Method, apparatus, device, and medium for determining the amount of resources required for delivery of a media item

Abstract

To provide a method for determining the quantity of requisite resources for placing a media item.SOLUTION: A method includes: extracting feature information of a target media item from relevant data of the target media item; acquiring predicted values of the quantity of requisite resources for a plurality of placements in competitive placement of the target media item, using a prediction model based at least on the feature information, the predicted values of the quantity of requisite resources for the plurality of placements respectively corresponding to a plurality of predetermined probabilities of the target media item being placed; determining the quantity of requisite resources for placing the target media item from the predicted values of the quantity of requisite resources for the plurality of placements, based on a plurality of placement efficiency measures associated with the predicted values of the quantity of requisite resources. Accordingly, the quantity of requisite resources for placement at various probabilities may be predicted, and then the quantity of requisite resources for placement with higher placement efficiency measures may be selected, thereby improving placement efficiency of the media item.SELECTED DRAWING: Figure 6

Description

[Technical Field]

[0001] FIELD Exemplary implementations of the present invention relate generally to the field of computer technology, and more particularly to methods, apparatus, devices, and computer-readable storage media for determining the amount of resources required for the delivery of a media item. [Background technology]

[0002] The Internet provides access to a wide variety of objects. For example, various applications, products, audio, video, and other data can be accessed via the Internet. Accessible data also includes specific media items related to each of the above-mentioned items, including, for example, advertisements. Object providers who own objects can offer media items to media distributors. Distribution of media items can be competitive. Successful distribution depends on the amount of resources, also known as bidding, required to distribute the media item. Summary of the Invention

[0003] In a first aspect of the present invention, a method for determining an amount of resources required for delivery of a media item is provided. The method includes extracting characteristic information of a target media item from related data for the target media item. Based on at least the characteristic information, a prediction model is utilized to obtain a predicted amount of resources required for multiple deliveries of the target media item during competitive delivery, the predicted amount of resources required for the multiple deliveries corresponding to multiple predetermined probabilities of the target media item being delivered, respectively. The amount of resources required for delivery of the target media item is determined from the predicted amount of resources required for the multiple deliveries based on multiple delivery efficiency measures respectively associated with the predicted amount of resources required for the multiple deliveries.

[0004] In a second aspect of the present invention, there is provided an apparatus for determining an amount of resources required for delivery of a media item, the apparatus comprising: an extraction module for extracting feature information of a target media item from related data of the target media item; an acquisition module for obtaining, based on at least the feature information, a predicted value of an amount of resources required for a plurality of deliveries during competitive delivery of the target media item using a prediction model, the predicted values ​​of the amount of resources required for the plurality of deliveries respectively corresponding to a plurality of predetermined probabilities that the target media item will be delivered; and a determination module for determining an amount of resources required for delivery of the target media item from the predicted values ​​of the amount of resources required for the plurality of deliveries, based on a plurality of delivery efficiency measures respectively associated with the predicted values ​​of the amount of resources required for the plurality of deliveries.

[0005] In a third aspect of the present invention, there is provided an electronic device comprising at least one processing unit and at least one memory coupled to the at least one processing unit and adapted to store instructions to be executed by the at least one processing unit, the instructions, when executed by the at least one processing unit, causing the electronic device to perform a method according to the first aspect of the present invention.

[0006] In a fourth aspect of the present invention, there is provided a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, causes the processor to implement the method of the first aspect of the present invention.

[0007] It should be understood that the contents described in the summary of the present invention are not intended to limit the main or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will be easily understood from the following description. [Brief explanation of the drawings]

[0008] The above and other features, advantages, and aspects of the embodiments of the present invention will become more apparent from the following detailed description taken in conjunction with the drawings, in which like or similar symbols indicate like or similar elements. [Figure 1] 1 illustrates a block diagram of an environment for distributing media items according to one exemplary implementation of the present invention. [Figure 2] 1 shows a block diagram for determining the amount of resources required for delivery of a media item according to some implementations of the present invention. [Figure 3] 1 illustrates a block diagram of media item feature information according to some implementations of the present invention. [Figure 4] 1 illustrates a block diagram of the relationship between a predictive model and multiple predetermined probabilities according to some implementations of the present invention. [Figure 5] FIG. 1 illustrates a block diagram of a predictive model according to some implementations of the present invention. [Figure 6] 1 shows a flowchart of a method for determining the amount of resources required for delivery of a media item according to some implementations of the present invention. [Figure 7] 1 shows a block diagram of an apparatus for determining the amount of resources required for delivery of a media item according to some implementations of the present invention. [Figure 8] 1 shows a block diagram of a device in which multiple implementations of the present invention can be practiced. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, the embodiments of the present invention will be described in more detail with reference to the drawings. Although the drawings show specific embodiments of the present invention, it should be understood that the present invention can be implemented in various forms and should not be construed as being limited to the embodiments described herein, but rather, these embodiments are provided for a more thorough and complete understanding of the present invention. It should be understood that the drawings and embodiments of the present invention are for illustrative purposes and are not used to limit the protection scope of the present invention.

[0010] In describing embodiments of the present invention, the term "comprises" and similar terms are used to mean an open-ended inclusion of "including, but not limited to." The term "based on" should be understood as "based at least in part on." The terms "one implementation" or "the implementation" should be understood as "at least one implementation." The term "some implementations" should be understood as "at least some implementations." Other explicit and implicit definitions may be included below. The term "model" as used herein may indicate a relationship between each piece of data. For example, the above-described relationship may be obtained based on various technical solutions currently known and / or developed in the future.

[0011] It is understood that data related to the present technical solution (including but not limited to the data itself, the acquisition of data, or the use of data) should comply with applicable laws and regulations and related designated requirements.

[0012] It is understood that before using the technical solutions disclosed in each embodiment of the present invention, users should be informed of the types, scope of use, use scenarios, etc. of personal information involved in the present invention in an appropriate manner in accordance with relevant laws and regulations, and their consent should be obtained.

[0013] For example, in response to receiving an unsolicited request from a user, presentation information may be sent to the user to explicitly indicate to the user that the requested operation requires the acquisition and use of the user's personal information, allowing the user to independently choose whether or not to provide the personal information to software or hardware, such as an electronic device, application, server, or storage medium, that performs the operation of the technical solution of the present invention, based on the presentation information.

[0014] In an optional, non-limiting implementation, the method for sending the user the offered information in response to receiving the user's unsolicited request may be, for example, a method using a pop-up window in which the offered information can be displayed in text form, and the pop-up window may further include a selection control for the user to select "agree" or "disagree" to providing personal information to the electronic device.

[0015] It is understood that the notification and user authorization process described above is merely a general outline and is not intended to limit the implementation of the present invention, and that other means that comply with relevant laws and regulations may also be applied to the implementation of the present invention.

[0016] The term "in response to" means that a corresponding event occurs or a condition is satisfied. It is understood that there is not necessarily a strong correlation between the timing sequence of execution of subsequent actions performed in response to the event or condition and the time of the event occurring or the time of the condition being satisfied. In certain cases, the subsequent action may be performed immediately upon the occurrence of the event or the fulfillment of the condition, and in other cases, the subsequent action may be performed a period of time after the event occurs or the condition is fulfilled.

[0017] Example Environment An overview according to one exemplary implementation of the present invention will be described with reference to FIG. 1. FIG. 1 illustrates a block diagram of an environment 100 for distributing media items according to one exemplary implementation of the present invention. As illustrated in FIG. 1, one or more media providers can use a media management system 120 to manage media provided on a media distribution platform 110. One or more terminal devices 130-1, 130-2, 130-3, etc. (collectively or individually referred to as terminal devices 130 for ease of description) are associated with the media distribution platform 110 and can access various media provided on the media distribution platform 110 based on, for example, corresponding audiences 132-1, 132-2, 132-3, etc. (collectively or individually referred to as audiences 132 for ease of description). By way of example, the media distribution platform 110 may be an application, a website, a netpage, or other accessible platform. The terminal devices 130 may have an application installed for accessing the media distribution platform 110 or may access the media distribution platform 110 in any suitable manner.

[0018] Media management system 120 may be configured to deliver (e.g., provide or display at terminal device 130) one or more specific media items associated with one or more objects to an audience group based on a corresponding strategy. The delivered media items may include, for example, one or more media items 142-1, 142-2, ..., 142-M (collectively or individually referred to as media items 142 for ease of explanation) in media database 140.

[0019] As used herein, an object may include various recommendable recommendation items, which may include, for example, an application, an entity product, a virtual product, audio / video media, etc. As used herein, a media item refers to a medium that is displayed to recommend a corresponding object. An example of a media item may include an advertisement. As used herein, an audience group may include one or more audience members, such as audience 132. An audience member may be any potential consumer of an object, such as a user, a group, an organization, an entity, etc.

[0020] In some implementations, the media management system 120 can distribute corresponding media items on the media distribution platform 110 based on requests from object providers 150-1, 150-2, 150-3, etc. (collectively or individually referred to as object providers 150 for ease of description). In some implementations, the media management system 120 can deliver media items 142 to corresponding audiences 132 on the media distribution platform 110 based on requests from at least each object provider 150-1, 150-2, 150-3, etc. (collectively or individually referred to as object providers 150 for ease of description). In advertising distribution scenarios, the object providers 150 may also be referred to as advertisers. In some implementations, the object providers may also pay fees to media providers based, for example, on the exhibition and subsequent conversion of the media items.

[0021] In some implementations, the media management system 120 can select media items to display on specific terminal devices 130 at media distribution opportunities (e.g., at specific times and specific locations) on the media distribution platform 110 based on bidding results. For example, the media management system 120 can receive a unit price (bId) from the object provider 150. In some implementations, the media management system 120 can allocate the media distribution opportunity to the highest bidder, meaning that the corresponding media item can successfully compete for distribution. The unit price may refer to the amount of resources (e.g., the amount of resources required) expended to compete for distribution of a specific media item at a specific media distribution opportunity. Successfully delivering a media item at a certain cost is referred to as a single send, and the cost is referred to as the unit price (rank bId) for this delivery or send.

[0022] In environment 100, terminal device 130 may be any type of mobile, fixed, or portable terminal, including a mobile cell phone, desktop computer, laptop computer, notebook computer, netbook computer, flat panel computer, media computer, multimedia flat panel, personal communication system (PCS) device, personal navigation device, personal digital assistant (PDA), personal navigation device, assistant (PDA), audio / video player, digital camera / camcorder, positioning device, television receiver, radio receiver, e-book device, gaming device, or any combination thereof, including accessories and peripherals of these devices or any combination thereof. In some implementations, terminal device 130 may also support any type of interface to a user (such as "wearable" circuitry). Media management system 120 may be, for example, any type of computing system / server capable of providing computing functionality, including, but not limited to, a mainframe, an edge computing node, a computing device in a cloud environment, etc. It should be understood that the structure and functionality of each element in environment 100 is described for illustrative purposes only and does not imply any limitation on the scope of the invention.

[0023] In competitive distribution, multiple object providers 150 can provide the amount of resources required for the distribution of each of their media items, i.e., bid according to the amount of resources required for the distribution. The media management system 120 can, for example, select the highest bid from among the bids and distribute the related media items 142 of the object provider 150 to the media distribution platform 110. While a higher bid may increase the chances of winning, a bid that is too high may increase the burden on the object provider 150. A bid that is too low may result in the object provider losing the opportunity to exhibit the media item. In this case, it is expected that the amount of resources required for distribution will be more rationally determined and bids will be made according to the amount of resources required for the distribution.

[0024] Overview of determining the amount of resources required for delivery To at least partially address the shortcomings of the prior art, one exemplary implementation of the present invention proposes a method for determining the amount of resources required for the delivery of a media item. An overview of one exemplary implementation of the present invention is provided with reference to FIG. 2, which shows a block diagram 200 for determining the amount of resources required for the delivery of a media item according to some implementations of the present invention. FIG. 2 illustrates a process for determining the amount of resources required for the delivery of a target media item 210 to be delivered. Here, the target media item 210 is delivered to promote a target object depicted in the target media item 210. For example, the target media item 210 may be an advertisement promoting a particular application software.

[0025] Specifically, the characteristic information 212 of the target media item 210 can be extracted from various related data of the target media item 210. According to one exemplary implementation of the present invention, a prediction model 220 can be pre-established, and based on at least the characteristic information 212, the prediction model 220 can be used to determine a predicted amount of resources required for multiple deliveries of the target media item 210 during competitive delivery. Here, the predicted amounts of resources required for multiple deliveries output by the prediction model 220 may correspond to a plurality of predetermined probabilities that the target media item 210 will be successfully delivered, respectively.

[0026] Here, the prediction model 220 can provide predicted values ​​of the amount of resources required for delivery associated with multiple (e.g., n) predetermined probabilities. The multiple predicted probabilities can be denoted, for example, as F1(i), where 0≦I≦n−1, and multiple corresponding predicted values ​​of the amount of resources required for delivery can be denoted, for example, as B1(i), where 0≦I≦n−1. In this case, I represents the I-th position, F1(i) can represent the I-th predetermined probability, and B1(i) can represent the predicted value of the amount of resources required for delivery for the I-th predetermined probability. For example, the multiple predetermined probabilities can be expressed in the form of quantiles, where, for example, predetermined probability 240 indicates a 5% probability of winning the bid according to the corresponding predicted value 230, ..., predetermined probability 242 indicates a 95% probability of winning the bid according to the corresponding predicted value 232.

[0027] Furthermore, the amount of resources required for the delivery of the target media item can be determined 260 from the predicted amounts of resources required for the multiple deliveries based on multiple delivery efficiency measures 250, ..., and 252, each associated with a respective predicted amount of resources required for the multiple deliveries. Here, the delivery efficiency measure may indicate, for example, revenue that may be generated by delivering the target media item, and the predicted amount of resources required for the multiple deliveries corresponding to the highest or higher delivery efficiency measure can be selected as the amount of resources required for the final delivery (i.e., bid). This can improve the delivery efficiency of the media item, for example, by expanding the impact of the target media item and enabling a larger audience to view the delivered media item.

[0028] A detailed process for determining the amount of resources required for delivery Having provided an overview of one exemplary implementation of the present invention, the following provides more details of determining the amount of resources required for the delivery of target media items. In the context of the present invention, historical data from historical delivery processes is used as evidence to generate a predictive model. It should be understood that although the media items involved in each historical delivery process are not completely identical, certain commonalities may exist between these media items, and therefore, using the common aspects as evidence makes it easier to extract a general basis for determining the amount of resources required for the delivery of media items.

[0029] The term "model" as used herein refers to a model that learns association relationships between corresponding inputs and outputs from training data, thereby generating a corresponding output for a given input after training is complete. The generation of the model can be based on machine learning techniques. Deep learning is a machine learning algorithm that uses multiple layer processing units to process inputs and provide a corresponding output. A neural network model is an example based on a deep learning model. As used herein, a "model" may also be referred to as a "machine learning model," a "learning model," a "machine learning network," or a "learning network," and these terms are used interchangeably herein.

[0030] A "neural network" is a machine learning network based on deep learning. A neural network can process inputs and provide corresponding outputs. It typically includes an input layer, an output layer, and one or more hidden layers between the input and output layers. Neural networks used in deep learning applications typically include many hidden layers, thereby increasing the depth of the network. Each layer of a neural network is connected sequentially so that the output of the previous layer provides the input to the subsequent layer. The input layer receives the input of the neural network, and the output of the output layer serves as the final output of the neural network. Each layer of a neural network includes one or more nodes (also called processing nodes or neurons), each of which processes the input from the previous layer.

[0031] Generally, machine learning consists of three main stages: training, testing, and application (also known as inference). In the training stage, a given model is trained using a large amount of training data, and parameter values ​​are continuously and iteratively updated until the model can obtain consistent inferences from the training data that meet the desired goal. Through training, the model can be thought of as learning the input-to-output association (also known as input-to-output mapping) from the training data. Parameter values ​​for the trained model are determined. In the testing stage, test inputs are applied to the trained model to test whether it can provide the correct output, thereby determining the model's performance. The testing stage is then combined with the training stage. In the application stage, the trained model is used to process actual model inputs and determine the corresponding model output based on the parameter values ​​obtained in training.

[0032] According to one exemplary implementation of the present invention, the target media item 210 may have associated data from which characteristic information 212 can be extracted. The characteristic information 212 may include, for example, at least one of the type of the target media item, the platform on which the target media item is distributed, the operating system of the platform on which the target media item is distributed, the client device of the platform on which the target media item is distributed, and the region in which the target media item is distributed.

[0033] FIG. 3 illustrates a block diagram 300 of media item characteristic information 212 according to some implementations of the present invention. As illustrated in FIG. 3, type 311 may indicate the type of the target media item, which may include, but is not limited to, application, product, audio, video, etc. Platform 312 may indicate the platform to which the target media item is distributed, which may include, but is not limited to, a social networking platform, a news push platform, a music platform, a video platform, a short video platform, a game platform, etc. Operating system 313 may indicate the operating system of the platform to which the target media item is distributed, which may include, but is not limited to, the Android system, the IOS system, various operating systems for desktop computers, etc. Client device 314 may indicate the client device of the platform to which the target media item is distributed, which may include, but is not limited to, a desktop computer device, a mobile device, etc. Region 315 may indicate the region to which the target media item is distributed, such as Region I, Region II, etc.

[0034] Using an exemplary implementation of the present invention, the feature information 212 may describe multiple aspects of the content of the target media item 210 in detail, thereby facilitating the representation of various information relevant to different distribution scenarios and thus improving the accuracy of the prediction model 220. Here, for example, the feature information 212 may be represented in the form of features (e.g., embedded), which are not visible from the outside in an encoded form and therefore do not reveal potentially sensitive information.

[0035] According to one exemplary implementation of the present invention, the feature information 212 may be used as input to a prediction model 220 to determine predicted amounts of resources required for multiple deliveries corresponding to multiple predetermined probabilities. More details of the prediction model 220 are described with reference to Figure 4, which shows a block diagram 400 of the relationship between a prediction model and multiple predetermined probabilities according to some implementations of the present invention.

[0036] As shown in FIG. 4 , the abscissa represents a plurality of probabilities of winning the competitive broadcast, and the plurality of predetermined probabilities of the predictive model 220 may be expressed, for example, in the form of quantiles. Also, as an example, quantiles may be determined at 5% (or other) intervals, and the plurality of predetermined probabilities may be represented as, for example, 5%, 10%, 15%, 20%, ..., 95%, etc. Alternatively and / or additionally, quantiles may be determined at 10% intervals, and the plurality of predetermined probabilities may be represented as, for example, 10%, 20%, ..., 90%, etc. The smaller the interval, the higher the accuracy represented by the predictive model 220, but the greater the amount of calculation involved. According to one exemplary implementation of the present invention, a balance between accuracy and amount of calculation is achieved, and an appropriate interval is determined accordingly.

[0037] According to one exemplary implementation of the present invention, in the process of determining a predicted value of the amount of resources required for the multiple streams, a prediction model 220 can be utilized to determine a distribution of the amount of resources required for the multiple streams, each associated with a plurality of predetermined probabilities, based at least on the characteristic information 212. Continuing with the above example, if the prediction model 220 includes N predetermined probabilities, the distribution may be represented in the form of an n-dimensional array. For example,

number

[0038] Exemplary implementations of the present invention allow for accurate and efficient description of the relationship between the amount of resources required for each stream and the corresponding probability of winning, thereby improving the accuracy of the prediction model. Furthermore, predictions of the amount of resources required for multiple streams can be determined based on the distributions of the amount of resources required for each stream.

[0039] It should be understood that the distribution of the amount of resources required for distribution herein may be obtained, for example, based on historical data during the historical distribution process. It should be understood that the historical data may include different scenarios, and the distribution may be determined based on the different scenarios. According to one exemplary implementation of the present invention, related bid data for a particular historical distribution may be obtained. For example, a particular media management system 120 may receive multiple bids from multiple object providers, with the highest bid winning, and the media management system 120 may distribute the media item of the highest bidder on the media distribution platform 110.

[0040] In one scenario, the media management system 120 transmits bid data (e.g., the smallest winning bid (mInbId)) as feedback to each object provider. In this case, this scenario is called a "scenario with feedback transmission," and the mInbId can represent an objective bidding format. If the object provider wins, the mInbId represents the second-highest bid among all object providers' unit prices; if the object provider loses, the mInbId represents the winning bid. Once the feedback bid data is obtained, this historical bidding data can be used to obtain the distribution of the amount of resources required for the distribution described above.

[0041] In another scenario, the media management system 120 does not feedback-transmit bid data to each object provider. In this case, this scenario is called a "no feedback transmission scenario." When the mInbId is not feedback-transmitted, only data on the associated send-show rate (SSR) of the media item is obtained. Here, the SSR data represents the rate at which a media item is sent and displayed. However, the actual display process of the media item is Send-Win-Show. However, the SSR data does not represent the influence of the bidding process and may be influenced by the bidding strategy, resulting in bias. In this case, the influence of the bidding process needs to be taken into account to eliminate potential bias in the SSR data.

[0042] According to one exemplary implementation of the present invention, the proposed prediction model 220 can implement a bias correction method for the bidding style estimation model based on mInbId post-verification by comprehensively considering both the scenarios with and without feedback transmission as described above. This allows the amount of resources required for delivery to be better predicted even in the scenario without feedback transmission by utilizing mInbId information from other similar cases. This allows for more accurate bidding style estimation even in the scenario without feedback transmission.

[0043] 5 illustrates a block diagram 500 of a predictive model according to some implementations of the present invention. As illustrated in FIG. 5, the predictive model 220 can receive feature information 212 of a target media item 210 and can obtain posterior validation distributions 542 associated with the feature information 212 from a database 540 for storing historical bidding data. Furthermore, a first network 510 and a second network 520 can be utilized to process the feature information and corresponding posterior validation distributions 542 to determine specific distributions associated with a plurality of predetermined probabilities.

[0044] While FIG. 5 illustrates database 540 for storing historical data as being located within predictive model 220, it should be understood that alternatively and / or additionally, database 540 may be located external to predictive model 220, such as in any location accessible to predictive model 220, as long as post-validation distribution 542 of historical bid data can be obtained from database 540.

[0045] According to one exemplary implementation of the present invention, the prediction model 220 may include a first network 510 that can describe an association relationship between a posterior validation distribution 542 of the amount of resources required for each predetermined probability winning related distribution and the characteristics of the media item in the historical bid. In other words, the first network 510 can describe a first association relationship between the amount of resources (i.e., historical unit price) required for the first reference distribution in a first reference competitive distribution (i.e., historical bid) to distribute the first reference media item (i.e., the historical media item) and first reference characteristic information (i.e., the characteristics of the historical media item) of the first reference media item.

[0046] According to one exemplary implementation of the present invention, in the process of determining a distribution of resource amounts required for a plurality of deliveries each associated with a plurality of predetermined probabilities, a first network 510 can be utilized to determine the first distribution of resource amounts required for a plurality of deliveries associated with a plurality of predetermined probabilities, thereby allowing the prediction model 220 to fully consider the post-validation distribution of true historical bid data and thus utilize the distribution trends of these true bid data to calibrate potential prediction biases.

[0047] According to one exemplary implementation of the present invention, the above-mentioned association relationship can be determined based on a statistical method, for example, a distribution of the amount of resources required for each delivery corresponding to each winning probability can be obtained from a large amount of historical bidding data based on a statistical method. Alternatively and / or additionally, the above-mentioned association relationship can be expressed based on a machine learning model. If the association relationship has already been obtained, the distribution of the amount of resources required for delivery based on statistics (also referred to as a first distribution) can be determined using feature data of the target media item.

[0048] Assuming that the characteristic information of the target media item is (application, social network platform, Android system, mobile device, region I), Value1=(0.21, 0.43, 0.65,...). Alternatively and / or additionally, if the historical bidding data in database 540 is different or if the characteristic data of the target media item is different, the distribution of the amount of resources required for delivery may have other values. Also, for example, assuming that the characteristic information of the target media item is (product, social network platform, iOS system, mobile device, region II), Value1=(0.20, 0.45, 0.60,...).

[0049] According to one exemplary implementation of the present invention, a predicted amount of resources required for delivery at different quantiles can be determined based on the post-validation distribution. Specifically, in advertising scenarios, ECPM (Expected Cost Per Mille, or ECPM) is typically used as a value measure for a target media item. For example, the ECPM of a target media item can be calculated using the following formula:

number

[0050] In the context of the present invention, a bid cpa_bId from an object provider can be received, and a bid can be submitted in a bidding manner with a predicted amount of resources required for delivery (e.g., represented as bId_prIce) on one or more third-party media distribution platforms. This can improve the exposure rate of the media item and expand the impact of the media item. In this case, it is expected that the relationship between the predicted amount of resources required for delivery and the value measure of the target media item satisfies a predetermined constraint tac. In the context of the present invention, the constraint tac can be expressed, for example, as a predetermined percentage (typically tac<1). For example, the predicted amount of resources required for delivery can be limited based on the following formula:

number

[0051] According to one exemplary implementation of the present invention, the distribution of the amount of resources required for delivery associated with each predetermined probability is

number

number

number

[0052] According to one exemplary implementation of the present invention, the predictive model 220 may include a second network 520, which may represent relevant knowledge extracted from the true SSR in historical distributions. Specifically, the second network 520 may describe a second relevant relationship between the display (i.e., historical SSR data) of the second reference media item in a second reference competitive distribution (i.e., historical bidding) of the distributed second reference media item (i.e., historical media item) and second reference feature information (i.e., features of the historical media item) of the second reference media item.

[0053] Here, the second network 520 may be trained based on historical truth data. Specifically, initial parameters of the second network 520 may be obtained and then updated in various currently known and / or future developed manners to minimize the difference between the output results of the second network 520 and the corresponding truth data. Furthermore, once the second network 520 is obtained, the second network 520 may be used to determine a second distribution of resource amounts required for multiple deliveries, each associated with a plurality of predetermined probabilities. Using the exemplary implementation of the present invention, knowledge of multiple aspects of media item delivery can be extracted from the truth data during the historical delivery process, further assisting in the subsequent bidding process.

[0054] Similar to the various steps of operating the first network 510 described above, a second distribution of the amount of resources required for the multiple deliveries associated with each predetermined probability determined based on the SSR data can be determined.

number

number

number

[0055] According to one exemplary implementation of the present invention, a plurality of delivery efficiency measures may be determined based on a predicted amount of resources required for a plurality of deliveries, a value measure of the target media item, and a plurality of predetermined probabilities. Specifically, the plurality of delivery efficiency measures may be determined based on the following formula:

number

number

[0056] Exemplary implementations of the present invention can be utilized to convert the complex steps of determining delivery efficiency measures into a simple mathematical process, thereby enabling the determination of delivery efficiency measures resulting from bids with each predetermined probability in a simple and effective manner based on mathematical operations.

[0057] According to an exemplary implementation of the present invention, the control parameters in the above formulas can be determined based on various methods. For example, the above control parameters can be determined based on a proportional integral differential (PID) control technique to adjust multiple delivery efficiency measures. The above control parameters can be set based on the general principle of a PID control algorithm, and in this case, multiple delivery efficiency measures can be determined based on the following formulas:

number

number

[0058] According to one exemplary implementation of the present invention, the above-described formula can be utilized to determine a delivery efficiency metric associated with each predetermined probability. Furthermore, the magnitude of each delivery efficiency metric can be compared to further select a predicted value for the amount of resources required for delivery with a larger delivery efficiency metric (e.g., higher revenue). In this case, if it is determined that a first delivery efficiency metric of the plurality of delivery efficiency metrics is higher than a second delivery efficiency metric, a first predicted value corresponding to the first delivery efficiency metric can be selected from the predicted values ​​for the amount of resources required for delivery, i.e., a predicted value corresponding to a higher revenue.

[0059] Specifically, it is possible to determine the respective delivery efficiency measures corresponding to probabilities such as 5%, 10%, 15%, etc. The delivery efficiency measures can be compared, and the amount of resources required for delivery corresponding to the largest delivery efficiency measure can be selected. For example, the amount of resources B required for delivery corresponding to max[profit1(i)] and max[profit2(i)] can be selected. 1,max and B 2,max and then determine the final bid for the target media item.

[0060] According to one exemplary implementation of the present invention, the above-mentioned Equations 9.1 and 9.2 can be summed to determine the probability of obtaining the largest delivery efficiency measure, and then the amount of resources required for delivery corresponding to the post-validation distribution and the amount of resources required for delivery based on SSR data can be obtained, respectively.

[0061] According to one exemplary implementation of the present invention, a predicted amount of resources required for multiple deliveries can be determined based on both the first network 510 and the second network 520. Specifically, weights for the first network 510 and the second network 520 are determined based on the posterior validation distribution and the importance of the SSR data, respectively, and a predicted amount of resources required for multiple deliveries can be determined based on the weighting of the first distribution and the second distribution. Assuming that the weights of the two networks are coef and (1-coef), respectively, then a predicted amount of resources required for final deliveries associated with multiple predetermined probabilities can be determined based on the following formula:

number

[0062] Exemplary implementations of the present invention can be used to predict the amount of resources required for delivery with various probabilities and to select, as a bid, the amount of resources required to generate a delivery with a higher delivery efficiency measure. This can improve the delivery efficiency of media items, for example, by expanding the impact of the target media item and allowing a larger audience to view the delivered media item. Furthermore, the post-validation distribution of historical bid data can be used to correct for the problem of the SSR data not containing objective bid information, thereby enabling the prediction model 220 to provide more accurate predictions of the amount of resources required for delivery.

[0063] Example Process 6 shows a flowchart of a method 600 for determining the amount of resources required for delivery of a media item according to some implementations of the present invention. At block 610, feature information of a target media item is extracted from data related to the target media item. At block 620, a prediction model is utilized based on at least the feature information to obtain a predicted amount of resources required for multiple deliveries during competitive delivery of the target media item, the predicted amount of resources required for the multiple deliveries corresponding to multiple predetermined probabilities of the target media item being delivered, respectively. At block 630, the amount of resources required for delivery of the target media item is determined from the predicted amount of resources required for the multiple deliveries based on multiple delivery efficiency measures respectively associated with the predicted amount of resources required for the multiple deliveries.

[0064] According to one exemplary implementation of the present invention, obtaining a predicted value of an amount of resources required for the plurality of deliveries includes: determining, based on at least the characteristic information, a distribution of amounts of resources required for the plurality of deliveries, each associated with a plurality of predetermined probabilities, using a prediction model; and determining a predicted value of an amount of resources required for the plurality of deliveries based on each of the distributions of amounts of resources required for the plurality of deliveries.

[0065] According to one exemplary implementation of the present invention, the predictive model includes a first network, the first network describing a first association relationship between an amount of resources required for a first reference delivery during a first reference competitive delivery of the first reference media item and first reference feature information of the first reference media item, and determining a distribution of the amount of resources required for multiple delivery associated with multiple predetermined probabilities respectively includes determining a first distribution of the amount of resources required for the multiple delivery associated with the multiple predetermined probabilities using the first network.

[0066] According to one exemplary implementation of the present invention, the predictive model includes a second network, the second network describing a second association relationship between the exhibition of the second reference media item during a second reference competitive distribution delivering the second reference media item and second reference feature information of the second reference media item, and determining a distribution of resource amounts required for the multiple distributions each associated with a plurality of predetermined probabilities includes determining a second distribution of resource amounts required for the multiple distributions each associated with a plurality of predetermined probabilities using the second network.

[0067] According to one exemplary implementation of the present invention, obtaining a predicted value of the amount of resources required for the multiple deliveries includes determining a predicted value of the amount of resources required for the multiple deliveries based on a first distribution and a second distribution.

[0068] According to one exemplary implementation of the present invention, a plurality of delivery efficiency measures are determined based on a predicted amount of resources required for the plurality of deliveries, a value measure for the target media item, and a plurality of predetermined probabilities.

[0069] According to one exemplary implementation of the present invention, the method further includes adjusting a plurality of delivery efficiency measures based on the proportional-integral-derivative control parameters.

[0070] According to one exemplary implementation of the present invention, the relationship between the predicted amount of resources required for multiple deliveries and the value measure of the target media item satisfies a predetermined constraint.

[0071] According to one exemplary implementation of the present invention, the value measure of a target media item is determined based on the amount of resources required to deliver the target media item, a predicted click-through rate for the target media item, and a predicted conversion rate for the target media item.

[0072] According to one exemplary implementation of the present invention, determining the amount of resources required for delivery based on the plurality of delivery efficiency measures includes, in response to determining that a first delivery efficiency measure of the plurality of delivery efficiency measures is higher than a second delivery efficiency measure, selecting a first predicted value corresponding to the first delivery efficiency measure from the plurality of predicted values ​​of the amount of resources required for delivery.

[0073] According to one exemplary implementation of the present invention, the characteristic information of the target media item includes at least one of the type of the target media item, the platform on which the target media item is distributed, the operating system of the platform on which the target media item is distributed, the client device of the platform on which the target media item is distributed, and the region in which the target media item is distributed.

[0074] Exemplary Apparatus and Devices 7 shows a block diagram of an apparatus 700 for determining an amount of resources required for delivery of a media item according to some implementations of the present invention. The apparatus includes an extraction module 710 for extracting feature information of a target media item from related data of the target media item, an acquisition module 720 for obtaining, based on at least the feature information, a predicted value of an amount of resources required for multiple deliveries during competitive delivery of the target media item using a prediction model, where the predicted values ​​of the amount of resources required for the multiple deliveries respectively correspond to multiple predetermined probabilities that the target media item will be delivered, and a determination module 730 for determining an amount of resources required for delivery of the target media item from the predicted values ​​of the amount of resources required for the multiple deliveries based on multiple delivery efficiency measures respectively associated with the predicted values ​​of the amount of resources required for the multiple deliveries.

[0075] According to one exemplary implementation of the present invention, the acquisition module includes: a distribution determination module for determining a distribution of amounts of resources required for a plurality of deliveries, each associated with a plurality of predetermined probabilities, based on at least the feature information and utilizing a prediction model; and a predicted value acquisition module for determining predicted values ​​of amounts of resources required for a plurality of deliveries based on each of the distributions of amounts of resources required for the plurality of deliveries.

[0076] According to one exemplary implementation of the present invention, the predictive model includes a first network, which describes a first association relationship between an amount of resources required for a first reference delivery during a first reference competitive delivery of the first reference media item and first reference feature information of the first reference media item, and a distribution determination module uses the first network to determine a first distribution of the amount of resources required for multiple deliveries associated with multiple predetermined probabilities.

[0077] According to one exemplary implementation of the present invention, the predictive model includes a second network, the second network describing a second association relationship between the exhibition of the second reference media item during a second reference competitive distribution distributing the second reference media item and second reference feature information of the second reference media item, and a distribution determination module determines a distribution of resource amounts required for multiple distributions each associated with a plurality of predetermined probabilities, and determines a second distribution of resource amounts required for multiple distributions each associated with a plurality of predetermined probabilities using the second network.

[0078] According to one exemplary implementation of the present invention, the predicted value determination module further determines a predicted value of the amount of resources required for the multiple distributions based on the first distribution and the second distribution.

[0079] According to one exemplary implementation of the present invention, a plurality of delivery efficiency measures are determined based on a predicted amount of resources required for the plurality of deliveries, a value measure for the target media item, and a plurality of predetermined probabilities.

[0080] According to one exemplary implementation of the present invention, the apparatus adjusts a plurality of delivery efficiency measures based on proportional-integral-derivative control parameters.

[0081] According to one exemplary implementation of the present invention, the relationship between the predicted amount of resources required for multiple deliveries and the value measure of the target media item satisfies a predetermined constraint.

[0082] According to one exemplary implementation of the present invention, the value measure of a target media item is determined based on the amount of resources required to deliver the target media item, a predicted click-through rate for the target media item, and a predicted conversion rate for the target media item.

[0083] According to one exemplary implementation of the present invention, the determination module includes a selection module for selecting a first predicted value corresponding to the first delivery efficiency measure from the predicted values ​​of the amount of resources required for the plurality of deliveries in response to determining that a first delivery efficiency measure among the plurality of delivery efficiency measures is higher than a second delivery efficiency measure.

[0084] According to one exemplary implementation of the present invention, the characteristic information of the target media item includes at least one of the type of the target media item, the platform on which the target media item is distributed, the operating system of the platform on which the target media item is distributed, the client device of the platform on which the target media item is distributed, and the region in which the target media item is distributed.

[0085] 8 illustrates a block diagram of a device 800 capable of implementing multiple implementations of the present invention. It should be understood that the computing device 800 illustrated in FIG. 8 is merely exemplary and does not constitute any limitation on the functionality and scope of the implementations described herein. The computing device 800 illustrated in FIG. 8 is capable of implementing the methods described above.

[0086] 8, computing device 800 is a form of general-purpose electronic device. Components of computing device 800 may include, but are not limited to, one or more processors or processing units 810, memory 820, storage device 830, one or more communication units 840, one or more input devices 850, and one or more output devices 860. Processing unit 810 may be a real or virtual processor and may perform various processes based on programs stored in memory 820. In a multiprocessor system, multiple processing units execute computer-executable instructions in parallel, thereby increasing the parallel processing capabilities of computing device 800.

[0087] Computing device 800 typically includes a number of computer storage media. Such media may be any obtainable media accessible by computing device 800, including, but not limited to, volatile and nonvolatile media, removable and non-removable media. Memory 820 may be volatile memory (e.g., registers, cache, random access memory (RAM)), non-volatile memory (e.g., read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory), or some combination thereof. Storage device 830 may be removable or non-removable media and may include machine-readable media, such as a flash memory drive, a magnetic disk, or any other media, that may be used to store information and / or data (e.g., training data for training) and that may be accessible within computing device 800.

[0088] Computing device 800 may further include other removable / non-removable, volatile / non-volatile storage media. Although not shown in FIG. 8, a magnetic disk drive for reading from or writing to a removable, non-volatile magnetic disk (e.g., a “floppy disk”) and an optical disk drive for reading from or writing to a removable, non-volatile optical disk may be provided. In these cases, each drive may be connected to a path (not shown) by one or more data media interfaces. Memory 820 includes a computer program product 825 having one or more program modules configured to perform various methods or operations of various implementations of the present invention.

[0089] The communications unit 840 implements communications with other computing devices over a communications medium. Additionally, the functionality of the components of computing device 800 may be implemented as a single computing cluster or multiple computing machines, which can communicate over a communications connection. Thus, computing device 800 can operate in a networked environment using logical connections to one or more other servers, network personal computers (PCs), or other network nodes.

[0090] The input device(s) 850 may be one or more input devices such as a mouse, keyboard, trackball, etc. The output device(s) 860 may be one or more output devices such as a display, speakers, printer, etc. The computing device 800 may further communicate with one or more external devices (not shown) such as a storage device, a display device, etc. via the communication unit 840, as needed, to communicate with one or more devices that allow a user to interact with the computing device 800, or the computing device 800 may communicate with any device (such as a net card, modem, etc.) that communicates with one or more other computing devices. Such communication may be performed via an input / output (I / O) interface (not shown).

[0091] According to an exemplary implementation of the present invention, a computer-readable storage medium having one or more computer instructions stored thereon is provided, the one or more computer instructions being executed by a processor to implement the above-described method. According to an exemplary implementation of the present invention, a computer program product is further provided, the computer program product being tangibly stored on a non-transitory computer-readable medium and including computer-executable instructions that are executed by a processor to implement the above-described method. An exemplary implementation of the present invention provides a computer program product having a computer program stored therein, the computer program product implementing the above-described method when the program is executed by a processor.

[0092] Aspects of the present invention have been described herein with reference to flowchart and / or block diagrams of methods, apparatus (systems), and computer program products implemented by the present invention. It will be understood that each box in the flowchart and / or block diagrams, and combinations of boxes in the flowchart and / or block diagrams, can all be implemented by computer-readable program instructions.

[0093] These computer-readable program instructions may be provided to a processing unit of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus to generate a machine such that, when the instructions are executed by the processing unit of the computer or other programmable data processing apparatus, it generates an apparatus for implementing the functions / acts specified in one or more boxes in the flowcharts and / or block diagrams. These computer-readable program instructions may be stored on a computer-readable storage medium such that the instructions cause the computer, programmable data processing apparatus, and / or other device to operate in a particular manner such that the computer-readable medium on which the instructions are stored constitutes an article of manufacture containing instructions that implement each aspect of the functions / acts specified in one or more boxes in the flowcharts and / or block diagrams.

[0094] Loading the computer-readable program instructions into a computer, other programmable data processing apparatus, or other device causes the computer, other programmable data processing apparatus, or other device to perform a series of operational steps to produce a computer-implemented process, such that the instructions executing on the computer, other programmable data processing apparatus, or other device implement the functions / acts specified in one or more boxes in the flowcharts and / or block diagrams.

[0095] The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of various possible systems, methods, and computer program products according to the present invention. In this regard, each box in a flowchart or block diagram may represent a module, program fragment, or portion of instructions, which includes one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions depicted in the boxes may occur in a different order than depicted in the figures. For example, two consecutive boxes may actually execute substantially in parallel, or may execute in reverse order, depending on the functionality involved. It should also be noted that each box in the block diagrams and / or flowcharts, and combinations of boxes in the block diagrams and / or flowcharts, may be implemented by a special-purpose hardware-based system that performs the specified function(s) or operation(s), or by a combination of special-purpose hardware and computer instructions.

[0096] Although various implementations of the present invention have been described above, the above descriptions are illustrative and not exhaustive, and are not limited to the disclosed implementations. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described implementations. The terminology used herein is intended to best interpret the principles, practical applications, or improvements to commercially available technologies of the implementations, or to enable those skilled in the art to understand the implementations disclosed herein.

Claims

1. A computer-implemented method for determining an amount of resources required for delivery of a media item, comprising: extracting feature information of the target media item from the associated data of the target media item; utilizing a prediction model to obtain, based on at least the characteristic information, a plurality of predicted values ​​of the amount of resources required for a plurality of deliveries of the target media item during competitive delivery, the plurality of predicted values ​​of the amount of resources required for the plurality of deliveries corresponding to a plurality of predetermined probabilities of the target media item being delivered, respectively; determining an amount of resources required for delivery of the target media item from the plurality of predicted amounts of resources required for the plurality of deliveries based on a plurality of delivery efficiency measures respectively associated with the plurality of predicted amounts of resources required for the plurality of deliveries; A method for determining the amount of resources required for delivery of a media item.

2. Obtaining the plurality of predicted values ​​of the amount of resources required for the plurality of deliveries includes: determining, based on at least the characteristic information, a plurality of distributions of resource amounts required for a plurality of deliveries, each distribution being associated with a plurality of predetermined probabilities, using the prediction model; determining the plurality of predicted values ​​of the amount of resources required for the plurality of distributions based on the plurality of distributions of the amount of resources required for the plurality of distributions, respectively; The method of claim 1.

3. The prediction model includes a first network, the first network describing a first association relationship between a resource amount required for a first reference distribution during a first reference competitive distribution of a first reference media item and first reference characteristic information of the first reference media item; determining the plurality of distributions of amounts of resources required for a plurality of deliveries associated with the plurality of predetermined probabilities, respectively, includes determining a plurality of first distributions of amounts of resources required for a plurality of deliveries associated with the plurality of predetermined probabilities using the first network. The method of claim 2.

4. the predictive model includes a second network, the second network describing a second association relationship between a display of the second reference media item during a second reference competitive distribution of the second reference media item and second reference characteristic information of the second reference media item; determining the plurality of distributions of amounts of resources required for a plurality of deliveries each associated with a plurality of predetermined probabilities includes determining, using the second network, a plurality of second distributions of amounts of resources required for a plurality of deliveries each associated with a plurality of predetermined probabilities; The method of claim 3.

5. obtaining the plurality of predicted values ​​of amounts of resources required for the plurality of deliveries includes determining the plurality of predicted values ​​of amounts of resources required for the plurality of deliveries based on the plurality of first distributions and the plurality of second distributions; The method of claim 4.

6. the plurality of delivery efficiency measures are determined based on the plurality of predicted values ​​of the amount of resources required for the plurality of deliveries, the value measure of the target media item, and the plurality of predetermined probabilities. The method of claim 1.

7. further comprising adjusting the plurality of delivery efficiency measures based on proportional-integral-derivative control parameters. The method of claim 6.

8. a relationship between the plurality of predicted amounts of resources required for the plurality of deliveries and the value measure of the target media item satisfies a predetermined constraint; The method of claim 6.

9. a value measure for the target media item is determined based on an amount of resources required for delivery of the target media item, a predicted click-through rate for the target media item, and a predicted conversion rate for the target media item; The method of claim 1.

10. determining the amount of resources required for the delivery based on the plurality of delivery efficiency measures includes, in response to determining that a first delivery efficiency measure among the plurality of delivery efficiency measures is higher than a second delivery efficiency measure, selecting a first predicted value corresponding to the first delivery efficiency measure from the plurality of predicted values ​​of the amount of resources required for the plurality of deliveries. The method of claim 1.

11. The characteristic information of the target media item includes at least one of the following: a type of the target media item; a platform on which the target media item is distributed; an operating system of the platform on which the target media item is distributed; a client device of the platform on which the target media item is distributed; and a region on which the target media item is distributed. The method of claim 1.

12. an extraction module for extracting feature information of the target media item from the related data of the target media item; an acquisition module for utilizing a prediction model to acquire, based on at least the feature information, a plurality of predicted values ​​of the amount of resources required for a plurality of deliveries of the target media item during competitive delivery, the plurality of predicted values ​​of the amount of resources required for the plurality of deliveries corresponding to a plurality of predetermined probabilities that the target media item will be delivered; a determination module for determining an amount of resources required for delivery of the target media item from the plurality of predicted amounts of resources required for the plurality of deliveries based on a plurality of delivery efficiency measures respectively associated with the plurality of predicted amounts of resources required for the plurality of deliveries; An apparatus for determining the amount of resources required for delivery of a media item.

13. 1. An electronic device comprising: at least one processing unit; and at least one memory coupled to said at least one processing unit and adapted to store instructions to be executed by said at least one processing unit, said instructions, when executed by said at least one processing unit, causing said electronic device to perform the method of any one of claims 1 to 11. Electronic devices.

14. a computer program stored therein, the computer program causing a processor to implement the method of any one of claims 1 to 11 when executed by the processor; A computer-readable storage medium.

Images