An advertisement delivery method and apparatus

Abstract

The application discloses an advertisement putting method and device, which comprises the following steps: predicting the first resource consumed by putting target advertisement in each time slice of a specified period; predicting the second resource consumed by the delay corresponding to each time slice in the process of putting advertisement; when putting target advertisement, determining whether the first sum of the total resource consumed at present and the second resource corresponding to the current time slice is greater than or equal to the total resource of putting target advertisement in the specified period; if yes, stopping putting target advertisement; if no, determining whether to execute the specified operation for accelerating resource consumption according to the first sum and the first resource of each time slice, and dynamically adjusting the first resource of the remaining time slice according to the total resource consumed. The embodiment of the application can improve the accuracy of the prediction result of advertisement resource, thereby avoiding over-putting, and ensuring that the advertisement resource can be consumed smoothly in the whole process of putting advertisement.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of Internet, and particularly relates to an advertisement putting method and device. BACKGROUND

[0002] When an advertisement putting system puts an advertisement, it usually needs to control the advertisement resources consumed by the advertisement to consume the advertisement resources smoothly. Currently, when the advertisement resources are controlled, the putting data of the day can be predicted according to historical putting data, and when the actually consumed resources are greater than the predicted resources, the putting speed can be slowed down, and vice versa.

[0003] However, in the above control strategy, when the putting data of the day is predicted according to the historical putting data, the accuracy of the prediction result is usually low, resulting in poor control effect when the advertisement resources are controlled according to the prediction result, and the advertisement resources cannot be consumed smoothly. SUMMARY

[0004] Embodiments of the present application provide an advertisement putting method and device, which are used to solve the problem that the current advertisement resource control strategy has poor control effect and cannot consume advertisement resources smoothly.

[0005] To solve the above technical problem, embodiments of the present application are implemented as follows:

[0006] In a first aspect, an advertisement putting method is provided, comprising:

[0007] predicting a first resource consumed by putting a target advertisement in each time slice in a specified period;

[0008] predicting a second resource consumed by a delay corresponding to each time slice in the process of putting the advertisement;

[0009] when putting the target advertisement, determining whether a first sum of a total resource currently consumed and the second resource corresponding to the current time slice is greater than or equal to a total resource of the target advertisement put in the specified period in each time slice;

[0010] if yes, stopping putting the target advertisement;

[0011] if no, determining whether to perform a specified operation according to the first sum and the first resource of each time slice, and dynamically adjusting the first resource of the remaining time slice according to the total resource consumed, the specified operation being used to speed up resource consumption.

[0012] In a second aspect, an advertisement putting device is provided, comprising:

[0013] a first prediction module configured to predict a first resource consumed by launching a target advertisement in each time slice of a specified time period;

[0014] a second prediction module configured to predict a second resource consumed by a time delay corresponding to each time slice in an advertisement launching process;

[0015] a determination module configured to determine, when launching the target advertisement, whether a first sum of a total resource already consumed in each time slice and the second resource corresponding to the current time slice is greater than or equal to a total resource for launching the target advertisement in the specified time period;

[0016] a first advertisement launching module configured to stop launching the target advertisement if the determination module determines that the first sum is greater than or equal to the total resource;

[0017] a second advertisement launching module configured to determine, if the determination module determines that the first sum is less than the total resource, whether to perform a specified operation for accelerating resource consumption according to the first sum and the first resource of each time slice, and to dynamically adjust the first resource of the remaining time slices according to the total resource already consumed;

[0018] In a third aspect, an electronic device is provided, the electronic device comprising:

[0019] a processor; and

[0020] a memory arranged to store computer-executable instructions that, when executed, cause the processor to perform the following operations:

[0021] predict a first resource consumed by launching a target advertisement in each time slice of a specified time period;

[0022] predict a second resource consumed by a time delay corresponding to each time slice in an advertisement launching process;

[0023] determine, when launching the target advertisement, whether a first sum of a total resource already consumed in each time slice and the second resource corresponding to the current time slice is greater than or equal to a total resource for launching the target advertisement in the specified time period;

[0024] if the first sum is greater than or equal to the total resource, stop launching the target advertisement;

[0025] if the first sum is less than the total resource, determine whether to perform a specified operation for accelerating resource consumption according to the first sum and the first resource of each time slice, and dynamically adjust the first resource of the remaining time slices according to the total resource already consumed.

[0026] In a fourth aspect, a computer-readable storage medium is provided, which stores one or more programs, and when the one or more programs are executed by an electronic device comprising a plurality of application programs, the electronic device is caused to perform the following method:

[0027] predicting a first resource consumed by delivering a target advertisement in each time slice of a specified period;

[0028] predicting a second resource consumed by a delay corresponding to each time slice in the process of delivering the advertisement;

[0029] when delivering the target advertisement, determining whether a first sum of a total resource currently consumed and the second resource corresponding to the current time slice is greater than or equal to a total resource for delivering the target advertisement in the specified period in each time slice;

[0030] if yes, stopping delivering the target advertisement;

[0031] if no, determining whether to perform a specified operation for accelerating resource consumption according to the first sum and the first resource of each time slice, and dynamically adjusting the first resource of the remaining time slice according to the total resource consumed.

[0032] The above at least one technical solution adopted by the embodiments of the present application can achieve the following beneficial effects:

[0033] When predicting the resource consumed by the target advertisement in each time slice, the embodiments of the present application can dynamically adjust the predicted resource in each remaining time slice in combination with the total resource consumed in the actual process of delivering the advertisement after predicting the resource consumed in each time slice, so as to improve the accuracy of the prediction result. In the process of delivering the advertisement, the resource consumed by the delay can be considered, and it can be determined whether to stop delivering according to the sum of the resource consumed by the delay and the total resource consumed in each time slice, so as to effectively control whether to over-deliver in each time slice, thereby avoiding the problem of over-delivery and the problem of resource remaining caused by stopping delivering too early. In the case of not over-delivering, it can be determined whether to perform a specified operation for accelerating resource consumption according to the total resource consumed and the predicted resource in each time slice, so as to make the consumption of the resource more in line with expectations. The embodiments of the present application dynamically adjust the prediction result of the advertisement resource, and determine whether to over-deliver and whether to perform a specified operation for accelerating resource consumption in each time slice, so as to ensure that the advertisement resource can be smoothly consumed in the whole process of delivering the advertisement. BRIEF DESCRIPTION OF DRAWINGS

[0034] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required by the embodiments or prior art description. Obviously, the drawings described below only show some embodiments of the present application, and the drawings can be obtained by those skilled in the art without any creative effort.

[0035] Figure 1 is a flowchart of an embodiment of the advertisement delivery method of the present application;

[0036] Figure 2 is a flowchart of an embodiment of the advertisement delivery method of the present application;

[0037] Figure 3 is a structural diagram of an embodiment of the electronic device of the present application;

[0038] Figure 4 is a structural diagram of an embodiment of the advertisement delivery device of the present application. DETAILED DESCRIPTION

[0039] In order to make the technical personnel in the art better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without any creative effort should be within the scope of protection of the present application.

[0040] At present, when the advertisement delivery system controls the advertisement resources, the core idea of the control strategy is: predicting the delivery data of the day according to the historical delivery data, and taking the predicted data as the basis to judge whether the actual consumed resources are greater than the predicted resources in the advertisement delivery process. If yes, it means that the resource consumption is too fast, and the delivery speed needs to be slowed down. If no, it means that the resource consumption is slow, and the delivery speed can be accelerated, so as to realize the smooth consumption of the advertisement resources.

[0041] However, when predicting the delivery data of the day according to the historical delivery data, the accuracy of the prediction result is usually low, especially for the advertisement delivery system with unstable flow (specifically, the flow stability of the advertisement delivery system can reflect the stability of the advertisement exposure to a certain extent. If the flow is unstable, it means that the exposure of the advertisement is unstable. In the case of unstable exposure, the accuracy of the predicted delivery data is low). The predicted delivery data obtained according to the historical delivery data is quite different from the actual delivery data. In this way, when the advertisement resources are controlled according to the prediction result, the control effect is poor, which leads to the inability to smoothly consume the advertisement resources.

[0042] To solve the above technical problems, the embodiment of the present application provides an advertisement putting method and device, which is suitable for not only the advertisement putting system with poor traffic stability, but also the advertisement putting system with good traffic stability. The advertisement putting method comprises: predicting a first resource consumed by putting a target advertisement in each time slice in a specified period; predicting a second resource consumed by a delay corresponding to each time slice in the process of putting the advertisement; when putting the target advertisement, determining whether a first sum of a total resource currently consumed and the second resource corresponding to the current time slice is greater than a total resource of the target advertisement put in the specified period; if yes, stopping putting the target advertisement; if no, determining whether to perform a specified operation for accelerating resource consumption according to the first sum and the first resource of each time slice, and dynamically adjusting the first resource of the remaining time slice according to the total resource consumed.

[0043] The technical scheme provided by the embodiment of the present application can achieve at least the following technical effects:

[0044] (1) When predicting the resource consumed by the target advertisement in each time slice, since the predicted resource in each time slice can be dynamically adjusted according to the total resource consumed in the actual process of putting the advertisement, the accuracy of the prediction result can be improved;

[0045] (2) In the process of putting the advertisement, since the resource consumed by the delay is considered, and whether to stop putting is determined according to the sum of the resource consumed by the delay and the total resource consumed in each time slice, the over-putting in each time slice can be effectively controlled, so as to avoid the problem of over-putting, and the problem of resource remaining caused by stopping putting too early can also be avoided;

[0046] (3) In the case of not over-putting, since whether to perform the specified operation for accelerating resource consumption is determined according to the total resource consumed and the predicted resource in each time slice, the resource consumption can be appropriately accelerated in the case of not over-putting, so that the resource consumption is more in line with the expectation. Alternatively, the specified operation can also be an advertisement bidding operation, and in the case that the specified operation is the advertisement bidding operation, the return on investment of the advertiser can also be improved.

[0047] The embodiment of the present application can smoothly consume the advertisement resource in the whole process of putting the advertisement by dynamically adjusting the prediction result of the advertisement resource, and determining whether to over-put and whether to perform the specified operation for accelerating resource consumption in each time slice.

[0048] It should be noted that the advertising resource in the embodiments of the present application can be an advertising budget provided by an advertiser, or an advertising flow provided by an advertising delivery system. The technical solutions provided by the embodiments of the present application can smoothly consume the advertising budget provided by the advertiser in the whole process of advertising delivery, or can smoothly consume the advertising flow provided by the advertising delivery system.

[0049] The technical solutions provided by the embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0050] Figure 1 is a flowchart of an advertising delivery method according to an embodiment of the present application. The method is described as follows.

[0051] S102: predicting a first resource consumed by delivering a target advertisement in each time slice of a specified period.

[0052] The advertising delivery system can predict a resource possibly consumed by delivering the target advertisement in each time slice of a specified period before delivering the target advertisement. Here, in order to facilitate distinction, the resource corresponding to each time slice obtained by prediction can be represented by a first resource.

[0053] The target advertisement can be an advertisement of a certain type or a certain class that the advertiser wants to deliver. For the target advertisement, the advertiser can set a corresponding advertising budget in the case of an advertising resource being an advertising budget, or the advertising delivery system can set a corresponding advertising flow in the case of an advertising resource being a flow. The embodiments of the present application control the advertising budget / advertising flow to achieve smooth consumption of the advertising budget / advertising flow in the whole process of advertising delivery.

[0054] The specified period can be a unit period corresponding to the advertising resource. For example, if the advertising resource is in units of a day (such as a daily budget or a daily flow), the specified period can be one day, and if the advertising resource is in units of two days, the specified period can be two days. For ease of understanding, the embodiments of the present application can be described by taking one day as the specified period.

[0055] The specified period can include a plurality of time slices, and the time length corresponding to each of the plurality of time slices can be the same or different. The embodiments of the present application do not limit the number of time slices in the specified period and the time length of each time slice. Preferably, in order to better control the advertising resource, the time length corresponding to each of the plurality of time slices can be the same, and one time slice can correspond to one hour. That is, if the specified time is one day, the specified period can be divided into 24 time slices according to 24 hours of one day, and each time slice corresponds to one hour of one day.

[0056] S104: predicting a second resource consumed by a delay corresponding to each time slice in the process of advertising delivery.

[0057] Generally, after the advertisement delivery system delivers an advertisement, the advertisement may not be displayed in time due to various reasons, but is displayed after a delay, and the resources consumed during the delay are the resources consumed by the delay. For example, in a possible scenario, after the advertisement delivery system delivers an advertisement to a client, the client does not display the advertisement in time due to network reasons, but displays the advertisement after a period of time, and the resources consumed during the display of the advertisement are the resources consumed by the delay. In the embodiments of the present application, each time slice can have corresponding delay consumption, and the resources consumed by the delay of each time slice in the advertisement delivery process will be included in the actual resources consumed by the next time slice.

[0058] In the embodiments of the present application, before delivering the target advertisement, the advertisement delivery system can predict the resources consumed by the delay corresponding to each time slice of the specified period during the delivery of the target target advertisement. Here, in order to facilitate distinction, the resources consumed by the delay can be represented as second resources, and the second resources consumed by the delay can be used for subsequent judgment of whether over-delivery occurs.

[0059] S106: When delivering the target advertisement, it is determined whether the first sum of the total resources consumed at present and the second resources corresponding to the current time slice is greater than or equal to the total resources of the target advertisement delivered in the specified period.

[0060] In S106, the judgment of whether over-delivery occurs can be performed in each time slice. When judging whether over-delivery occurs, in order to accurately judge whether over-delivery occurs, the second resources consumed by the delay of the current time slice can be considered, and in order to avoid the problem of resource remaining when over-delivery is determined and stopped, the total resources of the target advertisement delivered in the specified period can be used as a basis for judgment. The total resources of the target advertisement delivered in the specified period are resources expected to be consumed smoothly. Specifically, in the case of advertisement resources being an advertisement budget, the total resources of the target advertisement delivered in the specified period can be the total budget provided by the advertiser for delivering the target advertisement in the specified period, and in the case of advertisement resources being an advertisement flow, the total resources of the target advertisement delivered in the specified period can be the total flow provided by the advertisement delivery system for delivering the target advertisement in the specified period.

[0061] In the judgment of whether to overbid, specifically, the sum (i.e., the first sum) between the total resources that have been consumed and the second resources consumed by the delay corresponding to the current time slice can be judged whether it is greater than or equal to the total resources of the target advertisement in the specified period. The total resources consumed are the total resources that have actually been accumulated and consumed up to the current time slice, which can be equal to the sum of the resources consumed in the current time slice and the resources consumed in other time slices before the current time slice. The second resources consumed by the delay corresponding to the current time slice are obtained by S104 prediction. The total resources of the target advertisement in the specified period can be the total resources set for the target advertisement in the specified period, such as if the specified period is one day, the total resources can be the daily budget provided by the advertiser for the target advertisement, or the daily flow provided by the advertisement delivery system for the target advertisement.

[0062] Since the judgment of whether to overbid can be made in each time slice, the resource overbid can be effectively controlled in each time slice, thereby avoiding the problem of overbid, and also avoiding the problem of resource remaining caused by early stop of delivery.

[0063] It should be noted that in the judgment of whether to overbid, only the second resources consumed by the delay of the current time slice are considered, and the second resources consumed by the delay of other time slices before the current time slice are not considered, because the second resources consumed by the delay of other time slices have been calculated into the next time slice of other time slices, i.e., have been calculated into the total resources accumulated and consumed up to the current time slice, and only the second resources consumed by the delay of the current time slice are not calculated, therefore, only the second resources consumed by the delay of the current time slice need to be considered.

[0064] In each time slice, after the judgment of whether to overbid, if the first sum of the total resources that have been consumed and the second resources consumed by the delay corresponding to the current time slice is not greater than the total resources of the target advertisement in the specified period, it can be determined that there is no overbid, and S110 can be executed; otherwise, if the first sum of the total resources that have been consumed and the second resources consumed by the delay corresponding to the current time slice is greater than or equal to the total resources of the target advertisement in the specified period, it can be determined that overbid has occurred or will occur, and S108 can be executed.

[0065] For ease of understanding, a day can be taken as a specified period, and a day includes 24 time slices, each of which corresponds to one hour. In the first time slice (i.e., between 0-1 o'clock), it can be determined whether the sum of the first resource consumed in the first time slice and the second resource consumed in the first time slice delay is greater than or equal to the total resource (hereinafter referred to as daily resource) of the target advertisement launched in a day. In the second time slice (i.e., between 1-2 o'clock), it can be determined whether the sum of the total resource accumulated consumed up to the second time slice (i.e., the sum of the first resource consumed in the first time slice and the second resource consumed in the current second time slice) and the second resource consumed in the second time slice delay is greater than the daily resource. In the third time slice (i.e., between 2-3 o'clock), it can be determined whether the sum of the resource accumulated consumed up to the third time slice (i.e., the sum of the first resource consumed in the first time slice, the second resource consumed in the second time slice, and the third resource consumed in the current third time slice) and the second resource consumed in the third time slice delay is greater than the daily resource. In this way, the determination of whether the target advertisement is over-launched can be performed in each time slice. In each time slice, if the determination result is that the target advertisement is not over-launched, S110 can be performed, and the determination of whether the target advertisement is over-launched can be continued in the next time slice. If the determination result is that the target advertisement is over-launched, S108 can be performed, and the determination of whether the target advertisement is over-launched can be stopped.

[0066] S108: Stop launching the target advertisement.

[0067] Since the launching of the target advertisement can be stopped in a timely manner when it is determined that the target advertisement is over-launched or is about to be over-launched, the problem of over-launching of the target advertisement can be avoided, and the over-launching of the target advertisement can be effectively controlled.

[0068] It should be noted that when the launching of the target advertisement is stopped in the case where it is determined that the target advertisement is over-launched, the problem of over-launching of the target advertisement has actually occurred. However, since the determination of whether the target advertisement is over-launched is performed in each time slice, and the resource consumed in the delay is considered, even when the launching of the target advertisement is stopped, the consumption of the target advertisement is generally within an acceptable range and meets expectations. Compared with the current solution, the problem of over-launching of the target advertisement can still be avoided, and the over-launching of the target advertisement can be effectively controlled. Of course, in some possible scenarios, in order to strictly control the target advertisement and prevent the problem of over-launching of the target advertisement, when the determination of whether the target advertisement is over-launched is performed in S106, it can also be determined whether the first sum of the total resource currently consumed and the second resource consumed in the delay corresponding to the current time slice is greater than or equal to a preset ratio of the total resource of the target advertisement launched in the specified period. The preset ratio can be a value close to 1, for example, 98%, 99%, or the like. In this way, the launching of the target advertisement can be stopped when the target advertisement is about to be over-launched, so that the problem of over-launching of the target advertisement can be strictly controlled.

[0069] S110: determining whether to perform a specified operation for accelerating resource consumption according to the first sum and the first resource of each time slice, and dynamically adjusting the first resource of the remaining time slice according to the total consumed resource.

[0070] In S110, when the result of the judgment is that the target advertisement is not overcast, the target advertisement can be continued to be cast. In the case of continuing to cast the target advertisement, it can be further judged whether to perform a specified operation for accelerating resource consumption based on the first sum of the total resource consumed in S106 and the delay consumption second resource corresponding to the current time slice and the first resource corresponding to each time slice predicted in S102, so as to make the resource consumption more in line with expectations. The specified operation can be to accelerate the speed of casting the target advertisement. In a possible implementation, the specified operation can also be an advertisement bidding operation. The advertisement bidding can accelerate the resource consumption, specifically, a plurality of advertisements obtain the opportunity to display through bidding, and the higher the price of the advertisement, the greater the probability of obtaining the display opportunity, and usually the advertisement with the highest price can obtain the display opportunity. The embodiments of the present application can maximize the return on investment of the advertiser by judging whether to perform the advertisement bidding operation in the case of not overcasting.

[0071] In addition, in the case of determining that the target advertisement is not overcast, the predicted first resource in the remaining time slice can also be dynamically adjusted according to the total consumed resource, that is, after predicting the first resource in each time slice, the predicted first resource in each subsequent time slice can also be dynamically adjusted in combination with the total consumed resource in the actual process of casting the advertisement, so that the adjusted first resource can be more in line with the actual resource consumption, thereby improving the accuracy of the prediction result.

[0072] In the embodiments of the present application, after dynamically adjusting the first resource of the remaining time slice, when determining whether to perform a specified operation in each of the remaining time slices, the adjusted first resource can be used for judgment, so that the judgment result is more in line with the actual consumption of the advertisement resource.

[0073] For example, if the current time slice is the first time slice in the specified period, the first resource of the second and subsequent time slices can be adjusted according to the actual resource consumption of the first time slice. When the current time slice is the second time slice in the specified period, if the investment is not over, whether to perform the specified operation can be determined according to the first resource of the first time slice and the adjusted first resource of the second time slice, and the first resource of the third and subsequent time slices can also be adjusted according to the cumulative resource consumption up to the second time slice. When the current time slice is the third time slice in the specified period, if the investment is not over, whether to perform the specified operation can be determined according to the first resource of the first time slice, the adjusted first resource of the second time slice, and the first resource of the third time slice obtained after the adjustment of the first resource of the third time slice when the current time slice is the second time slice, and the first resource of the fourth and subsequent time slices can also be adjusted according to the cumulative resource consumption up to the third time slice. And so on. From the whole process of adjusting the first resource, the more times the first resource of the time slice is adjusted, the more times the first resource of the time slice is adjusted, and the last adjusted first resource can be used when using the first resource of the time slice.

[0074] Based on the above-described S102 to S110, it can be seen that the embodiments of the present application can dynamically adjust the prediction results of the advertising resources, and determine whether to over-invest and whether to perform the specified operation for accelerating resource consumption in each time slice, so as to ensure that the advertising resources can be smoothly consumed in the whole process of advertising investment. In addition, in the case where the specified operation is an advertising bidding operation, the return on investment of the advertiser can also be improved.

[0075] In an implementation manner, when predicting the first resource corresponding to each time slice in the specified period in S102, the prediction can be performed according to the historical investment data and the total resource of the target advertisement in the specified period, which can specifically include the following steps:

[0076] S1021: Obtain the historical investment data of the first advertisement from the advertising investment system.

[0077] The advertising delivery system stores historical delivery data of a plurality of advertisements. The embodiment of the present application can obtain the historical delivery data of the first advertisement from the historical delivery data. The first advertisement refers to an advertisement with a similarity between the target advertisement and the audience being greater than or equal to a preset threshold (which can be set according to actual conditions). That is, the first advertisement is an advertisement with high similarity to the target advertisement audience. For example, the target advertisement audience (i.e., the user portrait targeted by the target advertisement) is male and aged between 25 and 30 years old. Then, the historical delivery data similar to the audience can be extracted from the advertising delivery system. In this way, by obtaining the historical delivery data of the first advertisement similar to the target advertisement audience, the first resource predicted based on the historical delivery data can be more reasonable and more consistent with the actual exposure trend of the target advertisement, thereby improving the accuracy of the prediction result to a certain extent.

[0078] The historical delivery data of the first advertisement can be historical delivery data in the past six months, one year, or other time periods. The historical delivery data can specifically include the total exposure of the first advertisement in the historical specified period and the exposure of the first advertisement in each time slice in the historical specified period. The historical specified period can be the same as the specified period in which the target advertisement is delivered, and the division manner of each time slice in the historical specified period can be the same as the division manner of the time slice in the specified period. Preferably, the historical specified period can be one day, and each time slice in the historical specified period can be each hour. The historical delivery data of the first advertisement obtained by the embodiment of the present application can be the total exposure of the first advertisement in each day in the recent period and the exposure in each hour in each day. The exposure can be the number of displays of the advertisement.

[0079] S1022: predicting the first resource consumed for delivering the target advertisement in each time slice in the specified period according to the historical delivery data of the first advertisement and the total resource for delivering the target advertisement in the specified period.

[0080] In an implementation manner, the average total exposure (which can also be the median or root mean square value, etc.) of the first advertisement in each historical specified period can be determined according to the historical delivery data of the first advertisement, and then the average exposure (which can also be the median or root mean square value, etc.) in each time slice in the historical specified period is further determined. Finally, the first resource is predicted according to the average total exposure value of the historical specified period, the average exposure in each time slice, and the total resource of the target advertisement in the specified period, wherein the ratio of the average exposure of the first advertisement in each time slice to the average total exposure of the first advertisement in the historical specified period is equal to the ratio of the first resource of the target advertisement in each time slice to the total resource of the target advertisement in the specified period.

[0081] For example, assuming that the historical specified period and the specified period are both one day, each time slice is one hour, the advertising resource is an advertising budget, the average total exposure of the first advertisement in the historical period is 1000, the average exposure per hour in a day is 27, 25, 22, 23, 25, 28, 46, 51, 53, 54, 51, 40, 37, 35, 37, 40, 43, 46, 58, 68, 63, 59, 37, 32, the daily budget of the target advertisement is 1000 yuan, then the first resource of the target advertisement in each hour of the day (i.e. the first resource of the target advertisement in each time slice of the specified period) can be shown in Table 1:

[0082] Table 1

[0083]

[0084] In Table 1, 00 represents the first time slice, corresponding to the time period between 0 o'clock and 1 o'clock in the morning, 01 represents the second time slice, corresponding to the time period between 1 o'clock and 2 o'clock in the morning, …, 23 represents the twenty-fourth time slice, corresponding to the time period between 23 o'clock and 24 o'clock in the evening. As can be seen from Table 1, the ratio between the budget of the target advertisement in each hour (i.e. the first resource) and the daily budget (i.e. the total resource provided for the target advertisement in a day) is equal to the ratio of the average exposure of the first advertisement in each hour to the average total exposure of the first advertisement in a day.

[0085] Alternatively, considering that in special scenarios, the first resource in some time slices is higher than the first resource in normal situations, for example, during the promotion period, the first resource of the time slice corresponding to the promotion period (such as between 0 o'clock and 1 o'clock in the morning) is much higher than the first resource in normal situations, therefore, when predicting the first resource in each time slice according to the historical advertising data of the first advertisement and the total resource of the target advertisement in the specified period, the special scenarios can also be considered. Specifically, the weight of each time slice can be pre-set according to the actual application scenario, and when predicting the first resource of each time slice, the pre-set weight can be obtained, and then the first resource in each time slice can be predicted according to the obtained weight, the historical advertising data of the first advertisement and the total resource of the target advertisement in the specified period. The weight of each time slice can be set according to the actual application scenario, generally, in the promotion scenario, the weight of the time slice corresponding to the promotion period can be greater than the weight of the time slice corresponding to the non-promotion period.

[0086] It should be noted that the prediction result of the first resource shown in Table 1 above can be regarded as a prediction result obtained under the condition that the weights of each time slice are the same. That is, in a normal case, the weight of each time slice can be regarded as equal and 1, and in a special scenario, the weight of each time slice can be flexibly set according to the actual scenario, so that the first resource finally predicted is more in line with the actual application scenario, thereby improving the accuracy of the prediction result.

[0087] In an implementation manner, when predicting the second resource consumed by the delay corresponding to each time slice, taking one time slice in the plurality of time slices as an example, the following steps can be specifically included:

[0088] S1041: determining a delay proportion corresponding to the time slice according to historical delivery data of the advertisement delivery system and / or actual delivery data of the target advertisement.

[0089] The historical delivery data here can be the historical delivery data of the first advertisement, or optionally, the historical delivery data of other advertisements in the advertisement delivery system. The actual delivery data of the target advertisement can be delivery data generated in the process of actually delivering the target advertisement. The delay proportion can be a ratio between the resource consumed by the delay corresponding to a single time slice and the resource actually consumed in the time slice. The embodiment of the application can determine the delay proportion corresponding to the time slice according to at least one of the historical delivery data and the actual delivery data of the target advertisement.

[0090] Optionally, when predicting the second resource consumed by the delay of the first time slice in the specified period, considering that there is not enough actual delivery data to determine the delay proportion, the delay proportion can be determined preferentially according to the historical delivery data. When determining the second resource consumed by the delay of other time slices, considering that there is enough actual delivery data to determine the delay proportion, the delay proportion can be determined preferentially according to the actual delivery data, so that the determined delay proportion is more in line with the actual delivery situation of the target advertisement, thereby improving the accuracy of the determined delay proportion.

[0091] S1042: determining the second resource consumed by the delay corresponding to the time slice according to the delay proportion and the resource actually consumed in the time slice.

[0092] The resource actually consumed in the time slice can be determined by the actual exposure in the time slice and the charging mode of the target advertisement. Specifically, the resource actually consumed in the time slice is equal to the product of the actual exposure in the time slice and the charging mode of the target advertisement. The charging mode of the target advertisement can be CPM, or CPC or CPA, etc., which can be determined according to the actual scenario.

[0093] In the second resource corresponding to the time slice is determined according to the delay ratio and the resource actually consumed in the time slice, the product of the delay ratio and the resource actually consumed in the time slice can be determined as the second resource consumed by the delay corresponding to the time slice. For example, if the delay ratio of a certain time slice is 10%, and the resource actually consumed in the time slice is the advertising budget and is specifically 50 yuan, then the second resource consumed by the delay corresponding to the time slice is 5 yuan.

[0094] Based on the content described in S1041 and S1042, the second resource consumed by the delay corresponding to a single time slice can be determined by the following formula:

[0095] delayConsume = delayRate * preTimeSliceImp * chargingMode.

[0096] Wherein, delayConsume is the second resource consumed by the delay corresponding to a single time slice, delayRate is the delay ratio, preTimeSliceImp is the actual exposure in a single time slice, and chargingMode can be the charging mode set for the target advertisement, which can be CPM, CPC or CPA.

[0097] In an implementation mode, the determination of whether to perform the specified operation according to the first sum and the first resource of each time slice in S110 can specifically include the following steps:

[0098] S1101: Determine the second sum of the first resource from the first time slice to the current time slice according to the first resource of each time slice.

[0099] Specifically, if the current time slice is the first time slice in the specified period, the first resource of each time slice can be the first resource predicted in S102. If the current time slice is not the first time slice in the specified period, the first resource of each time slice can be the first resource dynamically adjusted in the above S110.

[0100] For example, if the current time slice is the first time slice in the specified period, the second sum value is specifically the first resource of the current time slice, which can be the first resource predicted in S102. If the current time slice is the second time slice in the specified period, the second sum value is specifically the sum of the first resource of the first time slice and the first resource of the second time slice, the first resource of the first time slice can be the first resource predicted in S102, and the first resource of the second time slice can be the first resource obtained by adjusting the first resource of the second time slice when the first time slice is taken as the current time slice in S110. If the current time slice is the third time slice in the specified period, the second sum value is specifically the sum of the first resource of the first time slice, the first resource of the second time slice, and the first resource of the third time slice, the first resource of the first time slice can be the first resource predicted in S102, the first resource of the second time slice can be the first resource obtained by adjusting the first resource of the second time slice when the first time slice is taken as the current time slice in S110, and the first resource of the third time slice can be the first resource obtained by adjusting the first resource of the third time slice when the second time slice is taken as the current time slice in S110. In this way, when the current time slice is any time slice in the specified period, the second sum value of the first resource from the first time slice to the current time slice can be determined.

[0101] S1102: Determine whether the ratio of the first sum value to the second sum value is less than or equal to a preset threshold value.

[0102] The first sum value is the sum of the total resource accumulated until the current time slice and the second resource consumed by the current time slice, and the second sum value is the second sum value of the first resource from the first time slice to the current time slice. The preset threshold value can be a natural number greater than 0 and less than 1, and optionally, the preset threshold value can be 80%. In the embodiments of the present application, it is determined whether the ratio of the first sum value to the second sum value is less than or equal to the preset threshold value, that is, it is determined whether the total resource actually consumed exceeds a certain proportion of the total resource predicted. If yes, it can be indicated that the difference between the total resource actually consumed and the total resource predicted is large, and the resource consumption can be accelerated and S1103 is executed; if not, it can be indicated that the difference between the total resource actually consumed and the total resource predicted is small, and S1104 can be executed.

[0103] S1103: Perform a specified operation.

[0104] The specified operation can be to accelerate the delivery speed of the target advertisement. Alternatively, the specified operation can also be an advertisement bidding operation. The execution of the specified operation by the embodiments of the present application can accelerate the consumption of resources, so that the consumption of resources is more in line with expectations. In addition, in the case where the specified operation is an advertisement bidding operation, the return on investment of the advertiser can also be improved.

[0105] S1104: Determine whether to perform the specified operation according to the ratio of the first sum and the second sum and a preset index.

[0106] In the case where the total consumed resources and the total predicted resources are not much different, in order to further determine whether to perform the specified operation, that is, whether to accelerate the consumption of resources, the preset index can be combined to make a comprehensive judgment. The preset index can represent the value of the advertisement traffic, which can be the request concurrency of the advertisement. The greater the request concurrency, the greater the traffic, and the greater the traffic, the more the exposure times of the advertisement, and accordingly, the consumption of resources will be accelerated, and the advertiser will also bring greater value. That is, the preset index can represent the value that the delivery of the target advertisement in the current time slice can bring to the advertiser, which is usually closely related to the return on investment of the advertiser. The embodiments of the present application can more scientifically consume advertisement resources by comprehensively judging whether to perform the specified operation through the preset index, and in the case where the specified operation is an advertisement bidding operation, the return on investment of the advertiser can also be maximized by comprehensively judging whether to perform the advertisement bidding operation through the preset index.

[0107] Alternatively, when the target of the advertisement is the user clicking the advertisement, the above-mentioned preset index can be the click rate. The click rate can be the ratio between the number of times of clicking the advertisement by the user and the number of times of displaying the advertisement, for example, 100 times of displaying the advertisement, and 10 times of clicking by the user, then the click rate is 10%. Alternatively, when the target of the advertisement is that the user needs to further perform other operations (such as installation, activation, etc.) after clicking the advertisement, the above-mentioned preset index can also be the product of the click rate and the conversion rate. The conversion rate can be the ratio between the number of times of performing other operations (such as the number of times of installing or activating) by the user after clicking the advertisement and the number of times of clicking by the user, for example, 10 times of clicking the advertisement by the user, and 1 time of installing or activating the application in the advertisement, then the conversion rate is 10%. Alternatively, the above-mentioned preset index can also be other indexes, which can be set according to the actual application scenario, which is not limited here.

[0108] In the case of determining whether to perform the specified operation according to the ratio of the first sum value and the second sum value and the preset index, specifically, the value of the preset index can be predicted according to a predetermined model. The model can be a model in the prior art for predicting click rate and conversion rate according to delivery data. After obtaining the value of the preset index, the execution probability corresponding to the ratio of the first sum value and the second sum value and the value of the preset index can be determined according to a preset correspondence relationship, and whether to perform the specified operation can be determined according to the execution probability. In the preset correspondence relationship, different execution probabilities corresponding to different ratios of the first sum value and the second sum value and different values of the preset index are included. In the case of a fixed value of the preset index, the smaller the ratio of the first sum value and the second sum value, the greater the execution probability. In the case of a fixed ratio of the first sum value and the second sum value, the greater the value of the preset index, the greater the execution probability. The execution probability represents the probability of performing the specified operation. In the case of the specified operation being an advertisement bidding operation, the execution probability can be a participation probability, i.e., representing the probability of the target advertisement participating in the advertisement bidding.

[0109] In a possible implementation, the above-mentioned preset correspondence relationship can be as shown in Table 2 below:

[0110] Table 2

[0111]

[0112] As can be seen from Table 2, in the case of knowing the ratio between the first sum value and the second sum value and the value of the preset index, the corresponding execution probability can be determined according to the preset correspondence relationship. For example, if the ratio of the first sum value and the second sum value is 85% and the value of the preset index is 0.09, the execution probability is 0.6.

[0113] After the execution probability is determined, in the case of determining whether to perform the specified operation according to the execution probability, in a possible implementation, the execution probability can be compared with a specified threshold to determine whether the execution probability is greater than or equal to the specified threshold. If yes, it can be determined to perform the specified operation, and if no, it can be determined to cancel the execution of the specified operation. The specified threshold can be set according to actual conditions, which is not limited here. In this way, the specified operation can be performed when the execution probability is large, so as to speed up the consumption of resources and make the consumption of resources more in line with expectations, and the specified operation can be cancelled when the execution probability is small, so as to avoid over-delivery caused by participating in bidding.

[0114] Optionally, in the case of the execution probability being a participation probability, since the advertisement bidding operation can be performed when the participation probability is large, the return on investment of the advertiser can be improved, and since the advertisement bidding operation is cancelled when the participation probability is small, the problem of over-delivery caused by participating in bidding and low return on investment of the advertiser can be avoided.

[0115] In another possible implementation, the execution probability can also be compared with the first variable, a difference between the execution probability and the first variable is determined, and it is judged whether the difference is greater than 0. If yes, the specified operation is executed. If no, the specified operation is cancelled. Optionally, the first variable can also be determined according to a ratio between the first sum and the second sum, and the greater the ratio between the first sum and the second sum, the greater the value of the first variable. In this way, when the ratio between the first sum and the second sum is large, the difference between the execution probability and the first variable can be made to be less than 0 by using a larger first variable, so that the specified operation is cancelled, and over-investment is avoided. When the ratio between the first sum and the second sum is small, the difference between the execution probability and the first variable can be made to be greater than 0 by using a smaller first variable, so that the specified operation is executed, to speed up resource consumption and make the resource consumption more in line with expectations. In addition, when the operation is an advertisement bidding operation, the return on investment of an advertiser can also be improved. Alternatively, the first variable can be a random number, which can be set according to historical experience. In this way, the first variable can be used to flexibly control whether the specified operation is executed.

[0116] In the above S110, when the first resource of the remaining time slice is dynamically adjusted according to the total consumed resource, specifically, the remaining resource can be determined according to the total consumed resource and the total resource of the target advertisement to be put in the specified period, and then the third resource consumed by the target advertisement in each time slice of the remaining time slice can be determined according to the historical delivery data of the first advertisement in the above S1021 and the remaining resource. The third resource is the first resource of the remaining time slice. When the third resource consumed by the target advertisement in each time slice of the remaining time slice is determined according to the historical delivery data of the first advertisement in the above S1021 and the remaining resource, the specific implementation can refer to the corresponding content described in the above S1022, which will not be described in detail here.

[0117] The embodiments of the present application dynamically adjust the prediction result of the advertisement resource, and judge whether over-investment occurs and whether the specified operation for speeding up resource consumption is executed in each time slice, so that the smooth consumption of the advertisement resource in the whole process of the advertisement delivery can be ensured. In addition, when the specified operation is an advertisement bidding operation, the return on investment of an advertiser can also be improved.

[0118] In order to facilitate understanding of the technical solutions provided by the embodiments of the present application, the embodiments shown in Figure 2 are referred to. Figure 2 is a flowchart of an embodiment method of the advertisement delivery of the present application, Figure 2 is an embodiment of the present application, Figure 1 is an embodiment of the present application, Figure 2The illustrated embodiments can include the following steps.

[0119] S201: predicting a first budget consumed for putting the target advertisement in each time slice in the specified period according to the historical putting data of the first advertisement in the advertisement putting system and the total budget for putting the target advertisement in the specified period.

[0120] S202: determining a second budget consumed in each time slice according to the actual consumed budget in each time slice and the historical putting data of the advertisement putting system and / or the actual putting data of the target advertisement.

[0121] S203: determining whether the first sum of the total budget actually consumed in the current time slice and the second budget corresponding to the current time slice is greater than or equal to the total budget for putting the target advertisement in the specified period.

[0122] If yes, S211 is executed; if no, S204 is executed.

[0123] S204: dynamically adjusting the first budget of the remaining time slice according to the total budget actually consumed.

[0124] S205: determining the second sum of the first budget from the first time slice to the current time slice according to the first budget of each time slice.

[0125] S206: determining whether the ratio of the first sum to the second sum is less than or equal to a preset threshold.

[0126] If yes, S210 is executed; if no, S207 is executed.

[0127] S207: predicting the value of the preset index according to the predetermined model.

[0128] S208: determining the participation probability corresponding to the ratio of the first sum to the second sum and the value of the preset index according to the preset corresponding relationship.

[0129] S209: determining whether the difference between the participation probability and the first variable is greater than 0.

[0130] If yes, S210 is executed; if no, the advertisement bidding operation is cancelled, and in the case of cancelling the advertisement bidding operation, when entering the next time slice, S203 can be returned to be executed.

[0131] S210: executing the advertisement bidding operation.

[0132] In the case of executing the advertisement bidding operation, when entering the next time slice, S203 can also be returned to be executed.

[0133] S211: stop putting the target advertisement.

[0134] The specific implementation of S201 to S211 can be seen from the specific implementation of the corresponding steps in the embodiments shown in the above Figure 1 The specific implementation of S201 to S211 can be seen from the specific implementation of the corresponding steps in the embodiments shown in the above Figure 2 As can be seen from the embodiments shown in the above, in the process of putting the target advertisement, the embodiments of the present application can cyclically execute S203 to S210 until the target advertisement is stopped in the case of over-putting.

[0135] Figure 2 As can be seen from the embodiments shown in the above, the advertisement putting system of different traffic scales can achieve true smooth consumption, avoid the problem of too much budget remaining or over-putting, and be more friendly to advertisers with low budgets. In addition, through the judgment of whether to execute the advertisement bidding, the advertiser can get more high-quality traffic at a lower price, and the return on investment of the advertiser is improved.

[0136] The technical scheme provided by the embodiments of the present application can improve the accuracy of the prediction result when predicting the resources consumed by the target advertisement in each time slice, because the total resources consumed in the actual advertisement putting process can be dynamically adjusted after the resources consumed in each time slice are predicted. In the advertisement putting process, the resources consumed by the delay can be considered, and whether to stop putting can be judged according to the sum of the resources consumed by the delay and the total resources consumed in each time slice, so that the over-putting can be effectively controlled in each time slice, thereby avoiding the problem of over-putting and the problem of resource remaining caused by stopping putting too early. In the case of not over-putting, whether to execute the specified operation for accelerating the consumption of resources can be judged according to the total resources consumed and the predicted resources in each time slice, so that the resource consumption can be more in line with the expectation. The embodiments of the present application dynamically adjust the prediction result of the advertisement resources, and judge whether to over-put and whether to execute the specified operation for accelerating the consumption of resources in each time slice, so as to ensure that the advertisement resources can be smoothly consumed in the whole process of putting the advertisement. In addition, in the case of the specified operation being the advertisement bidding operation, the return on investment of the advertiser can be improved.

[0137] The specific embodiments of the present application are described above. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in a different order and still achieve desirable results. Additionally, the processes depicted in the figures do not necessarily require the particular order shown or sequential order in order to achieve the desired results. In certain implementations, multitasking and parallel processing can be advantageous.

[0138] Figure 3 is a structural schematic diagram of an electronic device according to an embodiment of the present application. Please refer to Figure 3 At the hardware level, the electronic device comprises a processor, and optionally further comprises an internal bus, a network interface, and a memory. The memory can comprise a memory, such as a random-access memory (RAM), and can further comprise a non-volatile memory, such as at least one disk memory. Of course, the electronic device can further comprise other hardware required by a business.

[0139] The processor, the network interface, and the memory can be connected to each other through the internal bus, which can be an industry standard architecture (ISA) bus, a peripheral component interconnect (PCI) bus, or an extended industry standard architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, and a control bus, etc. For the convenience of representation, Figure 3 Only one bidirectional arrow is used in the figure, but it does not mean that there is only one bus or only one type of bus.

[0140] The memory is used to store a program. Specifically, the program can comprise program code, and the program code comprises computer operation instructions. The memory can comprise a memory and a non-volatile memory, and provides instructions and data to the processor.

[0141] The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs, and forms an advertisement delivery device at the logical level. The processor executes the program stored in the memory, and is specifically used for performing the following operations:

[0142] predicting a first resource consumed by delivering a target advertisement in each time slice of a specified period;

[0143] predicting a second resource consumed by a delay corresponding to each time slice in the advertisement delivery process;

[0144] when delivering the target advertisement, determining whether a first sum of a total resource currently consumed and the second resource corresponding to the current time slice is greater than or equal to a total resource for delivering the target advertisement in the specified period in each time slice;

[0145] if yes, stopping delivering the target advertisement;

[0146] If not, then based on the first sum and the first resource of each time slice, determine whether to perform the specified operation, and dynamically adjust the first resource of the remaining time slices based on the total resources consumed, the specified operation being used to accelerate resource consumption.

[0147] The above is as stated in this application. Figure 3 The advertising delivery device disclosed in the illustrated embodiments can be applied to a processor, or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method can be completed by integrated logic circuits in the processor's hardware or by instructions in software form. The processor can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software module can reside in a mature storage medium in the field, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory, and the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method.

[0148] The electronic device can also perform Figure 1 and Figure 2 The method, and to realize the advertising placement device in Figure 1 and Figure 2 The functions described in the illustrated embodiments will not be repeated here.

[0149] Of course, in addition to software implementation, the electronic device of this application does not exclude other implementation methods, such as logic devices or a combination of hardware and software, etc. In other words, the execution subject of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.

[0150] The embodiment of the present application further provides a computer readable storage medium storing one or more programs, the one or more programs comprising instructions which, when executed by a portable electronic device comprising a plurality of applications, enable the portable electronic device to perform the method of the embodiment shown in Figure 1 and Figure 2 The method of the embodiment shown in

[0151] predicting a first resource consumed for launching a target advertisement in each time slice of a specified period;

[0152] predicting a second resource consumed for a time delay corresponding to each time slice in the process of launching the advertisement;

[0153] when launching the target advertisement, determining whether a first sum of a total resource currently consumed and the second resource corresponding to a current time slice is greater than or equal to a total resource for launching the target advertisement in the specified period in each time slice;

[0154] if yes, stopping launching the target advertisement;

[0155] if no, determining whether to perform a specified operation for accelerating resource consumption according to the first sum and the first resource of each time slice, and dynamically adjusting the first resource of a remaining time slice according to the total resource consumed.

[0156] Figure 4 is a structural schematic diagram of an advertisement launching device 40 of an embodiment of the present application. Please refer to Figure 4 In a software implementation, the advertisement launching device 40 can comprise a first prediction module 41, a second prediction module 42, a determination module 43, a first advertisement launching module 44 and a second advertisement launching module 45, wherein:

[0157] The first prediction module 41 predicts a first resource consumed for launching a target advertisement in each time slice of a specified period;

[0158] The second prediction module 42 predicts a second resource consumed for a time delay corresponding to each time slice in the process of launching the advertisement;

[0159] The determination module 43, when launching the target advertisement, determines whether a first sum of a total resource currently consumed and the second resource corresponding to a current time slice is greater than or equal to a total resource for launching the target advertisement in the specified period in each time slice;

[0160] The first advertisement launching module 44 stops launching the target advertisement if the determination module 43 determines that the first sum is greater than or equal to the total resource;

[0161] The second advertisement delivery module 45, in the case that the determination result of the determination module 43 is no, determines whether to perform a specified operation according to the first sum value and the first resource of each time slice, the specified operation being used to accelerate resource consumption, and dynamically adjusts the first resource of the remaining time slice according to the total consumed resource.

[0162] Optionally, the first prediction module 41 predicts the first resource consumed for delivering the target advertisement in each time slice of a specified period, comprising:

[0163] obtaining historical delivery data of a first advertisement from an advertisement delivery system, the similarity between the audience of the first advertisement and the audience of the target advertisement being greater than or equal to a preset threshold, the historical delivery data comprising the total exposure of the first advertisement in a historical specified period and the exposure of the first advertisement in each time slice of the historical specified period;

[0164] predicting the first resource consumed for delivering the target advertisement in each time slice of the specified period according to the historical delivery data of the first advertisement and the total resource for delivering the target advertisement in the specified period.

[0165] Optionally, the first prediction module 41 predicts the first resource consumed for delivering the target advertisement in each time slice of the specified period according to the historical delivery data of the first advertisement and the total resource for delivering the target advertisement in the specified period, comprising:

[0166] obtaining a weight corresponding to each time slice of the specified period respectively;

[0167] predicting the first resource consumed for delivering the target advertisement in each time slice of the specified period according to the weight, the historical delivery data of the first advertisement and the total resource for delivering the target advertisement in the specified period.

[0168] Optionally, the second prediction module 42 predicts the second resource consumed by the delay corresponding to each time slice in the advertisement delivery process, comprising:

[0169] for each time slice in the specified period, determining the delay proportion corresponding to the time slice according to the historical delivery data of the advertisement delivery system and / or the actual delivery data of the target advertisement;

[0170] determining the second resource consumed by the delay corresponding to the time slice according to the delay proportion and the resource actually consumed in the time slice.

[0171] Optionally, the second advertisement delivery module 45 determines whether to perform a specified operation, comprising:

[0172] determining a second sum value of the first resources from a first time slice to a current time slice according to the first resources of each time slice;

[0173] determining whether a ratio of the first sum value and the second sum value is less than or equal to a preset threshold value;

[0174] if yes, performing the specified operation;

[0175] if no, determining whether to perform the specified operation according to the ratio and a preset index.

[0176] Optionally, the preset index comprises at least one of:

[0177] a click rate;

[0178] a product of the click rate and a conversion rate.

[0179] Optionally, the second advertisement launching module 45 determines whether to perform the specified operation according to the ratio and a preset index, comprising:

[0180] predicting a value of the preset index according to a predetermined model;

[0181] determining an execution probability corresponding to the ratio and the value of the preset index according to a preset corresponding relationship, wherein the preset corresponding relationship comprises different execution probabilities corresponding to different ratios and different values of the preset index, and the execution probability represents a probability of performing the specified operation;

[0182] determining whether to perform the specified operation according to the execution probability.

[0183] Optionally, the second advertisement launching module 45 determines whether to perform the specified operation according to the execution probability, comprising:

[0184] determining a difference value between the execution probability and a first variable;

[0185] determining whether the difference value is greater than 0;

[0186] if yes, performing the specified operation;

[0187] if no, canceling the specified operation.

[0188] Optionally, the second advertisement launching module 45 dynamically adjusts the first resources of the remaining time slices according to the total consumed resources, comprising:

[0189] determining a remaining resource according to the total consumed resources and a total resource of the target advertisement launched in the specified period;

[0190] determine third resources consumed for placing the target advertisement in each time slice of the remaining time slices according to the historical placement data of the first advertisement and the remaining resources;

[0191] determine the third resources as the first resources of the remaining time slices.

[0192] The advertisement placement device 40 provided by the embodiments of the present application can also execute the method of Figure 1 and Figure 2 and realize the functions of the advertisement placement device in the embodiments shown in Figure 1 and Figure 2 The embodiments of the present application will not be described here again.

[0193] In summary, the above only describes the preferred embodiments of the present application, and is not used to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

[0194] The system, device, module or unit illustrated in the above embodiments can be specifically implemented by a computer chip or entity, or by a product with certain functions. A typical implementation device is a computer. Specifically, the computer may, for example, be a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0195] The computer readable medium includes permanent and non-permanent, removable and non-removable media, which can be implemented by any method or technology to store information. The information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory (ROM), electrically erasable programmable read only memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette, magnetic tape magnetic disk storage or other magnetic storage device, or any other non-transmission medium that can be used to store information accessible by a computing device. According to the definition in this paper, computer readable medium does not include transitory computer readable medium, such as modulated data signal and carrier wave.

[0196] It is also to be noted that the terms "comprising", "including", and any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises a... " does not, without more constraints, exclude the existence of additional identical elements in the process, method, article, or apparatus that comprises the recited element.

[0197] The various embodiments in this application are described in a progressive manner, and the same or similar parts among the various embodiments can be mutually referred to. Each embodiment focuses on the difference from other embodiments. In particular, for system embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the part of the method embodiment.

Claims

1. An advertisement distribution method characterized by comprising: The method comprises the following steps: predicting a first resource consumed by a target advertisement in each time slice of a specified period; predicting a second resource consumed by a delay corresponding to each time slice in the process of the advertisement; when the target advertisement is launched, determining whether a first sum of a total resource currently consumed and the second resource corresponding to the current time slice is greater than or equal to a total resource of the target advertisement launched in the specified period in each time slice; if yes, stopping launching the target advertisement; if no, determining whether to perform a specified operation according to the first sum and the first resource of each time slice, and dynamically adjusting the first resource of the remaining time slice according to the total resource consumed, the specified operation being used to accelerate resource consumption. The method of predicting the second resource consumed by the delay corresponding to each time slice in the process of the advertisement comprises the following steps: for each time slice in the specified period, determining a delay ratio corresponding to the time slice according to historical launching data of an advertisement launching system and / or actual launching data of the target advertisement; the delay ratio is a ratio between the resource consumed by the delay corresponding to a single time slice and the resource actually consumed in the time slice; determining the second resource consumed by the delay corresponding to the time slice according to the delay ratio and the resource actually consumed in the time slice.

2. The method of claim 1, wherein, The method of predicting the first resource consumed by the target advertisement in each time slice of the specified period comprises the following steps: obtaining historical launching data of a first advertisement from an advertisement launching system, the similarity between the audience of the first advertisement and the audience of the target advertisement being greater than or equal to a preset threshold, the historical launching data comprising a total exposure of the first advertisement in a historical specified period and an exposure of the first advertisement in each time slice of the historical specified period; predicting the first resource consumed by the target advertisement in each time slice of the specified period according to the historical launching data of the first advertisement and a total resource of the target advertisement launched in the specified period.

3. The method of claim 2, wherein, The method of predicting the first resource consumed by the target advertisement in each time slice of the specified period according to the historical launching data of the first advertisement and the total resource of the target advertisement launched in the specified period comprises the following steps: obtaining a weight corresponding to each time slice of the specified period; predicting the first resource consumed by the target advertisement in each time slice of the specified period according to the weight, the historical launching data of the first advertisement and the total resource of the target advertisement launched in the specified period.

4. The method of claim 1, wherein, The method of determining whether to perform the specified operation comprises the following steps: determining a second sum of the first resource from a first time slice to a current time slice according to the first resource of each time slice; judging whether a ratio of the first sum to the second sum is less than or equal to a preset threshold; if yes, performing the specified operation; if no, determining whether to perform the specified operation according to the ratio and a preset index.

5. The method of claim 4, wherein, The preset index comprises at least one of the following: a click rate; a product of a click rate and a conversion rate.

6. The method of claim 4, wherein, The determining whether to perform the specified operation according to the ratio and the preset index comprises: predicting a value of the preset index according to a predetermined model; determining an execution probability corresponding to the ratio and the value of the preset index according to a preset corresponding relationship, wherein the preset corresponding relationship comprises different execution probabilities corresponding to different ratios and different values of the preset index, and the execution probability represents a probability of performing the specified operation; determining whether to perform the specified operation according to the execution probability.

7. The method of claim 6, wherein, The determining whether to perform the specified operation according to the execution probability comprises: determining a difference between the execution probability and a first variable; judging whether the difference is greater than 0; if yes, performing the specified operation; if no, canceling the execution of the specified operation.

8. The method of claim 2, wherein, The dynamically adjusting the first resource of the remaining time slice according to the total consumed resource comprises: determining a remaining resource according to the total consumed resource and a total resource of the target advertisement in the specified period; determining a third resource consumed by the target advertisement in each time slice of the remaining time slice according to historical delivery data of the first advertisement and the remaining resource; determining the third resource as the first resource of the remaining time slice.

9. An advertisement distribution device characterized by comprising: The method comprises: a first prediction module, which predicts a first resource consumed by a target advertisement in each time slice of a specified period; a second prediction module, which predicts a second resource consumed by a delay corresponding to each time slice in an advertisement delivery process; a determination module, which, when the target advertisement is delivered, determines whether a first sum of a total consumed resource and the second resource corresponding to a current time slice is greater than or equal to a total resource of the target advertisement in the specified period in each time slice; a first advertisement delivery module, which stops delivering the target advertisement when the determination module determines that the first sum is greater than or equal to the total resource; a second advertisement delivery module, which, when the determination module determines that the first sum is less than the total resource, determines whether to perform a specified operation according to the first sum and the first resource of each time slice, and dynamically adjusts the first resource of the remaining time slice according to the total consumed resource, wherein the specified operation is used to accelerate resource consumption. In the second prediction module, the second resource consumed by the delay corresponding to each time slice in the advertisement delivery process comprises: for each time slice in the specified period, determining a delay ratio corresponding to the time slice according to historical delivery data of an advertisement delivery system and / or actual delivery data of the target advertisement, wherein the delay ratio is a ratio between a resource consumed by a delay corresponding to a single time slice and a resource actually consumed in the time slice; determining the second resource consumed by the delay corresponding to the time slice according to the delay ratio and the resource actually consumed in the time slice.

10. An electronic device, comprising: The method comprises: a processor; and a memory arranged to store computer-executable instructions that, when executed, cause the processor to perform the following operations: predicting a first resource consumed by a target advertisement in each time slice of a specified period; ​ predicting second resources consumed by time delays corresponding to each time slice in an advertising process; when the target advertisement is being delivered, determining, in each time slice, whether a first sum value of a total resource currently consumed and the second resource corresponding to the current time slice is greater than or equal to a total resource for delivering the target advertisement in the specified period; if yes, stopping delivering the target advertisement; if no, determining whether to perform a specified operation for accelerating resource consumption according to the first sum value and the first resource of each time slice, and dynamically adjusting the first resource of the remaining time slices according to the total resource consumed; the predicting of the second resources consumed by time delays corresponding to each time slice in an advertising process comprises: for each time slice in the specified period, determining a time delay ratio corresponding to the time slice according to historical delivery data of an advertising delivery system and / or actual delivery data of the target advertisement; the time delay ratio is a ratio between resources consumed by a time delay corresponding to a single time slice and resources actually consumed in the time slice; determining the second resource consumed by the time delay corresponding to the time slice according to the time delay ratio and the resources actually consumed in the time slice.

11. A computer readable storage medium, characterized in that, The computer-readable storage medium stores one or more programs, which, when executed by an electronic device including a plurality of applications, cause the electronic device to perform the following method: predicting first resources consumed for delivering a target advertisement in each time slice in a specified period; predicting second resources consumed by time delays corresponding to each time slice in an advertising process; when the target advertisement is being delivered, determining, in each time slice, whether a first sum value of a total resource currently consumed and the second resource corresponding to the current time slice is greater than or equal to a total resource for delivering the target advertisement in the specified period; if yes, stopping delivering the target advertisement; if no, determining whether to perform a specified operation for accelerating resource consumption according to the first sum value and the first resource of each time slice, and dynamically adjusting the first resource of the remaining time slices according to the total resource consumed; the predicting of the second resources consumed by time delays corresponding to each time slice in an advertising process comprises: for each time slice in the specified period, determining a time delay ratio corresponding to the time slice according to historical delivery data of an advertising delivery system and / or actual delivery data of the target advertisement; the time delay ratio is a ratio between resources consumed by a time delay corresponding to a single time slice and resources actually consumed in the time slice; determining the second resource consumed by the time delay corresponding to the time slice according to the time delay ratio and the resources actually consumed in the time slice. The computer-readable storage medium stores one or more programs, which, when executed by an electronic device including a plurality of applications, cause the electronic device to perform the following method:

Images