Information processing method, electronic device and computer program product

Abstract

The invention discloses an information processing method, electronic equipment and a computer program product, and relates to the technical field of advertisement, and the method comprises the steps: constructing a first function between a price adjustment parameter and a consumption parameter change rate based on multiple groups of sample data; obtaining conversion parameters of the user under the action of the price adjustment parameters as historical conversion parameters of the user after price adjustment; constructing a second function between the consumption parameter change rate and future conversion parameters of the user after price adjustment under the action of the price adjustment parameters; the future conversion parameters comprise predicted conversion parameters, and the second function represents the relationship between the change rate of the cost corresponding to the successfully delivered advertisements under the action of the price adjustment parameters and the predicted number of the successfully delivered advertisements; and determining a target price adjustment parameter based on the first function and the second function, the target price adjustment parameter being used for maximizing the sum of the historical conversion parameter of the user after price adjustment and the future conversion parameter of the user after price adjustment under a target constraint condition. The accuracy of the obtained target price adjustment parameter can be improved.

Description

Technical Field

[0001] The present application relates to the field of advertising technology, and more specifically, to an information processing method, an electronic device, and a computer program product. Background Art

[0002] At present, with the development of advertising technology, it is possible to determine the price adjustment parameters, and then adjust the bids for placing advertisements by using the price adjustment parameters, thereby achieving higher returns. However, the accuracy of the price adjustment parameters currently determined is low, resulting in low returns. Summary of the invention

[0003] The present application proposes an information processing method, an electronic device and a computer program product.

[0004] In a first aspect, an embodiment of the present application provides an information processing method, comprising: constructing a first function between a price adjustment parameter and a consumption parameter change rate based on multiple groups of sample data, wherein each group of the sample data comprises a historical price adjustment parameter and a historical consumption parameter change rate corresponding to the historical price adjustment parameter, the consumption parameter change rate being used to characterize the ratio of the consumption parameter of an advertisement not under the effect of the price adjustment parameter to the consumption parameter of the advertisement under the effect of the price adjustment parameter, the consumption parameter being used to characterize the cost corresponding to a successfully delivered advertisement, and the first function being used to characterize the degree of influence of the price adjustment parameter on the change rate of the cost corresponding to a successfully delivered advertisement; obtaining a conversion parameter of a user under the effect of the price adjustment parameter as the user's conversion parameter Historical conversion parameters after price adjustment, wherein the conversion parameters are used to characterize the number of successfully delivered advertisements; constructing a second function between the consumption parameter change rate and the user's future conversion parameters after price adjustment under the action of the price adjustment parameters, wherein the future conversion parameters include predicted conversion parameters, and the second function is used to characterize the relationship between the rate of change of costs corresponding to successfully delivered advertisements under the action of the price adjustment parameters and the predicted number of successfully delivered advertisements after the current moment; determining a target price adjustment parameter based on the first function and the second function, wherein the target price adjustment parameter is used to maximize the sum of the user's historical conversion parameters after price adjustment and the user's future conversion parameters after price adjustment under the target constraint condition.

[0005] In a second aspect, an embodiment of the present application further provides an electronic device, comprising: one or more processors; a memory; one or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, and the one or more programs are configured to execute the method described in the first aspect.

[0006] In a third aspect, an embodiment of the present application further provides a computer program product, including a computer program / instruction, characterized in that when the computer program / instruction is executed by a processor, the method described in the first aspect is implemented.

[0007] The information processing method, electronic device and computer program product provided by the embodiments of the present application first construct a first function between a price adjustment parameter and a consumption parameter change rate based on multiple groups of sample data; then obtain the user's conversion parameter under the action of the price adjustment parameter as the user's post-price adjustment historical conversion parameter, wherein the conversion parameter is used to characterize the number of successfully delivered advertisements; then construct a second function between the consumption parameter change rate and the user's post-price adjustment future conversion parameter under the action of the price adjustment parameter, wherein the future conversion parameter includes a predicted conversion parameter, and the second function is used to characterize the relationship between the rate of change of the cost corresponding to the successfully delivered advertisements under the action of the price adjustment parameter and the predicted number of successfully delivered advertisements; thereby determining a target price adjustment parameter based on the first function and the second function, wherein the target price adjustment parameter is used to maximize the sum of the user's post-price adjustment historical conversion parameter and the user's post-price adjustment future conversion parameter under the target constraint condition. Among them, each group of sample data includes historical price adjustment parameters and the historical consumption parameter change rate corresponding to the historical price adjustment parameters, the consumption parameter change rate is used to characterize the ratio of the consumption parameters of the advertisement not under the effect of the price adjustment parameters to the consumption parameters of the advertisement under the effect of the price adjustment parameters, the consumption parameters are used to characterize the costs corresponding to the successfully delivered advertisements, and the first function is used to characterize the degree of influence of the price adjustment parameters on the rate of change of the costs corresponding to the successfully delivered advertisements. The target price adjustment parameters in this application can maximize the sum of the user's post-price adjustment historical conversion parameters and the user's post-price adjustment future conversion parameters under the target constraint conditions. It can be seen that the price adjustment parameters are not determined solely by the historical conversion parameters, but the target price adjustment parameters are determined by combining the post-price adjustment historical conversion parameters and the post-price adjustment future conversion parameters. And in this application, the second function includes the post-price adjustment future conversion parameters, so that the subsequent combination of the first function and the second function to determine the target price adjustment parameters can bring better benefits and improve the accuracy of the target price adjustment parameters.

[0008] Other features and advantages of the embodiments of the present application will be described in the subsequent description, and partly become apparent from the description, or can be understood by practicing the embodiments of the present application. The purposes and other advantages of the embodiments of the present application can be realized and obtained by the structures specifically pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0010] Figure 1 An application scenario diagram of the information processing method provided by an embodiment of the present application is shown;

[0011] Figure 2 A method flow chart of an information processing method provided by an embodiment of the present application is shown;

[0012] Figure 3 A method flow chart of an information processing method provided by another embodiment of the present application is shown;

[0013] Figure 4 A schematic diagram showing advertisement sorting provided by an embodiment of the present application is shown;

[0014] Figure 5 A method flow chart of an information processing method provided by yet another embodiment of the present application is shown;

[0015] Figure 6 A schematic diagram showing a historical time period provided by an embodiment of the present application is shown;

[0016] Figure 7 A method flow chart of an information processing method provided by another embodiment of the present application is shown;

[0017] Figure 8 A schematic diagram of a first function provided in an embodiment of the present application is shown;

[0018] Fig. 9 A system block diagram of an information processing system provided by an embodiment of the present application is shown;

[0019] Fig.10 A structural block diagram of an electronic device provided in an embodiment of the present application is shown;

[0020] Fig.11 A structural block diagram of a computer program product provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0021] In order to make those skilled in the art better understand the present application scheme, the technical scheme in the present application embodiment will be clearly and completely described below in conjunction with the drawings in the present application embodiment. Obviously, the described embodiment is only a part of the present application embodiment, rather than all the embodiments. The components of the present application embodiment usually described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the present application for protection, but merely represents the selected embodiment of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0022] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0023] At present, with the development of advertising technology, it is possible to determine the price adjustment parameters, and then adjust the bids for placing advertisements through the price adjustment parameters, thereby achieving higher profits. However, the accuracy of the price adjustment parameters currently determined is low, resulting in low profits. How to improve the accuracy of determining the price adjustment parameters is an urgent problem to be solved.

[0024] Currently, you can obtain the current cost of an ad, compare it with the target cost, and then determine the price adjustment parameters. Then, you can use the price adjustment parameters to adjust the bid for the ad, so that the current cost after the price adjustment is close to the target cost. For example, if the current cost of an ad is higher than the target cost, the price adjustment parameters will be used to lower the bid for the ad, so that the cost after the price adjustment is close to the target cost; if the current cost of an ad is lower than the target cost, the price adjustment parameters will be used to increase the bid for the ad, so that the cost after the price adjustment is close to the target cost.

[0025] For example, the price adjustment parameters can be adjusted by using the proportional-integral-derivative (PID) algorithm so that the current cost of the advertisement is close to the target cost. The PID control algorithm is a control algorithm that combines proportional, integral and derivative.

[0026] However, the inventors found in their research that the existing method of determining price adjustment parameters, when the current cost of an advertisement is seriously lower than the target cost, the adjustment strategy may still have a low cost after adjustment due to the upper limit of the adjustment strategy; when the current cost of an advertisement is seriously higher than the target cost, the price adjustment parameter will be adjusted to a smaller value, but this may cause the advertisement to bid too low to win, making the future advertising consumption close to 0, and thus making it impossible to achieve a price adjustment that approaches the target cost. It can be seen that the accuracy of the existing determined price adjustment parameters is low. Among them, the advertising consumption can represent the cost incurred by successfully delivered advertisements.

[0027] See also Figure 1 , Figure 1 An application scenario diagram of the information processing method provided in an embodiment of the present application is shown, namely, an information processing scenario 100, which may include a terminal device 110, an advertising server platform 120, and an advertising provider 130, wherein the advertising server platform 120 is connected to the terminal device 110 and the advertising provider 130, respectively.

[0028] The terminal device 110 and the advertisement provider 130 can access the Internet and thereby establish a connection with the advertisement server platform 120 which is also connected to the Internet. The terminal device 110 and the advertisement provider 130 can access the Internet wirelessly, such as through wireless communication technologies such as Wi-Fi and Bluetooth. The terminal device 110 and the advertisement provider 130 can also access the Internet in a wired manner, such as through an Rj45 network cable or optical fiber. The terminal device 110 can be an electronic device, and the advertisement provider 130 can also achieve communication connection with the advertisement server platform 120 through the electronic device.

[0029] The advertisement delivery company 130 can deliver the advertisements to be delivered to the terminal device 110 through the advertisement server platform 120. The advertisement delivery company 130 can pay the delivery fee to the advertisement server platform 120 so that the advertisement server platform 120 can deliver the advertisements to each terminal device 110. In addition, after the user successfully performs a specified operation on the delivered advertisement in the terminal device 110, it can be regarded as a successful delivery of the advertisement, and the advertisement delivery company 130 needs to pay the advertisement server platform 120 for each successful delivery of the advertisement. The sum of the delivery fee and the successful delivery fee paid by the advertisement delivery company 130 to the advertisement server platform 120 can be regarded as the cost corresponding to the advertisement.

[0030] It is understandable that the higher the delivery fee paid by the advertisement delivery company 130 to the advertisement server platform 120, the easier it is for the advertisement server platform 120 to deliver the advertisement to each terminal device 110, thereby resulting in a greater probability that the user will successfully perform a specified operation on the delivered advertisement in the terminal device 110; and the lower the delivery fee paid by the advertisement delivery company 130 to the advertisement server platform 120, the less likely it is for the advertisement server platform 120 to deliver the advertisement to each terminal device 110, thereby resulting in a lower probability that the user will successfully perform a specified operation on the delivered advertisement in the terminal device 110. Therefore, the advertisement delivery company 130 can determine the target price adjustment parameter through the information processing method provided in this application, thereby adjusting the delivery fee through the target price adjustment parameter, thereby achieving the adjustment of the cost of the advertisement, and the detailed description can be referred to the subsequent embodiments.

[0031] See also Figure 2 , Figure 2 A method flow chart of an information processing method provided by an embodiment of the present application is shown. The information processing method can be applied to Figure 1 The advertisement provider in the information recommendation scenario shown may specifically use the processor of the advertisement provider as the execution subject of the information processing method. The information processing method may include steps S110 to S140.

[0032] Step S110: Based on multiple groups of sample data, construct a first function between price adjustment parameters and consumption parameter change rates, wherein the sample data includes historical price adjustment parameters and historical consumption parameter change rates corresponding to the historical price adjustment parameters, the consumption parameter change rates are used to characterize the ratio of consumption parameters of advertisements not affected by price adjustment parameters to consumption parameters of advertisements affected by price adjustment parameters, the consumption parameters are used to characterize the costs corresponding to successfully delivered advertisements, and the first function is used to characterize the degree of influence of price adjustment parameters on the rate of change of costs corresponding to successfully delivered advertisements.

[0033] As can be seen from the above introduction, the advertising publisher can pay the advertising server platform for the advertising fee, and then deliver the advertisements to various terminal devices through the advertising server platform. The advertising publisher can adjust the advertising fee, for example, by adjusting the price parameter.

[0034] In some embodiments, when the advertiser detects that the price adjustment trigger condition is met, it can perform the subsequent steps of determining the price adjustment parameters, thereby obtaining the target price adjustment parameters. Exemplarily, the price adjustment trigger condition can be reaching a preset price adjustment moment, so that when the advertiser detects that the preset price adjustment moment is reached, it can be determined that the price adjustment trigger condition is met. In another exemplary embodiment, the price adjustment trigger condition can also include an interval of a specified duration. Thus, if the advertiser detects that the time difference between the current moment and the last moment when the price adjustment trigger condition was met reaches a specified duration, it can be determined that the price adjustment trigger condition is met.

[0035] It is understandable that after adjusting the delivery cost through the price adjustment parameters, the probability of the advertising server platform delivering the advertisement to the terminal device will change. Exemplarily, the advertising server platform will maintain a list of advertisements to be delivered, and generally select some advertisements from the list of advertisements to be delivered to the terminal device. The list of advertisements to be delivered includes multiple advertisements arranged in sequence based on a specified order, and the specified order is related to the delivery cost of each advertisement. Therefore, after the delivery cost changes, the order of advertisements in the list of advertisements to be delivered will change, and thus the advertisements selected from the list of advertisements to be delivered will also change. This will cause the number of terminal devices that receive the delivery of the advertisement to change, thereby affecting the number of successfully delivered advertisements.

[0036] Thus, in some embodiments, a first function between a price adjustment parameter and a consumption parameter change rate can be constructed, wherein the consumption parameter change rate is used to characterize the ratio of the consumption parameter of an advertisement not under the effect of the price adjustment parameter to the consumption parameter of an advertisement under the effect of the price adjustment parameter, and the first function is used to characterize the degree of influence of the price adjustment parameter on the rate of change of the cost corresponding to the successfully delivered advertisement. For example, the price adjustment parameter can be used as an independent variable and the consumption parameter change rate can be used as a dependent variable to construct the first function. The consumption parameter is used to characterize the cost corresponding to the successfully delivered advertisement. That is, for each successfully delivered advertisement, the cost that the advertiser needs to pay to the advertising server platform can be characterized by the consumption parameter.

[0037] Wherein, a first function between a price adjustment parameter and a consumption parameter change rate can be constructed based on multiple groups of sample data, each group of the sample data including a historical price adjustment parameter and a historical consumption parameter change rate corresponding to the historical price adjustment parameter. Exemplarily, each group of the sample data including the historical price adjustment parameter and the historical consumption parameter change rate corresponding to the historical price adjustment parameter can be fitted to obtain the first function, and a detailed description can be referred to in the subsequent embodiments.

[0038] It should be noted that the historical price adjustment parameter is the price adjustment parameter corresponding to a previous price adjustment, such as the price adjustment parameter corresponding to a price adjustment before the current moment. The historical consumption parameter change rate corresponding to the historical price adjustment parameter is the ratio of the consumption parameter of the advertisement under the effect of the historical price adjustment parameter to the consumption parameter of the advertisement under the effect of the historical price adjustment parameter.

[0039] The sample data can be obtained through online collection. For detailed introduction, please refer to the subsequent embodiments.

[0040] Step S120: Acquire the user's conversion parameter under the effect of the price adjustment parameter as the user's post-price adjustment historical conversion parameter, wherein the conversion parameter is used to characterize the number of successfully delivered advertisements.

[0041] From the above introduction, it can be known that adjusting the delivery cost through the price adjustment parameter will affect the probability of the subsequent user successfully performing a specified operation on the delivered advertisement in the terminal device. Therefore, the number of successfully delivered advertisements can be obtained first, such as the number of successfully delivered advertisements before the current moment. Among them, the successfully delivered advertisements are the advertisements for which the user successfully performs a specified operation in the terminal device.

[0042] Specifically, the conversion parameter of the user under the effect of the price adjustment parameter may be obtained as the user's post-price adjustment historical conversion parameter, wherein the conversion parameter is used to characterize the number of successfully delivered advertisements.

[0043] For some implementations, the conversion parameter of the user under the price adjustment parameter corresponding to the time period within the day before the current moment may be obtained as the user's post-price adjustment historical conversion parameter. For example, if the current moment is t, and the corresponding time period within the day is 0 to 24, the conversion parameter of the user under the price adjustment parameter corresponding to the time period between 0 and t may be used as the user's post-price adjustment historical conversion parameter. Wherein, t is greater than or equal to 0 and less than or equal to 24.

[0044] In some implementations, a request may be sent to the advertising server platform to obtain the conversion parameters of the user under the effect of the price adjustment parameters sent by the advertising server platform.

[0045] Step S130: Construct a second function between the consumption parameter change rate and the user's future conversion parameter after the price adjustment under the action of the price adjustment parameter, wherein the future conversion parameter includes the predicted conversion parameter, and the second function is used to characterize the relationship between the rate of change of the cost corresponding to the successfully delivered advertisements under the action of the price adjustment parameter and the predicted number of successfully delivered advertisements.

[0046] From the above introduction, it can be known that adjusting the delivery cost through the price adjustment parameters will affect the probability of subsequent users successfully performing designated operations on the delivered advertisements in the terminal device. Therefore, it is also possible to predict the future conversion parameters under the effect of the price adjustment parameters, so that the future conversion parameters can be considered when determining the target price adjustment parameters in the future, so as to maximize the accuracy of the obtained target price adjustment parameters.

[0047] In some embodiments, a second function can be constructed between the consumption parameter change rate and the user's future conversion parameter after the price adjustment under the influence of the price adjustment parameter, and the aforementioned constructed first function is used to characterize the degree of influence of the price adjustment parameter on the rate of change of the cost corresponding to the successfully delivered advertisement, so that the first function and the second function can be combined to determine the target price adjustment parameter later.

[0048] Wherein, the future conversion parameter includes a predicted conversion parameter, and the second function is used to characterize the relationship between the rate of change of the cost corresponding to the successfully delivered advertisement under the action of the price adjustment parameter and the predicted number of successfully delivered advertisements. Exemplarily, the time period corresponding to the future conversion parameter can be a time period within the day after the current moment. Exemplarily, if the current moment is t, and the time period corresponding to the day is 0 to 24, the time period between t and 24 can be the time period corresponding to the future conversion parameter, wherein t is greater than or equal to 0 and less than or equal to 24. That is, the conversion parameter of the time period between t and 24 is predicted.

[0049] The future conversion parameter after price adjustment can be determined by the future cost after price adjustment and the future consumption parameter after price adjustment. The future cost is the predicted cost corresponding to the advertisement, and the future consumption parameter after price adjustment is the predicted cost corresponding to the successfully placed advertisement. For detailed description, please refer to the subsequent embodiments.

[0050] It should be noted that, “after price adjustment” refers to some parameters or variables obtained under the effect of price adjustment parameters; while “before price adjustment” refers to some parameters or variables not obtained under the effect of price adjustment parameters.

[0051] Step S140: determining a target price adjustment parameter based on the first function and the second function, wherein the target price adjustment parameter is used to maximize the sum of the user's post-price adjustment historical conversion parameter and the user's post-price adjustment future conversion parameter under target constraints.

[0052] Thus, the target price adjustment parameter can be determined based on the first function and the second function. The target price adjustment parameter is used to maximize the sum of the user's post-price adjustment historical conversion parameter and the user's post-price adjustment future conversion parameter under the target constraint condition. In some embodiments, an equation for the sum of the user's post-price adjustment historical conversion parameter and the user's post-price adjustment future conversion parameter can be constructed, and then the target constraint condition is used as the constraint condition of the equation to obtain a solution corresponding to the maximum value that makes the equation satisfy the constraint condition, and the solution is the target price adjustment parameter. For a detailed description, please refer to the subsequent embodiments.

[0053] The target constraint condition may include a first constraint condition constructed based on a target cost, and the target cost may be a pre-set cost, such as a cost expected to be achieved pre-set by an advertiser, which may be used to represent the expected value of the advertiser.

[0054] Optionally, the target constraint condition may further include a second constraint condition constructed based on a preset target total consumption parameter, where the target total consumption parameter may be used to represent a maximum budget cost preset by the advertisement delivery provider for paying for successfully delivered advertisements.

[0055] It is understandable that the sum of the historical conversion parameter after price adjustment and the user's future conversion parameter after price adjustment can be used to characterize the revenue of the advertisement in the corresponding time period. Exemplarily, if the time period corresponding to the historical conversion parameter after price adjustment is a time period within the day before the current moment, and if the time period corresponding to the future conversion parameter is a time period within the day after the current moment, then the sum of the historical conversion parameter after price adjustment and the user's future conversion parameter after price adjustment can be used to characterize the revenue of the advertisement on that day.

[0056] Among them, the first function is used to characterize the influence of the price adjustment parameters on the rate of change of costs corresponding to successfully delivered advertisements, and the second function is used to characterize the relationship between the rate of change of costs corresponding to successfully delivered advertisements under the action of the price adjustment parameters and the predicted number of successfully delivered advertisements. Therefore, under the target constraint conditions, the target price adjustment parameters determined by the first function and the second function have higher accuracy, and the subsequent adjustment of the delivery costs by the target price adjustment parameters can maximize the revenue of the advertisement.

[0057] The information processing method provided in the embodiment of the present application first constructs a first function between a price adjustment parameter and a consumption parameter change rate based on multiple groups of sample data; then obtains the user's conversion parameter under the action of the price adjustment parameter as the user's post-price adjustment historical conversion parameter, wherein the conversion parameter is used to characterize the number of successfully delivered advertisements; then constructs a second function between the consumption parameter change rate and the user's post-price adjustment future conversion parameter under the action of the price adjustment parameter, wherein the future conversion parameter includes a predicted conversion parameter, and the second function is used to characterize the relationship between the rate of change of the cost corresponding to the successfully delivered advertisements under the action of the price adjustment parameter and the predicted number of successfully delivered advertisements; thereby determining a target price adjustment parameter based on the first function and the second function, wherein the target price adjustment parameter is used to maximize the sum of the user's post-price adjustment historical conversion parameter and the user's post-price adjustment future conversion parameter under the target constraint condition. Among them, each group of sample data includes historical price adjustment parameters and the historical consumption parameter change rate corresponding to the historical price adjustment parameters, the consumption parameter change rate is used to characterize the ratio of the consumption parameters of the advertisement not under the effect of the price adjustment parameters to the consumption parameters of the advertisement under the effect of the price adjustment parameters, the consumption parameters are used to characterize the costs corresponding to the successfully delivered advertisements, and the first function is used to characterize the degree of influence of the price adjustment parameters on the rate of change of the costs corresponding to the successfully delivered advertisements. The target price adjustment parameters in this application can maximize the sum of the user's post-price adjustment historical conversion parameters and the user's post-price adjustment future conversion parameters under the target constraint conditions. It can be seen that the price adjustment parameters are not determined solely by the historical conversion parameters, but the target price adjustment parameters are determined by combining the post-price adjustment historical conversion parameters and the post-price adjustment future conversion parameters. And in this application, the second function includes the post-price adjustment future conversion parameters, so that the subsequent combination of the first function and the second function to determine the target price adjustment parameters can bring better benefits and improve the accuracy of the target price adjustment parameters.

[0058] See also Figure 3 , Figure 3 A method flow chart of an information processing method provided by an embodiment of the present application is shown. The information processing method can be applied to Figure 1The advertisement provider in the information recommendation scenario shown may specifically use the processor of the advertisement provider as the execution subject of the information processing method. The information processing method may include steps S210 to S280.

[0059] Step S210: obtaining a plurality of historical price adjustment parameters corresponding to the advertisement, a reference consumption parameter before price adjustment of the user not affected by the historical price adjustment parameters, and a reference consumption parameter after price adjustment of the user affected by the historical price adjustment parameters.

[0060] In order to facilitate the subsequent construction of the first function through sample data, multiple sets of sample data can be constructed first. As can be seen from the above introduction, each set of sample data includes historical price adjustment parameters and the historical consumption parameter change rate corresponding to the historical price adjustment parameters. Therefore, multiple historical price adjustment parameters corresponding to the advertisement, the pre-price adjustment reference consumption parameters of the user not under the effect of the historical price adjustment parameters, and the post-price adjustment reference consumption parameters of the user under the effect of the historical price adjustment parameters can be obtained first.

[0061] The historical price adjustment parameters and the user's reference consumption parameters after price adjustment under the historical price adjustment parameters can be obtained through the advertising server platform, and the user's reference consumption parameters before price adjustment without the historical price adjustment parameters can be obtained after price adjustment processing. The reference consumption parameters can represent standard consumption parameters. Specifically, step S210 can also include step S211 and step S212.

[0062] Step S211: obtaining, through the advertisement server platform, a plurality of historical price adjustment parameters corresponding to the advertisement and a reference consumption parameter of the user after price adjustment under the effect of the historical price adjustment parameters.

[0063] In some implementations, multiple historical price adjustment parameters corresponding to the advertisement and the user's reference consumption parameters after price adjustment under the effect of the historical price adjustment parameters can be obtained through the advertising server platform. Exemplarily, the bidding atmosphere log can be obtained through the bidding atmosphere log playback technology, and then multiple historical price adjustment parameters and the user's reference consumption parameters after price adjustment under the effect of the historical price adjustment parameters can be obtained through the bidding atmosphere log. Among them, the bidding atmosphere log can be collected in real time by the advertising server platform and sent to the advertising provider.

[0064] Exemplarily, the multiple historical price adjustment parameters corresponding to an advertisement may include an advertisement corresponding to the advertisement provider, so that the advertisement may correspond to different historical price adjustment parameters. In another exemplary embodiment, the multiple historical price adjustment parameters corresponding to an advertisement may also include multiple advertisements, each with a corresponding historical price adjustment parameter.

[0065] Optionally, step S211 may also include steps S2111 to S2113.

[0066] Step S2111: obtaining, through the advertising server platform, a plurality of advertisements sorted from high to low based on the reference consumption parameters after price adjustment.

[0067] Step S2112: Select a specified number of advertisements ranked high as candidate advertisements.

[0068] Step S2113: Acquire the historical price adjustment parameters of the candidate advertisement and the reference consumption parameters of the user after the price adjustment under the effect of the historical price adjustment parameters.

[0069] First, multiple advertisements sorted from high to low based on the reference consumption parameter after price adjustment can be obtained through the advertising server platform. Exemplarily, a bidding atmosphere log can be obtained through the advertising server platform, and the bidding atmosphere log includes multiple advertisements. The multiple advertisements included in the bidding atmosphere log can be arranged in a specified order, for example, they can be sorted from high to low according to the reference consumption parameter after price adjustment.

[0070] In some implementations, the reference consumption parameter after price adjustment can be characterized by ecpm, so that the reference consumption parameter after price adjustment can be determined by the following formula (1).

[0071] ecpm=pctr*pcvr*tcpa*λ (1)

[0072] Among them, pctr is used to characterize the user's predicted click-through rate (Predict Click-Through Rate), pcvr is used to characterize the predicted conversion rate (Predict ConVersion Rate), tcpa is used to characterize the target bid (Target Cost Per Action), and λ is used to characterize the price adjustment parameter. It should be noted that the predicted click-through rate, predicted conversion rate, target bid, and price adjustment parameters are all obtained by the advertising server platform, and the advertising server platform generates the reference consumption parameters after price adjustment corresponding to each advertisement based on the above parameters, thereby determining the ranking of each advertisement based on the reference consumption parameters after price adjustment.

[0073] It is understandable that the higher the ranking of the reference consumption parameter after the price adjustment, the greater the consumption that may be generated. Therefore, the advertising server platform generally delivers the advertisements with the higher ranking of the reference consumption parameter to the terminal device. Therefore, a specified number of advertisements with the higher ranking can be selected as alternative advertisements. For example, the specified number can be 3.

[0074] Furthermore, the historical price adjustment parameters of the candidate advertisements and the reference consumption parameters of the user after the price adjustment under the historical price adjustment parameters can be obtained. Similar to the above steps, the bidding atmosphere log can also include the historical price adjustment parameters of each advertisement, so that after the candidate advertisements are determined, the historical price adjustment parameters of the candidate advertisements can be directly obtained through the bidding atmosphere log.

[0075] From the above introduction, it can be known that the bidding atmosphere log also includes the user's reference consumption parameters after the price adjustment of each advertisement under the effect of the historical price adjustment parameters. Therefore, the user's reference consumption parameters after the price adjustment of the alternative advertisement under the effect of the historical price adjustment parameters can be directly obtained through the bidding atmosphere log. The historical price adjustment parameters of the alternative advertisements obtained and the user's reference consumption parameters after the price adjustment under the effect of the historical price adjustment parameters are the multiple historical price adjustment parameters corresponding to the advertisement and the user's reference consumption parameters after the price adjustment under the effect of the historical price adjustment parameters.

[0076] Step S212: De-pricing the post-price adjustment reference consumption parameters of each advertisement using the historical price adjustment parameters of the advertisement to obtain the pre-price adjustment reference consumption parameters of the advertisement that are not affected by the historical price adjustment parameters.

[0077] Furthermore, after obtaining multiple historical price adjustment parameters corresponding to the advertisement and the user's post-price adjustment reference consumption parameters under the influence of the historical price adjustment parameters, the post-price adjustment reference consumption parameters of each advertisement can be de-priced using the historical price adjustment parameters of the advertisement to obtain the pre-price adjustment reference consumption parameters of the advertisement that were not influenced by the historical price adjustment parameters.

[0078] Exemplarily, the reference consumption parameter after price adjustment of the user under the historical price adjustment parameter corresponding to the advertisement may be divided by the historical price adjustment parameter of the advertisement, thereby obtaining the reference consumption parameter before price adjustment of the user not under the historical price adjustment parameter corresponding to the advertisement. In some embodiments, the reference consumption parameter before price adjustment of the user not under the historical price adjustment parameter may be represented by ecpm′.

[0079] Optionally, the advertisements may also be sorted from high to low based on the reference consumption parameters of the user before the price adjustment without the effect of the historical price adjustment parameters. Figure 4 , Figure 4 A schematic diagram showing advertisement sorting in an embodiment of the present application is shown. Figure 4 4 shows a ranking 410 based on the reference consumption parameter after price adjustment and a ranking 420 based on the reference consumption parameter before price adjustment. The ranking 410 based on the reference consumption parameter after price adjustment includes advertisement 411, advertisement 412, advertisement 413, advertisement 414, advertisement 415 and advertisement 416. Figure 4The number shown in each advertisement from advertisement 411 to advertisement 416 is the reference consumption parameter after price adjustment corresponding to the advertisement. For example, the reference consumption parameter after price adjustment corresponding to advertisement 412 is 47, which shows that advertisement 412 is ranked second in the ranking 410 of reference consumption parameters after price adjustment. The ranking 420 based on the reference consumption parameters before price adjustment includes advertisement 421, advertisement 422, advertisement 423, advertisement 424, advertisement 425 and advertisement 426. Similarly, Figure 4 The numbers shown in each advertisement from advertisement 421 to advertisement 426 are the reference consumption parameters before price adjustment corresponding to the advertisement. Among them, advertisement 412 and advertisement 424 are the same advertisement, and the reference consumption parameter before price adjustment corresponding to advertisement 424 is 19. It can be seen that at this time, advertisement 424 ranks fourth in the ranking 420 of reference consumption parameters before price adjustment.

[0080] Optionally, the sorting of the above-mentioned advertisements can also be the sorting of advertisement groups, so that the obtained multiple historical price adjustment parameters, the reference consumption parameters before price adjustment of users not affected by the historical price adjustment parameters, and the reference consumption parameters after price adjustment of users under the influence of the historical price adjustment parameters all correspond to each advertisement group, and each advertisement group can include multiple advertisements.

[0081] Step S220: constructing a continuous price adjustment range divided into multiple segments based on the historical price adjustment parameters of each advertisement.

[0082] It is understandable that multiple historical price adjustment parameters of an advertisement may be the same or different. To facilitate the subsequent construction of multiple groups of sample data, a continuous price adjustment range divided into multiple segments may be constructed based on the historical price adjustment parameters of each advertisement.

[0083] Exemplarily, first, based on the historical price adjustment parameters of each of the advertisements, the maximum value and the minimum value of all the historical price adjustment parameters can be determined to obtain the interval range between the minimum value and the maximum value, and then the interval range between the minimum value and the maximum value can be divided to construct a continuous price adjustment interval range divided into multiple segments. For example, the interval range can be divided by a specified length. It can be seen that each price adjustment interval range belongs to a part of the interval range between the minimum value and the maximum value. For example, if the maximum value of all the historical price adjustment parameters is determined to be 20 and the minimum value is 1 based on the historical price adjustment parameters of each of the advertisements, the interval range between the minimum value and the maximum value can be characterized by (1, 20). If the specified length is 0.01, the interval range can be divided to obtain (20-1) / 0.01=1900 price adjustment interval ranges, and the 1900 price adjustment interval ranges are continuous.

[0084] Step S230: based on the pre-price adjustment reference consumption parameter and the post-price adjustment reference consumption parameter corresponding to each of the historical price adjustment parameters matching the price adjustment interval, obtain the historical consumption parameter change rate matching the price adjustment interval.

[0085] Therefore, based on the pre-price adjustment reference consumption parameter and the post-price adjustment reference consumption parameter corresponding to the historical price adjustment parameter that matches each price adjustment interval, the change rate of the historical consumption parameter that matches the price adjustment interval can be obtained.

[0086] It is understandable that the price adjustment interval range in which the historical price adjustment parameter is located is the price adjustment interval range that matches the historical price adjustment parameter. Therefore, the same price adjustment interval range may match multiple historical price adjustment parameters, and each historical price adjustment parameter also corresponds to a reference consumption parameter before price adjustment and a reference consumption parameter after price adjustment.

[0087] In some implementations, each historical price adjustment parameter matched by the price adjustment interval may correspond to a historical consumption parameter change rate, so that the historical consumption parameter change rates corresponding to each historical price adjustment parameter matched by the price adjustment interval may be averaged to serve as the historical consumption parameter change rate matched by the price adjustment interval. Specifically, step S230 may further include steps S231 to S233.

[0088] Step S231: summing the pre-price adjustment reference consumption parameters corresponding to each of the historical price adjustment parameters matching the price adjustment interval range to obtain a first value matching the price adjustment interval range.

[0089] Step S232: summing up the price-adjusted reference consumption parameters corresponding to each of the historical price adjustment parameters matching the price adjustment interval range to obtain a second value matching the price adjustment interval range.

[0090] Step S233: taking the ratio of the second numerical value matched by each price adjustment interval to the first numerical value matched by the price adjustment interval as the historical consumption parameter change rate matched by the price adjustment interval.

[0091] First, the pre-price adjustment reference consumption parameters corresponding to the historical price adjustment parameters matching each price adjustment interval range can be summed to obtain a first value matching the price adjustment interval range. For example, if the pre-price adjustment reference consumption parameters corresponding to the historical price adjustment parameters matching each price adjustment interval range are represented by ecpm′, the first value matching the price adjustment interval range can be represented as ∑ecpm′.

[0092] Similarly, the price adjustment reference consumption parameters corresponding to the historical price adjustment parameters matching each price adjustment interval range may be summed to obtain a second value matching the price adjustment interval range. For example, if the price adjustment reference consumption parameters corresponding to the historical price adjustment parameters matching each price adjustment interval range are represented by ecpm, the second value matching the price adjustment interval range may be represented as ∑ecpm.

[0093] Furthermore, the ratio of the second value of each price adjustment interval to the first value of the price adjustment interval is used as the historical consumption parameter change rate of the price adjustment interval. For example, if F(λ) is used to characterize the historical consumption parameter change rate of the price adjustment interval, then To characterize the rate of change of historical consumption parameters matching the price adjustment range.

[0094] Step S240: taking the historical price adjustment parameter and the historical consumption parameter change rate that matches the price adjustment interval that matches the historical price adjustment parameter as sample data.

[0095] The above λ represents the historical price adjustment parameter, and the historical consumption parameter change rate that matches the price adjustment interval of the historical price adjustment parameter is F(λ). Therefore, the sample data can be represented by (λ, F(λ)). And through multiple historical price adjustment parameters, multiple groups of sample data can be constructed.

[0096] Step S250: Based on multiple groups of sample data, construct a first function between price adjustment parameters and consumption parameter change rates, wherein the sample data includes historical price adjustment parameters and historical consumption parameter change rates corresponding to the historical price adjustment parameters, the consumption parameter change rates are used to characterize the ratio of consumption parameters of advertisements not affected by price adjustment parameters to consumption parameters of advertisements affected by price adjustment parameters, the consumption parameters are used to characterize the costs corresponding to successfully delivered advertisements, and the first function is used to characterize the degree of influence of price adjustment parameters on the rate of change of costs corresponding to successfully delivered advertisements.

[0097] Step S260: Acquire the user's conversion parameter under the effect of the price adjustment parameter as the user's post-price adjustment historical conversion parameter, wherein the conversion parameter is used to characterize the number of successfully delivered advertisements.

[0098] For some implementations, the conversion parameter of the user under the effect of the price adjustment parameter corresponding to the time period within the day before the current moment may be obtained as the user's post-price adjustment historical conversion parameter.

[0099] Step S270: Construct a second function between the consumption parameter change rate and the user's future conversion parameter after the price adjustment under the action of the price adjustment parameter, wherein the future conversion parameter includes the predicted conversion parameter, and the second function is used to characterize the relationship between the rate of change of the cost corresponding to the successfully delivered advertisements under the action of the price adjustment parameter and the predicted number of successfully delivered advertisements.

[0100] In some implementations, the time period corresponding to the future conversion parameter may be a time period within the same day after the current moment.

[0101] Step S280: determining a target price adjustment parameter based on the first function and the second function, wherein the target price adjustment parameter is used to maximize the sum of the user's post-price adjustment historical conversion parameter and the user's post-price adjustment future conversion parameter under target constraints.

[0102] If the time period corresponding to the historical conversion parameter after price adjustment is within the same day, and if the time period corresponding to the future conversion parameter is within the same day after the current moment, the sum of the historical conversion parameter after price adjustment and the user's future conversion parameter after price adjustment can be used to characterize the revenue of the advertisement on the same day. Therefore, in the embodiment provided by the present application, the overall control of the revenue of the whole day can be achieved, taking into account the influence of the consumption, conversion parameters, costs, etc. throughout the day, and then more accurate target price adjustment parameters can be obtained under the target constraint conditions.

[0103] The detailed description of steps S250 to S280 can be found in the aforementioned embodiments and will not be repeated here.

[0104] The information processing method provided in the embodiment of the present application obtains the historical price adjustment parameters corresponding to multiple advertisements, the pre-price adjustment reference consumption parameters of users who are not under the effect of the historical price adjustment parameters, and the post-price adjustment reference consumption parameters of users under the effect of the historical price adjustment parameters; based on the historical price adjustment parameters of each of the advertisements, a continuous price adjustment interval range divided into multiple segments is constructed; based on the pre-price adjustment reference consumption parameters and the post-price adjustment reference consumption parameters corresponding to the historical price adjustment parameters matching each of the price adjustment interval ranges, the historical consumption parameter change rate matching the price adjustment interval range is obtained; the historical price adjustment parameters and the historical consumption parameter change rates matching the price adjustment interval range matching the historical price adjustment parameters are used as sample data. If the sample data is obtained by directly collecting information such as the historical consumption parameter change rate from the online network, the cost is relatively high. In the embodiment of the present application, the sample data is not obtained by directly collecting information such as the historical consumption parameter change rate from the online network, but by obtaining the bidding atmosphere log to construct the sample data, and then the first function is constructed through the sample data to determine the target price adjustment parameters, thereby reducing the cost.

[0105] See also Figure 5 , Figure 5 A method flow chart of an information processing method provided by an embodiment of the present application is shown. The information processing method can be applied to Figure 1 The advertisement provider in the information recommendation scenario shown may specifically use the processor of the advertisement provider as the execution subject of the information processing method. The information processing method may include steps S310 to S360.

[0106] Step S310: Based on multiple groups of sample data, construct a first function between price adjustment parameters and consumption parameter change rates, wherein the sample data includes historical price adjustment parameters and historical consumption parameter change rates corresponding to the historical price adjustment parameters, the consumption parameter change rates are used to characterize the ratio of consumption parameters of advertisements not affected by price adjustment parameters to consumption parameters of advertisements affected by price adjustment parameters, the consumption parameters are used to characterize the costs corresponding to successfully delivered advertisements, and the first function is used to characterize the degree of influence of price adjustment parameters on the rate of change of costs corresponding to successfully delivered advertisements.

[0107] Step S320: Acquire the user's conversion parameter under the effect of the price adjustment parameter as the user's post-price adjustment historical conversion parameter, wherein the conversion parameter is used to characterize the number of successfully delivered advertisements.

[0108] Among them, step S310 and step S320 have been introduced in detail in the above embodiments and will not be repeated here.

[0109] Step S330: Based on the pre-acquired historical cost after price adjustment under the price adjustment parameter, predict the future cost after price adjustment under the price adjustment parameter, wherein the future cost includes the predicted cost.

[0110] From the above introduction, we can know that the second function is used to characterize the relationship between the rate of change of the cost corresponding to the successfully placed advertisements under the price adjustment parameter and the predicted number of successfully placed advertisements, wherein the predicted number of successfully placed advertisements is the future conversion parameter after the price adjustment. The future conversion parameter after the price adjustment can be determined by the future cost after the price adjustment and the future consumption parameter after the price adjustment, so the future cost after the price adjustment can be obtained first.

[0111] In some implementations, based on pre-acquired historical costs after price adjustment under the price adjustment parameters, future costs after price adjustment under the price adjustment parameters may be predicted, wherein the future costs include predicted costs.

[0112] The historical cost after price adjustment under the price adjustment parameter may be the cost that has been incurred by the advertisement under the price adjustment parameter, and the future cost after price adjustment under the price adjustment parameter may be the cost that is predicted to be incurred by the advertisement under the price adjustment parameter. Similar to the above embodiment, the historical cost after price adjustment corresponding to the time period before the current moment may be obtained; and the time period corresponding to the future cost after price adjustment may be the time period after the current moment.

[0113] The adjusted historical cost can be directly obtained from the advertising server platform. For example, the advertising provider can send a request to the advertising server platform to obtain the adjusted historical cost, and the advertising server platform can respond to the request and send the adjusted historical cost to the advertising provider.

[0114] Specifically, step S330 may include steps S331 to S333.

[0115] Step S331: Based on the pre-acquired historical costs after price adjustment under the price adjustment parameters and the cost change rates before and after the price adjustment, determine the historical costs before price adjustment without the price adjustment parameters.

[0116] Step S332: predicting the future cost before price adjustment without the price adjustment parameter based on the historical cost before price adjustment.

[0117] Step S333: Based on the future cost before the price adjustment and the cost change rate before and after the price adjustment, predict the future cost after the price adjustment under the action of the price adjustment parameters.

[0118] The cost change rate is used to represent the ratio of the post-price adjustment cost under the price adjustment parameter to the pre-price adjustment cost without the price adjustment parameter. In some embodiments, the pre-price adjustment cost without the price adjustment parameter can be represented by the following formula (2).

[0119]

[0120] Among them, ω is used to characterize the probability of winning the bid, which may be positively correlated with the reference consumption parameter ecpm after price adjustment. That is, the higher the ranking of the advertisement in ecpm, the greater the probability of winning the bid.

[0121] The adjusted cost under the influence of the price adjustment parameters can be represented by the following formula (3).

[0122]

[0123] Therefore, the cost change rate before and after the price adjustment can be obtained by the ratio of formula (3) to formula (2), which can be specifically represented by the following formula (4).

[0124]

[0125] Among them, C(λ) is the cost change rate before and after the price adjustment.

[0126] Specifically, the ratio of the historical cost after price adjustment to the cost change rate can be used as the historical cost before price adjustment without the effect of the price adjustment parameter. h,λ′ To characterize the historical cost after price adjustment, through cpq λ＝1 To characterize the historical cost before price adjustment, the historical cost before price adjustment can be characterized by the following formula (5).

[0127]

[0128] Among them, λ′ is used to characterize the average value of the historical price adjustment parameters. Furthermore, the future cost before price adjustment without the effect of the price adjustment parameters can be predicted based on the historical cost before price adjustment. In some embodiments, the historical cost before price adjustment can be regarded as the average cost within a period of time, so that for the convenience of calculation, the historical cost before price adjustment can be approximated as the average cost within a period of time in the future after the current moment, so that the historical cost before price adjustment can be directly regarded as the future cost before price adjustment without the effect of the price adjustment parameters.

[0129] Then, based on the future cost before price adjustment and the cost change rate before and after price adjustment, the future cost after price adjustment under the effect of price adjustment parameters can be predicted. Specifically, the product of the future cost before price adjustment and the cost change rate before and after price adjustment can be used as the future cost after price adjustment under the effect of price adjustment parameters. e,λ The future cost after price adjustment can be determined by the following formula (6).

[0130]

[0131] Step S340: Construct a third function between the future consumption parameters after the price adjustment under the action of the price adjustment parameters and the change rate of the historical consumption parameters, wherein the future consumption parameters include the consumption parameters predicted after the current moment, and the third function is used to characterize the relationship between the change rate of the cost corresponding to the advertisement successfully delivered under the action of the price adjustment parameters and the predicted cost corresponding to the advertisement successfully delivered under the action of the price adjustment parameters after the current moment.

[0132] In addition, a third function between the future consumption parameter after price adjustment and the change rate of the historical consumption parameter under the action of the price adjustment parameter can also be constructed. Similar to the above-mentioned embodiment, the future consumption parameter can include the consumption parameter predicted after the current moment. The third function is used to characterize the relationship between the change rate of the cost corresponding to the advertisement successfully placed under the action of the price adjustment parameter and the predicted cost corresponding to the advertisement successfully placed under the action of the price adjustment parameter after the current moment. Among them, the change rate of the historical consumption parameter can be used to characterize the change rate of the historical consumption parameter.

[0133] In some implementations, step S340 may include steps S341 to S345.

[0134] Step S341: Obtain the price-adjusted historical consumption parameters of the advertisement under the price adjustment parameters.

[0135] Step S342: Based on the post-price adjustment historical consumption parameters and the consumption parameter change rate, determine the pre-price adjustment historical consumption parameters of the advertisement that are not affected by the price adjustment parameters.

[0136] Step S343: Obtain the average consumption of the historical specified time period without the effect of the price adjustment parameters.

[0137] Step S344: taking the difference between the average consumption of the historical specified time period and the historical consumption parameter of the advertisement before the price adjustment as the future consumption parameter before the price adjustment of the advertisement without the effect of the price adjustment parameter.

[0138] Step S345: Based on the future consumption parameter before price adjustment and the consumption parameter change rate, construct a function representing the future consumption parameter after price adjustment of the advertisement under the price adjustment parameter as a third function between the future consumption parameter after price adjustment and the historical consumption parameter change rate.

[0139] First, the post-price adjustment historical consumption parameters of the advertisement under the effect of the price adjustment parameters can be obtained. The post-price adjustment historical consumption parameters can be directly obtained from the advertising server platform. For example, the advertising provider can send a request to the advertising server platform to obtain the post-price adjustment historical consumption parameters, and the advertising server platform can respond to the request and send the post-price adjustment historical consumption parameters to the advertising provider. In some embodiments, the cost can be obtained by h,λ′ Characterizes the historical consumption parameters after price adjustment.

[0140] Then, based on the historical consumption parameters after price adjustment and the consumption parameter change rate, the historical consumption parameters of the advertisement before price adjustment can be determined. Specifically, the ratio of the historical consumption parameters after price adjustment of the advertisement to the historical consumption parameter change rate can be used as the historical consumption parameters of the advertisement before price adjustment.

[0141] It can be understood that if F(λ) is used to represent the consumption parameter change rate, it can be represented by formula (7):

[0142]

[0143] The meaning of each parameter in formula (7) can be found in the introduction of the above embodiments, which will not be repeated here. It can be seen that the consumption parameter change rate is a function related only to λ. Therefore, the relationship between λ and the consumption parameter change rate can be fitted through the historical price adjustment parameters, so as to obtain the influence of the price adjustment parameters on the future consumption parameters, so that the final target price adjustment parameters have higher accuracy.

[0144] Therefore, when λ is λ′, the change rate of the historical consumption parameter can be represented by F(λ′), and the historical consumption parameter can be obtained. Specifically, the historical consumption parameter before the price adjustment of the advertisement can be represented by formula (8).

[0145]

[0146] Furthermore, the average consumption of a specified historical period without the effect of the price adjustment parameter can be obtained. For example, the average consumption of a specified historical period can be the average daily consumption in a week before the current moment. In another example, the average consumption of a specified historical period can also be the consumption corresponding to an adjacent day before the current moment. In some embodiments, the cost day To characterize the average consumption during a specified period of history.

[0147] Thus, the future consumption parameter of the advertisement before the price adjustment under the effect of the price adjustment parameter can be obtained. Specifically, the difference between the average consumption of the historical specified time period and the historical consumption parameter of the advertisement before the price adjustment can be used as the future consumption parameter of the advertisement before the price adjustment without the effect of the price adjustment parameter, which can be specifically represented by the following formula (9).

[0148] cost e,λ＝1 =cost day -cost h,λ＝1 (9)

[0149] Optionally, the future consumption parameters before price adjustment can also be characterized in combination with the number of times the advertisement is displayed. Specifically, G(t) can be used to characterize the ratio of the exposure corresponding to the advertisement in the historical time period of the day before the current moment to the total exposure, where the total exposure is used to characterize the average daily exposure of the advertisement before the current moment. Exposure is used to characterize the number of advertisements delivered to the terminal device. Exemplarily, the exposure corresponding to the advertisement in the historical time period of the day before the current moment can be represented by (0, t), and the total exposure can be represented by (0, 24), then G(t) = (0, t) / (0, 24).

[0150] Therefore, the above formula (9) can be further represented as the following formula (10).

[0151]

[0152] Therefore, based on the future consumption parameters before price adjustment and the consumption parameter change rate, a function representing the future consumption parameters after price adjustment of the advertisement under the price adjustment parameter can be constructed as a third function between the future consumption parameters after price adjustment and the historical consumption parameter change rate.

[0153] It is understandable that there may be a problem that the historical time period represented by (0, t) before the current moment is shorter, resulting in a lower exposure accuracy of the advertisement in the historical time period. Therefore, optionally, the time period corresponding to (0, t) can be adjusted to the time period corresponding to (-k, t), so that G(t) = (-k, t) / (-k, 24) can be adjusted to obtain a more accurate G(t). Among them, -k represents k hours before the current day.

[0154] For details, please refer to Figure 6 , Figure 6 A schematic diagram of a historical time period in an embodiment of the present application is shown. Figure 6 6 shows a historical time period 610 and a historical time period 620, wherein the historical time period 610 corresponds to the time period (0, t), and the historical time period 620 corresponds to the time period (-k, t). The time t is the current time, and 0 to 24 o'clock corresponds to the time period of the day.

[0155] Optionally, historical price adjustment parameters, historical consumption parameters, and historical costs after price adjustment, etc., can all be adjusted from a time period within the day before the corresponding current moment to a time period corresponding to (-k, t).

[0156] Specifically, the product of the future consumption parameter before the price adjustment and the consumption parameter change rate can be used as the future consumption parameter after the price adjustment. The third function can be specifically represented by the following formula (11).

[0157] cost e，λ =cost e,λ＝1 *F(λ) (11)

[0158] It can be seen that the third function shown in formula (11) is a function constructed by the future consumption parameter before price adjustment and the rate of change of the consumption parameter.

[0159] Optionally, formula (11) can also be expressed in the form of the following formula (12).

[0160]

[0161] Step S350: constructing a function representing the future conversion parameter after the price adjustment based on the future cost after the price adjustment, the future consumption parameter after the price adjustment and the third function as a second function between the consumption parameter change rate and the future conversion parameter after the price adjustment.

[0162] After obtaining the future cost after price adjustment, the future consumption parameter after price adjustment and the third function through the above steps, a function characterizing the future conversion parameter after price adjustment can be constructed as the second function between the consumption parameter change rate and the future conversion parameter after price adjustment.

[0163] For some implementations, the future consumption parameter after price adjustment may be divided by the future cost after price adjustment to obtain the future conversion parameter after price adjustment. The future conversion parameter after price adjustment is then characterized by the third function to obtain the second function. Specifically, the second function may be characterized by the following formula (13).

[0164]

[0165] Step S360: Determine a target price adjustment parameter based on the first function and the second function, wherein the target price adjustment parameter is used to maximize the sum of the user's post-price adjustment historical conversion parameter and the user's post-price adjustment future conversion parameter under target constraints.

[0166] Among them, step S360 has been introduced in detail in the above embodiment and will not be repeated here.

[0167] The information processing method provided in the embodiment of the present application predicts the future cost after the price adjustment under the action of the price adjustment parameter based on the pre-acquired historical cost after the price adjustment under the action of the price adjustment parameter, wherein the future cost includes the predicted cost; constructs a third function between the future consumption parameter after the price adjustment under the action of the price adjustment parameter and the change rate of the historical consumption parameter, wherein the future consumption parameter includes the predicted consumption parameter, and the third function is used to characterize the relationship between the change rate of the cost corresponding to the advertisement successfully placed under the action of the price adjustment parameter and the predicted cost corresponding to the advertisement successfully placed under the action of the price adjustment parameter; constructs a function characterizing the future conversion parameter after the price adjustment based on the future cost after the price adjustment, the future consumption parameter after the price adjustment, and the third function, as the second function between the change rate of the consumption parameter and the future conversion parameter after the price adjustment, the change rate of the consumption parameter is a function only related to λ. Therefore, in the embodiment provided in the present application, the relationship between the change rate of the consumption parameter and the change rate of the consumption parameter can be fitted by the historical price adjustment parameter, so as to obtain the influence of the price adjustment parameter on the future consumption parameter, so that the target price adjustment parameter finally determined has higher accuracy.

[0168] See also Figure 7 , Figure 7A method flow chart of an information processing method provided by an embodiment of the present application is shown. The information processing method can be applied to Figure 1 The advertisement provider in the information recommendation scenario shown may specifically use the processor of the advertisement provider as the execution subject of the information processing method. The information processing method may include steps S410 to S4110.

[0169] Step S410: Construct an alternative function by presetting the function type, taking the price adjustment parameter as the independent variable and the consumption parameter change rate as the dependent variable.

[0170] Step S420: Determine the parameters in the candidate function by using the historical price adjustment parameters in each set of sample data and the historical consumption parameter change rates corresponding to the historical price adjustment parameters.

[0171] Step S430: generating a first function between the price adjustment parameter and the consumption parameter change rate based on the parameter and the alternative function.

[0172] For some embodiments, multiple preset function types can be pre-set, and the preset function types include multiple types of functions. Subsequently, one type of function can be selected from the preset function types for fitting. For example, the selected type of function can be fitted based on historical price adjustment parameters and the historical consumption parameter change rate corresponding to the historical price adjustment parameters to obtain the first function.

[0173] Exemplarily, the preset function type may include a linear function, a polynomial function, an exponential function, a logarithmic function, or the like.

[0174] For example, an exponential function is selected from a preset function type. At this time, the price adjustment parameter can be used as an independent variable and the consumption parameter change rate can be used as a dependent variable to construct an alternative function. The alternative function can be represented by the following formula (14).

[0175] F(λ)=a*λ b (14)

[0176] Among them, the parameter a and parameter b in formula (14) are parameters to be determined.

[0177] Furthermore, the parameters in the alternative function can be determined according to the historical price adjustment parameters in each set of sample data and the historical consumption parameter change rate corresponding to the historical price adjustment parameters, that is, the parameters a and b in the above formula (14) are determined.

[0178] Exemplarily, the historical price adjustment parameters in each set of sample data and the historical consumption parameter change rates corresponding to the historical price adjustment parameters may be fitted by means of parameter fitting, thereby determining the parameters in the candidate function.

[0179] Thus, the obtained fitting parameters a and b are substituted into formula (14) to generate a first function between the price adjustment parameter and the consumption parameter change rate based on the parameters and the alternative function. Specifically, for example, the parameters a and b can be solved by the least square method.

[0180] For example, see Figure 8 , Figure 8 A schematic diagram of a first function provided in an embodiment of the present application is shown. Figure 8 The two-dimensional coordinate system 800 shown in the figure includes an x-axis and a y-axis, wherein the x-axis is used to represent the price adjustment parameter, and the y-axis is used to represent the consumption parameter change rate. Each discrete point 820 in the two-dimensional coordinate system 800 is determined by the historical price adjustment parameters in each group of sample data and the historical consumption parameter change rate corresponding to the historical price adjustment parameters, thereby fitting to obtain a curve 810, which is the first function.

[0181] Step S440: Acquire the user's conversion parameter under the effect of the price adjustment parameter as the user's post-price adjustment historical conversion parameter, wherein the conversion parameter is used to characterize the number of successfully delivered advertisements.

[0182] Step S450: construct a second function between the consumption parameter change rate and the user's future conversion parameters after the price adjustment under the action of the price adjustment parameters, wherein the future conversion parameters include predicted conversion parameters, and the second function is used to characterize the relationship between the rate of change of the cost corresponding to the successfully delivered advertisements under the action of the price adjustment parameters and the predicted number of successfully delivered advertisements.

[0183] Among them, successful delivery includes the user performing a specified operation on the advertisement delivered by the advertiser. For example, the specified operation may be a click operation, and the user clicks the advertisement on the terminal device where the advertisement is delivered, which can be regarded as a successful delivery of the advertisement. For another example, the specified operation may also be a download operation, and the user downloads the advertisement on the terminal device where the advertisement is delivered, which can be regarded as a successful delivery of the advertisement.

[0184] Among them, step S440 and step S450 have been introduced in detail in the above embodiments and will not be repeated here.

[0185] Step S460: Based on the first function and the second function, a plurality of candidate price adjustment parameters and a future conversion parameter of the user after the price adjustment corresponding to each of the candidate price adjustment parameters under the action of the price adjustment parameter are determined.

[0186] Step S470: Construct a ratio of the third value to the fourth value equal to a preset target cost as a first constraint condition, wherein the third value is the sum of the historical consumption parameters of the advertisement after the price adjustment under the price adjustment parameters and the future consumption parameters of the advertisement after the price adjustment under the price adjustment parameters, and the fourth value is the sum of the historical conversion parameters of the user after the price adjustment under the price adjustment parameters and the future conversion parameters of the user after the price adjustment under the price adjustment parameters, and the future consumption parameters include predicted consumption parameters.

[0187] Step S480: construct the third value to be less than or equal to a preset target total consumption parameter as a second constraint condition.

[0188] Step S490: construct the price adjustment parameter greater than 0 as the third constraint condition, and the first constraint condition, the second constraint condition and the third constraint condition are used as the target constraint condition.

[0189] Step S4100: Under the target constraint condition, searching among the plurality of candidate price adjustment parameters for a candidate price adjustment parameter that maximizes the sum of the user's historical conversion parameter after the price adjustment and the user's future conversion parameter after the price adjustment, and using the candidate price adjustment parameter as the target price adjustment parameter.

[0190] In some implementations, firstly, based on the first function and the second function, multiple alternative price adjustment parameters and the future conversion parameters of the user after the price adjustment under the action of the price adjustment parameters corresponding to each of the alternative price adjustment parameters can be determined. Specifically, since the first function is used to characterize the degree of influence of the price adjustment parameters on the rate of change of the cost corresponding to the successfully delivered advertisements, and the second function is used to characterize the relationship between the rate of change of the cost corresponding to the successfully delivered advertisements under the action of the price adjustment parameters and the predicted number of successfully delivered advertisements.

[0191] Exemplarily, different pricing parameters can be set as alternative pricing parameters, and then the consumption parameter change rate corresponding to each alternative pricing parameter can be obtained by bringing the alternative pricing parameters into the first function; and then the consumption parameter change rate corresponding to each alternative pricing parameter can be brought into the second function to obtain the user's future conversion parameter after the price adjustment under the action of the pricing parameter corresponding to each alternative pricing parameter.

[0192] Then, the target constraint condition may be used to search for a candidate price adjustment parameter from the plurality of candidate price adjustment parameters that maximizes the sum of the historical conversion parameter of the user after the price adjustment and the future conversion parameter of the user after the price adjustment, and use the candidate price adjustment parameter as the target price adjustment parameter.

[0193] In some implementations, the ratio of the third value to the fourth value may be first constructed to be equal to a preset target cost as a first constraint condition. The third value is the sum of the historical consumption parameter of the advertisement after the price adjustment under the price adjustment parameter and the future consumption parameter of the advertisement after the price adjustment under the price adjustment parameter, and the fourth value is the sum of the historical conversion parameter of the user after the price adjustment under the price adjustment parameter and the future conversion parameter of the user after the price adjustment under the price adjustment parameter. Similar to the aforementioned embodiment, the future consumption parameter includes a predicted consumption parameter, such as a consumption parameter corresponding to a time period within the day after the current moment, which can be used as a future consumption parameter.

[0194] The third value can be represented by the following formula (15), and the fourth value can be represented by the following formula (16).

[0195] cost h +cost e，λ (15)

[0196] conv h +conv e,λ (16)

[0197] If the pre-set target cost is represented by tcpa, the first constraint can be represented as the following formula (17).

[0198]

[0199] The target cost may be a pre-set cost, such as a cost that an advertiser expects to achieve and is pre-set, and may be used to represent the expected value of the advertiser.

[0200] Furthermore, the third value may be less than or equal to a preset target total consumption parameter as a second constraint. The target total consumption parameter is used to characterize the maximum consumption parameter within a specified time period. For example, the specified time period may be one day, and the target total consumption parameter may be the maximum daily consumption parameter preset by the advertising provider. In some embodiments, the target total consumption parameter may be characterized by a budget, and thus the second constraint may be characterized by the following formula (18).

[0201] cost h +cost e，λ ≤budget (18)

[0202] In addition, the price adjustment parameter is generally a number greater than 0, so the price adjustment parameter greater than 0 can be constructed as the third constraint condition. Specifically, the following formula (19) can be used as the third constraint condition.

[0203] λ>0(19)

[0204] Thus, the first constraint condition, the second constraint condition and the third constraint condition may be used as the target constraint condition, so as to search, based on the target constraint condition, from a plurality of the alternative price adjustment parameters for an alternative price adjustment parameter that maximizes the sum of the historical conversion parameter of the user after the price adjustment and the future conversion parameter of the user after the price adjustment.

[0205] However, under the target constraint, a candidate price adjustment parameter that maximizes the sum of the historical conversion parameter of the user after the price adjustment and the future conversion parameter of the user after the price adjustment can be found among the multiple candidate price adjustment parameters to be used as the target price adjustment parameter. As the goal, formula (17), formula (18) and formula (19) are respectively used as the target constraint conditions to obtain the candidate price adjustment parameter that satisfies the maximum sum of the historical conversion parameters of the user after the price adjustment and the future conversion parameters of the user after the price adjustment. In some implementations, a genetic algorithm can be used to solve the candidate price adjustment parameter that satisfies the maximum sum of the historical conversion parameters of the user after the price adjustment and the future conversion parameters of the user after the price adjustment.

[0206] Step S4110: When it is detected that the price adjustment trigger condition is met, return to execute the first function between the price adjustment parameter and the consumption parameter change rate based on multiple groups of sample data and subsequent steps until the specified time is reached.

[0207] In some embodiments, the price adjustment trigger condition may include a specified time interval, such as 3 minutes, 5 minutes, etc. Therefore, if the advertising provider detects that the time difference between the current moment and the last moment when the price adjustment trigger condition was met reaches the specified time, it can be determined that the price adjustment trigger condition is met. Then, the first function between the price adjustment parameter and the consumption parameter change rate based on multiple groups of sample data and subsequent steps can be returned to execute until the specified time is reached. For some embodiments, the specified time can be 24:00 on the day of the current moment.

[0208] In the embodiment of the present application, in the time period after the current moment, as long as the price adjustment trigger condition is met, the aforementioned steps can be triggered to re-determine the new target price adjustment parameters, so that the sum of the user's post-price adjustment historical conversion parameters and the user's post-price adjustment future conversion parameters is maximized, and the price adjustment range is more accurate. In addition, in the embodiment of the present application, the target price adjustment parameters are determined by constructing the first constraint condition, the second constraint condition, and the third constraint condition. The target price adjustment parameters can be determined multiple times so that when they are less than or equal to the target total consumption parameters, they are close to the pre-set target cost, avoiding large deviations from the target cost.

[0209] The inventors also verified through experiments that the information processing method provided by the embodiment of the present application has a certain improvement in the achievement rate of the pre-set target cost compared with the prior art. Specifically, in the alliance scenario, the target cost achievement rate increased by 0.8%; in the information flow scenario, the target cost achievement rate increased by 0.82%; in the distribution scenario, the target cost achievement rate increased by 0.85%; in the search scenario, the target cost achievement rate increased by 3.92%. It can be seen that it can approach the target cost to a certain extent.

[0210] See also Fig. 9 , Fig. 9 The information processing system 900 includes a target price adjustment parameter acquisition unit 901 , an advertisement ranking unit 902 , a user operation acquisition unit 903 , a first storage unit 904 , a function fitting unit 905 , and a second storage unit 906 .

[0211] The price adjustment parameter acquisition unit 901 is connected to the second storage unit 906 and the advertisement ranking unit 902 respectively, the advertisement ranking unit 902 is also connected to the user operation acquisition unit 903, the user operation acquisition unit 903 is also connected to the first storage unit 904 and the second storage unit 906 respectively, the first storage unit is also connected to the function fitting unit 905, and the function fitting unit 905 is also connected to the second storage unit 906. It can be understood that the target price adjustment parameter acquisition unit 901, the user operation acquisition unit 903, the first storage unit 904, the function fitting unit 905 and the second storage unit 906 can all belong to the advertisement delivery provider, and the advertisement ranking unit 902 can belong to the advertisement server platform.

[0212] Among them, the price adjustment parameter acquisition unit can obtain the fitted first function, second function, third function, etc. from the second storage unit 906, and then determine the target price adjustment parameter in combination with the aforementioned method embodiments. The advertisement sorting unit 902 can obtain the target price adjustment parameter from the price adjustment parameter acquisition unit 901, so as to sort the advertisements, for example, based on the reference consumption parameter after price adjustment. Then, the advertisement is delivered to the terminal device, and the user's operation can be obtained through the user operation acquisition unit 903, and then it is judged whether the advertisement is successfully delivered according to whether the user performs the specified operation.

[0213] In some implementations, the first storage unit 904 may be a Hadoop Distributed File System (HDFS), which can store a large amount of data and support streaming access to data. The second storage unit 906 may be a Remote Dictionary Server (Redis).

[0214] After the user operation acquisition unit 903 acquires the user's operation, historical conversion parameters and historical consumption parameters may be generated based on the user's operation storage, so that the historical conversion parameters and historical consumption parameters may be respectively stored in the first storage unit 904 and the second storage unit 906. Optionally, the first storage unit 904 and the second storage unit 906 may also store parameters such as the historical conversion parameter change rate.

[0215] The function fitting unit 905 can retrieve historical conversion parameters and historical consumption parameters from the first storage unit 904 to perform function fitting, obtain the first function and the second function, and output the first function and the second function to the second storage unit 906 for storage.

[0216] Among them, the specific method steps executed by the above-mentioned units can refer to the introduction in the above-mentioned method embodiment, and will not be repeated here.

[0217] See also Fig.10 , Fig.10 The structural block diagram of an electronic device provided in an embodiment of the present application is shown. The electronic device 1000 may be a desktop computer, a vehicle-mounted computer, a cloud server or a tablet computer, etc. The electronic device 1000 in the present application may include one or more of the following components: a processor 1001, a memory 1002 and one or more application programs, wherein the processor 1001 is electrically connected to the memory 1002, and the one or more program configurations are used to execute the methods described in the aforementioned embodiments.

[0218] The processor 1001 may include one or more processing cores. The processor 1001 uses various interfaces and lines to connect the various parts of the entire electronic device 1000, and executes various functions and processes data of the electronic device 1000 by running or executing instructions, programs, code sets or instruction sets stored in the memory 1002, and calling data stored in the memory 1002. Optionally, the processor 1001 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 1001 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and computer programs; the GPU is responsible for rendering and drawing display content; and the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 1001, but may be implemented separately through a communication chip. Specifically, the method described in the above embodiments may be executed by one or more processors 1001 .

[0219] For some embodiments, the memory 1002 may include a random access memory (RAM) or a read-only memory (ROM). The memory 1002 may be used to store instructions, programs, codes, code sets, or instruction sets. The memory 1002 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for implementing at least one function, instructions for implementing the following various method embodiments, etc. The data storage area may also store data created by the electronic device 1000 during use, etc.

[0220] See also Fig.11 , which shows a structural block diagram 1100 of a computer program product provided by an embodiment of the present application. The computer program product 1100 includes a computer program / instruction 1110, and when the computer program / instruction 1110 is executed by a processor, the steps of the above method are implemented.

[0221] The computer program / instructions 1110 may be stored in a computer-readable storage medium, which may be an electronic memory such as a flash memory, an EEPROM (electrically erasable programmable read-only memory), an EPROM, a hard disk, or a ROM. Optionally, the computer-readable storage medium includes a non-transitory computer-readable storage medium. The computer-readable storage medium has storage space for the computer program / instructions 1110 for executing any of the method steps of the above method. These computer programs / instructions 1110 may be read from or written to one or more computer program products.

[0222] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. An information processing method, characterized in that: include: Based on multiple groups of sample data, a first function between a price adjustment parameter and a consumption parameter change rate is constructed, wherein the sample data includes a historical price adjustment parameter and a historical consumption parameter change rate corresponding to the historical price adjustment parameter, the consumption parameter change rate is used to characterize the ratio of the consumption parameter of the advertisement not under the effect of the price adjustment parameter to the consumption parameter of the advertisement under the effect of the price adjustment parameter, the consumption parameter is used to characterize the cost corresponding to the successfully delivered advertisement, and the first function is used to characterize the degree of influence of the price adjustment parameter on the change rate of the cost corresponding to the successfully delivered advertisement; Acquire the user's conversion parameter under the effect of the price adjustment parameter as the user's post-price adjustment historical conversion parameter, wherein the conversion parameter is used to characterize the number of successfully delivered advertisements; Constructing a second function between the consumption parameter change rate and the user's future conversion parameter after the price adjustment under the price adjustment parameter, wherein the future conversion parameter includes the predicted conversion parameter, and the second function is used to characterize the relationship between the change rate of the cost corresponding to the successfully delivered advertisement under the price adjustment parameter and the predicted number of successfully delivered advertisements; A target price adjustment parameter is determined based on the first function and the second function, wherein the target price adjustment parameter is used to maximize the sum of the user's post-price adjustment historical conversion parameter and the user's post-price adjustment future conversion parameter under a target constraint condition.

2. The method according to claim 1, characterized in that Before constructing the first function between the price adjustment parameter and the consumption parameter change rate based on the multiple groups of sample data, the method further includes: Obtain multiple historical price adjustment parameters corresponding to the advertisement, reference consumption parameters before price adjustment of users who are not affected by the historical price adjustment parameters, and reference consumption parameters after price adjustment of users who are affected by the historical price adjustment parameters; Based on the historical price adjustment parameters of each advertisement, construct a continuous price adjustment interval range divided into multiple segments; Based on the pre-price adjustment reference consumption parameter and the post-price adjustment reference consumption parameter corresponding to the historical price adjustment parameter matched by each price adjustment interval range, a change rate of the historical consumption parameter matched by the price adjustment interval range is obtained; The historical price adjustment parameters and the historical consumption parameter change rates that match the price adjustment intervals that match the historical price adjustment parameters are used as sample data.

3. The method according to claim 2, characterized in that The obtaining of the historical consumption parameter change rate matching the price adjustment interval range based on the pre-price adjustment reference consumption parameter and the post-price adjustment reference consumption parameter corresponding to the historical price adjustment parameter matching the price adjustment interval range includes: Sum the pre-price adjustment reference consumption parameters corresponding to each of the historical price adjustment parameters matching the price adjustment interval range to obtain a first value matching the price adjustment interval range; Sum the price-adjusted reference consumption parameters corresponding to each of the historical price adjustment parameters matching the price adjustment interval to obtain a second value matching the price adjustment interval; The ratio of the second value matched by each price adjustment interval to the first value matched by the price adjustment interval is used as the historical consumption parameter change rate matched by the price adjustment interval.

4. The method according to claim 2, characterized in that: The obtaining of multiple historical price adjustment parameters corresponding to the advertisement, the reference consumption parameters of the user before the price adjustment without the historical price adjustment parameters, and the reference consumption parameters of the user after the price adjustment under the historical price adjustment parameters includes: Obtain multiple historical price adjustment parameters corresponding to the advertisement and the user's price adjustment reference consumption parameters under the effect of the historical price adjustment parameters through the advertisement server platform; The post-price adjustment reference consumption parameters of each advertisement are de-priced using the historical price adjustment parameters of the advertisement to obtain the pre-price adjustment reference consumption parameters of the advertisement that are not affected by the historical price adjustment parameters.

5. The method according to claim 4, characterized in that The method of obtaining multiple historical price adjustment parameters corresponding to the advertisement and the user's price adjustment reference consumption parameters under the effect of the historical price adjustment parameters through the advertisement server platform includes: Obtain multiple advertisements sorted from high to low based on the reference consumption parameters after price adjustment through the advertisement server platform; Select a specified number of advertisements ranked top as candidate advertisements; The historical price adjustment parameters of the candidate advertisement and the reference consumption parameters of the user after the price adjustment under the effect of the historical price adjustment parameters are obtained.

6. The method according to claim 1, characterized in that The second function between the consumption parameter change rate and the user's future conversion parameter after the price adjustment under the price adjustment parameter is constructed, including: Based on the previously acquired historical cost after price adjustment under the price adjustment parameter, predict the future cost after price adjustment under the price adjustment parameter, wherein the future cost includes the predicted cost; Constructing a third function between the future consumption parameter after price adjustment and the change rate of the historical consumption parameter under the price adjustment parameter, wherein the future consumption parameter includes the predicted consumption parameter, and the third function is used to characterize the relationship between the change rate of the cost corresponding to the advertisement successfully delivered under the price adjustment parameter and the predicted cost corresponding to the advertisement successfully delivered under the price adjustment parameter; A function characterizing the future conversion parameter after the price adjustment is constructed based on the future cost after the price adjustment, the future consumption parameter after the price adjustment and the third function as a second function between the consumption parameter change rate and the future conversion parameter after the price adjustment.

7. The method according to claim 6, characterized in that The method of predicting the future cost after price adjustment under the effect of the price adjustment parameter based on the pre-acquired historical cost after price adjustment under the effect of the price adjustment parameter includes: Based on the pre-acquired historical costs after price adjustment under the price adjustment parameters and the cost change rates before and after the price adjustment, determine the historical costs before price adjustment that are not affected by the price adjustment parameters; Predicting the future cost before price adjustment without the price adjustment parameter based on the historical cost before price adjustment; Based on the future cost before the price adjustment and the cost change rate before and after the price adjustment, the future cost after the price adjustment under the effect of the price adjustment parameters is predicted.

8. The method according to claim 6, characterized in that The third function constructed between the future consumption parameter after price adjustment and the change rate of the historical consumption parameter under the action of the price adjustment parameter includes: Get the historical consumption parameters of the advertisement after price adjustment under the price adjustment parameters; Based on the post-price adjustment historical consumption parameter and the historical consumption parameter change rate, determining the pre-price adjustment historical consumption parameter of the advertisement not affected by the price adjustment parameter; Get the average consumption of a specified period of history without the effect of price adjustment parameters; The difference between the average consumption of the historical specified time period and the historical consumption parameter of the advertisement before the price adjustment is used as the future consumption parameter before the price adjustment of the advertisement without the price adjustment parameter; Based on the future consumption parameter before price adjustment and the consumption parameter change rate, a function representing the future consumption parameter after price adjustment of the advertisement under the price adjustment parameter is constructed as a third function between the future consumption parameter after price adjustment and the historical consumption parameter change rate.

9. The method according to claim 1, characterized in that: The determining the target price adjustment parameter based on the first function and the second function includes: Based on the first function and the second function, determining a plurality of candidate price adjustment parameters and a future conversion parameter of the user after the price adjustment under the action of the price adjustment parameter corresponding to each of the candidate price adjustment parameters; Constructing a ratio of the third value to the fourth value equal to a preset target cost as a first constraint condition, wherein the third value is the sum of the price-adjusted historical consumption parameter of the advertisement under the price adjustment parameter and the price-adjusted future consumption parameter of the advertisement under the price adjustment parameter, the fourth value is the sum of the price-adjusted historical conversion parameter of the user under the price adjustment parameter and the price-adjusted future conversion parameter of the user under the price adjustment parameter, and the future consumption parameter includes the predicted consumption parameter; Constructing the third value to be less than or equal to a preset target total consumption parameter as a second constraint condition; Constructing the price adjustment parameter to be greater than 0 as the third constraint condition, and the first constraint condition, the second constraint condition and the third constraint condition as the target constraint condition; Under the target constraint condition, a candidate price adjustment parameter that maximizes the sum of the historical conversion parameter of the user after the price adjustment and the future conversion parameter of the user after the price adjustment is searched among the multiple candidate price adjustment parameters as the target price adjustment parameter.

10. The method according to claim 1, characterized in that The step of constructing a first function between the price adjustment parameter and the consumption parameter change rate based on the multiple groups of sample data includes: By presetting the function type, the alternative function is constructed with the price adjustment parameter as the independent variable and the consumption parameter change rate as the dependent variable; Determine the parameters in the candidate function by using the historical price adjustment parameters in each set of sample data and the historical consumption parameter change rate corresponding to the historical price adjustment parameters; A first function between the price adjustment parameter and the consumption parameter change rate is generated based on the parameter and the candidate function.

11. The method according to claim 1, characterized in that: After determining the target price adjustment parameter based on the first function and the second function, the method further includes: When it is detected that the price adjustment trigger condition is met, the execution returns to the first function between the price adjustment parameter and the consumption parameter change rate based on multiple groups of sample data and subsequent steps until the specified time is reached.

12. The method according to claim 11, characterized in that The price adjustment triggering condition includes a specified interval duration.

13. The method according to claim 1, characterized in that The successful delivery includes the user performing a specified operation on the advertisement delivered by the advertisement delivery provider.

14. An electronic device, characterized in that: include: one or more processors; Memory; One or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs are configured to execute the method according to any one of claims 1-13.

15. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the method according to any one of claims 1 to 13 is implemented.

Images