Wherein, the estimated value of the delivery principal j to the resource bit i, that is, the real value of the resource bit i considered by the delivery principal j when bidding for the resource bit i.

[0057] In an optional embodiment, as shown in Figure 2 , the resource bit allocation method provided by the embodiment of the present application further comprises:

[0058] S104: When the clicking behavior is generated, the Myerson deduction fee scheme is used to deduct the delivery principal j of the creative script object.

[0059] The resource bit allocation method provided by the embodiment of the present application uses the Myerson deduction fee scheme to encourage the delivery principal of the creative script object to report the real value when bidding for the resource bit, thereby ensuring the stability of the effect of the creative script object bidding mechanism and realizing the dominant strategy incentive compatible characteristic (DSIC) of the creative script object bidding mechanism.

[0060] In an optional embodiment, the above is based on preset parameters. The bid b of the advertising entity j for resource slot i ja The probability of a creative script object being clicked. Creative script object to promote product conversion probability Product prices for products promoted using creative script objects. Determine the set of available resource bits I τa Each resource bit i in the set is assigned to the delivery entity set J. τa Performance score for each delivery subject j in the data. Specifically, it can be calculated in the following way:

[0061] The bid b of the campaign subject j targeting ad slot i using the creative script object. ja Multiply by the click probability of the creative script object Obtain the parameter components of the social benefit SW;

[0062] Use preset parameters The first eigenvalue θ1 in * Multiply by the click probability of the creative script object Obtain the CTR parameter components of the creative script object;

[0063] Use preset parameters The second eigenvalue θ2 * The probability of a creative script object being clicked. And the conversion rate of products promoted by creative script objects. Multiply them to obtain the parameter components of the Creative Script Object's Conversion Rate (CVR);

[0064] Use preset parameters The third eigenvalue θ3 * The probability of a creative script object being clicked. Creative script object to promote product conversion probability And the product price of the product promoted by the creative script object. Multiplying these components yields the parameter component of the sales revenue of the goods in the benefits of the investment entity j.

[0065] Add the parameter components of social benefit (SW), the parameter components of click-through rate (CTR) of the creative script object, the parameter components of conversion rate (CVR) of the creative script object, and the parameter component of product sales revenue in the benefit of the advertising subject (j) to obtain the available resource slot set I. τa One resource slot i is allocated to the set of delivery entities J τa Performance score when one of the delivery entities j is selected.

[0066] In an optional embodiment, the matching algorithm selected in step S103 may be the maximum weighted matching algorithm, and the set limiting conditions may be:

[0067] Available resource bit set I τa No resource bit i in the list can be assigned to multiple delivery entities j;

[0068] The set of entities participating in the delivery of creative script objects J τa No single entity j in the delivery process can occupy multiple resource slots;

[0069] Available resource bit set I τa Any resource bit i in the creative script object can be assigned to the set of delivery entities J. τa Any delivery entity j in the process;

[0070] Using the maximum weighted matching algorithm, the set of available resource bits I is determined under the above constraints. τa The resource allocation vector is obtained by finding the resource allocation scheme that maximizes the sum of the effects of all creative script objects in all resource slots. That is, to obtain A valid vector

[0071] The following detailed analysis yields the result that... A valid vector This yields the specific implementation process of the resource allocation scheme. In this embodiment of the invention, it is made that... A valid vector In essence, it aims to maximize the social benefit (SW) of the creative script object's click-through rate (CTR), conversion rate (CVR), and gross merchandise volume (GMV) while satisfying the constraints.

[0072] To achieve the goals of media outlets participating in the bidding mechanism for ad placement allocation, the audience of the creative scripts, and the advertisers of the creative scripts—namely, the long-term revenue of the media, the audience satisfaction of the creative scripts, and the profitability of the advertisers—the ad placement allocation process requires simultaneous optimization of four target parameters: Social Revenue (SW), Click-Through Rate (CTR) of the creative scripts, Conversion Rate (CVR) of the creative scripts, and Gross Merchandise Volume (GMV). The CTR of the creative scripts can be calculated based on the click-through rate (CTN), and the CVR can be calculated based on the conversion rate (CVN). Assume that N searches are submitted on the media outlet within a certain period, with search keywords τ1, ..., τ2. a , ..., τ N Then, for any search 'a' in the media, given the search keyword information τ aResource bit allocation vector and satisfy the following equations (1)-(4):

[0073]

[0074]

[0075]

[0076]

[0077] Since the target parameters are usually calculated by averaging the performance of all resource bit allocation instances in a period of time (such as a day), rather than being limited to certain types of instances, the media should consider the overall satisfaction of the three parties in the resource bit allocation process from a global perspective. The resource bit allocation mechanism designer inevitably has to weigh a large number of different resource bit allocation instances, and due to the global consideration of the target, each resource bit allocation instance cannot be treated as an isolated instance. Therefore, in order to protect the interests of the target audience of the creative script object, the subject of the placement of the creative script object and the media, the present inventors have set the lower limit values T1, T2 and T3 of the average click volume CTN of the creative script object, the average conversion volume CVN of the creative script object and the average sales volume GMV of the goods in a predetermined period of time in the media, and converted the optimization problem of multiple target parameters into the following conditions: the average click volume CTN of the creative script object, the average conversion volume CVN of the creative script object and the average sales volume GMV of the goods are greater than or equal to T1, T2 and T3 respectively; the matching algorithm used is the maximum weighted matching algorithm; and according to the established conditions: any resource bit i in the available resource bit set I τa cannot be allocated to multiple placement subjects j; any placement subject j in the set of placement subjects J τa cannot occupy multiple resource bits; any resource bit i in the available resource bit set I τa can be allocated to any placement subject j in the set of placement subjects J τa , the social benefit SW function G(T) is maximized, where any search a≤N in the media takes the resource bit allocation vector as a variable, and specifically refers to the following function G(T), where T=(T1, T2, T3):

[0078]

[0079]

[0080]

[0081]

[0082]

[0083]

[0084]

[0085] It should be noted that, in the embodiment of the application, in order to facilitate calculation, the constraint condition (6) of the function G(T) is replaced by X ija ≥0. ija

[0086] Since the number of resource bit allocation occurring in any time period on the media is huge, which can reach tens of millions or hundreds of millions, the calculation amount for obtaining the resource bit allocation result for different search keyword information τ is very huge, in order to obtain the optimal solution of the function G(T), the constraint conditions (1)-(3) are added to the function G(T) by using the Lagrange multiplier method, and the Lagrange function is obtained under the premise of meeting the above limited conditions

[0087]

[0088]

[0089]

[0090] Wherein, is a vector value meeting the constraint conditions (4)-(6); θ1, θ2 and θ3 are three eigenvalues of the preset parameter when the Lagrange multiplier method is used, that is Since the constraint conditions (1)-(3) are inequalities, the values of the three eigenvalues θ1, θ2 and θ3 of the parameter will have multiple values. In order to obtain the optimal solution of the function G(T), the vector of the three eigenvalues of the parameter needs to be obtained when the optimal value is takenThe inventor of the application, by applying the strong duality to the function G(T) under the linear constraint condition, obtains the following formula (5):

[0091]

[0092] Wherein, S represents a set of vector values meeting the constraint conditions (4)-(6) From the formula (5), it can be seen that when the function G(T) is maximized, the minimum value of the value of the parameter is the value of the parameter optimal solution Right now:

[0093]

[0094] In this embodiment of the invention, substituting equations (1)-(4) into equation (5), it can be seen that the social benefit SW parameter component can be represented by the bid b of the creative script object's placement subject j for resource position i. ja Click probability of creative script objects The product representation; the click-through rate (CTR) parameter components of the creative script object can be represented by preset parameters. θ1 and the click probability of the creative script object The product representation; the conversion rate (CVR) parameter components of the creative script object can be represented by preset parameters. θ2, the click probability of the creative script object. And the conversion rate of products promoted by creative script objects. The product representation of merchandise sales GMV can be expressed using preset parameters. θ3, the click probability of the creative script object. Creative script objects promote the conversion rate of products. And the product price of the product promoted by the creative script object. The product representation.

[0095] By adding the parameter components representing the social benefit (SW), the parameter components representing the click-through rate (CTR) of the creative script object, the parameter components representing the conversion rate (CVR) of the creative script object, and the parameter component representing the sales revenue of the merchant in the advertising entity of the creative script object, we can determine the effectiveness score of the advertising entity j occupying the resource position i. For details, please refer to the following formula (7):

[0096]

[0097] To facilitate online resource allocation for media, the inventors of this invention further experimented by substituting equations (6) and (7) into... In the process, the following function is obtained.

[0098]

[0099]

[0100]

[0101]

[0102] By function It can be seen that for any search 'a' in the media, where a≤N, the following function can be obtained.

[0103]

[0104]

[0105]

[0106]

[0107] Using the maximum weighted matching algorithm, under the premise of satisfying the above constraints (7)-(9), the function is calculated. The optimal solution yields the resource allocation vector. That is, when submitting search keyword information τ to the media for the a-th search, the resource allocation scheme is obtained, and at this time, the available resource set I is... τa Performance scores for all resource slots The sum is maximized. Among them, constraint (7) means preventing a resource bit i from being allocated to multiple creative script objects' delivery entities; constraint (8) means preventing a creative script object's delivery entity j from occupying multiple resource bits; constraint (9) means that the set of available resource bits I can be used. τa Any resource bit i in the creative script object is assigned to the set of delivery subjects J. τa Any delivery entity j in the process.

[0108] In this embodiment of the invention, according to preset parameters The set of entities for delivering creative script objects J τa The bidding entity j in the middle bids b for the resource slots ja The probability of a creative script object being clicked. Creative script object promotion product conversion rate Product prices for products promoted using creative script objects. Maximize the function under constraints (7)-(9) Obtain the resource bit allocation vector It achieves the effect of maximizing function G(T) under constraints (1)-(6).

[0109] Because media resource allocation is done online, when a search keyword information τ is submitted to the media in real time, the media cannot receive the parameters in real time. The optimal solution, i.e., the preset parameters Therefore, when allocating resource slots in media, a function can be set periodically according to a set time length (e.g., 24 hours). Parameters in For example, at 12 midnight every day, the parameters are adjusted. optimal solution Make adjustments in the function The adjusted version is applied in China. Set new parameters until midnight the following night. optimal solution

[0110] In one embodiment, the parameters can be determined empirically. The optimal values ​​of parameters θ1, θ2, and θ3 are: θ1 * θ2 * and θ3 * To obtain the optimal preset parameters

[0111] In one embodiment, it can also involve acquiring historical creative script object data within a preset media time period (e.g., 24 hours), determining the average click-through rate (CTN), average conversion rate (CVN), and average sales revenue (GMV) of the creative script objects within the preset time period, and determining the lower limits T1, T2, and T3 for the average click-through rate (CTN), average conversion rate (CVN), and average sales revenue (GMV) of the creative script objects, respectively, and determining parameters. The optimal value θ1 * θ2 * and θ3 * That is, to determine the preset parameters

[0112] In this embodiment of the invention, it can be seen from equation (6) that the parameters optimal solution That is, to maximize the Lagrange function At that time, it made The value is minimized. For simplicity, we assume that the frequency of each search type is the same in each resource allocation cycle, and that the bid of the creative script object's target audience is fixed. This means that the parameter... optimal solution The variation throughout the entire cycle is very small. For ease of representation, given the parameters... The global objectives appearing in the constraints are represented by the following formula (8): average click-through rate (CTN) of the creative script object, average product conversion rate (CVN) of the products promoted by the creative script object, and average product sales (GMV).

[0113]

[0114] We assume that constraints (1)-(3) are bindable, then in the constraints... Next, there is an optimal parameter vector Make This optimal parameter That is, the preset parameter

[0115] In one embodiment, the Trust Region Method can be used to solve the minimum value of the following formula The parameter The optimal eigenvalue θ1 * , θ2 * and θ3 * , determine the optimal solution of the parameter Make

[0116] Specifically, the Trust Region Method can be used to iteratively approximate the target For example, an initial value of θ1, θ2 and θ3 is preset respectively, and the minimum Solve

[0117] Based on the same inventive concept, the embodiments of the present application also provide a resource bit allocation device, a related storage medium and equipment. Since the principles of the problems solved by these devices, related storage media and equipment are similar to the aforementioned resource bit allocation method, the implementation of the device, related storage medium and equipment can be referred to the implementation of the aforementioned method, and the repeated parts will not be described again.

[0118] Referring to Figure 3 The embodiments of the present application provide a resource bit allocation device, which comprises:

[0119] The acquisition module 31 is configured to acquire a set of available resource bits and a set of delivery subject objects of creative script objects according to search keyword information.

[0120] The score determination module 32 is configured to acquire an effect score of each resource bit in the set of available resource bits when each resource bit is allocated to each delivery subject in the set of delivery subject objects based on a value parameter of the resource bit, an influence parameter of an audience of the creative script object and a feature parameter of the creative script object.

[0121] The allocation module 33 is configured to obtain a resource bit allocation scheme that maximizes the sum of effect scores of all resource bits in the set of available resource bits based on the effect scores of each resource bit allocated to each delivery subject according to a set of limiting conditions.

[0122] In one embodiment, referring to Figure 4 ​As shown, the resource bit allocation device further comprises a charge deduction module 34 configured to deduct the bid of the creative script object from the bid of the resource bit when a click action occurs.

[0123] In one embodiment, the value parameter of the resource bit comprises a bid of the bid subject for the resource bit.

[0124] The influence parameter of the audience of the creative script object comprises a click probability of the creative script object of the bid subject when the resource bit is allocated to the bid subject.

[0125] The characteristic parameter of the creative script object comprises a conversion probability of a promoted product of the creative script object and a product price of the promoted product.

[0126] The score determination module 32 is specifically configured to determine an effect score when each resource bit in the available resource bit set is allocated to each bid subject in the bid subject set according to the preset parameter, the bid, the click probability, the conversion probability and the product price.

[0127] In one embodiment, the bid of the bid subject for the resource bit is equal to an estimated value of the resource bit of the bid subject when the creative script object is clicked through the search keyword information.

[0128] In one embodiment, the bid of the bid subject for the resource bit is equal to an estimated value of the resource bit of the bid subject when the creative script object is clicked through the search keyword information. Figure 4 As shown, the resource bit allocation device further comprises a parameter determination module 30 configured to obtain historical creative script object data in a preset time length of the media, determine an average click amount of the creative script object, an average product conversion amount of a promoted product of the creative script object and a product average sales amount in the preset time length.

[0129] The preset parameter is determined according to the average click amount of the creative script object, the average product conversion amount and the product average sales amount, and preset lower limit values T1, T2 and T3 of the three, wherein the preset parameter comprises a first characteristic value, a second characteristic value and a third characteristic value, and each characteristic value is greater than or equal to 0.

[0130] In one embodiment, the parameter determination module 30 is specifically configured to transform a Lagrange function by using a Lagrange multiplier method to maximize a social benefit function and a constraint condition, and obtain the preset parameter when the Lagrange function is maximized; the constraint condition comprises that the average click amount of the creative script object is greater than the lower limit value T1, the average product conversion amount is greater than the lower limit value T2 and the product average sales amount is greater than the lower limit value T3.

[0131] The maximum social benefit function represents a function of maximizing social benefits representing long-term benefits of the media.

[0132] In one embodiment, the score determination module 32 is specifically configured to multiply the bid by the click probability to obtain a parameter component of the social benefit.

[0133] The first characteristic value in the preset parameter is multiplied by the click probability to obtain a parameter component of the click rate of the creative script object.

[0134] The second characteristic value in the preset parameter, the click probability, and the conversion probability are multiplied to obtain a parameter component of the conversion rate of the creative script object.

[0135] The third characteristic value in the preset parameter, the click probability, the conversion probability, and the product price are multiplied to obtain a parameter component of the product sales in the benefit of the delivery subject.

[0136] The parameter component of the social benefit, the parameter component of the click rate of the creative script object, the parameter component of the conversion rate of the creative script object, and the parameter component of the product sales in the benefit of the delivery subject are added to obtain an effect score when one resource site in the available resource site set is allocated to one delivery subject in the delivery subject set.

[0137] In one embodiment, the allocation module 33 is specifically configured to establish a limited condition that any resource site in the available resource site set cannot be allocated to multiple delivery subjects, any delivery subject in the delivery subject set participating in the creative script object cannot occupy multiple resource sites, and any resource site in the available resource site set can be allocated to any delivery subject in the delivery subject set participating in the creative script object.

[0138] A maximum weighted matching algorithm is used to determine a resource site allocation scheme that maximizes the sum of the effect scores of all resource sites in the available resource site set under the limited condition.

[0139] The resource site allocation apparatus provided by the embodiment of the present application can be arranged on a server side, and can obtain submitted search keyword information τ from a database of the media in real time to realize allocation of resource sites. When an audience of a creative script object clicks the creative script object, a Myerson deduction scheme is used to deduct a delivery subject of the creative script object.

[0140] The embodiment of the present application further provides a computer readable storage medium having computer instructions stored thereon, and the instructions are executed by a processor to implement the above-mentioned resource site allocation method.

[0141] The embodiment of the present application further provides a computer device, comprising: a processor, a memory for storing processor-executable commands; wherein the processor is configured to execute the resource bit allocation method described above.

[0142] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt a form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage and optical storage, etc.) containing computer usable program codes.

[0143] The present application is described with reference to flowcharts and / or block diagrams of the method, device (system), and computer program product according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be realized by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a means for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 an apparatus for performing the functions specified in one or more flows and / or blocks.

[0144] These computer program instructions can also be stored in a computer readable memory capable of guiding a computer or other programmable data processing devices to work in a specific way, so that the instructions stored in the computer readable memory produce a product including instruction apparatus, which implements the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 an apparatus for performing the functions specified in one or more flows and / or blocks.

[0145] These computer program instructions can also be loaded into a computer or other programmable data processing device, so that a series of operation steps are performed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide a process for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 an apparatus for performing the functions specified in one or more flows and / or blocks.

[0146] Obviously, many modifications and variations of the present application are possible in light of the above teachings. It is, therefore, to be understood that within the scope of the appended claims and their equivalents, the application can be practiced otherwise than as specifically described.

Claims

1. A method for allocating resource bits, comprising: Based on the search keyword information, obtain the set of available resource slots and the set of target audiences for creative script objects; Based on the value parameters of the resource slots, the audience impact parameters of the creative script objects, and the feature parameters of the creative script objects, obtain the effect score when each resource slot in the set of available resource slots is allocated to each advertising entity in the set of advertising entities; Based on the performance score of each resource slot allocated to each advertising entity, a resource slot allocation scheme that maximizes the sum of the performance scores of all resource slots in the available resource slot set is obtained according to the set constraints. The value parameters of the resource slot include: the bid made by the offering entity for the resource slot; The audience impact parameters of the creative script object include: the click probability of the creative script object placed by the advertising entity when the resource slot is allocated to the advertising entity; The characteristic parameters of the creative script object include: the conversion probability of the product promoted by the creative script object and the product price of the promoted product; The method of obtaining the performance score when each resource slot in the available resource slot set is allocated to each advertiser in the advertiser set, based on the value parameters of the resource slot, the audience impact parameters of the creative script object, and the feature parameters of the creative script object, includes: Based on preset parameters, the bid, the click probability, the conversion probability, and the product price, determine the parameter components of social benefit, the parameter components of the click rate of the creative script object, the parameter components of the conversion rate of the creative script object, and the parameter components of product sales revenue in the benefit of the advertising entity. The social benefit parameter component, the click-through rate parameter component of the creative script object, the conversion rate parameter component of the creative script object, and the product sales revenue parameter component of the benefit of the advertising entity are added together to obtain the effect score when a resource slot in the available resource slot set is allocated to an advertising entity in the advertising entity set.

2. The method according to claim 1, further comprising: Acquire historical creative script object data within a preset time period of the media, and determine the average click-through rate, average product conversion rate, and average product sales of the creative script objects within the preset time period; The preset parameters are determined based on the average click-through rate of the creative script object, the average product conversion rate, and the average product sales, as well as the preset lower limits T1, T2, and T3 of the three; wherein the preset parameters include a first feature value, a second feature value, and a third feature value, and each feature value is greater than or equal to 0.

3. The method according to claim 2, wherein the preset parameter is specifically obtained in the following manner: The Lagrange multiplier method is used to transform the function maximizing social benefits and its constraints to obtain the Lagrange function. When maximizing the Lagrange function, the preset parameters are obtained. The constraints include: The average click-through rate of the creative script object is greater than the lower limit T1, the average product conversion rate is greater than the lower limit T2, and the average product sales revenue is greater than the lower limit T3; The social benefit maximization function represents a function that maximizes the social benefits that characterize the long-term benefits of the media.

4. The method according to claim 3, wherein determining the parameter components of social benefit, the parameter component of click-through rate of creative script object, the parameter component of conversion rate of creative script object, and the parameter component of sales revenue in the benefit of the advertising entity based on preset parameters, the bid, the click probability, the conversion probability, and the product price, comprises: Multiply the bid by the click probability to obtain the parameter components of the social benefit; Multiply the first feature value in the preset parameters by the click probability to obtain the parameter components of the click rate of the creative script object; The conversion rate parameter components of the creative script object are obtained by multiplying the second feature value of the preset parameters, the click probability, and the conversion probability. The third feature value in the preset parameters, the click probability, the conversion probability, and the product price are multiplied together to obtain the parameter component of the product sales revenue in the benefits of the advertising entity.

5. The method according to claim 1, wherein the step of obtaining a resource allocation scheme that maximizes the sum of the effect scores of all resource positions in the available resource position set based on the effect score of each resource position allocated to each delivery entity, according to set limiting conditions, comprises: Establish the following constraints: No single resource slot in the available resource slot set can be allocated to multiple advertising entities; No single advertising entity in the set of advertising entities participating in the creative script object can occupy multiple resource slots; Any single resource slot in the available resource slot set can be allocated to any single advertising entity in the set of advertising entities participating in the creative script object. Using the maximum weighted matching algorithm, under the aforementioned constraints, a resource allocation scheme that maximizes the sum of the effect scores of all resource bits in the available resource bit set is determined.

6. The method according to claim 1, wherein the bid is equal to the estimated value of the ad placement by the advertiser when the creative script object is clicked through the search keyword information.

7. A resource bit allocation device, comprising: The acquisition module is used to obtain the set of available resource slots and the set of advertising entities for creative script objects based on search keyword information; The scoring determination module is used to obtain the effect score when each resource in the set of available resource positions is allocated to each advertising entity in the set of advertising entities, based on the value parameter of the resource position, the audience impact parameter of the creative script object, and the feature parameter of the creative script object. The value parameters of the resource slot include: the bid made by the offering entity for the resource slot; The audience impact parameters of the creative script object include: the click probability of the creative script object placed by the advertising entity when the resource slot is allocated to the advertising entity; The characteristic parameters of the creative script object include: the conversion probability of the product promoted by the creative script object and the product price of the promoted product; The method of obtaining the performance score when each resource slot in the available resource slot set is allocated to each advertiser in the advertiser set, based on the value parameters of the resource slot, the audience impact parameters of the creative script object, and the feature parameters of the creative script object, includes: Based on preset parameters, the bid, the click probability, the conversion probability, and the product price, determine the parameter components of social benefit, the parameter components of the click rate of the creative script object, the parameter components of the conversion rate of the creative script object, and the parameter components of product sales revenue in the benefit of the advertising entity. The social benefit parameter component, the click-through rate parameter component of the creative script object, the conversion rate parameter component of the creative script object, and the product sales revenue parameter component of the benefit of the advertising entity are added together to obtain the effect score when a resource slot in the available resource slot set is allocated to an advertising entity in the advertising entity set. The allocation module is used to allocate each resource slot to each advertising entity based on the performance score, and obtain a resource slot allocation scheme that maximizes the sum of the performance scores of all resource slots in the available resource slot set according to the set constraints.

8. A computer-readable storage medium having stored thereon computer instructions that, when executed by a processor, implement the resource bit allocation method as described in any one of claims 1-6.

9. A computer device, comprising: A processor; a memory for storing processor-executable commands; wherein the processor is configured to execute the resource bit allocation method as described in any one of claims 1-6.

Images