Information processing device, prediction model generation method, prediction method, and program

Abstract

This invention provides an information processing device, a method for generating a prediction model, a prediction method, and a program that can predict the environmental impact associated with bidding on performance-based advertising. [Solution] One aspect of the information processing device of the present disclosure includes: a load amount acquisition unit that acquires a first load amount indicating the environmental load incurred in connection with the bidding of programmatic advertising; a performance data acquisition unit that acquires performance data identified from the delivery performance of the programmatic advertising; and a prediction model generation unit that generates a prediction model for predicting a second load amount indicating the environmental load incurred in connection with a new bid for programmatic advertising, using the first load amount and the performance data.

Description

Technical Field

[0001] The present disclosure relates to an information processing apparatus, a prediction model generation method, a prediction method, and a program.

Background Art

[0002] In recent years, environmental problems have been worsening, and the importance of efforts to reduce environmental impact has been increasing not only in fields such as manufacturing that handle physical products but also in all fields such as non-manufacturing.

[0003] For example, Patent Document 1 discloses a technique for calculating the amount of greenhouse gas emissions required for advertising production and distribution in the field of advertising, calculating the credit amount that offsets the calculated emissions, and selecting greenhouse gas reduction activities corresponding to the calculated credit amount. According to the technique disclosed in Patent Document 1, advertising agencies and advertising media providers are said to be able to easily perform carbon footprint to carbon offset.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] [[ID=...]] [[ID=...]]

[0006] [[ID=...]] [[ID=4...]]

[0006] [[ID=4...]]

[0006] By the way, in the recent advertising market, the proportion of digital advertising conducted on the Internet has been increasing, and among them, operation-based advertising that distributes advertisements in an auction format has become the mainstream. Therefore, efforts to reduce the environmental impact of operation-based advertising are expected. <...]] [Means for solving the problem]

[0007] One aspect of the information processing device of the present disclosure includes: a load amount acquisition unit that acquires a first load amount indicating the environmental burden incurred in connection with the bidding of programmatic advertising; a performance data acquisition unit that acquires performance data identified from the delivery performance of the programmatic advertising; and a prediction model generation unit that generates a prediction model for predicting a second load amount indicating the environmental burden incurred in connection with a new bidding of programmatic advertising using the first load amount and the performance data.

[0008] One aspect of the information processing device of the present disclosure includes: a receiving unit that receives input of expected performance data indicating expected delivery performance in a new programmatic advertisement; a prediction unit that inputs the expected performance data into a prediction model generated using a first load amount indicating the environmental burden incurred in connection with past programmatic advertisement bidding and performance data identified from the delivery performance of said past programmatic advertisement, in order to predict a second load amount indicating the environmental burden incurred in connection with the new programmatic advertisement bidding; and an output unit that outputs the predicted second load amount.

[0009] One aspect of the predictive model generation method of this disclosure includes: a load amount acquisition step in which a load amount acquisition unit acquires a first load amount indicating the environmental load incurred in connection with the bidding of programmatic advertising; a performance data acquisition step in which a performance data acquisition unit acquires performance data identified from the delivery performance of the programmatic advertising; and a predictive model generation step in which a predictive model is generated by a predictive model using the first load amount and the performance data to predict a second load amount indicating the environmental load incurred in connection with the bidding of new programmatic advertising.

[0010] One aspect of the prediction method of this disclosure includes: a reception step in which a reception unit receives input of expected performance data indicating expected delivery performance for a new programmatic advertisement; a prediction step in which a prediction unit inputs the expected performance data into a prediction model generated using a first load amount indicating the environmental burden incurred in connection with past programmatic advertisement bidding and performance data identified from the delivery performance of said past programmatic advertisement, in order to predict a second load amount indicating the environmental burden incurred in connection with the new programmatic advertisement bidding; and an output step in which an output unit outputs the predicted second load amount.

[0011] One aspect of the program of this disclosure is for causing a computer to perform the above method. [Effects of the Invention]

[0012] According to this disclosure, it is possible to provide an information processing device, a method for generating a prediction model, a prediction method, and a program that can predict the environmental impact associated with bidding on programmatic advertising. [Brief explanation of the drawing]

[0013] [Figure 1] Figure 1 is a block diagram showing an example of the system configuration of the advertising distribution system of this embodiment. [Figure 2] Figure 2 is a flowchart showing an example of RTB (Real Time Bidding), which is an advertising delivery process performed in the advertising delivery system of this embodiment. [Figure 3] Figure 3 is a block diagram showing an example of the system configuration of the environmental load prediction system of this embodiment. [Figure 4] Figure 4 is a block diagram showing an example of the hardware configuration of the information processing device of this embodiment. [Figure 5] Figure 5 is a block diagram showing an example of the functional configuration of the information processing device according to this embodiment. [Figure 6] Figure 6 shows an example of actual data from this embodiment. [Figure 7] Figure 7 shows an example of the first load amount in this embodiment. [Figure 8] FIG. 8 is a flowchart showing an example of prediction model generation processing performed by the information processing apparatus of the present embodiment. [Figure 9] FIG. 9 is a flowchart showing an example of prediction processing performed by the information processing apparatus of the present embodiment. MODE FOR CARRYING OUT THE INVENTION

[0014] Hereinafter, embodiments of the present disclosure (hereinafter simply referred to as "the present embodiment") will be described in detail with reference to the drawings. Note that the present disclosure is not limited to the following embodiments.

[0015] First, taking an operation-based advertisement that distributes advertisements in an auction format as an example, an advertisement distribution system that is a premise of the present embodiment will be described.

[0016] FIG. 1 is a block diagram showing an example of the system configuration of an advertisement distribution system 1 of the present embodiment. As shown in FIG. 1, the advertisement distribution system 1 includes an SSP (Supply Side Platform) 10, DSPs (Demand Side Platforms) 20-1 to 20-n (n is a natural number), and media servers 30-1 to 30-m (m is a natural number).

[0017] The SSP 10, DSPs 20-1 to 20-n, and media servers 30-1 to 30-m are connected via a network 2. The network 2 can be realized by, for example, at least any one of the Internet and a LAN (Local Area Network). The network 2 may be a wired network, a wireless network, or a mixture of a wired network and a wireless network.

[0018] In the following description, when it is not necessary to distinguish between the DSPs 20-1 to 20-n, they may be simply referred to as DSP 20, and when it is not necessary to distinguish between the media servers 30-1 to 30-m, they may be simply referred to as media server 30.

[0019] SSP10 is an order-receiving platform that hosts auctions for operational advertisements, and can be implemented by, for example, one or more server devices. Specifically, SSP10 holds an auction for the right to distribute (publish) advertisements to the advertisement slots owned by the media, and determines the winning bidder for the right to distribute advertisements in an auction format from the bidders who wish to distribute advertisements.

[0020] Examples of media include, but are not limited to, Web media that transmit some information viewed from the Internet, such as video distribution sites and SNS (Social Networking Service). Examples of bidders include, but are not limited to, the order-placing side of advertisements, such as advertisers and advertising agencies.

[0021] DSP20-1 to 20-n are multiple order-placing platforms for participating in auctions for operational advertisements. DSP20 can be implemented by, for example, one or more server devices. Specifically, DSP20 places a bid with SSP10 in order to win the right to distribute (publish) advertisements to the advertisement slots owned by the media.

[0022] The media server 30 is a server device for providing Web media. When the right to distribute advertisements to the advertisement slots of the Web media is won, the media server 30 displays the advertisements distributed from the DSP20 used by the winning bidder in the advertisement slots. In this embodiment, it is assumed that the content and operator of the Web media provided by each media server 30 are different, but it is not limited to this.

[0023] FIG. 2 is a flowchart showing an example of RTB (Real Time Bidding), which is an advertisement distribution process performed in the advertisement distribution system 1 of this embodiment.

[0024] First, when a user is browsing a website of a web media provided by the media server 30 using a user terminal (not shown), and comes into contact with an advertisement space on the web media, the media server 30 requests an advertisement from the SSP 10 along with information such as the user's attributes (step S101).

[0025] Next, SSP10 transmits information such as user attributes to DSP20-1 to 20-n and requests their participation in the bidding for the right to deliver advertisements to advertising slots on web media (step S103).

[0026] Next, each DSP20 solicits bidders and, based on information such as the bid amounts of the bidders and the attributes of the users received, conducts a pre-auction within the DSP20. The bidder who submits the highest bid amount is selected as the bidder to submit to SSP10, and submits a bid to SSP10 at that amount (step S105).

[0027] Next, SSP10 selects the bidder with the highest bid from among the bidders from each DSP20 and determines that they have won the right to deliver advertisements to the advertising space on the web media. SSP10 then communicates various information, such as reporting the successful bid to the DSP20 used by the successful bidder (not shown in the diagram), and reports the information of the DSP20 to the media server 30 (step S107).

[0028] Next, the media server 30 requests the DSP 20 used by the successful bidder to deliver the advertisement (step S109).

[0029] Next, the DSP20 used by the successful bidder delivers the requested advertisement to the media server 30 (step S111). As a result, the advertisement delivered from the DSP20 used by the successful bidder is displayed in the advertising space of the web media and viewed by users.

[0030] In this way, with programmatic advertising, when a user comes across an ad space on a web media, an auction to determine who will deliver the ad is conducted automatically and in real time. However, with the increase in programmatic advertising and the increasing complexity of the auction process, the amount of data traffic associated with bidding in the auction is increasing, leading to an environmental burden.

[0031] Next, an environmental load prediction system including the information processing device of this embodiment will be described. The information processing device of this embodiment predicts the environmental load generated in the advertising distribution system described above. Environmental loads include, but are not limited to, emissions of greenhouse gases such as carbon dioxide (CO2).

[0032] Specifically, the information processing device of this embodiment predicts the environmental burden caused by communication and other activities associated with bidding on programmatic advertising by the DSP20. This makes it possible to provide consulting services using predicted environmental burden values ​​to advertisers, advertising agencies, and other parties ordering advertising using the DSP20.

[0033] Figure 3 is a block diagram showing an example of the system configuration of the environmental load prediction system 100 of this embodiment. As shown in Figure 3, the environmental load prediction system 100 comprises an SSP 10, a DSP 20-1, and an information processing device 40.

[0034] SSP10, DSP20-1, and information processing device 40 are connected via network 102. Network 102 can be implemented using the same method as network 2, and may be the same network as network 2 or a different network. Note that the information processing device 40 may also be in a standalone configuration and not connected to network 102.

[0035] SSP10 and DSP20-1 are as described above. Note that DSP20-1 can be any DSP that is the target of environmental load prediction, and may be any DSP other than DSP20-1, such as any of DSP20-2 to 20-n.

[0036] The information processing device 40 is used to predict the environmental impact generated by communications, etc., for the DSP 20-1 to bid on the SSP 10, and can be implemented using one or more computers. In this embodiment, it is assumed that the DSP 20-1 and the information processing device 40 are operated by the same operator, but it is not limited to this. In this configuration, the operator can predict the environmental impact that will occur when bidding on programmatic advertising with its DSP 20-1, and can provide consulting services to advertisers, advertising agencies, and other parties ordering advertising, such as advertising strategies that take environmental impact into consideration.

[0037] Figure 4 is a block diagram showing an example of the hardware configuration of the information processing device 40 of this embodiment. As shown in Figure 4, the information processing device 40 comprises a control device 41, a main memory 42, an auxiliary storage device 43, a communication device 44, an input device 45, a display device 46, and various buses 49. The control device 41, main memory 42, auxiliary storage device 43, communication device 44, input device 45, and display device 46 are connected via various buses 49. Thus, the information processing device 40 of this embodiment has a general hardware configuration using a normal computer.

[0038] The control device 41 controls the overall operation of the information processing device 40. The control device 41 may be, for example, at least one of a CPU (Central Processing Unit) and a GPU (Graphics Processing Unit), but is not limited to these. There may be one or more CPUs or GPUs, and they may be single-core or multi-core.

[0039] Examples of main memory 42 include, but are not limited to, ROM (Read Only Memory) and RAM (Random Access Memory). ROM stores various programs, such as a program for controlling the information processing device 40, a program for generating a prediction model for predicting environmental impact, and a program for predicting the environmental impact that occurs when the DSP 20 bids on programmatic advertising using the prediction model. RAM is used as a work area when the control device 41 performs various controls based on the programs stored in ROM.

[0040] The auxiliary storage device 43 stores the various programs mentioned above, as well as various data used to generate prediction models and predict environmental load. The various programs mentioned above only need to be stored in at least one of the main storage device 42 and the auxiliary storage device 43. Examples of auxiliary storage devices 43 include, but are not limited to, at least one of existing storage devices capable of magnetic, electrical, or optical storage, such as HDDs (Hard Disk Drives), SSDs (Solid State Drives), and DVDs (Digital Versatile Discs). The auxiliary storage device 43 may be built into the information processing device 40 or externally connected to the information processing device 40 via an interface such as USB (Universal Serial Bus). Furthermore, the auxiliary storage device 43 may be a NAS (Network Attached Storage) connected via a network such as a LAN or WAN (Wide Area Network).

[0041] The communication device 44 is used to communicate with the SSP10 and DSP20-1, etc., via the network 102. Examples of the communication device 44 include, but are not limited to, a communication device for a wired LAN or a wireless communication device for a wireless LAN.

[0042] The input device 45 is used for various inputs, selections, and specifications for generating a predictive model or predicting environmental load, and serves as a user interface with the user. Examples of input devices 45 include, but are not limited to, a keyboard, mouse, and touch panel. The input device 45 may be built into the information processing device 40 or be externally connected to the information processing device 40 via an interface such as USB.

[0043] The display device 46 displays various screens and serves as a user interface between the user and the device. Examples of the display device 46 include, but are not limited to, liquid crystal displays, organic electro-luminescence (OLED) displays, and touch panel displays. The display device 46 may be an internal display built into the information processing device 40, or an external display connected to the information processing device 40 via a display interface such as HDMI®.

[0044] In addition to the above configuration, the information processing device 40 may further include hardwired circuits such as ICs (Integrated Circuits), ASICs (Application Specific Integrated Circuits), and FPGAs (Field-Programmable Gate Arrays) that are specific to the information processing device 40.

[0045] Figure 5 is a block diagram showing an example of the functional configuration of the information processing device 40 in this embodiment. As shown in Figure 5, the information processing device 40 includes a performance data acquisition unit 401, a load amount acquisition unit 403, a prediction model generation unit 405, a prediction model storage unit 407, a reception unit 411, a prediction unit 413, and an output unit 415. The performance data acquisition unit 401, the load amount acquisition unit 403, the prediction model generation unit 405, the reception unit 411, the prediction unit 413, and the output unit 415 can be realized, for example, by the control device 41, main memory 42, and communication device 44 described in Figure 4. The prediction model storage unit 407 can be realized, for example, by the auxiliary storage device 43 described in Figure 4.

[0046] For example, the control device 41 reads a program for generating a prediction model stored in the main memory 42 (ROM) or auxiliary memory 43 and loads it into the main memory 42 (RAM). The control device 41 then executes various processes according to the loaded program to realize the aforementioned actual data acquisition unit 401, load amount acquisition unit 403, and prediction model generation unit 405. Alternatively, for example, the control device 41 reads a program for predicting environmental load stored in the main memory 42 (ROM) or auxiliary memory 43 and loads it into the main memory 42 (RAM). The control device 41 then executes various processes according to the loaded program to realize the aforementioned receiving unit 411, prediction unit 413, and output unit 415.

[0047] Here, we have explained the case where each of the above-mentioned functional units is implemented as software, but at least a part of each of the above-mentioned functional units may be implemented as hardware. In this case, the functional unit to be implemented as hardware can be implemented, for example, by the hardwired circuit described above. Alternatively, any of the above-mentioned functional units may be implemented through the cooperation of software and hardware.

[0048] The following describes each functional unit in detail. First, we will explain the actual data acquisition unit 401, the load amount acquisition unit 403, and the prediction model generation unit 405, which are used to generate the prediction model.

[0049] The performance data acquisition unit 401 acquires performance data identified from the delivery performance of programmatic advertising. Specifically, the performance data acquisition unit 401 acquires performance data from DSP20-1, which is one of the ordering platforms among the multiple ordering platforms DSP20-1~20-n that participate in the bidding for programmatic advertising.

[0050] The performance data of this embodiment shows the advertising delivery performance for each media outlet providing advertising space in DSP20-1. The items that can constitute the performance data include, but are not limited to, actual advertising delivery values ​​that are also used as operational indicators for advertising delivery, such as impressions, clicks, cost per click (CPC), click-through rate (CTR), amount, conversion rate (CV), cost per action (CPA), conversion rate (CVR), cost per mille (CPM), and bid success rate.

[0051] The number of impressions is the number of times an advertisement (performance-based advertising) was displayed in the advertising space of a web media, and in this embodiment, it is synonymous with the number of times the right to deliver the advertisement was successfully bid on. The number of clicks is the number of times the displayed advertisement was clicked. CPC is the cost per click, which indicates the cost to acquire one click. CTR is the click-through rate, which indicates the percentage of displayed advertisements that were clicked. The amount indicates the cost to display the advertisement for the number of impressions. CV is the number of times the desired outcome (e.g., product purchase) was achieved on the web. CPA is the cost per conversion. CVR is the conversion rate, which indicates the percentage of clicked advertisements that resulted in a conversion. CPM indicates the advertising cost per 1000 impressions. The successful bid rate indicates the ratio of successful bids to the number of bids for the right to deliver the advertisement.

[0052] In this embodiment, the performance data includes, but is not limited to, the number of impressions and the successful bid rate of the programmatic advertisement. Figure 6 shows an example of the performance data in this embodiment. In the example shown in Figure 6, the performance data consists of the media domain, the number of impressions, and the successful bid rate, and shows the number of impressions and successful bid rate for each media within a certain period, as performed by DSP20-1. The media domain indicates the domain of the media and is used as media identification information. For example, in the case of the media AAA.co.jp, the delivery performance of DSP20-1 within a certain period is 16,000 impressions and a successful bid rate of 0.03.

[0053] In this embodiment, the performance data acquisition unit 401 is assumed to acquire performance data from the DSP 20-1 via the network 102, but it is not limited to this configuration. Performance data may be pre-stored in the auxiliary storage device 43 of the information processing device 40, and the performance data acquisition unit 401 may acquire the performance data from the auxiliary storage device 43.

[0054] The load acquisition unit 403 acquires a first load amount that indicates the environmental burden generated in connection with the bidding for programmatic advertising. Specifically, the load acquisition unit 403 acquires the first load amount from the SSP 10, which is the receiving platform that hosts the bidding for programmatic advertising.

[0055] In this embodiment, the first load is a load measured based on the communication volume associated with the bidding of programmatic advertising, and is assumed to be measured on the SSP10 side based on the communication volume with DSP20-1 associated with the bidding of programmatic advertising. The first load may be measured by the operator of SSP10 or by an external contractor. In this embodiment, the first load shows the load amount for each media that represents the environmental load generated by the bidding of each media by DSP20-1.

[0056] Figure 7 shows an example of the first load amount in this embodiment. In the example shown in Figure 7, the first load amount consists of the media domain and CO2 emissions, and for each media, it shows the CO2 emissions generated in connection with bidding by DSP20-1 within a certain period. The certain period is assumed to be the same period as the actual data. For example, in the case of the AAA.co.jp media, the CO2 emissions generated in connection with bidding by DSP20-1 within a certain period is 0.007.

[0057] In this embodiment, the load amount acquisition unit 403 is assumed to acquire the first load amount from the SSP 10 via the network 102, but it is not limited to this, and the first load amount may be stored in advance in the auxiliary storage device 43 of the information processing device 40, and the load amount acquisition unit 403 may acquire the first load amount from the auxiliary storage device 43.

[0058] The prediction model generation unit 405 uses the first load amount acquired by the load amount acquisition unit 403 and the actual data acquired by the actual data acquisition unit 401 to generate a prediction model for predicting the second load amount, which represents the environmental burden associated with bidding on new performance-based advertising.

[0059] Any method can be used to generate the predictive model. For example, a predictive model expressed as a regression equation may be generated (derived) by analyzing the first loading and actual data using statistical analysis methods such as simple linear regression or multiple linear regression, or a trained predictive model may be generated by machine learning using the first loading and actual data as training data.

[0060] The following describes an example of generating a predictive model by performing multiple regression analysis on actual data as shown in Figure 6 and the first loading as shown in Figure 7, but the method for generating a predictive model is not limited to this. Note that the parameters obtained by the multiple regression analysis mentioned below are merely examples and are not parameters obtained using the actual data shown in Figure 6 or the first loading as shown in Figure 7, but this does not limit the content of this disclosure in any way.

[0061] For example, the prediction model generation unit 405 defines the regression equation Z = αX + βY. Here, X is the number of displays / winning bid rate, Y is the number of displays, Z is the CO2 emissions, and α and β are the regression coefficients. Under these conditions, the prediction model generation unit 405 performs a multiple regression analysis using the actual data shown in Figure 6 and the first loading shown in Figure 7, deriving the values ​​α = 3.876 and β = 291.0, and obtains the regression equation represented by the following formula (1) as the prediction model.

[0062] Z = (3.876X + 291.0Y) × 10 -9 …(1)

[0063] Furthermore, calculating the p-values ​​for α and β, we get α = 0.000442 and β = 3.59 × 10⁻¹⁰. -13 This is the result. Furthermore, calculating the 5% significance level, we get 0.025, so we can conclude that both α and β are statistically significant.

[0064] As described above, the prediction model generation unit 405 generates a prediction model using the first load amount and actual data, and stores it in the prediction model storage unit 407.

[0065] Next, we will describe the reception unit 411, the prediction unit 413, and the output unit 415, which are used to predict environmental load using a prediction model.

[0066] The reception unit 411 accepts input of projected performance data that indicates the expected delivery performance for the new programmatic advertising. Specifically, the reception unit 411 accepts projected performance data entered by the operators of the information processing device 40 and DSP 20-1, the advertising client, etc., using the input device 45.

[0067] The projected performance data represents the expected delivery performance for any media in the new programmatic advertising campaign, and consists of the same items as the actual performance data. Therefore, in this embodiment, the projected performance data shows the projected number of impressions and projected bid success rate for any media, as determined by DSP20-1.

[0068] For example, using the CPM included in the performance data, the expected number of impressions can be derived from the budget allocated to a new programmatic advertising campaign. Furthermore, the expected bid success rate is the same as the bid success rate included in the performance data. Therefore, the expected number of impressions and expected bid success rate for any given media can be determined in advance using DSP20-1.

[0069] The prediction unit 413 inputs the assumed performance data received by the reception unit 411 into the prediction model generated by the prediction model generation unit 405 to predict the second load amount, which indicates the environmental burden that will be incurred as a result of bidding on new programmatic advertising. As described above, the prediction model is generated using the first load amount, which indicates the environmental burden incurred as a result of past programmatic advertising bidding, and performance data identified from the delivery performance of said past programmatic advertising, and is stored in the prediction model storage unit 407.

[0070] Specifically, the prediction unit 413 calculates the assumed CO2 emissions, which are the second load, as Z by substituting X = assumed number of displays / assumed bid success rate and Y = assumed number of displays into the regression equation represented by formula (1) above.

[0071] The output unit 415 outputs the second load amount predicted by the prediction unit 413. For example, the output unit 415 displays the second load amount predicted by the prediction unit 413 on a display device 46 or the like.

[0072] For example, suppose the target audiences of AAA.co.jp and BBB.co.jp are similar (e.g., Gen Z and Millennials). In such a case, by predicting and comparing the secondary load for each media outlet, it becomes possible to provide consulting services to advertisers that select media outlets based not only on advertising effectiveness but also on environmental impact reduction.

[0073] Furthermore, as is clear from the regression equation shown in formula (1), the higher the bid success rate, the fewer unnecessary bids there are, and therefore the greater the effect of reducing environmental impact. For this reason, even for the same media, it is possible to provide consulting services to clients that derive and propose an equilibrium point that maximizes the cost-effectiveness of advertising and the effect of reducing environmental impact by adjusting the assumed bid success rate, that is, by adjusting the assumed bid amount.

[0074] Figure 8 is a flowchart showing an example of the predictive model generation process performed in the information processing device 40 of this embodiment.

[0075] First, the performance data acquisition unit 401 acquires performance data identified from the delivery performance of performance-based advertising (step S201).

[0076] Next, the load acquisition unit 403 acquires a first load amount indicating the environmental burden generated in connection with the bidding for programmatic advertising (step S203).

[0077] Next, the prediction model generation unit 405 uses the actual data acquired by the actual data acquisition unit 401 and the first load amount acquired by the load amount acquisition unit 403 to generate a prediction model for predicting a second load amount that indicates the environmental burden associated with bidding on new performance-based advertising (step S205).

[0078] Figure 9 is a flowchart showing an example of the prediction process performed in the information processing device 40 of this embodiment.

[0079] First, the reception unit 411 accepts input of projected performance data that shows the expected delivery performance for the new programmatic advertising (step S301).

[0080] Next, the prediction unit 413 inputs the assumed performance data received by the reception unit 411 into the prediction model generated by the prediction model generation unit 405 to predict a second load amount that indicates the environmental burden associated with bidding on new performance-based advertising (step S303).

[0081] Next, the output unit 415 outputs the second load amount predicted by the prediction unit 413 (step S305).

[0082] As described above, in this embodiment, it is possible to predict in advance the environmental impact generated by programmatic advertising, so that measures to reduce environmental impact can be considered before launching new programmatic advertising (advertising campaigns). Therefore, according to this embodiment, it is possible to provide consulting services such as advertising placement strategies that take into account not only advertising effectiveness but also environmental impact reduction effects.

[0083] In particular, in this embodiment, even when launching a new programmatic advertising campaign, CO2 emissions can be predicted from easily foreseeable factors such as the number of impressions and the bid rate, making it possible to easily predict the environmental impact of the new programmatic advertising.

[0084] (program) The program executed by the information processing device of the above embodiment is provided in an installable or executable file format stored on a computer-readable storage medium such as a CD-ROM, CD-R, memory card, DVD, or flexible disk (FD).

[0085] Furthermore, the program executed by the information processing device of the above embodiment may be stored on a computer connected to a network such as the Internet and provided by downloading it via the network. Alternatively, the program executed by the information processing device of the above embodiment may be provided or distributed via a network such as the Internet. Alternatively, the program executed by the information processing device of the above embodiment may be pre-installed in ROM or the like and provided.

[0086] The program executed by the information processing device of the above embodiment has a modular configuration for realizing each of the above-described parts on a computer. In actual hardware, for example, the CPU reads the program from the HDD into RAM and executes it, thereby realizing each of the above-described parts on the computer.

[0087] The embodiments described above are merely examples of how this disclosure may be implemented, and they do not restrict the technical scope of this disclosure. Therefore, this disclosure can be implemented in various ways without departing from its essence or its main features. For example, in the embodiments described above, some components may be removed from the entire set of components.

[0088] This disclosure includes the following aspects:

[0089] (1) A load amount acquisition unit that acquires a first load amount indicating the environmental burden incurred in connection with the bidding of programmatic advertising, A performance data acquisition unit that acquires performance data identified from the performance data of the aforementioned performance-based advertising, A prediction model generation unit generates a prediction model for predicting a second load amount that indicates the environmental burden associated with bidding on new programmatic advertising, using the first load amount and the performance data. An information processing device equipped with the following features.

[0090] (2) A reception unit that accepts input of projected performance data showing the expected delivery performance in the new programmatic advertising, A prediction unit that inputs the assumed actual data into the prediction model to predict the second load, The system further comprises an output unit that outputs the predicted second load amount, The information processing device described in (1) above.

[0091] (3) The performance data includes at least one of the number of impressions and the bid success rate of the performance-based advertisement, The information processing device described in (1) or (2) above.

[0092] (4) The performance data is the delivery performance for each media outlet that provides advertising space on one of the multiple ordering platforms that participate in the bidding for the performance-based advertising. The first load amount indicates the amount of environmental load generated for each of the media by the ordering platform of the first, as a result of bidding for each of the media. The information processing device described in (1) above.

[0093] (5) The performance data is the delivery performance for each media outlet that provides advertising space on one of the multiple ordering platforms that participate in the bidding for the performance-based advertising. The first load amount indicates the amount of environmental load generated by the bidding for each of the media by the ordering platform of the first, and is shown for each of the media. The aforementioned projected performance data represents the delivery performance for any media expected in the new programmatic advertising. The information processing device described in (2) above.

[0094] (6) The first load is the load measured on the receiving platform that hosts the bidding for the programmatic advertising, based on the amount of communication with the ordering platform in the aforementioned 1 in connection with the bidding for the programmatic advertising. The information processing device described in (4) or (5) above.

[0095] (7) The prediction model is expressed by a regression equation, The information processing device described in (1) or (2) above.

[0096] (8) A reception unit that accepts input of projected performance data showing the expected delivery performance in the new programmatic advertising, A prediction unit that inputs the assumed performance data into a prediction model generated using a first load amount indicating the environmental burden associated with past programmatic advertising bids and performance data identified from the delivery performance of said past programmatic advertising, in order to predict a second load amount indicating the environmental burden associated with the new programmatic advertising bid, An output unit that outputs the predicted second load amount, An information processing device equipped with the following features.

[0097] (9) Load acquisition step, in which the load acquisition unit acquires a first load amount that indicates the environmental load generated in connection with the bidding of programmatic advertising, The performance data acquisition unit performs a performance data acquisition step in which it acquires performance data identified from the delivery performance of the aforementioned performance-based advertising, A prediction model generation step in which the prediction model generation unit generates a prediction model for predicting a second load amount that indicates the environmental load associated with bidding on new performance-based advertising, using the first load amount and the performance data, A method for generating predictive models that includes this.

[0098] (10) A reception step in which the reception department accepts input of expected performance data that shows the expected delivery performance for the new programmatic advertising, The prediction step involves the prediction unit inputting the assumed performance data into a prediction model generated using a first load amount indicating the environmental burden associated with past programmatic advertising bids and performance data identified from the past programmatic advertising delivery performance, in order to predict a second load amount indicating the environmental burden associated with the new programmatic advertising bid. The output unit performs an output step that outputs the predicted second load amount, A prediction method that includes this.

[0099] (11) A program that causes a computer to perform the method described in (9) or (10) above. [Explanation of Symbols]

[0100] 1. Ad delivery system 2 Network 10 SSP 20-1~20-n, 20 DSP 30-1~30-m, 30 media servers 40 Information Processing Devices 100 Environmental Load Prediction System 102 Network 401 Performance Data Acquisition Department 403 Load amount acquisition part 405 Predictive Model Generation Unit 407 Predictive Model Memory Unit 411 Reception Department 413 Prediction Section Output section of 415

Claims

1. A load amount acquisition unit acquires a first load amount that indicates the environmental burden incurred as a result of bidding for performance-based advertising, A performance data acquisition unit that acquires performance data identified from the performance data of the aforementioned performance-based advertising, A prediction model generation unit generates a prediction model for predicting a second load amount that indicates the environmental burden associated with bidding on new programmatic advertising, using the first load amount and the performance data. An information processing device equipped with the following features.

2. A reception unit that accepts input of projected performance data showing the expected delivery performance in the aforementioned new programmatic advertising, A prediction unit that inputs the assumed actual data into the prediction model to predict the second load, The system further comprises an output unit that outputs the predicted second load amount, The information processing apparatus according to claim 1.

3. The aforementioned performance data includes at least one of the number of impressions and the bid success rate of the performance-based advertisement. The information processing apparatus according to claim 1 or 2.

4. The aforementioned performance data represents the delivery performance for each media outlet providing advertising space on one of the multiple client platforms participating in the bidding for the aforementioned programmatic advertising. The first load amount indicates the amount of environmental load generated for each of the media by the ordering platform of the first, as a result of bidding for each of the media. The information processing apparatus according to claim 1.

5. The aforementioned performance data represents the delivery performance for each media outlet providing advertising space on one of the multiple client platforms participating in the bidding for the aforementioned programmatic advertising. The first load amount indicates the amount of environmental load generated by the bidding for each of the media by the ordering platform of the first, and is shown for each of the media. The aforementioned projected performance data represents the delivery performance for any media expected in the new programmatic advertising. The information processing apparatus according to claim 2.

6. The first load is the load measured on the receiving platform that hosts the bidding for the programmatic advertising, based on the amount of communication with the ordering platform in the first instance associated with the bidding for the programmatic advertising. The information processing apparatus according to claim 4 or 5.

7. The aforementioned prediction model is expressed by a regression equation. The information processing apparatus according to claim 1 or 2.

8. A reception desk that accepts input of projected performance data showing the expected delivery performance for the new programmatic advertising, A prediction unit that inputs the assumed performance data into a prediction model generated using a first load amount indicating the environmental burden associated with past programmatic advertising bids and performance data identified from the delivery performance of said past programmatic advertising, in order to predict a second load amount indicating the environmental burden associated with the new programmatic advertising bid, An output unit that outputs the predicted second load amount, An information processing device equipped with the following features.

9. The load amount acquisition unit performs a load amount acquisition step in which it acquires a first load amount that indicates the environmental burden generated in connection with the bidding of performance-based advertising, The performance data acquisition unit performs a performance data acquisition step in which it acquires performance data identified from the delivery performance of the aforementioned performance-based advertising, A prediction model generation step in which the prediction model generation unit generates a prediction model for predicting a second load amount that indicates the environmental load associated with bidding on new performance-based advertising, using the first load amount and the performance data, A method for generating predictive models that includes this.

10. The reception department accepts input of projected performance data that shows the expected delivery performance for the new programmatic advertising, and The prediction step involves the prediction unit inputting the assumed performance data into a prediction model generated using a first load amount indicating the environmental burden associated with past programmatic advertising bids and performance data identified from the past programmatic advertising delivery performance, in order to predict a second load amount indicating the environmental burden associated with the new programmatic advertising bid. The output unit performs an output step of outputting the predicted second load amount, A prediction method that includes this.

11. A program for causing a computer to perform the method described in claim 9 or 10.

Images