CN111756827A - Information pushing method and device and server - Google Patents

Abstract

The application discloses an information pushing method, an information pushing device and a server, and relates to the technical field of computers. The method comprises the following steps: determining a first time period corresponding to target pushing information; acquiring the accumulated pushing consumption of the target pushing information in a second period of the current cycle, wherein the second period of the current cycle is a time period from the starting time of a first period of the current cycle to the current time, and the current time is the time within the first period of the current cycle; and if the accumulated pushing consumption of the target pushing information in the second period of the current period does not reach the target consumption, increasing the price adjusting coefficient of the target pushing information, determining the current exposure price of the target pushing information based on the increased price adjusting coefficient, and selecting the information to be pushed for pushing based on the sequencing of the target pushing information and the current exposure prices of other pushing information. Therefore, the exposure probability of the target push information can be improved, and the stability of the push consumption of the target push information is improved.

Description

Information pushing method and device and server

Technical Field

The present application relates to the field of computer technologies, and in particular, to an information pushing method, an information pushing apparatus, and a server.

Background

With the rapid development of internet technology, the internet becomes one of the important ways for people to obtain information. The information provider can push information to the Internet platform through the information push system. However, in the current information push mode, the click rate and the conversion amount of information pushed to the internet platform are generally difficult to keep stable. For some pushed information, once the click rate and the conversion amount decrease, the bid ordering may need to be adjusted in a manual setting manner, which is difficult to adjust quickly and accurately on one hand, and on the other hand, the number of pushed information is large, and the workload is very large.

Disclosure of Invention

The application provides an information pushing method, an information pushing device and a server, which can solve the problems.

In one aspect, an embodiment of the present application provides an information pushing method, including: determining a first time period corresponding to target pushing information, wherein the change condition of pushing consumption of the target recommending information in the first time period of a previous cycle meets a target condition; acquiring the accumulated pushing consumption of the target pushing information in a second period of the current cycle, wherein the second period of the current cycle is a time period from the starting time of a first period of the current cycle to the current time, and the current time is a time within the first period of the current cycle; if the accumulated pushing consumption of the target pushing information in the second period of the current period does not reach the target consumption, increasing the price adjusting coefficient of the target pushing information, and determining the current exposure price of the target pushing information based on the increased price adjusting coefficient; and sequencing the current exposure prices of the target push information and other push information, and selecting information to be pushed from the target push information and other push information according to the sequencing for pushing.

On the other hand, an embodiment of the present application provides an information pushing apparatus, including: the device comprises a determining module, an obtaining module and an adjusting module. The determining module is used for determining a first time period corresponding to the target pushing information, wherein the change condition of the pushing consumption of the target pushing information in the first time period of the previous cycle meets the target condition. The acquisition module is used for acquiring accumulated push consumption of the target push information in a second period of the current cycle, wherein the second period of the current cycle is a time period from the starting time of the first period of the current cycle to the current time, and the current time is a time within the first period of the current time. The adjusting module is used for increasing the price adjusting coefficient of the target pushing information when the accumulated pushing consumption of the target pushing information in the second period of the current period does not reach the target consumption, and determining the exposure price of the target pushing information based on the increased price adjusting coefficient. The pushing module is used for sequencing the current exposure prices of the target pushing information and other pushing information, and selecting the information to be pushed from the target pushing information and other pushing information according to the sequencing for pushing.

In another aspect, an embodiment of the present application provides a server, including: one or more processors; a memory; one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs configured to perform the methods described above.

In another aspect, an embodiment of the present application provides a computer-readable storage medium, on which program code is stored, and the program code can be called by a processor to execute the method described above.

According to the scheme provided by the application, a first time period corresponding to the target push information is determined according to the change condition of the push consumption of the target push information in the previous period, and the accumulated push consumption of the target push information in a second time period of the current period is obtained, wherein the second time period of the current period is a time period from the starting time of the first time period of the current period to the current time, and the current time is the time within the first time period of the current period. And if the current price adjusting coefficient of the recommended information is pushed by the target pushing information, determining the current exposure price of the target pushing information based on the increased price adjusting coefficient. And sequencing the current exposure prices of the target push information and other push information, and selecting the information to be pushed from the target push information and other push information according to the sequencing for pushing. Therefore, when the target pushing information is low in pushing consumption due to the fact that the click rate and the conversion amount are too low, the price adjusting coefficient of the target pushing information is automatically increased in the first period of the current cycle, so that the competitiveness of the target pushing information in the information bidding sorting stage is improved, the exposure probability of the target pushing information is effectively increased, and the click rate, the conversion amount and the pushing consumption of the target pushing information can be timely recovered and stabilized. Because the process is triggered automatically, the workload is greatly reduced, and compared with a manual setting mode, the method is quicker, more accurate and more timely.

These and other aspects of the present application will be more readily apparent from the following description of the embodiments.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

Fig. 1 shows an architecture diagram of an information push system provided by an embodiment of the present application.

Fig. 2 shows a schematic diagram of an application environment suitable for the embodiment of the present application.

Fig. 3 shows a flow chart of an information pushing method provided according to an embodiment of the present application.

Fig. 4 shows a schematic diagram of the substeps of step S302 shown in fig. 3.

Fig. 5 shows another flow chart of the information push method provided by the embodiment of the application.

Fig. 6 shows a schematic time period diagram in an example provided by the embodiment of the present application.

Fig. 7 shows another flow chart of the information push method provided by the embodiment of the application.

Fig. 8 shows another flow chart of the information push method provided by the embodiment of the application.

Fig. 9 shows a schematic diagram of the substeps of step S306 shown in fig. 3.

Fig. 10 shows a graph of the change in the actual conversion unit price in one example.

Fig. 11 shows another flow chart of the information push method provided in the embodiment of the present application.

Fig. 12 shows a further flowchart of the information push method according to the embodiment of the present application.

Fig. 13 shows a schematic time period diagram in another example provided by the embodiment of the present application.

Fig. 14 shows a block diagram of an information pushing apparatus according to an embodiment of the present application.

Fig. 15 is a block diagram of a server for executing an information push method according to an embodiment of the present application.

Fig. 16 is a storage unit for storing or carrying program codes for implementing an information pushing method according to an embodiment of the present application.

Detailed Description

In order to make the technical solutions better understood by those skilled in the art, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application.

With the rapid development of internet technology, the internet gradually becomes one of the important ways for people to obtain information, and the demand of people on information push is more and more extensive. In practical application, the information provider can pay a certain fee to the internet platform so as to push the specific information to the terminal user through the internet platform, and the terminal user can perform corresponding operation based on the specific information. Wherein, if the operation expected by the information provider is performed based on the push information, the push information can be considered to realize the one-time conversion behavior. The cost paid by the information provider can then be considered the push consumption spent by the information provider for the particular information.

One typical application of information push is advertisement push, and currently, there is a demand for pushing advertisements on an internet platform to attract users, regardless of internet products or physical products. In this case, the information provider is the advertiser.

For the internet platform, information which the user is interested in is pushed, so that profits can be obtained, and the user viscosity can be improved to a certain extent. Based on this, the internet platform is usually connected with a corresponding information push system, and the information push system can sort the information based on a certain mechanism, and then select the information required to be pushed according to the sorting.

On one hand, the internet platform pursues long-term experience of platform products, namely, a high-quality information provider can be reserved; on the other hand, it is sought to maximize the flow efficiency. The total income of the internet platform in one period of information push can be obtained through the following expression:

the total yield is the total flow x the flow fill rate x the price per exposure,

the period of information push can be flexibly set by the internet platform, and can be, for example, one day, two days, one week, one month, and the like. The embodiments of the present application do not limit this. The total traffic may be a total browsing number (or a total click number) of information pushed by the internet platform in a period, the traffic filling rate may be a ratio of a number of information pushed by the internet platform in the period to a number of information pushing opportunities, and the number of information pushing opportunities in the period may also be understood as a number of information pushing requests received in the period. The one-time exposure can be understood as one-time information pushing or one-time information displaying of the internet platform, and the one-time exposure price is the price of the one-time information pushing or displaying of the internet platform. In some scenarios, for statistical purposes, the exposure price is usually expressed in terms of a thousand exposure prices or a thousand show prices (CPM), i.e., the cost of a thousand shows on the internet platform is paid.

The information provider pursues maximization of the information push effect, that is, obtaining a sufficient amount of conversion by a sufficiently low exposure price. The conversion may be understood as an operation intended by an information provider performed by a user of the internet platform based on information pushed by the internet platform. For example, when the information provider is the internet application a1, the internet platform may push the link information of the internet application a1, and accordingly, the operation expected by the information provider is registration and activation of the internet application a 1. In this case, the operation of registering and activating the internet application a1 by the user of the internet platform based on the information pushed by the internet platform may be regarded as a conversion Action (Action). For another example, when the information provider is the electronic store a2, the internet platform may push the link information of the electronic store a2, and accordingly, the operation expected by the information provider is the ordering operation at the electronic store a 2. In this case, a ordering operation performed by the user of the internet platform at the electronic shop a2 based on the information pushed by the internet platform may be regarded as a conversion behavior.

The information provider may give a specific optimization goal, i.e. the conversion behavior it wishes to obtain, when using the services of the information push system. Further, the information provider gives the average price that it is willing to pay for one conversion action, which is the target conversion unit price (also referred to as the expected conversion unit price of the information provider), and also referred to as the target consumption per conversion (TargetCPA). Further, the information provider may also provide timely conversion data to the information push system, the conversion data including a number of conversions obtained based on information pushed by the information push system during use of a service of the information push system.

Referring to fig. 1, fig. 1 shows a schematic architecture diagram of an information push system. The information pushing process of the information pushing system 10 will be described with reference to the schematic diagram.

The information pushing system 10 may select information to be pushed from a pushing information library, retrieve a user interested in the information to be pushed based on a user figure of the internet platform, and determine the retrieved user as a target user, thereby pushing the information to be pushed to an internet platform client corresponding to the target user. The internet platform client may be understood as the traffic access layer 20, and the user profile of the internet platform may be obtained from the traffic access layer 20 or may be obtained from a specific storage server, which is not limited in this embodiment.

The information push system 10 selects a process for pushing information to be pushed to the traffic access layer, which is generally referred to as a bid ordering process. During the bid sorting process, the information push system 10 may calculate an exposure cost per thousand (eCPM) of the push information in the push information base by the following expression:

eCPM＝SmartBid×pCVR×pCTR×1000，(1)

wherein SmartBid is an intelligent bid provided by the information pushing system 10, and the intelligent bid is a predicted conversion unit price which can enable the final actual conversion unit price not to exceed the target conversion unit price given by the information provider. The pCVR is the estimated conversion rate obtained by a conversion rate estimation model based on the conversion data of an information provider, and the conversion data can be obtained through a conversion log. The pCTR is an estimated click rate obtained by the click rate estimation model based on click data of the pushed information. The conversion rate pre-estimation model and the click rate pre-estimation model can be obtained through the pre-estimation model training link shown in fig. 1.

Further, in some cases, information push system 10 may also modify the pCVR based on historical data. In this case, the calculation expression of eCPM may be as follows: eCPM ═ TargetCPA × (pCVR × pCVR correction coefficient) × pCTR × 1000.

eCPM is the bid of the pushed information participating in the bid ordering process. In detail, after obtaining the thousands of exposure prices eCPM of each piece of push information, the information push system 10 may sort the eCPM of each piece of push information, and then select a certain number of pieces of push information as the information to be pushed according to the descending order of the eCPM. The certain number can be flexibly set according to needs, for example, the number of information pushing positions available for display can be set.

In some embodiments, the selection process of the information to be pushed may include a rough selection process and a fine selection process. Correspondingly, in the rough selection process, the estimated conversion rate can be calculated through a lightweight conversion rate estimation model (LiteCVR), and the estimated click rate can be calculated through a lightweight click rate estimation model (LiteCTR), so that the rough selection eCPM of each piece of push information can be obtained. And then a certain amount of pushed information can be selected as the roughly selected information to be pushed according to the sequence of the roughly selected eCPMs of the information to be pushed from large to small.

Then, the estimated conversion rate can be calculated through a conversion rate estimation model with a large magnitude (for example, a million magnitude and a million magnitude), the estimated click rate is calculated through a click rate estimation model with a large magnitude, so as to obtain a selected eCPM of each roughly selected information to be pushed, and then a certain amount of information to be pushed is selected as selected information to be pushed according to a sequence of the selected eCPM from large to small, namely the information to be pushed which needs to be pushed to a flow access layer finally.

In this embodiment, SmartBid may be calculated by the following expression:

SmartBid TargetCPA x charge ratio coefficient x cost achievement adjustment coefficient, (2)

Wherein the billing ratio coefficient may be a ratio between the actual consumption of a single click and the bid of the information provider for the single click. The cost achievement adjusting coefficient, which may also be referred to as a price adjusting coefficient or a price adjusting factor, may be used to adjust the TargetCPA to the intelligent bid, and the final actual conversion unit price of the pushed information may not exceed the target conversion unit price of the information provider under the condition that the pushed information participates in the bidding sorting process according to the intelligent bid, that is, the cost requirement of the information provider is met. The actual conversion unit price of one piece of push information refers to the ratio of the cost actually paid by the information provider for the push information in one period to the conversion quantity actually obtained. The actual conversion unit price may also be referred to in some embodiments as the unit cost of the conversion.

In some examples, the information provider's expected conversion behavior is application activation, which gives a target conversion unit price of 10 dollars, and if the information provider pays 3000 dollars in a day and obtains 200 application activations, the actual unit price per application activation is 3000 ÷ 200 ═ 15 dollars, and the actual conversion unit price 15 is greater than 10, meaning that the information provider spends push consumption to not obtain the expected amount of conversion behavior. If the information provider pays 3000 yuan in one day and the obtained application activation number is 300, the actual conversion unit price is 10 yuan, and the target conversion unit price is not more than 10 yuan, which is in accordance with the expectation of the information provider. Therefore, if the unit cost of the pushed information can be controlled to be close to 10 yuan as far as possible under the condition that the unit cost does not exceed 10 yuan, the information provider can be willing to continuously utilize the internet platform to push the information under the condition that the benefit of the internet platform is guaranteed. This is advantageous for both internet platforms and information providers.

The information push system 10 typically charges the information provider for the actual number of clicks of the push information, which depends on the number of exposures and the click rate of the push information. When the number of exposures is stable, the number of clicks may be kept stable basically, and accordingly, the fee (push consumption) paid by the information provider for the push information is also stable basically. However, in some scenarios, it is easy to cause some push information with stable push consumption in a continuous period, and a situation that the push consumption is greatly reduced suddenly occurs. In the case of push information, the number of clicks, the number of exposures, the amount of conversion, and the like are substantially reduced. For example, the pushing consumption of a certain piece of pushing information in the previous day is 1.4 ten thousand yuan, the conversion amount is large, and the actual conversion unit price approaches the target conversion unit price; the pushing consumption of the current pushing is suddenly reduced to 1.2 thousand yuan, the conversion amount is greatly reduced, and the actual conversion unit price is increased.

At this time, the information provider cannot obtain the expected number of conversion behaviors, on one hand, the information push service of the internet platform may not be used any more, and the internet platform will also lose corresponding benefits; on the other hand, new push information may need to be created again, the new push information needs to undergo a new learning process, and the conversion rate estimation model and the click rate estimation model both lack historical data required by learning, so that an early-stage error is large.

Research shows that for pushing information with stable pushing consumption in a plurality of continuous periods, the pushing consumption is suddenly reduced, which is probably caused by the reduced competitiveness of the pushing information in the bidding sorting process, that is, the bidding (eCPM) in the bidding sorting process is too low, so that the pushing information is difficult to be determined as information to be pushed and further difficult to be pushed to a traffic access layer, the number of exposures is small, and correspondingly, the click rate and the conversion amount are also reduced.

Based on this, the embodiment of the application provides an information pushing method, an information pushing device and a server, which can improve the stability of pushing consumption of target pushed information. This is described in detail below.

Referring to fig. 2, fig. 2 is a schematic view of an application environment according to an embodiment of the present disclosure. The server 100 may be communicatively connected to the internet platform server 200 through a network, and the internet platform server 200 may be communicatively connected to the terminal device 300 through a network. The terminal device 300 may have a client 310 running therein, and may log in to the internet platform server 200 through the client 310 or display information pushed by the internet platform server 200.

In this embodiment of the application, the information push system may be deployed in the server 100, and the server 100 may be an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, or a cloud server providing basic cloud computing services such as cloud computing, big data, and an artificial intelligence platform. The internet platform server 200 may be a server deployed with an internet platform, and may be configured to push information to the client 310 in the terminal device 300, where the internet platform deployed in the internet platform server 200 may be any internet application platform such as a page browsing platform, a payment platform, a social platform, a content interaction platform, an education platform, and a video sharing platform.

The terminal device 300 may be, but is not limited to, a smart phone, a tablet Computer, a notebook Computer, a Personal Computer (PC), a portable wearable device, and the like. The client 210 may be any application having an information display function, such as a client of the above-mentioned page browsing platform, payment platform, social platform, content interaction platform, education platform, video sharing platform, and the like.

Referring to fig. 3, fig. 3 is a flowchart illustrating an information pushing method according to an embodiment of the present application, where the method may be applied to the server 100 shown in fig. 2, and steps of the method are described below.

S302, a first time period corresponding to target pushing information is determined, wherein the change condition of pushing consumption of the target pushing information in the first time period of the previous cycle meets a target condition.

The period refers to a period of information pushing, and as described above, the period can be flexibly set, for example, it may be one day, two days, one week, half a month, one month, and the like. Two adjacent cycles may be consecutive in time, and for example, a cycle spans one week, the former of two adjacent cycles is 11 days at 5 months in 2020 to 17 days at 5 months in 2020, and the latter may be 18 days at 5 months in 2020 to 24 days at 5 months in 2020. Taking the example that the span of one cycle is one day, the former of two adjacent cycles can be 24 days 5 and 24 days 2020, and the latter can be 25 days 5 and 25 days 2020.

In this embodiment, the server 100 may execute S302 after entering a new period. The new period may be regarded as a current period, and a period that is before and adjacent to the new period may be regarded as the previous period. For example, in the above example, if the new cycle is 5/25/2020, the previous cycle is 5/24/2020.

The push consumption of the push information refers to the fee paid by the information provider for pushing the push information. For the target push information ad-x, the information provider accumulates the payment for the target push information ad-x from the starting time of the period T to the time T in the period T, which may be referred to as the accumulated push consumption of the target push information ad-x at the time T of the period T. At the end of the period T, the total fee paid by the information provider for the target push information ad-x in the period T may be referred to as the total push consumption of the target push information ad-x in the period T.

The variation of the push consumption of the target push information may be understood as a variation range of a ratio between an accumulated push consumption of the target push information in a previous period and a total push consumption of the target push information in the previous period, which varies with time. Correspondingly, the target condition may be: the ratio of the cumulative push consumption of the target push information in the previous period to the total push consumption of the target push information in the previous period varies within the target range over time. This target range may be flexibly set, for example, may be a range of the third ratio to the fourth ratio.

Correspondingly, the first period when the variation of the push consumption meets the target condition may be understood as: the cumulative push consumption of the target push information in the previous cycle varies from a third proportion to a fourth proportion of a total push consumption of the target push information in the previous cycle. Wherein the fourth ratio is greater than the third ratio. In order to distinguish from other conditions mentioned later, in the following description, the target condition will be described as a first condition.

Assume that the current period is T_i(i is an integer greater than 1), the previous cycle is T_i-1If the target push information is in the previous period T_i-1Total push consumption of cost_i-1The third ratio is denoted as p3 and the fourth ratio is denoted as p4, then the first condition (i.e., the target condition) may be understood as: target push information in previous period T_i-1Cumulative push consumption of from p₃×cost_i-1Is incremented to p₄×cost_i-1The time period of (a). Based on this, S302 may include the steps shown in fig. 4.

S302-1, determining a moment when the accumulated push consumption of the target push information in the previous period reaches a third proportion of the total push consumption in the previous period, and taking the determined moment as a first moment.

S302-2, determining a moment when the accumulated push consumption of the target push information in the previous period reaches a fourth proportion of the total push consumption in the previous period, and taking the determined moment as a second moment, wherein the fourth proportion is larger than the third proportion.

Illustratively, if the target push information is in the previous period T_i-1Time t1 of (1) reaches the total push consumption cost of the target push information in the previous period_i-1May be determined as the first time instant t 1. If the target push information is in the previous period T_i-1Time t2 of (1) reaches the total push consumption cost of the target push information in the previous period_i-1May be determined as the second time instant t 2. The second time is a time after the first time.

S302-3, determining a time period from the first time to the second time in the previous cycle as the first time period.

Correspondingly, the cumulative push consumption of the target push informationIs from a third proportional cost during the time period between the first time and the second time_i-1Cost changed to fourth ratio_i-1Therefore, the previous cycle T can be divided by S302-3_i-1Is determined as a first period corresponding to the target push information, the period from time t1 to time t 2.

Optionally, in this embodiment, the difference between the fourth ratio and the third ratio may be greater than or equal to 50%. For example, the third proportion may be 10%, the fourth proportion may be 60%; for another example, the third ratio may be 5%, and the fourth ratio may be 60%; for another example, the third proportion may be 15% and the fourth proportion may be 70%.

The first time period determined in this embodiment is a time period in which the accumulated push consumption of the target push information in the previous cycle is high, and the subsequent process of this embodiment is to detect whether the adjustment coefficient of the target push information needs to be adjusted in the first time period of the current cycle, which is equivalent to predicting a time period in which the demand of the user for the target push information is high in the current cycle, and then adjust the adjustment coefficient of the target push information in the time period in which the demand is high, so that after the exposure rate of the target push information is increased, the click rate and the conversion amount of the target push information can be more effectively increased, and the stability of the target push information is more effectively maintained.

In this embodiment of the application, the target push information may be push information for which the internet platform side desires to keep the push consumption stable, for example, the target push information may be push information in which both the push consumption and the conversion amount reach set values, and the actual conversion unit price does not exceed the target conversion unit price. Illustratively, before executing S302, the target push information may be determined through the flow shown in fig. 5, which is described in detail below.

S301-1, acquiring the total pushing consumption, the conversion amount, the expected conversion unit price and the actual conversion unit price of each piece of pushing information in the previous period.

Each piece of push information refers to information available for pushing in the push information base. As described above, the total push consumption of one push information in the previous period refers to the total amount of the fee the information provider paid for the push information in the previous period. The conversion amount of a piece of push information in the previous period refers to the amount of expected conversion behaviors obtained by the information provider based on the push information pushed by the internet platform in the previous period.

The expected conversion unit price of one piece of push information in the previous period refers to the fee which is given by the information provider in the previous period and is willing to pay for one conversion action of the push information. The actual conversion unit price of a piece of push information in the previous period refers to the ratio of the total push consumption of the piece of push information in the previous period to the conversion amount of the piece of push information in the previous period, namely how much the information provider actually pays for one conversion action of the piece of push information in the previous period.

S301-2, determining the push information, which is in the previous period, the total push consumption and the conversion amount meet the second condition and the actual conversion unit price does not exceed the expected conversion unit price, as the target push information.

The second condition may include thresholds respectively set for the total push consumption and the conversion amount, for example, the second condition may be that the total push consumption of the push information in the previous cycle reaches a consumption threshold, and the conversion amount of the push information in the previous cycle reaches a conversion amount threshold. The consumption threshold and the conversion amount threshold can be flexibly set according to statistical data or experience, for example, the consumption threshold can be 2500 yuan to 3500 yuan, for example 3000 yuan. The conversion amount threshold varies with the consumption threshold, and is intended to define the magnitude of the actual conversion unit price, for example, when the consumption threshold is 3000 yuan, the conversion amount threshold may be 200; for another example, where the consumption threshold is 3500 dollars, the conversion amount threshold may be 235 dollars. The present embodiment does not limit this.

In the implementation process, the push information meeting the second condition may be determined from the push information as candidate push information, and then the candidate push information having the actual conversion unit price less than or equal to the expected conversion unit price is determined from the candidate push information, and the determined candidate push information is used as the target push information.

S304, obtaining the accumulated pushing consumption of the target pushing information in a second period of the current cycle, wherein the second period of the current cycle is a time period from the starting time of the first period of the current cycle to the current time, and the current time is a time within the first period of the current cycle.

After determining the first period, a start time and an end time of the first period of the current cycle may be further determined. Referring to fig. 6, a schematic diagram of the first period is exemplarily shown, in which the span of one cycle is one day, the previous cycle is day1, and the current cycle is day 2. In this example, if it is determined that the cumulative push consumption of the target push information at 10 o 'clock of day1 reaches the third proportion of its total push consumption at day1 and the cumulative push consumption at 14 o' clock of day1 reaches the fourth proportion of its total push consumption at day1, it may be determined that the time period intv1-1 between 10 o 'clock and 14 o' clock of day1 is the first period corresponding to the target push information. Further, it may be determined that the first period of the current cycle is: time period between 10 and 14 points of day2, intv 2-1.

It is worth noting that, since the trend of the target push information's push consumption increases in different periods are not necessarily the same, the cumulative push consumption of the target push information at 10 o' clock of day2 may not reach the third proportion of its total push consumption at day2, but may also reach the third proportion; the cumulative push consumption of the target push information at 14 o' clock of day2 may not reach its fourth proportion of the total push consumption at day2, but may also reach.

In an implementation process, after determining that the first period of the current cycle is entered, when the tuning coefficient of the target push information is obtained in each bidding sorting process, the server 100 may determine a time period taking the starting time of the first period of the current cycle as the starting time and taking the current time as the ending time, where the time period is the second period. Therefore, the current time in S304 can be understood as: the server 100 obtains the price adjustment coefficient of the target push information each time in the first period of the current cycle. Correspondingly, the current time is a certain time within the first period of the current cycle and may be the same as the starting or ending time of the first period of the current cycle.

After determining the second period, the server 100 may obtain the total fee paid by the information provider for the push of the target push information in the second period of the current cycle, i.e., the cumulative push consumption of the target push information in the second period of the current cycle.

S306, if the accumulated pushing consumption of the target pushing information in the second period of the current period does not reach the target consumption, increasing the current price adjustment coefficient of the target pushing information, and determining the current exposure price of the target pushing information based on the increased price adjustment coefficient.

Wherein, if the cumulative push consumption of the target push information in the second period of the current cycle reaches the target consumption, the difference between the trend of the cumulative push consumption of the target push information in the second period of the previous cycle and the trend of the cumulative push consumption in the second period of the current cycle is within an acceptable range, that is, the situation is not a great reduction. Based on this, the target consumption can be flexibly set, and in one embodiment, can be a fixed value set based on statistical data or experience. In another embodiment, the target consumption may be determined dynamically, for example, the target consumption may be determined through the flow shown in fig. 7 before S306 is executed.

S305-1, acquiring the total pushing consumption of the target pushing information in the previous period, and determining the total pushing consumption of the first proportion as the expected consumption of the target pushing information in the first period of the current period.

The expected consumption is a product of a total push consumption of the target push information in a previous period and a first proportion, and if the accumulated push consumption of the target push information in the first period of the current period can reach the expected consumption, a difference between the accumulated push consumption of the target push information in the first period of the current period and the accumulated push consumption of the target push information in the first period of the previous period is smaller, that is, a probability of greatly reducing the push consumption is reduced.

Based on the above characteristics of expected consumption, the first ratio may be related to the fourth ratio and the third ratio used to define the first period. For example, the first ratio may be in a positive correlation with the difference between the fourth ratio and the third ratio. For example, when the difference between the fourth ratio and the third ratio is 50%, the first ratio may be 30%; the first ratio may be greater than 30% when the difference between the fourth ratio and the third ratio is greater than 50%, for example, the first ratio may be 35% when the difference between the fourth ratio and the third ratio is 55%.

S305-2, acquiring the accumulated push consumption of the target push information in the first period of the previous cycle as a first accumulated push consumption.

S305-3, acquiring the accumulated push consumption of the target push information in the second period of the previous period as a second accumulated push consumption.

In this embodiment, the execution sequence of S305-2 and S305-3 is not limited, and the two may be executed in parallel or sequentially according to a certain sequence. The second period of the previous cycle may be a period between a start time of the first period of the previous cycle to a target time of the previous cycle, the target time being a time corresponding to the current time in the previous cycle. For example, as shown in fig. 6, the first period of the previous cycle is a period intv1-1 from 10 to 14 points of day1, the first period of the current cycle is a period intv1-2 from 10 to 14 points of day2, if the current time is 10:30 of day2, 10:30 of day1 may be regarded as the target time, and correspondingly, the second period of the previous cycle may be a period intv2-2 from 10 to 10:30 of day 2.

S305-4, determining the proportion of the second accumulated push consumption in the first accumulated push consumption, taking the determined proportion as a second proportion, and determining the expected consumption of the second proportion as the target consumption.

Since the expected consumption is the cost expected to be paid by the information provider for the target push information in the whole first period of the current cycle, and the second period is only a part of the first period, in order to smoothly distribute the expected consumption to each section of the first period, the cost expected to be paid by the information provider for the target push information in the second period of the current cycle, that is, the target consumption, may be determined according to the ratio of the cumulative push consumption of the target push information in the second period of the previous cycle to the cumulative push consumption of the target push information in the first period of the previous cycle.

For example, in the scenario shown in fig. 6, a ratio of the accumulated push consumption of the target push information in the time period intv1-2 to the accumulated push consumption thereof in the time period intv1-1 may be obtained, and the ratio is the second ratio. Correspondingly, the product of the ratio and the expected consumption is the target consumption corresponding to the time period intv 2-2.

In detail, the target consumption may be determined by the following expression:

wherein yesterday _ cost represents a total push consumption of the target push information in a previous period, p1 represents a first ratio, p3 represents a third ratio, p4 represents a fourth ratio, and rate represents an accumulated push consumption of the target push information at a target time in the previous period. Correspondingly, rate-p3 indicates a ratio of the cumulative push consumption of the target push information in the second period of the previous period to the total push consumption (yesterday _ cost) of the target push information in the previous period, and p4-p3 indicates a ratio of the cumulative push consumption of the target push information in the first period of the previous period to yesterday _ cost, and (rate-p3)/(p4-p3) indicates a ratio of the second cumulative push consumption to the first cumulative push consumption, i.e., a second ratio.

Thus, the above expression represents: when the second ratio is a value within the interval (0,1), the target consumption is equal to the product of p1 × yesterday _ cost (i.e., expected consumption) and the second ratio; when the second ratio is 0, the target consumption is 0, i.e. the product of 0 and the expected consumption; when the second ratio is 1, the target consumption is the expected consumption, i.e. the product of 1 and the expected consumption. Thus, based on the above expression, it is possible to achieve the determination of the expected consumption of the second proportion as the target consumption.

After determining the target consumption, it may be determined whether an accumulated push consumption of the target push information in the second period of the current cycle reaches the target consumption. In one embodiment, the accumulated push consumption of the second period (e.g., the time period intv2-2 shown in fig. 6) of the current cycle of the target push information may be directly obtained and compared with the target consumption.

In another embodiment, the accumulated push consumption of the target push information in the time period from the start time of the current cycle to the target time (e.g., the time period intv2-3 from 0 point to 10:30 of day 2) may also be compared with the target consumption. In detail, this comparison can be realized by the flow shown in fig. 8.

S305-5, acquiring a third accumulated push consumption of the target push information, where the third accumulated push consumption is an accumulated push consumption of the target push information from the starting time of the current period to the current time.

Wherein, taking the scenario shown in fig. 6 as an example, if the current time is 10:30, the third cumulative push consumption may be the cumulative push consumption in the time period between 0 point of day2 and 10: 30.

S305-6, if the third cumulative push consumption of the target push information is less than the target consumption, determining that the cumulative push consumption of the target push information in the second period of the current cycle does not reach the target consumption.

In the present embodiment, in a case where the third accumulated push consumption does not reach the target consumption, in a second period (e.g., intv2-2) within a period corresponding to the third accumulated push consumption (e.g., a period between 0 point and 10:30 of day2 in fig. 6), the accumulated push consumption in which the target push information is present is less likely to reach the target consumption. That is, based on the comparison of the third cumulative push consumption with the target consumption, it has been roughly determined whether the cumulative push consumption of the target push information in the current period is small. And, the above-mentioned S306 aims to adjust the price adjustment coefficient of the target push information in the current period when the cumulative push consumption of the target push information in the current period is less, and there may be no precise limit on how much the cumulative push consumption of the target push information in the current period is less. Therefore, when the third cumulative push consumption does not reach the target consumption, it may be determined that the cumulative push consumption of the target push information in the second period of the current cycle does not reach the target consumption, so that the current price adjustment coefficient of the target push information may be increased. By the aid of the judging mode, accumulated pushing consumption of the second time period does not need to be counted separately, and data processing amount is reduced.

The magnitude or proportion of the increase of the pricing factor can be flexibly set, for example, the magnitude or proportion can be flexibly set according to statistical data or experience, and the embodiment has no limitation. After the current price adjustment coefficient of the target push information is increased to obtain the increased price adjustment coefficient, the increased price adjustment coefficient can be calculated through the expressions (1) and (2), so that the current exposure price of the target push information can be obtained, and the bid ordering can be participated on the basis of the current exposure price.

S308, sequencing the current exposure prices of the target push information and other push information.

The current exposure prices of the target push information and other push information can be obtained through expressions (1) and (2) above. It should be noted that the price adjustment coefficient used in the process of calculating the current exposure price of the other pieces of pushed information is a price adjustment coefficient obtained based on the optimization target. The adjustment coefficient used in the process of calculating the current exposure price of the target push information is the adjustment coefficient obtained by increasing the adjustment coefficient obtained based on the optimization target through S306.

After the current exposure prices of the push information are obtained, the exposure prices of the push information can be sorted in the order from large to small or from small to large. Due to the fact that the price adjusting coefficient of the target pushing information is increased, the competitiveness of the exposure price of the target pushing information is also increased, and therefore the target pushing information can be determined as information to be pushed more easily in the bidding sorting process and then pushed to the client side of the internet platform.

S310, selecting information to be pushed from the target pushing information and the other pushing information according to the sequence to push.

In this embodiment, the server 100 may send a given amount of push information from the target push information and other push information in the order from the large exposure price to the small exposure price, and send the selected given amount of push information as information to be pushed, and push the information to be pushed to the internet platform. The given number can be flexibly set, for example, the number of positions where push information can be shown on an internet platform can be set. The present embodiment does not limit this.

In the information pushing method provided by this embodiment, in each cycle, a first time period in which a change condition meets a target condition is determined according to a change condition of pushing consumption of target pushing information in a previous cycle, and further, when the pushing consumption of the target pushing information is low due to too low click rate and conversion amount, a price adjustment coefficient of the target pushing information is increased in the first time period of the current cycle, so that bids (eCPM) of the target pushing information in a bidding sorting process in the first time period can be timely promoted, and thus competitiveness of the target pushing information in the bidding sorting process in the first time period is improved, a probability of the target pushing information in the sorting process is higher, and a probability of being exposed is higher, and correspondingly, a high click rate and a high conversion amount are more easily obtained. The pushing consumption is calculated according to the click rate, and under the condition that the click rate is improved, the pushing consumption is also improved in time, so that the disadvantages of an internet platform and an information provider, which are caused by the fact that the pushing consumption of target pushing information is greatly reduced, are avoided. Because the process is triggered automatically, the workload can be greatly reduced, and the method is quicker, more accurate and more timely compared with a manual setting mode.

Further, each period determines the target pushing information based on the total pushing consumption, the conversion amount, the actual conversion unit price and the expected conversion unit price of each piece of pushing information in the previous period, so that the competitiveness of the target pushing information in the bidding sorting process in the first period of the current period is improved, and the steps are repeated in such a way, finally, the competitiveness of the high-quality pushing information is stronger and stronger, and the competitiveness of the low-frequency pushing information is gradually reduced, so that the overall quality of the pushing information is improved.

Referring to fig. 3 and 9 together, the steps shown in fig. 3 will be further described in detail.

In S306, the increase ratio of the pricing factor can be flexibly set. In order to make the pushing consumption of the target pushing information rise smoothly in the first period, the step of increasing the price adjustment coefficient of the target pushing information in S306 may include the substeps shown in fig. 9.

S306-1, obtaining a first price adjustment coefficient of the target pushing information, wherein the first price adjustment coefficient is the price adjustment coefficient of the target pushing information at the current moment.

S306-2, if the first price adjustment coefficient is not larger than the second price adjustment coefficient, weighting and summing the first price adjustment coefficient and the second price adjustment coefficient, and updating the price adjustment coefficient of the target push information at the current moment into the sum.

In this embodiment, the adjustment coefficient used in the process of calculating the current exposure price of other pushed information is similar to the calculation method of the first adjustment coefficient, and can be solved through the target optimization problem, and the process of obtaining the first adjustment coefficient is described in detail below with reference to fig. 10.

In detail, the objective to be optimized is that the final actual conversion unit price of the pushed information can be close to the target conversion unit price, that is:

wherein, cost_hRepresenting the cumulative push consumption of the time period between 0 and the current time t, conv_hRepresents the cumulative amount of conversion, λ, of the time period between 0 and the current time t_hRepresenting a historical rating factor (which may also be referred to as a historical cost achievement rating factor). All three can be obtained from historical data as known quantities.

cost_eRepresenting the cumulative push consumption, conv, of the time period between the current time t and 24_eRepresenting the cumulative amount of conversion, λ, of the time period between the current time t and 24_eRepresenting a future price adjustment factor (which may also be referred to as a future cost achievement adjustment factor). Since it is currently at time t, all three are unknowns.

Carrying out the processBased on the history data of the time period from 0 point to the current time t, a cost can be fitted_hAnd cost_{h_1.0}Is (λ ═ λ)_h) And historical price adjustment coefficient lambda_hThe association relationship between them is as follows:

wherein, cost_{h_1.0}Representing the cumulative push consumption for the time period between 0 and the current time t without a price adjustment (i.e., a historical price adjustment factor of 1).

Further, a cost can also be fitted based on the historical data_{h_1.0}Duty ratio in total push consumption throughout the day as a function of time relation:

cost_{e_1.0}＝cost_{all day}-cost_{h_1.0}， (6)

Wherein, cost_{e_1.0}Representing the cumulative push consumption for the time period between the current time t and 24 without a price adjustment (i.e., a future price adjustment factor of 1).

Based on future price-adjusting coefficient lambda_eThe future push consumption cost can be calculated by the following expression (7)_e:

Optionally, an average conversion unit price CPA can also be fitted based on historical data_1.0The following relation (8) exists between the valence regulating coefficient lambda and the valence regulating coefficient:

correspondingly, the future conversion amount conv_eCan be obtained by the following expression (9):

substituting the above expressions (6) and (8) into the above expression (3) can obtain:

based on the expression (3), the objective to be optimized is E (CPA)_final) Is equal to TargetCPA, which is known, and therefore, the future price adjustment coefficient λ can be solved based on expression (10)_e。

The processing flow of S306-2 may be understood as weighting and summing the second price adjustment coefficient larger than the first price adjustment coefficient and the first price adjustment coefficient, and using the obtained sum as a new price adjustment coefficient of the target push information at the current time, which is equivalent to amplifying a part of the first price adjustment coefficient, where the obtained sum is necessarily larger than the first price adjustment coefficient, that is, increasing the current price adjustment coefficient (first price adjustment coefficient) of the target push information is achieved.

In one embodiment, the second tuning coefficient may be an empirically set threshold. In another embodiment, since the price adjustment coefficient is essentially a cost achievement adjustment coefficient, which is related to the target conversion unit price that the information provider wishes to achieve, and the target conversion unit price of the target push information in the previous period and the target conversion unit price in the current period may be different, a coefficient obtained by scaling the price adjustment coefficient of the target push information at the target time of the previous period may be used as the second price adjustment coefficient, and the scaling process may be performed based on the target conversion unit price of the previous period and the target conversion unit price of the current period.

In detail, as shown in fig. 11, the second tuning coefficient may be obtained through the steps shown in fig. 11 before performing S306-2.

S305-7, obtaining a third price adjustment coefficient of the target pushing information, wherein the third price adjustment coefficient is the price adjustment coefficient of the target pushing information at the target moment in the previous period.

As described above, the target time refers to a time in the previous cycle corresponding to the current time of the current cycle. For example, in the scenario shown in FIG. 6, where the current time is 10:30 of day2, it may be determined that 10:30 of day1 is the target time. Correspondingly, the actual pricing factor of 10:30 of the target push information at day1 is the third pricing factor.

S305-8, carrying out scaling processing on the third price adjustment coefficient according to the proportional relation between the expected conversion unit price in the previous period and the expected conversion unit price in the current period of the target push information.

S305-9, determining the third adjusting coefficient subjected to the scaling processing as the second adjusting coefficient.

In implementation, the scaling process of the third tuning coefficient can be implemented by the following expression (11):

wherein, the factor on the left side of equal sign_N-1The third adjusting coefficient after scaling processing is shown, and the factor on the right side of the equal sign_N-1The original third adjustment factor is shown. TargetCPA_NAnd the target conversion unit price of the target push information in the current period is represented, namely the fee willing to be paid by the information provider for one conversion action of the target push information is represented, namely the expected conversion unit price. TargetCPA_N-1A target conversion unit price (i.e., an expected conversion unit price) representing the previous period of the target push information.

Optionally, before performing the step of weighting and summing the first price adjustment coefficient and the second price adjustment coefficient in S306-2, the information pushing method provided in the embodiment of the present application may further include the step shown in fig. 12. The detailed description is as follows.

S305-10, obtaining a ratio of a third accumulated push consumption of the target push information to the target consumption.

As described above, the third cumulative push consumption refers to a time period between the start time of the current cycle and the current time of the targeted push informationCumulative push consumption of. For example, in the scenario shown in fig. 6, if the current time is 10:30 of day2, the third cumulative push consumption refers to the cumulative push consumption cost of the target push information in the time period intv2-3 between 0 and 10:30 of day2_intv2-3. Correspondingly, assume that the cumulative push consumption of the target push information for the time period intv2-2 is cost_intv2-2The cumulative push consumption at the time period intv2-1 is cost_intv2-1Then the following expression (12) can represent the ratio in S305-10.

S305-11, and comparing the magnitude relation between the ratio and 1. If the ratio is less than 1, executing S318; if the ratio is not less than 1, S319 may be performed.

S305-12, determining the ratio as the weight of the first adjusting coefficient, and determining the difference value between 1 and the ratio as the weight of the second adjusting coefficient.

S305-13, determining the weight of the first adjusting coefficient as 1, and determining the weight of the second adjusting coefficient as 0.

In this embodiment, assuming that the weight of the first tuning coefficient is β and the weight of the second tuning coefficient is α, S1202-S1204 can be implemented by the following expression:

where cost represents the third cumulative push consumption described in S1101, c is the target consumption, and min () represents the minimum of cost/c and 1.0. It is to be appreciated that in some cases, the cost in expression (13) may also be replaced with the accumulated push consumption of the target push information in the second period of the current cycle.

After determining the weights of the first adjusting coefficient and the second adjusting coefficient, the first adjusting coefficient and the second adjusting coefficient may be weighted and summed through S303-2, so that:

factor_N＝α×factor_N-1+β×factor_N， (14)

wherein, the factor on the left side of equal sign_NAnd the result of the weighted summation, namely the increased price adjustment coefficient of the target push information at the current moment is shown. Constant-sign right-side factor_NA first price adjustment coefficient representing the target push information. factor_N-1And the second price adjustment coefficient is represented, namely, a third price adjustment coefficient of the target push information subjected to the scaling processing.

The price adjustment coefficient of the target push information in the first period of the current cycle can be increased proportionally and smoothly in the manner shown in fig. 12, so that the push consumption of the target push information in the first period of the current cycle can be increased smoothly, the information provider can obtain enough conversion amount at the expected conversion unit price, and the internet platform can obtain stable and good information push benefits.

In order to make those skilled in the art understand the scheme of the embodiment of the present application more clearly, the following explains the information push method provided by the embodiment of the present application with reference to the scenario shown in fig. 13 by taking the push information as the advertisement information as an example.

Assuming that 20 advertisements exist in the advertisement library, each advertisement is represented as adi (i is more than or equal to 1 and less than or equal to 20, and i is an integer), it is determined that the total push consumption reaches 3000 yuan in the previous day, and the advertisement information with the conversion amount reaching 200 is ad1, ad3 and ad8, and the advertisement information ad1, ad3 and ad8 can be used as target push information. Correspondingly, the server 100 may implement the information pushing method provided by the present embodiment according to the following flow.

S1, when determining that today' S0 point is reached, the server 100 determines that t is t when the cumulative push consumption of adj (j is 1,3,8) in yesterday reaches 10% of the total push consumption yesterday _ cost of adj in yesterday₀Determining the time at which the cumulative push consumption of adj reaches 60% of yesterday _ cost as t₁Will be t of yesterday₀To t₁The period 1-1 in between is determined as the first period.

S2, the server 100 can judge the current time t after obtaining the current adjusting rate of the advertisement information adj in the process of bidding sorting_xWhether or not to positionAt time t of today₀To time t₁In the period 2-1 in between, if yes, the advertisement information adj is acquired at the time t of yesterday_xThe ratio rate of the cumulative push consumption to yesterday _ cost. Wherein, yesterday's time t_xMay be regarded as the target time in the above-described embodiment.

S3, obtaining the difference between rate and 10% as a first difference, obtaining the difference between 60% and 10% as a second difference, and obtaining the ratio of the first difference to the second difference. Wherein the ratio of the first difference to the second difference is represented by: the cumulative push consumption of the advertisement information adj in the time period 2-2 accounts for the proportion of the cumulative push consumption of the advertisement information adj in the time period 2-1. I.e. the second ratio in the above embodiment.

S4, 30% of yesterday-cost is determined as the expected consumption of the advertisement information adj in today' S time period 2-1. Of these, 30% may be regarded as the first ratio in the above embodiment.

S5, a second percentage of expected consumption is determined as the target consumption cj of the advertisement information adj for time period 2-2 today.

S6, obtaining advertisement information adj at time t of yesterday_xFactor of adjusting price_-1And at time t of today_xFactor of adjusting price₀. Factor in S6_-1It can be regarded as the third adjustment coefficient in the above embodiment, the factor in S6₀May be regarded as the first adjustment coefficient in the above embodiment.

S7, acquiring expected conversion unit price TargetCPA of advertisement information adj in yesterday_-1And the expected conversion unit price TargetCPA today₀。

S8, according to the TargetCPA_-1And TargetCPA₀Ratio of (a) to factor_-1Scaling to obtain scaled factor_-1. Wherein, the scaling factor_-1Can be regarded as the second adjustment coefficient in the above embodiment.

S9, if the scaling factor is processed_-1Less than factor₀Then the factor can be set₀Set the weight β of (1), and the scaled factor_-1Is set to 0.

S10, if the scaling factor is processed_-1Greater than factor₀Then the ratio of the cumulative push consumption of the advertisement information adj in today's time period 2-2 to the target consumption cj may be obtained, and if the ratio is less than 1, the factor may be set₀Set weight β to the ratio, and scale the scaled factor_-1α may be set to 1- β, otherwise, the weight β may be set to 1 and the weight α may be set to 0.

Alternatively, the ratio of the accumulated push consumption of the advertisement information adj in today's time period 2-3 to the target consumption cj may be obtained, and if the ratio is less than 1, the weight β may be set to the ratio, and the weight α may be set to 1- β; otherwise, the weight β may be set to 1 and the weight α may be set to 0.

S11, using α as the scaled factor_-1β as factor₀To the scaled factor_-1And factor₀And carrying out weighted summation, and taking the obtained sum as the current price adjustment coefficient of the advertisement information adj.

And S12, obtaining the current exposure price of the advertisement information adj based on the current price adjusting coefficient of the advertisement information adj.

And S13, acquiring the current exposure price of other advertisement information except the advertisement information adj in the advertisement library.

And S14, sequencing the current exposure price of each advertisement message, and selecting a given amount of advertisement messages as the advertisements to be pushed according to the sequence from big to small.

S15, the determined advertisement to be pushed is pushed to the Internet platform, so that the Internet platform pushes the advertisement to be pushed to the corresponding client.

Through the process, the competitiveness of the advertisement information adj in the bidding sorting process can be improved, so that the exposure rate and the click rate of the advertisement information adj are improved, and the pushing consumption of the advertisement information adj is further improved. Further, by executing S1-S15 in a cyclic manner, the competitiveness of the high-quality advertisement information can be gradually enhanced, and the competitiveness of the low-frequency advertisement information can be gradually reduced, i.e., the low-frequency advertisement can be filtered. In addition, the achievement rate of the whole advertisement information can be improved, namely the proportion of the advertisement information of which the actual conversion unit price does not exceed the target conversion unit price in all the advertisement information can be increased.

Referring to fig. 14, a block diagram of an information pushing apparatus 1400 according to an embodiment of the present disclosure is shown. The apparatus 1400 may include a determination module 1410, an acquisition module 1420, an adjustment module 1430, and a push module 1440.

The determining module 1410 is configured to determine a first time period corresponding to target push information, where a change condition of push consumption of the target push information in the first time period of a previous cycle meets a target condition.

The obtaining module 1420 is configured to obtain accumulated push consumption of the target push information in a second period of the current cycle, where the second period of the current cycle is a period from a start time of the first period of the current cycle to a current time, and the current time is a time within the first period of the current cycle.

The adjusting module 1430 is configured to increase the current price adjustment coefficient of the target push information when the accumulated push consumption of the target push information in the second period of the current cycle does not reach the target consumption, and determine the exposure price of the target push information based on the increased price adjustment coefficient.

The pushing module 1440 is configured to rank the current exposure prices of the target pushing information and other pushing information, and select information to be pushed from the target pushing information and the other pushing information according to the ranking to push the information.

Alternatively, the determination module 1410 may also determine the target consumption by: acquiring the total pushing consumption of the target pushing information in the previous period, and determining a first proportion of the total pushing consumption as the expected consumption of the target pushing information in a first period of the current period; acquiring a cumulative push consumption of the target push information in a first period of the previous cycle as a first cumulative push consumption, and acquiring a cumulative push consumption of the target push information in a second period of the previous cycle as a second cumulative push consumption, the second period of the previous cycle being a period from a start time of the first period of the previous cycle to a target time of the previous cycle, the target time being a time within the previous cycle corresponding to the current time; determining a proportion of the second cumulative push consumption in the first cumulative push consumption, taking the determined proportion as a second proportion, and determining the expected consumption of the second proportion as the target consumption.

Optionally, before the adjusting module 1430 increases the price adjustment coefficient of the target push information, the obtaining module 1420 may further be configured to:

acquiring third accumulated push consumption of the target push information, wherein the third accumulated push consumption is accumulated push consumption of the target push information from the starting time of the current period to the current time; if the third cumulative push consumption of the target push information is less than the target consumption, determining that the cumulative push consumption of the target push information in the second period of the current cycle does not reach the target consumption.

Optionally, the adjusting module 1430 may increase the price adjustment coefficient of the target push information by:

acquiring a first price adjustment coefficient of the target pushing information, wherein the first price adjustment coefficient is the price adjustment coefficient of the target pushing information at the current moment; and if the first price adjusting coefficient is not larger than the second price adjusting coefficient, weighting and summing the first price adjusting coefficient and the second price adjusting coefficient, and updating the price adjusting coefficient of the target push information at the current moment into the sum.

Alternatively, the adjustment module 1430 may obtain the second adjustment coefficient by:

acquiring a third price adjustment coefficient of the target push information, wherein the third price adjustment coefficient is the price adjustment coefficient of the target push information at the target moment of the previous period; scaling the third price adjustment coefficient according to a proportional relation between the expected conversion unit price in the previous period and the expected conversion unit price in the current period of the target push information; and determining the third adjusting coefficient subjected to the scaling processing as the second adjusting coefficient.

Optionally, the adjusting module 1430 may further be configured to: before weighted summation of the first price adjustment coefficient and the second price adjustment coefficient, obtaining a ratio of third accumulated push consumption of the target push information to the target consumption; if the ratio is smaller than 1, determining the ratio as the weight of the first price adjusting coefficient, and determining the difference value between 1 and the ratio as the weight of the second price adjusting coefficient; and if the ratio is not less than 1, determining the weight of the first adjusting coefficient as 1, and determining the weight of the second adjusting coefficient as 0.

Optionally, the manner of determining, by the determining module 1410, the first time period corresponding to the target push information may be:

determining a moment when the accumulated push consumption of the target push information in the previous period reaches a third proportion of the total push consumption in the previous period, and taking the determined moment as a first moment; determining a moment when the accumulated push consumption of the target push information in the previous period reaches a fourth proportion of the total push consumption in the previous period, and taking the determined moment as a second moment, wherein the fourth proportion is larger than the third proportion; determining a time period between the first time and the second time within the previous cycle as the first period.

Optionally, the target condition may be a first condition, and correspondingly, the determining module 1410 may be further configured to determine the target push information by:

acquiring the total pushing consumption, the conversion amount, the expected conversion unit price and the actual conversion unit price of each piece of pushing information in the previous period; and determining the target push information from the push information, wherein the total push consumption and the conversion amount in the previous period meet a second condition, and the actual conversion unit price does not exceed the expected conversion unit price.

It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described apparatuses and modules may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.

In the several embodiments provided in the present application, the coupling or direct coupling or communication connection between the modules shown or discussed may be through some interfaces, and the indirect coupling or communication connection between the devices or modules may be in an electrical, mechanical or other form.

In addition, functional modules in the embodiments of the present application may be integrated into one processing module, or each of the modules may exist alone physically, or two or more modules are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode.

Referring to fig. 15, a block diagram of a server according to an embodiment of the present application is shown. The server 100 in the present application may include one or more of the following components: a processor 110, a memory 120, and one or more applications, wherein the one or more applications may be stored in the memory 120 and configured to be executed by the one or more processors 110, the one or more programs configured to perform a method as described in the aforementioned method embodiments.

Processor 110 may include one or more processing cores. The processor 110 connects various parts within the overall server 100 using various interfaces and lines, performs various functions of the server 100 and processes data by executing or executing instructions, programs, code sets, or instruction sets stored in the memory 120, and calling data stored in the memory 120. Alternatively, the processor 110 may be implemented in hardware using at least one of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), and Programmable Logic Array (PLA). The processor 110 may integrate one or more of a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), a modem, and the like. Wherein, the CPU mainly processes an operating system, a user interface, an application program and the like; the GPU is used for rendering and drawing display content; the modem is used to handle wireless communications. It is understood that the modem may not be integrated into the processor 110, but may be implemented by a communication chip.

The Memory 120 may include a Random Access Memory (RAM) or a Read-Only Memory (Read-Only Memory). The memory 120 may be used to store instructions, programs, code sets, or instruction sets. The memory 120 may include a stored program area and a stored data area, wherein the stored program area may store instructions for implementing an operating system, instructions for implementing at least one function (such as a touch function, a sound playing function, an image playing function, etc.), instructions for implementing various method embodiments described below, and the like. The storage data area may also store data (such as push information) created by the terminal 100 in use, and the like.

Referring to fig. 16, a block diagram of a computer-readable storage medium according to an embodiment of the present application is shown. The computer readable medium 1600 has stored therein program code that can be invoked by a processor to perform the methods described in the method embodiments above.

The computer-readable storage medium 1600 may be an electronic memory such as a flash memory, an EEPROM (electrically erasable programmable read only memory), an EPROM, a hard disk, or a ROM. Optionally, the computer-readable storage medium 1600 includes a non-transitory computer-readable storage medium. The computer readable storage medium 1600 has storage space for program code 1610 for performing any of the method steps of the method described above. The program code can be read from or written to one or more computer program products. Program code 1610 may be compressed, for example, in a suitable form.

Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit the same; although the present application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not necessarily depart from the spirit and scope of the corresponding technical solutions in the embodiments of the present application.

Claims (12)

1. An information pushing method, comprising:

determining a first time period corresponding to target pushing information, wherein the change condition of pushing consumption of the target pushing information in the first time period of a previous period meets a target condition;

acquiring the accumulated pushing consumption of the target pushing information in a second period of the current cycle, wherein the second period of the current cycle is a time period from the starting time of the first period of the current cycle to the current time, and the current time is a time within the first period of the current cycle;

if the accumulated pushing consumption of the target pushing information in the second period of the current period does not reach the target consumption, increasing the current price adjustment coefficient of the target pushing information, and determining the current exposure price of the target pushing information based on the increased price adjustment coefficient;

and sequencing the current exposure prices of the target push information and other push information, and selecting information to be pushed from the target push information and other push information according to the sequencing for pushing.

2. The method of claim 1, wherein the target consumption is determined by:

acquiring the total pushing consumption of the target pushing information in the previous period, and determining a first proportion of the total pushing consumption as the expected consumption of the target pushing information in a first period of the current period;

acquiring a cumulative push consumption of the target push information in a first period of the previous cycle as a first cumulative push consumption, and acquiring a cumulative push consumption of the target push information in a second period of the previous cycle as a second cumulative push consumption, the second period of the previous cycle being a period from a start time of the first period of the previous cycle to a target time of the previous cycle, the target time being a time within the previous cycle corresponding to the current time;

determining a proportion of the second cumulative push consumption in the first cumulative push consumption, taking the determined proportion as a second proportion, and determining the expected consumption of the second proportion as the target consumption.

3. The method of claim 1 or 2, wherein prior to the increasing the tuning coefficient of the target push information, the method further comprises:

acquiring third accumulated push consumption of the target push information, wherein the third accumulated push consumption is accumulated push consumption of the target push information from the starting time of the current period to the current time;

if the third cumulative push consumption of the target push information is less than the target consumption, determining that the cumulative push consumption of the target push information in the second period of the current cycle does not reach the target consumption.

4. The method of claim 1 or 2, wherein the increasing the price adjustment coefficient of the target push information comprises:

acquiring a first price adjustment coefficient of the target pushing information, wherein the first price adjustment coefficient is the price adjustment coefficient of the target pushing information at the current moment;

and if the first price adjusting coefficient is not larger than the second price adjusting coefficient, weighting and summing the first price adjusting coefficient and the second price adjusting coefficient, and updating the price adjusting coefficient of the target push information at the current moment into the sum.

5. The method of claim 4, wherein the second tuning coefficient is obtained by:

acquiring a third price adjustment coefficient of the target push information, wherein the third price adjustment coefficient is the price adjustment coefficient of the target push information at the target moment of the previous period;

scaling the third price adjustment coefficient according to a proportional relation between the expected conversion unit price in the previous period and the expected conversion unit price in the current period of the target push information;

and determining the third adjusting coefficient subjected to the scaling processing as the second adjusting coefficient.

6. The method of claim 4, wherein prior to the weighted summation of the first and second pacing coefficients, the method further comprises:

obtaining a ratio of a third accumulated push consumption of the target push information to the target consumption;

if the ratio is smaller than 1, determining the ratio as the weight of the first price adjusting coefficient, and determining the difference between 1 and the ratio as the weight of the second price adjusting coefficient.

7. The method of claim 6, wherein prior to the weighted summation of the first and second pacing coefficients, the method further comprises:

and if the ratio is not less than 1, determining the weight of the first adjusting coefficient as 1, and determining the weight of the second adjusting coefficient as 0.

8. The method according to claim 1 or 2, wherein the determining a first time period corresponding to the target push information comprises:

determining a moment when the accumulated push consumption of the target push information in the previous period reaches a third proportion of the total push consumption in the previous period, and taking the determined moment as a first moment;

determining a moment when the accumulated push consumption of the target push information in the previous period reaches a fourth proportion of the total push consumption in the previous period, and taking the determined moment as a second moment, wherein the fourth proportion is larger than the third proportion;

determining a time period between the first time and the second time within the previous cycle as the first period.

9. The method according to claim 1 or 2, wherein the target condition is a first condition, and the target push information is determined by:

acquiring the total pushing consumption, the conversion amount, the expected conversion unit price and the actual conversion unit price of each piece of pushing information in the previous period;

and determining the target push information from the push information, wherein the total push consumption and the conversion amount in the previous period meet a second condition, and the actual conversion unit price does not exceed the expected conversion unit price.

10. An information pushing apparatus, comprising:

the device comprises a determining module, a judging module and a judging module, wherein the determining module is used for determining a first time period corresponding to target push information, and the change condition of the push consumption of the target push information in the first time period of the previous period meets a target condition;

an obtaining module, configured to obtain accumulated push consumption of the target push information in a second period of a current cycle, where the second period of the current cycle is a period from a start time of a first period of the current cycle to a current time, and the current time is a time within the first period of the current cycle;

the adjusting module is used for increasing the current price adjustment coefficient of the target pushing information when the accumulated pushing consumption of the target pushing information in the second period of the current period does not reach the target consumption, and determining the exposure price of the target pushing information based on the increased price adjustment coefficient;

and the pushing module is used for sequencing the current exposure prices of the target pushing information and other pushing information, and selecting information to be pushed from the target pushing information and other pushing information according to the sequencing for pushing.

11. A server, comprising:

one or more processors;

a memory;

one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs configured to perform the method of any of claims 1-9.

12. A computer-readable storage medium, characterized in that the computer-readable storage medium stores program code that can be invoked by a processor to perform the method according to any one of claims 1 to 9.

Images