CN113723983A - Information processing method and device, server and storage medium - Google Patents

Abstract

The embodiment of the application provides an information processing method and device, a server and a storage medium, relates to an artificial intelligence technology, and can improve the accuracy of pre-estimated exposure. The specific scheme comprises the following steps: counting exposure logs of the target advertisement space within a first preset time before the current time to obtain a plurality of estimated samples; obtaining an influence parameter influencing the exposure of the target advertisement space within a second preset time after the current moment; obtaining estimated exposure information within a second preset time length according to the estimated samples and the influence parameters; the estimated exposure information comprises estimated exposure of exposure objects with various attributes; responding to a plurality of received advertisement putting requests aiming at the target advertisement space, and obtaining exposure distribution information according to the estimated exposure information and the exposure requirement information included in each advertisement putting request; the exposure requirement information includes a target exposure amount of the exposure object of the target attribute; the exposure allocation information is used for indicating an advertisement placement strategy of a target advertisement space made for a plurality of advertisement placement requests.

Description

Information processing method and device, server and storage medium

Technical Field

The present application relates to the field of artificial intelligence technologies, and in particular, to an information processing method and apparatus, a server, and a storage medium.

Background

With the rise of personalized recommendation technology and artificial intelligence technology, information flow (Feeds) products such as short videos, information and social media are on the rise, and people spend more and more time on the information flow products. Thus, more and more advertisers select information streams for advertisement placement.

In order to improve the yield generated by advertisement putting, a platform side to which the information abortion products belong proposes a Guaranteed Delivery (GD) advertisement. The GD advertisement means that an advertiser and a platform party contract exposure amount, exposure object, price and the like of the advertisement in advance and contract. The platform side needs to play advertisements aiming at exposure objects required by advertisers, and reserves the exposure required by the advertisers, namely, the exposure of the advertisements is ensured.

In order to ensure the exposure of the advertisement, the platform side needs to estimate, search and lock the exposure available in the future. For this reason, the stage side may count the historical exposure for a period of time to estimate the exposure that will be available in the future, i.e., estimate the exposure. Then, the platform side can distribute the estimated exposure to the advertisements delivered by each advertiser according to the exposure required by each advertiser, and generate a distribution order of the estimated exposure. And finally, the platform side can provide the distribution order to the online release engine, and the online release engine finishes the amount preservation according to the distribution order.

However, the amount of exposure that can be provided in the future (i.e., the estimated amount of exposure) is estimated based on the historical amount of exposure over time, and the influence of the flow fluctuation of the stage on the amount of exposure over time in the future is not taken into consideration, reducing the accuracy of the estimated amount of exposure.

Disclosure of Invention

The embodiment of the application provides an information processing method and device, a server and a storage medium, which can improve the accuracy of the estimated exposure.

In order to achieve the technical purpose, the embodiment of the application adopts the following technical scheme:

in a first aspect, an embodiment of the present application provides an information processing method, where the method includes: and counting exposure logs of the target advertisement space within a first preset time before the current time to obtain a plurality of estimated samples. The exposure log records identification information of an exposure object of the target advertisement space, and the identification information is used for indicating the attribute of the exposure object. Each predictive sample of the plurality of predictive samples includes a historical exposure of the exposure object of one attribute divided by the identification information. And then obtaining the influence parameters influencing the exposure of the target advertisement space within a second preset time after the current moment. And then, obtaining estimated exposure information within a second preset time according to the plurality of estimated samples and the influence parameters. The estimated exposure information includes estimated exposure amounts of the exposure object of various attributes. Finally, responding to a plurality of received advertisement putting requests aiming at the target advertisement space, and obtaining exposure distribution information according to the estimated exposure information and the exposure requirement information included in each advertisement putting request; wherein the exposure requirement information includes a target exposure amount of the exposure object of the target attribute; the exposure allocation information is used for indicating an advertisement placement strategy of a target advertisement space made for a plurality of advertisement placement requests.

In one possible implementation, after obtaining the exposure allocation information according to the estimated exposure information and the exposure requirement information included in each advertisement placement request in response to receiving a plurality of advertisement placement requests for the target advertisement placement, the method further includes: and according to the exposure distribution information, delivering the advertisement to the target advertisement position through the online delivery engine.

In another possible implementation, the counting exposure logs of the target advertisement spots within a first preset time before the current time to obtain a plurality of estimated samples includes: and determining the attribute of each exposure object in the exposure log according to the preset object attribute and the identification information of the exposure object in the exposure log. Wherein the preset object attribute comprises at least one of: gender, age, address, operating system type, and date. And then counting the number of the exposure objects of each attribute in the exposure log to obtain the historical exposure of the exposure objects of each attribute. And finally, obtaining a plurality of estimated samples according to the historical exposure of the exposure object of each attribute.

In another possible implementation, the obtaining estimated exposure information within a second preset time period according to the plurality of estimated samples and the influence parameter includes: and performing vector conversion on the plurality of estimated samples to obtain a plurality of estimated vectors. The plurality of estimated vectors correspond to the plurality of estimated samples one by one, and each estimated vector in the plurality of estimated vectors is used for representing the corresponding estimated sample by adopting characters. And carrying out vector conversion on the influence parameters to obtain influence vectors. The influence vector is used for indicating whether each day in the second preset time length has an influence parameter. And taking the plurality of estimated vectors and the influence vectors as input, and operating a preset linear regression model to obtain the estimated exposure.

In another possible implementation, after obtaining exposure allocation information according to the estimated exposure information and the exposure requirement information included in each advertisement delivery request in response to receiving a plurality of advertisement delivery requests for a target advertisement slot, before delivering an advertisement to the target advertisement slot through an online delivery engine according to the exposure allocation information, the method further includes: and sending a confirmation message to the terminal. The confirmation message is used for indicating the estimated exposure amount allocated to the advertisement requested to be delivered by the terminal. And then receives the confirmation response from the terminal. The confirmation response is indicative of acceptance of the estimated exposure allocated for the advertisement requested for delivery by the terminal.

In another possible implementation, the delivering advertisements to the target advertisement spots through the online delivery engine according to the exposure allocation information includes: an exposure event for a target ad slot is received. The exposure event includes identification information of a currently exposed object of the target ad slot. And determining the attribute of the current exposure object according to the preset object attribute and the identification information of the current playing object. Wherein the preset object attribute comprises at least one of: gender, age, address, operating system type, and date. Then, according to the attribute of the current exposure object and the exposure distribution information, a target advertisement is selected from a plurality of advertisements to be delivered indicated by a plurality of advertisement delivery requests. And finally, delivering the target advertisement to the target advertisement position through an online delivery engine.

In another possible implementation, after selecting a target advertisement from a plurality of advertisements to be delivered indicated by a plurality of advertisement delivery requests according to the attribute of the current exposure object and the exposure allocation information, the method further includes: and acquiring at least one attribute of the current exposure object and at least one target advertisement in each preset period according to the preset period. And updating the estimated exposure information and the exposure requirement information according to the attribute of at least one current exposure object and at least one target advertisement. Then, based on the updated estimated exposure information and the updated exposure requirement information, the exposure distribution information is updated. And finally, delivering advertisements to the target advertisement positions through an online delivery engine according to the updated exposure distribution information.

In a second aspect, an embodiment of the present application further provides an information processing apparatus, including: and the log processing module is used for counting the exposure logs of the target advertisement positions within a first preset time before the current time to obtain a plurality of estimated samples. The exposure log records identification information of an exposure object of the target advertisement space, and the identification information is used for indicating the attribute of the exposure object; each predictive sample of the plurality of predictive samples includes a historical exposure of the exposure object of one attribute divided by the identification information. And the information acquisition module is used for acquiring influence parameters which influence the exposure of the target advertisement space within a second preset time after the current moment. And the exposure estimation module is used for obtaining estimated exposure information within a second preset time length according to the plurality of estimated samples and the influence parameters. The estimated exposure information includes estimated exposure amounts of the exposure object of various attributes. And the exposure distribution module is used for responding to a plurality of received advertisement putting requests aiming at the target advertisement positions and obtaining exposure distribution information according to the estimated exposure information and the exposure requirement information included by each advertisement putting request. Wherein the exposure requirement information includes a target exposure amount of the exposure object of the target attribute; the exposure allocation information is used for indicating an advertisement placement strategy of a target advertisement space made for a plurality of advertisement placement requests.

In one possible embodiment, the apparatus further comprises: and the advertisement delivery module is used for responding to a plurality of received advertisement delivery requests aiming at the target advertisement positions, obtaining exposure distribution information according to the estimated exposure information and the exposure requirement information included by each advertisement delivery request, and delivering advertisements to the target advertisement positions through the online delivery engine according to the exposure distribution information.

In another possible implementation manner, the log processing module is specifically configured to: and determining the attribute of each exposure object in the exposure log according to the preset object attribute and the identification information of the exposure object in the exposure log. Wherein the preset object attribute comprises at least one of: gender, age, address, operating system type, and date. And then counting the number of the exposure objects of each attribute in the exposure log to obtain the historical exposure of the exposure objects of each attribute. And finally, obtaining a plurality of estimated samples according to the historical exposure of the exposure object of each attribute.

In another possible implementation, the exposure estimation module is specifically configured to: and performing vector conversion on the plurality of estimated samples to obtain a plurality of estimated vectors. The plurality of estimated vectors correspond to the plurality of estimated samples one by one, and each estimated vector in the plurality of estimated vectors is used for representing the corresponding estimated sample by adopting characters. And then carrying out vector conversion on the influence parameters to obtain influence vectors. The influence vector is used for indicating whether each day in the second preset time length has an influence parameter. And then, taking the plurality of estimated vectors and the influence vectors as input, and operating a preset linear regression model to obtain the estimated exposure.

In another possible implementation manner, the exposure allocation module is further configured to send a confirmation message to the terminal after obtaining the exposure allocation information according to the estimated exposure information and the exposure requirement information included in each advertisement delivery request in response to receiving a plurality of advertisement delivery requests for the target advertisement placement and before delivering an advertisement to the target advertisement placement through the online delivery engine according to the exposure allocation information. The confirmation message is used for indicating the estimated exposure amount allocated to the advertisement requested to be delivered by the terminal. And then receives the confirmation response from the terminal. The confirmation response is indicative of acceptance of the estimated exposure allocated for the advertisement requested for delivery by the terminal.

In another possible implementation manner, the advertisement delivery module is specifically configured to: an exposure event for a target ad slot is received. The exposure event includes identification information of a currently exposed object of the target ad slot. And determining the attribute of the current exposure object according to the preset object attribute and the identification information of the current playing object. Wherein the preset object attribute comprises at least one of: gender, age, address, operating system type, and date. Then, according to the attribute of the current exposure object and the exposure distribution information, a target advertisement is selected from a plurality of advertisements to be delivered indicated by a plurality of advertisement delivery requests. And finally, delivering the target advertisement to the target advertisement position through an online delivery engine.

In another possible implementation manner, the exposure allocation module is further configured to, after selecting a target advertisement from the multiple advertisements to be delivered indicated by the multiple advertisement delivery requests according to the attribute of the current exposure object and the exposure allocation information, obtain, according to a preset period, at least one attribute of the current exposure object and at least one target advertisement in each preset period. And updating the estimated exposure information and the exposure requirement information according to the attribute of at least one current exposure object and at least one target advertisement. Then, based on the updated estimated exposure information and the updated exposure requirement information, the exposure distribution information is updated. And the advertisement putting module is also used for putting advertisements to the target advertisement positions through the online putting engine according to the updated exposure distribution information.

In a third aspect, an embodiment of the present application further provides a server, including: a processor and a memory for storing processor-executable instructions;

wherein the processor is configured to execute the instructions such that the server performs the information processing method as the first aspect and any one of its possible embodiments.

In a fourth aspect, the present application further provides a computer-readable storage medium, on which computer instructions are stored, where the computer instructions, when executed on a server, cause the server to perform an information processing method according to the first aspect and any possible implementation manner thereof.

In a fifth aspect, embodiments of the present application further provide a computer program product, which includes one or more instructions that can be executed on a server, so that the server executes an information processing method according to the first aspect and any possible implementation manner thereof.

It can be understood that, the method provided by the embodiment of the present application may count the exposure log of the target advertisement space within the first preset duration to obtain a plurality of estimated samples. Each pre-estimated sample in the plurality of pre-estimated samples comprises the historical exposure of the target advertisement space aiming at the exposure object with one attribute. That is, the historical exposure of an exposure object of an attribute, such as the number of exposure objects (i.e., users) who viewed the advertisement played by the target advertisement slot, can be determined from a predictive sample.

And then, obtaining an influence parameter in a second preset time length, and obtaining estimated exposure information in the second preset time length according to the estimated samples and the influence parameter. The estimated exposure information comprises estimated exposure of exposure objects with various attributes. That is to say, in the embodiment of the present application, when obtaining the estimated exposure information within the second preset duration, not only the historical exposure amount of the target advertisement space for the exposure objects with various attributes within the first preset duration is referred to, but also the influence parameter is referred to. The influence parameter may bring about flow fluctuation of the platform within a second preset time period; therefore, the exposure amount which is possibly generated by the target advertisement space in the second preset time period aiming at the exposure objects with various attributes is estimated by referring to the historical exposure amount of the target advertisement space aiming at the exposure objects with various attributes and the influence parameters in the second preset time period, and the accuracy of the estimated exposure amount can be improved.

Finally, according to exposure requirement information (namely the target exposure and the target attribute required by each advertisement putting request) included by each advertisement putting request in the plurality of advertisement putting requests and estimated exposure information (namely the exposure which may be generated by the target advertisement position aiming at the exposure objects with various attributes within the second preset time length), the exposure is distributed to each advertisement putting request, the completion degree of the distributed exposure can be improved, and the reliability of the exposure distribution information is further improved.

Drawings

Fig. 1 is a schematic diagram of an implementation environment related to an information processing method provided in an embodiment of the present application;

fig. 2 is a first flowchart of an information processing method provided in an embodiment of the present application;

fig. 3 is a structural diagram of an exposure prediction model according to an embodiment of the present application;

fig. 4 is a schematic diagram of exposure allocation information provided in an embodiment of the present application;

fig. 5 is a second flowchart of an information processing method according to an embodiment of the present application;

fig. 6 is a flowchart three of an information processing method according to an embodiment of the present application;

fig. 7 is a fourth flowchart of an information processing method provided in an embodiment of the present application;

fig. 8 is a fifth flowchart of an information processing method provided in an embodiment of the present application;

FIG. 9 is a schematic structural diagram of an information processing system according to an embodiment of the present application;

fig. 10 is a schematic structural diagram of an information processing apparatus according to an embodiment of the present application;

fig. 11 is a schematic structural diagram of a server according to an embodiment of the present application.

Detailed Description

In the following, the terms "first", "second" are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the present embodiment, "a plurality" means two or more unless otherwise specified.

The embodiment of the application provides an information processing method, and the accuracy of the estimated exposure can be improved through the method.

Embodiments of the present application will be described in detail below with reference to the accompanying drawings.

Please refer to fig. 1, which illustrates an implementation environment diagram of an information processing method according to an embodiment of the present application. As shown in fig. 1, the implementation environment may include a server 101 and a plurality of terminals, such as a terminal 102, a terminal 103, and the like.

The terminal 102 or the terminal 103 may be installed with various client applications, such as a video playing application, an information application, a social media application, and the like. Various client applications provide different information flow products to users, such as short videos, information, social information, and the like. As users spend more and more time on the information stream products, and the form of advertising interspersed in the information stream is more acceptable to the users. Thus, more and more advertisers have proposed placing advertisements in information flow products. The server 101 may provide a remote invocation interface to multiple advertisers to receive multiple ad placement requests issued by the multiple advertisers through the ad placement platform. And generating an advertisement putting strategy based on the plurality of advertisement putting requests, and putting advertisements requested by a plurality of advertisers in an information flow product of the terminal according to the advertisement putting strategy.

Specifically, the server 101 may be provided with an advertisement system, and the advertisement system executes the information processing method provided in the embodiment of the present application, and the advertisement system includes an offline estimation system and an online delivery engine. A user browses the client application by using the terminal 102 or the terminal 103, and when a target advertisement space in the client application is exposed, an exposure log is generated. The target ad slot may be any one of the ad slots in the client application. The offline prediction system may then collect a log of exposures for the targeted ad slots. The exposure log may be all exposure logs within a first preset time before the current time. And the off-line estimation system analyzes and processes the exposure log and generates exposure distribution information by combining the exposure requirement information of a plurality of advertisers. The exposure distribution information is used for indicating an advertisement putting strategy of the target advertisement space within a second preset time length after the current time. And finally, the off-line estimation system sends the exposure distribution information to an on-line delivery engine. The online delivery engine, upon detecting that a user is about to browse a target ad spot in a client application, selects an advertisement from a plurality of advertisements requested by a plurality of advertisers based on exposure allocation information and delivers to the target ad spot.

For example, the terminal in the embodiment of the present application may be a mobile phone, a tablet computer, a desktop computer, a laptop computer, a handheld computer, a notebook computer, a vehicle-mounted device, an ultra-mobile personal computer (UMPC), a netbook, a Personal Digital Assistant (PDA), an augmented reality device, a virtual reality device, and the like, and the embodiment of the present application does not particularly limit the specific form of the terminal.

It should be noted that the information processing method provided in the embodiment of the present application may be applied to the server 101. The executing body of the information processing method provided by the embodiment of the present application may also be an information processing apparatus, and the information processing apparatus may be the server 101 described above. The information processing apparatus may also be an Application (APP) installed with a function of providing information processing; alternatively, the information Processing apparatus may be a Central Processing Unit (CPU) in the server 101; or a control module in the server 101 for executing the information processing method.

It should be noted that the number of terminals and servers in fig. 1 is merely illustrative. There may be any number of terminals and servers, according to practical needs.

Currently, advertisements are mainly classified into effect advertisements and brand advertisements. An effect advertisement is an advertisement aimed at promoting sales or other consumer actions. Brand advertising is advertising that shapes the image of an enterprise with the goal of creating a certain idea. The brand advertisement comprises a GD advertisement, and the GD advertisement requires a platform side to which the advertiser and the information flow product belong to contract the exposure amount, the exposure object, the price and the like of the advertisement in advance and to sign a contract. The platform side needs to play advertisements aiming at exposure objects required by advertisers, and reserves the exposure required by the advertisers, namely, the exposure of the advertisements is ensured. For GD advertising, the platform side needs to estimate the amount of exposure that will be available in the future (i.e., estimate exposure) before contracting with the advertiser.

The related art proposes to obtain a historical exposure of an ad spot over a period of time. Then, summing and averaging the historical exposure to obtain the estimated exposure; or inputting historical exposure, running a prediction algorithm and outputting the prediction exposure. Because the above schemes do not consider the influence of the flow fluctuation of the platform on the exposure of the advertisement space in a future period of time, such as the increase of the platform flow caused by the influence factors such as holidays and the like, the accuracy of the estimated exposure is reduced. When the estimated exposure is far lower than the actual exposure, the advertisement space is wasted, and the advertisement income of the platform side is reduced. When the estimated exposure is higher than the actual exposure, the advertisement position is overdissured, the exposure required by an advertiser cannot be completed, and the reliability of the guaranteed amount is further reduced.

Secondly, the related technology also distributes the estimated exposure amount by using an HWM algorithm or a SHALE algorithm according to the exposure requirement information of a plurality of advertisers to generate exposure distribution information. And then, deducting the estimated exposure according to the exposure distribution information. And delivering the advertisements of the plurality of advertisers within a second preset time length after the current moment according to the exposure distribution information through the online delivery engine. Because the online putting condition in the second preset time length is not considered when the exposure distribution information is generated offline. And controlling the advertisement delivery within the second preset time according to the exposure distribution information, which can cause the advertisement missing delivery or the advertisement overtaking delivery, and can not complete the exposure required by the advertiser, thereby reducing the reliability of the guarantee quantity.

In summary, the accuracy of obtaining the estimated exposure is low, and the reliability of completing the exposure guarantee is also low. In contrast, the embodiment of the present application provides an information processing method, which can solve the above problems in the related art, improve the accuracy of the estimated exposure, complete the exposure required by the advertiser, and improve the reliability of the guaranteed amount.

Please refer to fig. 2, which is a flowchart illustrating an information processing method according to an embodiment of the present disclosure. As shown in fig. 2, the method may include steps 201-204.

Step 201: and counting exposure logs of the target advertisement space within a first preset time before the current time to obtain a plurality of estimated samples. The exposure log records identification information of an exposure object of the target advertisement space, the identification information indicating an attribute of the exposure object. Each predictive sample of the plurality of predictive samples includes a historical exposure of the exposure object of one attribute divided by the identification information.

The information processing device (such as a server) can receive exposure logs sent by a plurality of terminals within a first preset time period before the current time. And then filtering the information of each received exposure log to obtain the filtered exposure information and storing the filtered exposure information. And then, carrying out exposure statistics on all the filtered exposure information to obtain a plurality of estimated samples. Multiple pre-estimated samples may also be stored in a database. Wherein each prediction sample comprises an attribute and its corresponding historical exposure. The attributes are different between the plurality of predicted samples.

In some embodiments, the information processing apparatus may perform the information processing method for a client application that provides an ad slot. The target ad slot may be any one of the ad slots in the client application. And the information processing device receives and stores the exposure logs of the target advertisement space from all terminals running the client application until all the exposure logs within a first preset time length are obtained.

In some embodiments, the first preset duration may be set to be one week, one month, one quarter, or the like before the current time; the first preset time length can also be set according to a second preset time length to be estimated after the current time.

In some embodiments, each of the exposure logs includes an exposure time of the advertisement, identification information of the exposure object, and the like. The identification information of the exposure object may include, among others, identity information of the exposure object (e.g., sex, age, address, etc.), information of a terminal used by the exposure object (e.g., brand, operating system), and the like.

Illustratively, the content of an exposure log includes 19:30 on 01/02/2020, Zhang three browses advertisements on target spots using a mobile phone of the iOS system, identity information of Zhang three is male, 23 years old, and the address is the sunny district in Beijing City.

In some embodiments, the information processing apparatus determines the attribute of each exposure object in the exposure log according to the preset object attribute and based on the identification information of the exposure object in the exposure log. Wherein the preset object attribute comprises at least one of: gender, age, address, operating system type, and date. Then, the information processing apparatus may count the number of exposure objects of each attribute in the exposure log, obtaining the historical exposure amount of the exposure object of each attribute. The information processing apparatus may obtain a plurality of estimated samples from the historical exposure amount of the exposure object for each attribute.

The information processing device screens the identification information of the exposure object in each exposure log according to the preset object attribute to obtain the attribute of the exposure object of each exposure log. And then counting the number of the exposure objects of each attribute based on the attributes of the exposure objects of all the exposure logs, namely the historical exposure of the exposure objects of each attribute. And finally, forming a plurality of estimated samples by all attributes and the historical exposure of the exposure objects with all attributes.

The preset object attribute may include at least one of gender, age, address, operating system type and date, and may further include other attributes, such as marital status, fertility status, interest, occupation, and the like. Wherein, the age can refer to a plurality of age groups, the address can refer to a living city, and the date can refer to the year, month and day.

In some embodiments, the information processing apparatus may count only the number of exposure objects of one attribute that is identical among the attributes of the exposure objects of all the exposure logs, to obtain the historical exposure amount of the exposure object of the attribute.

Illustratively, one attribute is male, and another attribute is male and Beijing. When the information processing apparatus counts the number of exposure subjects corresponding to males, the information processing apparatus does not count the number of exposure subjects corresponding to males and Beijing. Similarly, when the number of exposure subjects corresponding to men and beijing is counted, the number of exposure subjects corresponding to men alone is not counted.

In some embodiments, the historical exposure may be a value in thousands set in terms of Cost of thousands (CPM), e.g., 20CPM, which represents a 20 x 1000 exposure count.

Illustratively, the preset object attributes include gender, 4 age groups, city, operating system type, and year, month, and day; the 4 age groups include 1-17, 18-25, 26-55, 56-80. The content of an exposure log comprises 19:30 on 01 d 19/02/2020, Zhang III browses advertisements on target advertisement spots by using a mobile phone of an iOS system, identity information of Zhang III is male, 23 years old, and the address is a sunny district in Beijing City. The information processing device filters the exposure log according to preset object attributes, and the obtaining of the attributes of the exposure object of the exposure log comprises the following steps: male, 18-25, beijing, iOS, 20200201. And counting the number of the attributes (male, 18-25, Beijing, iOS, 20200201) in the attributes of the exposure objects of all the exposure logs, namely the historical exposure of the exposure objects with the attributes. Suppose that an 18-25 year old male browses 100 x 1000 times a targeted ad slot in a client application on a terminal employing an iOS system the day 20200201. Then, the historical exposure of the exposure object for which such an attribute can be derived is 100 × 1000 or 100 CPM. Further, one predictor sample was obtained including male, 18-25, Beijing, iOS, 20200201 and 100 CPM. The estimated samples are shown in table 1 below:

TABLE 1

	Properties of the object of exposure	Historical exposure (in days)
			Prediction sample	Male, 18-25, Beijing, iOS, 20200201	100CPM

Step 202: and obtaining the influence parameters influencing the exposure of the target advertisement space within a second preset time after the current time.

The information processing device acquires the influence parameters of each day in a second preset time length. The impact parameters may include holiday information; the holiday information may be at least one of: statutory holiday information, network holiday information (e.g., network promotion days).

The second preset time length may include an advertisement playing time of the target advertisement position after the current time. For example, a week, a month, a quarter, etc. after the current time, the embodiments of the present application are not limited. The influence parameter may be information indicating whether or not there is a factor that influences the exposure amount of the target advertisement space, for example, holiday information may be information indicating whether or not each day is holiday.

For example, taking the holiday information as an example, the second preset duration is one week after the current time, and the holiday information of the week may include that the first day is a holiday, the second day is a holiday, the third day is a holiday, the fourth day is not a holiday, the fifth day is not a holiday, the sixth day is not a holiday, and the seventh day is not a holiday.

Step 203: obtaining estimated exposure information within a second preset time length according to the estimated samples and the influence parameters; the estimated exposure information includes estimated exposure amounts of the exposure object of various attributes.

The information processing device utilizes the plurality of estimated samples and the influence parameters to estimate the exposure of the exposure object with various attributes in the second preset duration to obtain estimated exposure information.

The estimated exposure information may include estimated exposure amounts of the exposure objects with multiple attributes for each day within a second preset time period.

In some embodiments, the information processing apparatus performs vector transformation on the plurality of predicted samples to obtain a plurality of predicted vectors; the plurality of estimated vectors correspond to the plurality of estimated samples one by one, and each estimated vector in the plurality of estimated vectors is used for representing the corresponding estimated sample by adopting characters; carrying out vector conversion on the influence parameters to obtain influence vectors; the influence vector is used for indicating whether each day has one influence parameter; and taking the plurality of estimated vectors and the influence vectors as input, and operating a preset linear regression model to obtain the estimated exposure of the exposure object with multiple attributes within a second preset time length.

The information processing device carries out vector conversion on each estimated sample in the plurality of estimated samples to obtain a plurality of estimated vectors. The plurality of estimated vectors may be vectors of the same dimension. And then, carrying out vector conversion on the influence parameters to obtain an influence vector. And then, taking the plurality of estimated vectors and the influence vectors as input, operating a preset linear regression model, and outputting the estimated exposure of the exposure object with various attributes within a second preset time length.

The influence parameter may be holiday information, and the information processing device performs vector conversion on the holiday information within the second preset duration to obtain a holiday vector. The holiday vector is used to indicate whether each day is a holiday. And taking the plurality of estimated vectors and the holiday and festival vectors as input, and operating a preset linear regression model to obtain the estimated exposure of the exposure object with multiple attributes within a second preset time length.

Illustratively, the second preset duration is one month after the current time. Outputting the estimated exposure amount of the exposure object with the plurality of attributes within the second preset time period may include: the estimated exposure amounts of the exposure subjects having the plurality of attributes on day 1, the estimated exposure amounts of the exposure subjects having the plurality of attributes on day 2, …, and the estimated exposure amounts of the exposure subjects having the plurality of attributes on day 30. The predicted exposures for the exposure subjects of the various attributes at day 1 may include: the estimated exposure for male, age 18 is 400CPM, the estimated exposure for male, Beijing, age 18 is 100CPM, the estimated exposure for female, age 25, iOS is 300CPM, and so on. The predicted exposure for the exposure subjects of the day 1 attributes is shown in table 2 below:

TABLE 2

Illustratively, the second preset time period includes 11 months and 11 days (i.e. twenty-one), and the predicted exposure amount of the exposure object with the plurality of attributes of twenty-one may include: the estimated exposure for male, 25 years old is 400CPM, the estimated exposure for male, Beijing, 18 years old is 600CPM, and the estimated exposure for female, 25 years old, iOS is 1000000CPM, and so on. The predicted exposure for the exposure subject for the various attributes of dieleven is shown in table 3 below:

TABLE 3

It can be seen that, since the twenty one is a shopping hasty festival, the estimated exposure of the exposure object of each attribute of the day of twenty one is improved, and particularly, the exposure of women is improved greatly.

It is understood that, when obtaining the estimated exposure information within the second preset time period, the information processing apparatus uses the influence parameter within the second preset time period in addition to the plurality of estimated samples. The influence parameter may bring about flow fluctuation of the platform within a second preset time period; therefore, the exposure amount which is possibly generated by the target advertisement space in the second preset time length aiming at the exposure objects with various attributes is estimated by referring to the historical exposure amount of the target advertisement space aiming at the exposure objects with various attributes and the influence parameters in the second preset time length, so that the accuracy of the estimated exposure amount of the exposure objects with various attributes can be improved.

In some embodiments, the information processing apparatus may perform vector conversion on each of the plurality of predicted samples according to an order corresponding to the preset object attribute and a character corresponding to the preset object attribute to obtain a plurality of predicted vectors. And carrying out vector conversion on the influence parameters according to characters corresponding to the preset influence parameters to obtain an influence vector. The characters corresponding to the preset object attributes may include a plurality of numeric values, a plurality of pinyins or a plurality of letters corresponding to each attribute and representing different conditions. The characters corresponding to the preset influence parameters may include numerical values, pinyin or letters indicating whether one influence parameter exists, for example, 1 indicates that one influence parameter exists, and 0 indicates that one influence parameter does not exist.

The influence parameter may be holiday information, and the information processing device may perform vector conversion on the holiday information according to a character corresponding to preset holiday information to obtain a holiday vector. The characters corresponding to the preset holiday information may include a plurality of numeric values, a plurality of pinyins, or a plurality of letters indicating whether the holiday is, for example, 1 indicates holiday and 0 indicates not holiday.

Illustratively, the preset object attribute may correspond to a sequence of gender, age group, address, operating system type and date. The characters corresponding to the preset object attributes may include: the numerical value corresponding to the gender is 1 and 2, the numerical value corresponding to the age group is 1-4, and the pinyin corresponding to the address is the abbreviation of each province of China. Wherein 1 indicates male and 2 indicates female according to sex. The corresponding values 1-4 for the age groups indicate 1-17 years, 18-25 years, 26-55 years, and 56-80 years, respectively. The pinyin corresponding to the address comprises Jing, E, Qin and the like, wherein the Jing represents Beijing, the E represents Hubei, and the Qin represents Shaanxi.

Illustratively, the content of a pre-estimated sample includes: male, 18-25, Beijing, iOS, 20200201 and 100 CPM. The estimated vector obtained by vector conversion of the estimated sample may be {1, 2, Jing, 1, 20200201, 100 }. Wherein the number from left to right in the prediction vector is 1 for male, 2 for 18-25 years old, jin for Beijing, 1 for iOS, 20200201 for time, and 100 for exposure in thousand units.

It should be noted that, when a prediction sample includes a part of object attributes in the preset object attributes, and vector transformation is performed on the prediction sample, the object attribute that does not exist in the prediction sample is marked as 0. For example, the content of a pre-estimated sample includes: male, iOS, 20200201 and 100 CPM. The estimated vector obtained by vector-converting the estimated sample may be {1, 0, 0, 1, 20200201, 100 }. Wherein, the number from left to right in the estimated vector is 1 for male, 0 for unknown age group, 0 for unknown address, 1 for iOS, 20200201 for time, and 100 for exposure in thousand units.

Illustratively, the holiday information for one month may include that the first day is a holiday, the second day is a holiday, …, and the thirty th day is not a holiday. The holiday vector obtained by vector converting the holiday information may be {1,1, …, 0}, and the number of elements of the holiday vector is equal to 30.

In some embodiments, the information processing apparatus combines each of the plurality of eigenvectors and the influence vector to obtain a plurality of combined vectors. The plurality of combined vectors and the plurality of estimated vectors correspond one to one. Each of the plurality of combination vectors is used to represent the prediction samples and the impact parameters using characters. And taking the plurality of combined vectors as input, operating a preset linear regression model, and outputting the estimated exposure of the exposure object with multiple attributes within a second preset time length. The preset linear regression model has the capability of predicting the estimated exposure of the exposure object with multiple attributes within the second preset duration according to the multiple combined vectors.

For example, the influence parameter may be holiday information, and the influence vector may be a holiday vector, as exemplified by the above prediction vector {1, 2, sting, 1, 20200201, 100}, holiday vector {1,1, …, 0 }. And combining the preset vector and the holiday vector to obtain a combined vector of {1, 2, sting, 1, 20200201, 100, 1,1, …, 0 }.

It should be noted that the predictor vector in the combined vector may be before the influence vector or after the influence vector, and the embodiment of the present application is not limited.

In some embodiments, the information processing apparatus may include an exposure prediction model. The information processing device takes a plurality of estimated samples and the influence parameters as input, operates the exposure estimation model and outputs estimated exposure information. The exposure estimation model has the capability of generating estimated exposure information based on the estimation sample and the influence parameters.

Specifically, the influence parameter may be holiday information, and the information processing device operates the exposure estimation model with a plurality of estimation samples and holiday information as inputs, and outputs estimated exposure information. The exposure prediction model has the capability of generating prediction exposure information based on prediction samples and holiday information.

In some embodiments, before step 203, the information processing apparatus obtains an exposure log of the target advertisement spots within a third preset duration, i.e., the first exposure log. And then, counting the first exposure logs to obtain a plurality of first estimated samples. And acquiring an exposure log of the target advertisement position within a fourth preset time length, namely a second exposure log. And then, counting the second exposure logs to obtain a plurality of second estimated samples. Influence parameters (e.g., holiday and festival information) within a fourth preset duration are also acquired. And the third preset time length is before the fourth preset time length, and the fourth preset time length is before the current moment. Then, the information processing device takes the plurality of first estimated samples and the influence parameters (such as holiday information) within a fourth preset time length as input samples, takes the plurality of second estimated samples as output samples, trains the initial exposure estimated model, and obtains the exposure estimated model.

The initial exposure estimation model may be a recurrent neural network, a Long Short-Term Memory (LSTM) network, or an Autoregressive model with differential Integrated Moving Average (ARIMA). Tests show that the exposure estimation model obtained by training the LSTM is superior to the exposure estimation model obtained by training the ARIMA model. And the LSTM can solve the long-term dependence problem of the recurrent neural network. Therefore, the exposure estimation model obtained by LSTM training can be adopted in the embodiment of the application.

Illustratively, referring to fig. 3, the exposure prediction model 3 trained using LSTM may include an input layer 31, a hidden layer 32, and an output layer 33. The input layer 31 is connected to the hidden layer 32, the hidden layer 32 is connected to the tandem operation, and the tandem operation is connected to the output layer 33. Wherein the input layer 31 has the ability to obtain a plurality of pre-estimated samples and influencing parameters (e.g. holiday information). The concealment layer 32 has the ability to vector convert each of the plurality of pre-estimated samples and the impact parameters (e.g., holiday information). The concatenation operation may include merging the plurality of predictor vectors and the influence vector (e.g., a holiday-festival vector) in a predetermined order. The predetermined sequence includes either the predictor vector preceding the influence vector or the influence vector preceding the predictor vector. The output layer 33 has the ability to derive an estimated exposure based on a plurality of combined vectors. The output layer 33 includes a linear regression model.

Specifically, the information processing apparatus inputs a plurality of estimated samples and influence parameters (for example, holiday information), and operates the input layer 31, the hidden layer 32, and the output layer 33 at the same time, to obtain estimated exposure information within a second preset time period.

It should be noted that the process of acquiring the first exposure log and the second exposure log by the information processing apparatus is the same as the process of acquiring the exposure log within the first preset duration, and details are not repeated here. Secondly, the process of obtaining a plurality of first estimated samples by counting the first exposure log and the process of obtaining a plurality of second estimated samples by counting the second exposure log are also the same as the process of obtaining a plurality of estimated samples by counting the exposure log within the first preset time duration, and the description is omitted here. Finally, the process of obtaining the impact parameter (e.g., holiday information) in the fourth preset duration is the same as the process of obtaining the impact parameter (e.g., holiday information) in the second preset duration, and is not described herein again.

In some embodiments, after step 203, the information processing apparatus acquires a new exposure log in each log update period in accordance with a preset log update period. And (5) counting the new exposure log to obtain a new estimated sample. And replacing the estimated sample with the new estimated sample in the plurality of estimated samples, which corresponds to the earliest exposure time and is in the same duration, to obtain a plurality of updated estimated samples. And acquiring new influence parameters (such as new holiday information) within the same time length as the preset log updating period after the second preset time length. And replacing the influence parameter which corresponds to the earliest time and is in the same time length in the influence parameters by the new influence parameter to obtain the updated influence parameter. And obtaining updated estimated exposure information according to the updated estimated samples and the updated influence parameters.

The preset log updating period may be one day, one week, etc.

Illustratively, the first predetermined period is from 1 month, 1 day, to 1 month, 10 days. The second preset time period is from 11 days in 1 month to 20 days in 1 month. The preset logging period is one day. The information processing apparatus acquires a new exposure log within 16 days of 1 month. And counting the new exposure log to obtain a new estimated sample. And replacing the estimated samples corresponding to 1 month and 1 day in the plurality of estimated samples by the new estimated samples. New holiday information is acquired within 21 days of 1 month. And replacing the holiday information corresponding to the 1 month and 11 days in the holiday information by the new holiday information to obtain updated holiday information. And obtaining updated estimated exposure information within 12 days in 1 month to 21 days in 1 month according to the updated estimated samples and the updated holiday information.

It should be noted that the process of acquiring a new exposure log in each log update period by the information processing apparatus is the same as the process of acquiring an exposure log in the first preset duration in step 201. The process of counting new exposure logs to obtain new estimated samples is the same as the process of counting new exposure logs to obtain multiple estimated samples in step 201. Obtaining the updated estimated exposure information according to the updated estimated samples and the updated influence parameters, which is the same as the process of step 203. Are not described in detail herein.

It should be noted that, after obtaining the updated estimated exposure information, the information processing apparatus replaces the estimated exposure information with the updated estimated exposure information, and executes the relevant steps.

It can be understood that, the method provided in the embodiment of the present application further updates the estimated exposure information according to a preset log update period (for example, one day), so as to further improve the accuracy of the estimated exposure information. And then, the exposure is distributed based on more accurate estimated exposure information, so that the utilization rate of the target advertisement space can be improved.

Step 204: responding to a plurality of received advertisement putting requests aiming at the target advertisement space, and obtaining exposure distribution information according to the estimated exposure information and the exposure requirement information included in each advertisement putting request; wherein the exposure requirement information includes a target exposure amount of the exposure object of the target attribute; the exposure allocation information is used for indicating an advertisement placement strategy of a target advertisement space made for a plurality of advertisement placement requests.

The information processing device acquires a plurality of exposure request information from a plurality of advertisement placement requests when receiving the plurality of advertisement placement requests for the target advertisement placement sent by a plurality of advertisers from the advertisement placement platform. The plurality of advertisement delivery requests and the plurality of exposure requirement information correspond to one another. And obtaining exposure distribution information according to the estimated exposure information and the plurality of exposure requirement information.

Wherein, the exposure allocation information is characterized by the exposure amount and/or the exposure sequence allocated by the plurality of advertisement putting requests, etc.

In some embodiments, the plurality of exposure requirement information may include a plurality of target attributes and a plurality of target exposures. The plurality of exposure requirement information, the plurality of target attributes, and the plurality of target exposures are in one-to-one correspondence. Each target attribute of the plurality of target attributes includes at least one attribute. The information processing device utilizes various target attributes to inquire the estimated exposure information to obtain the undetermined exposure information. The pending exposure information includes exposure information having attributes including any one of a plurality of target attributes. And generating exposure distribution information according to the information of the undetermined exposure, the attributes of the various targets and the exposure of the various targets.

In some embodiments, the plurality of exposure requirement information may further include a plurality of target exposure time periods. The plurality of exposure requirement information and the plurality of target exposure durations correspond one to one. The plurality of target exposure time periods may be equal. The information processing device can utilize various target attributes and various target exposure durations to inquire the estimated exposure information to obtain the undetermined exposure information. The undetermined exposure information comprises exposure information with the attribute including any one of the multiple target attributes in the multiple target exposure durations.

Wherein each of the plurality of target exposure durations may be equal to or within a second preset duration.

In some embodiments, the information processing apparatus may generate the exposure allocation information using the intended exposure information, the plurality of target attributes, and the plurality of target exposures using a HWM algorithm or a SHALE algorithm. Wherein, the HWM algorithm is more accurate and mature compared with the SHALE algorithm.

In some embodiments, when the information processing apparatus generates the exposure allocation information by using the HWM algorithm, a plurality of total estimated exposures corresponding to a plurality of target attributes are determined from the information to be determined. And calculating the total estimated exposure and the target exposure to obtain a plurality of supply ratios corresponding to the advertisement putting requests. The plurality of supply ratios are ranked to obtain a priority order. And obtaining a plurality of distribution service rates according to the priority order, the information of the undetermined exposure, the attributes of the various targets and the exposure of the various targets. The exposure allocation information is generated from a plurality of allocation service rates and a plurality of target attributes and a plurality of target exposure amounts corresponding thereto.

The information processing device searches the exposure amount of the exposure object with the corresponding attribute including each target attribute from the undetermined exposure information according to each target attribute in the multiple target attributes, and sums the exposure amounts to obtain the total estimated exposure amount corresponding to each target attribute. And substituting the total estimated exposure corresponding to each target attribute and the target exposure corresponding to each target attribute into a supply ratio calculation model, outputting a supply ratio corresponding to one advertisement putting request, and further obtaining a plurality of supply ratios corresponding to a plurality of advertisement putting requests. The supply ratio calculation model has the capability of calculating the supply ratio for the total estimated exposure amount and the target exposure amount. Then, a plurality of supply ratios corresponding to the plurality of advertisement placement requests may be sorted in ascending order to obtain a priority order. And calculating the exposure information to be determined, the multiple target attributes and the multiple target exposure according to the priority order to obtain multiple distribution service rates. Wherein, a lower supply ratio of an advertisement placement request indicates that the advertisement placement request needs to be processed earlier, i.e. the advertisement of the advertisement placement request is shown earlier.

In some embodiments, the supply ratio calculation model is represented by the following equation (1):

Yj＝Sj/Dj，j∈J (1)

where J represents a set of a plurality of ad placement requests, and J represents one of the plurality of ad placement requests. Sj represents the total estimated exposure corresponding to the target attribute in the advertisement delivery request. Dj represents the target exposure in this one advertisement placement request. Yj represents the serving rate of this one ad placement request.

In some embodiments, the information processing apparatus may further obtain a plurality of request times from a plurality of advertisement placement requests. The plurality of advertisement delivery requests correspond to the plurality of request times one to one. And sequencing the multiple request moments to obtain a priority order. The earlier the request time of an advertisement placement request is, the earlier the advertisement placement request is processed, that is, the earlier the advertisement of the advertisement placement request is displayed.

In some embodiments, the information processing apparatus initializes the remaining exposure amounts of the exposure object of all attributes in the pending exposure information in accordance with the initialization model. And then according to the priority order, for the advertisement putting request with the lowest supply ratio, determining the estimated exposure amount of which the corresponding attribute comprises the target attribute in the advertisement putting request from the undetermined exposure information, and determining the residual exposure amount of which the corresponding attribute comprises the target attribute in the advertisement putting request from the residual exposure amounts of the exposure objects with all attributes. And inputting the determined estimated exposure, the determined residual exposure and the target exposure in the advertisement putting request into a service rate calculation model to obtain the distribution service rate corresponding to the advertisement putting request. The service rate calculation model has the capability of calculating the distribution service rate of the estimated exposure, the residual exposure and the target exposure. And then inputting the determined estimated exposure, the determined residual exposure, the target exposure in the advertisement putting request and the distribution service rate corresponding to the advertisement putting request into a residual exposure updating model to obtain the updated residual exposure. The residual exposure update model has the capability of updating the residual exposure with the estimated exposure, the residual exposure, the target exposure, and the distribution service rate. And continuing to determine the estimated exposure amount with the corresponding attribute being the same as the target attribute in the advertisement putting request from the undetermined exposure information for the advertisement putting request with the supply rate being lower, and determining the residual exposure amount with the corresponding attribute being the same as the target attribute in the advertisement putting request from the updated residual exposure amount according to the priority order until the distribution service rate corresponding to each of the plurality of advertisement putting requests is obtained.

In some embodiments, the initialization model is shown in equation (2) below:

Ri＝Si，i∈I (2)

the service rate calculation model is shown in the following formula (3):

the remaining exposure amount update model is as shown in the following equation (4):

Ri＝Ri-min{Ri,Si*aj}，i∈Γ(j) (4)

wherein, I represents the set of all attributes in the undetermined exposure information, and I represents one attribute in the undetermined exposure information. Ri is the residual exposure of the exposure object with one attribute in the information of the undetermined exposure. Si is the estimated exposure of an exposure object with one attribute in the undetermined exposure information. aj is the allocated service rate corresponding to an advertisement putting request. Γ (j) represents the set of all attributes in I that include the target attribute in an ad placement request j.

Illustratively, a schematic diagram of exposure allocation information is shown in fig. 4. The information processing apparatus receives 3 advertisement delivery requests, which correspond to 3 blocks in fig. 4, respectively. The exposure requirement information in the 1 st advertisement placement request (j ═ 1) includes: the target attributes are { male }, and the target exposure amount D1 ═ 200K (200 × 1000). The exposure requirement information in the 2 nd advertisement placement request (j ═ 2) includes: the target attribute is { CA }, and the target exposure amount D2 is 200K. Where CA denotes an address. The exposure requirement information in the 3 rd advertisement placement request (j ═ 3) includes: the target attribute is { Age ═ 5}, and the target exposure amount D3 ═ 1M (1 × 10 } ⁶). Determining pending exposure information including any target attribute in the 3 advertisement putting requests from the estimated exposure informationThere are 6 types, which correspond to 6 circles in fig. 4, respectively, and the 1 st type of information to be exposed (i ═ 1) has an attribute of { male, Age ═ 5} and its corresponding estimated exposure amount S1 ═ 400K (400 × 1000); the 2 nd pending exposure information (i ═ 2) is given by the attribute { male, WA, Age ═ 5} and the corresponding estimated exposure amount S2 ═ 400K; the 3 rd pending exposure information (i ═ 3) is given by the attribute { male, CA, Age ═ 5} and its corresponding estimated exposure amount S3 ═ 100K (100 × 1000); the 4 th type of pending exposure information (i ═ 4) is given with an attribute of { CA, Age ═ 5} and its corresponding estimated exposure amount S4 ═ 100K; the 5 th type of pending exposure information (i ═ 5) is given by the attribute { NV, Age ═ 5} and the corresponding estimated exposure amount S5 ═ 500K (500 × 1000); the 6 th type of pending exposure information (i ═ 6) has an attribute of { Age ═ 5} and its corresponding estimated exposure amount S6 ═ 300K (300 × 1000). Wherein each box representing an ad placement request is connected to all circles representing attributes including the target attribute. Age 5 means Age 5. WA, CA, and NV represent addresses.

First, the information processing apparatus calculates a supply ratio for each of 3 advertisement placement requests using a supply ratio calculation model, the supply ratio calculation model for the 3 advertisement placement requests being expressed by the following expression (5):

Here, the supply ratio Y1 of the 1 st ad placement request is 4.5, the supply ratio Y2 of the 1 st ad placement request is 1, and the supply ratio Y3 of the 1 st ad placement request is 1.8. The supply ratios Y1, Y2, and Y3 are sorted in ascending Order, and the Order of priority of the 1 st ad placement request is Order 1-2, the Order of priority of the 2 nd ad placement request is Order 2-1, and the Order of priority of the 3 rd ad placement request is Order 3-3.

Next, the information processing apparatus initializes the remaining exposure amount of the exposure target of 6 attributes according to an initialization model shown by the following expression (6):

then, the information processing apparatus determines, for the 2 nd advertisement placement request, the estimated exposure amount whose corresponding attribute includes { CA } from the estimated exposure amounts of the exposure objects of the 6 kinds of attributes in order of priority, and the determined estimated exposure amounts include S3 and S4. The remaining exposure amounts corresponding to the attributes including { CA } are determined from the remaining exposure amounts of the exposure subjects of the 6 attributes, and the determined remaining exposure amounts include R3 and R4. Inputting S3, S4, R3, R4 and D2 into a service rate calculation model, which is shown in the following formula (7), to obtain an allocated service rate a2 corresponding to the 2 nd ad placement request as 1:

D2＝min(R3,S3*a2)+min(R4，S4*a2) (7)

The S3, S4, R3, R4, D2 and a2 are input to a residual exposure amount update model, which is expressed by the following formula (8), to obtain updated residual exposure amounts R3 and R4:

then, the information processing apparatus executes the above-described procedure for the 2 nd advertisement placement request for the 1 st advertisement placement request and the 3 rd advertisement placement request in order of priority, and obtains an allocation service rate a 1-1/4 corresponding to the 1 st advertisement placement request and an allocation service rate a 3-5/8 corresponding to the 3 rd advertisement placement request.

Finally, the exposure assignment information generated by the information processing apparatus includes: the target attributes of the 1 st advertisement placement request are { male }, the target exposure D1 is 200K, and the distribution service rate a1 is 1/4. The target attribute of the 2 nd advertisement placement request (j ═ 2) is { CA }, the target exposure amount D2 is 200K, and the distribution service rate a2 is 1. The target attribute of the 3 rd advertisement placement request is { Age ═ 5}, the target exposure amount D3 is 1M, and the distribution service rate a3 is 5/8.

In some embodiments, referring to fig. 5, after step 204, the information processing method further includes:

step 205: and according to the exposure distribution information, delivering the advertisement to the target advertisement position through the online delivery engine.

The information processing apparatus includes an online delivery engine. And delivering advertisements requested by a plurality of advertisers to the target advertisement positions within a second preset time length through the online delivery engine according to the exposure distribution information. Wherein the advertisement requested by each of the plurality of advertisers may be a brand advertisement.

In some embodiments, referring to FIG. 6, step 205 comprises steps 205 a-205 d.

Step 205 a: and receiving an exposure event of the target advertisement space, wherein the exposure event comprises identification information of a current exposure object of the target advertisement space.

An online delivery engine in an information processing device receives an exposure event of a target advertisement space from a terminal. Wherein, when receiving the exposure event of the target advertisement space, it indicates that the current exposure object (or user) is about to browse the target advertisement space.

When a user runs a client application through a terminal and enters a page where a target advertisement position is located, the user can be regarded as an exposure event. Or, when the user runs the client application through the terminal, enters the page where the target advertisement position is located, and the page is scrolled to a specified position above the target advertisement position, the page can be regarded as an exposure event. And then, the user is used as a current exposure object, and the terminal records the identification information of the current exposure object in an exposure event and sends the identification information to the online release engine.

Step 205 b: and determining the attribute of the current exposure object according to the preset object attribute and the identification information of the current playing object.

And the information processing device screens the identification information of the current exposure object in the exposure event according to the preset object attribute to obtain the attribute of the current exposure object corresponding to the exposure event. Wherein the preset object attribute may include at least one of: gender, age, address, operating system type and date, etc.

It should be noted that, the information processing apparatus performs a process of screening the identification information of the current exposure object in the exposure event according to the preset object attribute to obtain the attribute of the current exposure object, and similarly, performs a process of screening the identification information of the exposure object in each exposure log according to the preset object attribute to obtain the attribute of the exposure object in each exposure log, which is not described herein again.

Step 205 c: and selecting a target advertisement from a plurality of advertisements to be delivered indicated by the plurality of advertisement delivery requests according to the attribute of the current exposure object and the exposure distribution information.

The exposure allocation information includes a target attribute, a target exposure amount, and an allocation service rate corresponding to each advertisement placement request of the plurality of advertisement placement requests. The information processing device selects at least one advertisement placement request having the same target attribute as the attribute of the currently exposed object from the plurality of advertisement placement requests. And determining a target advertisement putting request from the at least one advertisement putting request according to the respective corresponding distribution service rate of the selected at least one advertisement putting request. And then, acquiring the target advertisement corresponding to the target advertisement putting request. The targeted advertisement may be a brand advertisement.

The information processing device may determine, from the at least one advertisement placement request, a target advertisement placement request with a maximum distribution service rate according to a distribution service rate corresponding to each of the selected at least one advertisement placement request.

Step 205 d: and delivering the target advertisement to the target advertisement position through the online delivery engine.

An online delivery engine in the information processing device transmits the target advertisement to the terminal so that the terminal displays the target advertisement on the target advertisement slot.

In some embodiments, referring to fig. 7, after step 205c, the information processing method further includes steps 205e to 205 h.

Step 205 e: and acquiring at least one attribute of the current exposure object and at least one target advertisement in each preset period according to the preset period.

The information processing device firstly obtains the attributes of all current exposure objects and all target advertisements in each preset period according to the preset period.

Wherein the preset period may be set to one hour, one day, etc. Each preset period belongs to a second preset duration.

Illustratively, the preset period is one hour. The information processing apparatus acquires attributes of all currently exposed objects, all targeted advertisements, per hour.

Step 205 f: and updating the estimated exposure information and the exposure requirement information according to the attribute of at least one current exposure object and at least one target advertisement.

The information processing apparatus counts the total number of attributes of each of the current exposure objects among the attributes of all the current exposure objects and the total number of each of the target advertisements among all the target advertisements. Wherein, the total number of the attributes of each kind of current exposure object represents the exposure amount of the exposure object of the attributes, which is provided by the target advertisement space in each preset period. The total number of each kind of target advertisement represents the exposure of the target advertisement on the target advertisement position in each preset period. And then, subtracting the total number of the corresponding attributes from the estimated exposure in the estimated exposure information to obtain the updated preset exposure. And subtracting the total number of the corresponding target advertisements from the target exposure amount in the exposure requirement information to obtain updated exposure requirement information.

Step 205 g: and updating the exposure distribution information based on the updated estimated exposure information and the updated exposure requirement information.

The information processing device updates the exposure allocation information based on the updated estimated exposure information and the updated exposure request information, and generates updated exposure allocation information.

In some embodiments, the information processing apparatus generates updated exposure assignment information according to the updated estimated exposure information and the updated exposure requirement information, similarly to the process of generating exposure assignment information according to the estimated exposure information and the exposure requirement information in step 204. Are not described in detail herein.

It can be understood that, the method provided by the embodiment of the present application updates the estimated exposure information and the exposure requirement information by using the attributes of all current exposure objects and all target advertisements in each preset period. The attributes of all current exposure objects and all target advertisements in each preset period respectively reflect the exposure event which actually occurs and the advertisements which are actually delivered, namely the online delivery condition. That is, the updated estimated exposure information and the updated exposure requirement information are the latest information after the online release in the preset period. And then, updating exposure distribution information formulated for the plurality of advertisement putting requests by using the updated estimated exposure information and the updated exposure requirement information. The updated exposure distribution information refers to the online release condition in the preset period, and the influence of the flow fluctuation of the platform in the preset period on the distribution information can be avoided. Finally, the updated exposure distribution information is used for controlling advertisement putting, so that the missing putting or the excess putting of the advertisement can be reduced, and the reliability of the guarantee amount is improved.

In some embodiments, the information processing apparatus may further obtain a new advertisement placement request for the target advertisement space every preset period. And obtaining updated exposure distribution information according to the updated estimated exposure information, the updated exposure requirement information and the exposure requirement information included in the new advertisement putting request. The new advertisement placement request may include one advertisement placement request or a plurality of advertisement placement requests.

It should be noted that, the information processing apparatus obtains updated exposure allocation information according to the updated estimated exposure information, the updated exposure request information, and the exposure request information included in the new advertisement delivery request. The process of generating the exposure distribution information according to the estimated exposure information and the exposure requirement information in step 204 is the same. Are not described in detail herein.

It is to be understood that, after the information processing apparatus updates the estimated exposure information and the exposure requirement information, a new advertisement placement request in each preset period is also acquired. And adding the exposure requirement information included in the new advertisement putting request into the updated estimated exposure information and the updated exposure requirement information to generate updated exposure distribution information. The updated exposure allocation information allocates exposure to the original plurality of advertisement placement requests and the new advertisement placement request. Therefore, the utilization rate of the target advertisement space is improved, or the income of a platform side to which the target advertisement space belongs is improved.

Step 205 h: and delivering advertisements to the target advertisement space through the online delivery engine according to the updated exposure distribution information.

It should be noted that the process of step 205h is the same as the process of steps 205a to 205d, and is not described herein again.

It should be noted that step 205d and steps 205 e-205 h may be performed simultaneously. Alternatively, step 205d may be performed first, and then steps 205 e-205 h may be performed.

In some embodiments, referring to fig. 8, after step 204 and before step 205, the information processing method further includes: step 206 and step 207.

Step 206: and sending a confirmation message to the terminal, wherein the confirmation message is used for indicating the exposure amount allocated to the advertisement requested to be delivered by the terminal.

The information processing apparatus transmits a confirmation message to the terminal. Wherein, this terminal station can include: the advertisement ordering platform is used for receiving the advertisement putting request of each advertiser, and the terminal specified by each advertiser in the advertisement putting request. The advertisement requested to be delivered by the terminal may be an advertisement requested to be delivered by each advertiser in the advertisement delivery request.

In some embodiments, the acknowledgement message may include: the amount of exposure allocated for the advertisement requested to be delivered by the terminal, a second preset time period, an allocation service rate, and the like.

Step 207: and receiving a confirmation response from the terminal, wherein the confirmation response is used for indicating that the estimated exposure amount allocated for the advertisement requested to be delivered by the terminal is accepted.

The advertiser checks the confirmation message through the terminal and sends a confirmation response to the information processing device through the terminal when agreeing. Further, the information processing device delivers an advertisement to the target advertisement space through the online delivery engine according to the exposure distribution information.

In some embodiments, an information processing system, such as that shown in FIG. 9, includes an information processing device 41, an advertising placement platform 42, and a terminal 43. The information processing apparatus 41 includes an offline prediction system 44 and an online delivery engine 45. The offline estimation system 44 includes an exposure log processing module 441, an estimation sample repository 442, an estimation module 443, a directed repository 444, a memory index storage module 445, a volume searching module 446, a distribution module 447, and an exposure update module 448. The online placement engine 45 includes a distribution information base 451 and an advertisement placement server 452.

The information processing device 41 may interact with the advertisement placement platform 42 through a remote call interface, for example, receive an advertisement placement request and a confirmation response sent by an advertiser from the advertisement placement platform 42. The remote invocation interface includes a Proxy rpc interface. Prediction sample store 442 may be a Hadoop Distributed File System (HDFS). The directed repository 444 and the assignment information repository 451 may be MySQL databases. A client application may be installed on the terminal 43, and a user (exposure object) opens the client application through the terminal 43 and enters the interface 431 of the client application. The interface 431 of the client application may include: an information flow product (e.g., video, picture), at least one ad spot (including a targeted ad spot). The information processing apparatus 41 may provide an advertisement service for a client application on the terminal 43, for example, control advertisement placement of a targeted advertisement spot in the client application.

The information processing apparatus 41 executes an information processing method of an embodiment of the present application including: first, the exposure log processing module 441 obtains an exposure log of a target advertisement slot within a first preset duration. The exposure log is counted to obtain a plurality of estimated samples, and the plurality of estimated samples are stored in the estimated sample storage 442. The estimation module 443 obtains the influence parameters (e.g., holiday and festival information) within the second predetermined time period, and reads a plurality of estimation samples from the estimation sample repository 442. And obtaining estimated exposure information within a second preset time according to the estimated samples and the influence parameters, and storing the estimated exposure information to the directional repository 444. Because the estimated exposure information includes the estimated exposure amount of the exposure object with various attributes, and various attributes can reach millions of dimensions in practical application, the read-write pressure of the direct read-write directional library 444 is very large, and therefore, based on various attributes in the estimated exposure information, a memory index is generated, and the memory index is stored in the memory index storage module 445 (for example, an online machine memory). The memory index includes attributes of all types in the pre-estimated exposure information.

Then, a plurality of advertisers perform ad volume targeting operations on the ad placement platform 42, and the ad placement platform 42 sends a plurality of volume targeting requests for targeted ad spots to the volume targeting module 446. The plurality of volume seeking requests correspond to the plurality of advertisers one to one. Each of the plurality of volume requests may include a target attribute. The volume finding module 446 reads and writes the memory index storage module 445 first, and searches the memory index for an index including the target attribute in each volume finding request. And then, reading the estimated exposure information of the target advertisement position in the directional repository 444 by using all the searched indexes to obtain all the estimated exposure values of which the corresponding attributes in the second preset duration comprise the target attributes. The total estimated exposure is summed up and carried in the volume finding response to the advertisement placement platform 42.

Second, a plurality of advertisers perform ad placement operations on the ad placement platform 42 based on the volume response, and the ad placement platform 42 sends a plurality of ad placement requests for targeted ad spots to the assignment module 447. The plurality of exposure requirement information in the plurality of advertisement placement requests may include a plurality of target attributes and a plurality of target exposures. The allocation module 447 obtains the information to be exposed by interacting with the memory index storage module 445 and the directional repository 444 according to various target attributes. The pending exposure information includes exposure information that the attribute of the exposure object includes any one of a plurality of target attributes. And then generates exposure distribution information according to the undetermined exposure information, the multiple target attributes and the multiple target exposure quantities, and stores the exposure distribution information to the distribution information base 451 through a remote calling interface. The allocated estimated exposure amount in the directional repository 444 is also locked (e.g., marked as allocated) according to the exposure allocation information, and the index of the memory index storage module 445, to which the estimated exposure amount of the corresponding attribute has been allocated, is marked (e.g., set to 0).

Then, when receiving the exposure event transmitted from the terminal 43, the advertisement delivery server 452 determines a target advertisement based on the exposure allocation information. The targeted advertisement may be a brand advertisement. The targeted advertisement is sent to the terminal 43 so that the terminal 43 displays the targeted advertisement on a targeted advertisement slot in the client application, and the currently exposed object can view the targeted advertisement. Attributes of the currently exposed object and the targeted advertisement are also saved to the assignment information base 451. The advertisement delivery server 452 may deliver targeted advertisements, and may also deliver targeted advertisements and other types of advertisements in a fusion manner. Other types of advertisements include effectiveness advertisements and bid advertisements. The fusion delivery refers to delivering other types of advertisements on the premise that a plurality of target exposure quantities in a plurality of advertisement delivery requests can be completed.

Then, the exposure amount update module 448 obtains the attributes of all currently exposed objects and all target advertisements in each preset period from the distribution information base 451 through the remote call interface according to the preset period. And then counting the total number of the attributes of each current exposure object in the attributes of all current exposure objects and the total number of each target advertisement in all target advertisements. And acquiring at least one attribute of the current exposure object and at least one target advertisement in each preset period according to the preset period. And updating the estimated exposure information and the exposure requirement information according to the attribute of at least one current exposure object and at least one target advertisement. Based on the updated estimated exposure information and the updated exposure requirement information, the exposure allocation information is updated, and the updated exposure allocation information is stored in the allocation information base 451 through the Proxy rpc interface. And locking the allocated estimated exposure in the directional library 444 according to the updated exposure allocation information, and marking the allocated index of the estimated exposure with the corresponding attribute in the memory index storage module 445. When receiving the exposure event transmitted from the terminal 43, the advertisement delivery server 452 determines a target advertisement based on the updated exposure allocation information.

It is understood that the information processing apparatus more accurately obtains the estimated exposure amount of the exposure object of the attributes up to the million dimensions by introducing the influence parameters (e.g., holiday information). And the attributes up to millions of dimensionalities are generated into memory indexes, so that the read-write speed of the estimated exposure information is improved. The response to the volume seeking request and the advertisement putting request is realized in millisecond. And the duration of the second preset duration can reach 150 days or longer, and the exposure of the target advertisement space within 150 days or longer is searched and preset in advance.

It is to be understood that the above-described method may be implemented by an information processing apparatus. In order to realize the above functions, the information processing apparatus includes a hardware configuration and/or a software module that performs each function. Those of skill in the art will readily appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as hardware or combinations of hardware and computer software. Whether a function is performed as hardware or computer software drives hardware depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the embodiments of the present application.

In the embodiment of the present application, the information processing apparatus and the like may be divided into functional modules according to the method example, for example, each functional module may be divided for each function, or two or more functions may be integrated into one processing module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. It should be noted that, in the embodiment of the present application, the division of the module is schematic, and is only one logic function division, and there may be another division manner in actual implementation.

In the case of adopting a division of the respective functional modules corresponding to the respective functions, fig. 10 shows a schematic diagram of a possible configuration of the information processing apparatus according to the above-described embodiment, the information processing apparatus 5 includes: a log processing module 51, an information acquisition module 52, an exposure prediction module 53 and an exposure allocation module 54. The log processing module 51 is configured to count an exposure log of the target advertisement space within a first preset duration before the current time to obtain a plurality of estimated samples. The exposure log records identification information of an exposure object of the target advertisement space, and the identification information is used for indicating the attribute of the exposure object; each predictive sample of the plurality of predictive samples includes a historical exposure of the exposure object of one attribute divided by the identification information. And the information acquisition module 52 is configured to acquire an influence parameter that influences the exposure of the target advertisement space within a second preset time period after the current time. And the exposure estimation module 53 is configured to obtain estimated exposure information within a second preset time according to the plurality of estimated samples and the influence parameter. The estimated exposure information includes estimated exposure amounts of the exposure object of various attributes. And the exposure allocation module 54 is configured to, in response to receiving a plurality of advertisement delivery requests for the target advertisement slot, obtain exposure allocation information according to the estimated exposure information and the exposure requirement information included in each advertisement delivery request. Wherein the exposure requirement information includes a target exposure amount of the exposure object of the target attribute; the exposure allocation information is used for indicating an advertisement placement strategy of a target advertisement space made for a plurality of advertisement placement requests.

In one possible embodiment, the information processing apparatus 5 further includes: and the advertisement delivery module 55 is configured to, in response to receiving a plurality of advertisement delivery requests for the target advertisement spots, obtain exposure allocation information according to the estimated exposure information and the exposure requirement information included in each advertisement delivery request, and deliver advertisements to the target advertisement spots through the online delivery engine according to the exposure allocation information.

In another possible implementation, the log processing module 51 is specifically configured to: and determining the attribute of each exposure object in the exposure log according to the preset object attribute and the identification information of the exposure object in the exposure log. Wherein the preset object attribute comprises at least one of: gender, age, address, operating system type, and date. And then counting the number of the exposure objects of each attribute in the exposure log to obtain the historical exposure of the exposure objects of each attribute. And finally, obtaining a plurality of estimated samples according to the historical exposure of the exposure object of each attribute.

In another possible implementation, the exposure estimation module 53 is specifically configured to: and performing vector conversion on the plurality of estimated samples to obtain a plurality of estimated vectors. The plurality of estimated vectors correspond to the plurality of estimated samples one by one, and each estimated vector in the plurality of estimated vectors is used for representing the corresponding estimated sample by adopting characters. And then carrying out vector conversion on the influence parameters to obtain influence vectors. The influence vector is used for indicating whether each day in the second preset time length has an influence parameter. And then, taking the plurality of estimated vectors and the influence vectors as input, and operating a preset linear regression model to obtain the estimated exposure.

In another possible embodiment, the exposure allocation module 54 is further configured to, in response to receiving a plurality of advertisement delivery requests for the target advertisement slot, obtain exposure allocation information according to the estimated exposure information and the exposure requirement information included in each advertisement delivery request, and send a confirmation message to the terminal before delivering an advertisement to the target advertisement slot through the online delivery engine according to the exposure allocation information. The confirmation message is used for indicating the estimated exposure amount allocated to the advertisement requested to be delivered by the terminal. And then receives the confirmation response from the terminal. The confirmation response is indicative of acceptance of the estimated exposure allocated for the advertisement requested for delivery by the terminal.

In another possible implementation, the advertisement delivery module 55 is specifically configured to: an exposure event for a target ad slot is received. The exposure event includes identification information of a currently exposed object of the target ad slot. And determining the attribute of the current exposure object according to the preset object attribute and the identification information of the current playing object. Wherein the preset object attribute comprises at least one of: gender, age, address, operating system type, and date. Then, according to the attribute of the current exposure object and the exposure distribution information, a target advertisement is selected from a plurality of advertisements to be delivered indicated by a plurality of advertisement delivery requests. And finally, delivering the target advertisement to the target advertisement position through an online delivery engine.

In another possible implementation manner, the exposure allocation module 54 is further configured to, after selecting a target advertisement from the multiple advertisements to be delivered indicated by the multiple advertisement delivery requests according to the attribute of the current exposure object and the exposure allocation information, obtain, according to a preset period, at least one attribute of the current exposure object and at least one target advertisement in each preset period. And updating the estimated exposure information and the exposure requirement information according to the attribute of at least one current exposure object and at least one target advertisement. Then, based on the updated estimated exposure information and the updated exposure requirement information, the exposure distribution information is updated. And the advertisement putting module 55 is further configured to put an advertisement to the target advertisement space through the online putting engine according to the updated exposure allocation information.

Of course, the information processing apparatus 5 includes, but is not limited to, the above-listed unit modules. For example, the information processing apparatus 5 may further include a storage module. The storage module may be configured to maintain the preconfigured rate table. Moreover, the functions that can be specifically realized by the functional units also include, but are not limited to, the functions corresponding to the method steps described in the above examples, and the detailed description of the corresponding method steps may be referred to for the detailed description of other modules of the information processing apparatus 5, which is not described herein again in this embodiment of the present application.

In the case of an integrated unit, fig. 11 shows a possible structural diagram of the server 101 involved in the above-described embodiment. The server 101 may include: a processor 1001, a memory 1002, and a communication module 1003. The processor 1001 is configured to control and manage the operation of the server. The memory 1002 is used for storing program codes and data of the server, such as an information processing method, a method for counting exposure logs to obtain a plurality of estimated samples, and the like. The communication module 1003 is configured to support the server to communicate with other network entities to implement functions such as data interaction, for example, the communication module 1003 supports the server to communicate with a terminal to implement a data interaction function.

Among other things, the processor 1001 may include one or more processing cores, such as a 4-core processor, an 8-core processor, and so on. Processor 1001 may include an Application Processor (AP), a modem processor, a Graphics Processing Unit (GPU), an Image Signal Processor (ISP), a controller, a memory, a video codec, a Digital Signal Processor (DSP), a baseband processor, and/or a neural-Network Processing Unit (NPU), among others. The different processing units may be separate devices or may be integrated into one or more processors.

Memory 1002 may include one or more computer-readable storage media, which may be non-transitory. The memory 1002 may also include high-speed random access memory, as well as non-volatile memory, such as one or more magnetic disk storage devices, flash memory storage devices. In some embodiments, the non-transitory computer readable storage medium in the memory 1002 is configured to store at least one instruction, which is configured to be executed by the processor 1001, so that the server 101 executes the information processing method provided by the embodiment of the present application.

Embodiments of the present application further provide a computer storage medium, where the computer storage medium includes computer instructions, and when the computer instructions are executed on the server, the server is caused to perform the functions or steps in the foregoing method embodiments. For example, the computer-readable storage medium may be a Read-Only Memory (ROM), a Random Access Memory (RAM), a Compact Disc Read-Only Memory (CD-ROM), a magnetic tape, a floppy disk, an optical data storage device, and the like.

Embodiments of the present application further provide a computer program product, which when run on a server, causes the server to perform the functions or steps of the above method embodiments.

Through the description of the above embodiments, it is clear to those skilled in the art that, for convenience and simplicity of description, the foregoing division of the functional modules is merely used as an example, and in practical applications, the above function distribution may be completed by different functional modules according to needs, that is, the internal structure of the device may be divided into different functional modules to complete all or part of the above described functions.

In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described embodiments of the apparatus are merely illustrative, and for example, a module or a unit may be divided into only one logic function, and may be implemented in other ways, for example, a plurality of units or components may be combined or integrated into another apparatus, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.

Units described as separate parts may or may not be physically separate, and parts displayed as units may be one physical unit or a plurality of physical units, may be located in one place, or may be distributed to a plurality of different places. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

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

The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a readable storage medium. Based on such understanding, the technical solutions of the embodiments of the present application may be essentially or partially contributed to by the prior art, or all or part of the technical solutions may be embodied in the form of a software product, where the software product is stored in a storage medium and includes several instructions to enable a device (which may be a single chip, a chip, or the like) or a processor (processor) to execute all or part of the steps of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a U disk, a removable hard disk, a ROM, a RAM, a magnetic disk, or an optical disk.

The above description is only an embodiment of the present application, but the scope of the present application is not limited thereto, and any changes or substitutions within the technical scope of the present disclosure should be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims (10)

1. An information processing method, characterized in that the method comprises:

counting exposure logs of the target advertisement space within a first preset time before the current time to obtain a plurality of estimated samples; wherein the exposure log records identification information of an exposure object of the target advertisement space, and the identification information is used for indicating the attribute of the exposure object; each of the plurality of estimated samples comprises historical exposure of an exposure object with one attribute divided according to the identification information;

obtaining an influence parameter influencing the exposure of the target advertisement space within a second preset time after the current moment;

obtaining estimated exposure information within the second preset time according to the estimated samples and the influence parameters; the estimated exposure information comprises estimated exposure of exposure objects with various attributes;

Responding to a plurality of received advertisement putting requests aiming at the target advertisement space, and obtaining exposure distribution information according to the estimated exposure information and exposure requirement information included in each advertisement putting request; wherein the exposure requirement information includes a target exposure amount of an exposure object of a target attribute; the exposure allocation information is used for indicating the advertisement putting strategy of the target advertisement space made aiming at the plurality of advertisement putting requests.

2. The method of claim 1, wherein after obtaining exposure allocation information according to the estimated exposure information and exposure requirement information included in each advertisement placement request in response to receiving a plurality of advertisement placement requests for the target advertisement placement, the method further comprises:

and delivering advertisements to the target advertisement positions through an online delivery engine according to the exposure distribution information.

3. The method according to claim 1 or 2, wherein the counting the exposure log of the target advertisement space within a first preset time period before the current time to obtain a plurality of pre-estimated samples comprises:

determining the attribute of each exposure object in the exposure log according to the preset object attribute and the identification information of the exposure object in the exposure log; wherein the preset object attribute comprises at least one of: gender, age, address, operating system type, and date;

Counting the number of the exposure objects of each attribute in the exposure log to obtain the historical exposure of the exposure objects of each attribute;

and obtaining the plurality of estimated samples according to the historical exposure of the exposure object with each attribute.

4. The method of claim 1, wherein obtaining the estimated exposure information within the second predetermined duration according to the estimated samples and the influence parameters comprises:

performing vector conversion on the plurality of estimated samples to obtain a plurality of estimated vectors; the plurality of estimated vectors correspond to the plurality of estimated samples one by one, and each estimated vector in the plurality of estimated vectors is used for representing the corresponding estimated sample by adopting characters;

carrying out vector conversion on the influence parameters to obtain influence vectors; the influence vector is used for indicating whether each day in the second preset time length has an influence parameter or not;

and taking the plurality of estimated vectors and the influence vector as input, and operating a preset linear regression model to obtain the estimated exposure.

5. The method of claim 2, wherein after obtaining exposure allocation information according to the estimated exposure information and exposure requirement information included in each advertisement placement request in response to receiving a plurality of advertisement placement requests for the target advertisement placement, before placing an advertisement to the target advertisement placement through an online placement engine according to the exposure allocation information, the method further comprises:

Sending a confirmation message to the terminal; the confirmation message is used for indicating the estimated exposure amount distributed for the advertisement requested to be delivered by the terminal;

receiving a confirmation response from the terminal; the confirmation response is used for indicating the acceptance of the estimated exposure amount allocated for the advertisement requested to be delivered by the terminal.

6. The method of claim 2, wherein the delivering advertisements to the target ad spots via an online delivery engine according to the exposure allocation information comprises:

receiving an exposure event of the target ad slot; the exposure event comprises identification information of a current exposure object of the target advertisement space;

determining the attribute of the current exposure object according to the preset object attribute and the identification information of the current playing object; wherein the preset object attribute comprises at least one of: gender, age, address, operating system type, and date;

selecting a target advertisement from a plurality of advertisements to be launched indicated by the plurality of advertisement launching requests according to the attribute of the current exposure object and the exposure distribution information;

and delivering the target advertisement to the target advertisement position through the online delivery engine.

7. The method according to claim 6, wherein after the selecting a target advertisement from the plurality of advertisements to be delivered indicated by the plurality of advertisement delivery requests according to the attribute of the current exposure object and the exposure allocation information, the method further comprises:

acquiring at least one attribute of a current exposure object and at least one target advertisement in each preset period according to the preset period;

updating the estimated exposure information and the exposure requirement information according to the attribute of the at least one current exposure object and the at least one target advertisement;

updating the exposure distribution information based on the updated estimated exposure information and the updated exposure requirement information;

and delivering advertisements to the target advertisement space through the online delivery engine according to the updated exposure distribution information.

8. An information processing apparatus characterized in that the apparatus comprises:

the log processing module is used for counting exposure logs of the target advertisement position within a first preset time before the current time to obtain a plurality of estimated samples; wherein the exposure log records identification information of an exposure object of the target advertisement space, and the identification information is used for indicating the attribute of the exposure object; each of the plurality of estimated samples comprises historical exposure of an exposure object with one attribute divided according to the identification information;

The information acquisition module is used for acquiring influence parameters influencing the exposure of the target advertisement space within a second preset time length after the current moment;

the exposure pre-estimation module is used for obtaining pre-estimated exposure information within the second preset duration according to the pre-estimated samples and the influence parameters; the estimated exposure information comprises estimated exposure of exposure objects with various attributes;

the exposure distribution module is used for responding to a plurality of received advertisement putting requests aiming at the target advertisement space and obtaining exposure distribution information according to the estimated exposure information and the exposure requirement information included by each advertisement putting request; wherein the exposure requirement information includes a target exposure amount of an exposure object of a target attribute; the exposure allocation information is used for indicating the advertisement putting strategy of the target advertisement space made aiming at the plurality of advertisement putting requests.

9. A server, comprising: a processor and a memory for storing processor-executable instructions;

wherein the processor is configured to execute the instructions to cause the server to perform the information processing method of any one of claims 1 to 7.

10. A computer-readable storage medium having stored thereon computer instructions, which, when run on a server, cause the server to execute the information processing method according to any one of claims 1 to 7.

Images