Information processing method, program, and information processing device

Abstract

When guaranteeing display metrics in advertising, including television commercials, advertising space must be allocated appropriately. [Solution] The information processing method involves an information processing device acquiring viewing history information for each program for each viewer and viewing forecast information, including future viewing forecasts for each program for each viewer based on the viewing history information, at predetermined timings; inputting at least advertising inventory, predetermined constraints regarding advertising restrictions, and indicators regarding display guaranteed by contract to a mathematical model that allocates advertising slots to satisfy the predetermined constraints, and at predetermined timings inputting viewing history information, viewing forecast information, and constraints regarding users to whom advertisements are displayed, identified based on the viewing history information, and updating the allocation of advertising slots, including advertisements for the contract period, to satisfy the constraints using the viewing forecast information; and acquiring the updated allocation data output by the mathematical model.

Description

Technical Field

[0001] The disclosed technology relates to an information processing method, a program, and an information processing apparatus.

Background Art

[0002] Conventionally, in Internet advertising, the delivery of advertisement data is controlled by managing the upper limit of the number of times the advertisement data is displayed (the upper limit of impressions) and the upper limit of the number of times the advertisement is displayed on the terminal (see, for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in television advertising, since the advertising distribution device cannot simply obtain the number of times the advertisement is displayed on the terminal, the impression guarantee technology in Internet advertising cannot be applied as it is. To address this issue, the applicant of this application has been considering providing impression guarantees for advertisements including television advertisements.

[0005] Therefore, when guaranteeing the display index for advertisements including television advertisements, the disclosed technology aims to provide an information processing method, a program, and an information processing apparatus that can appropriately perform the framing of advertisements.

Means for Solving the Problems

[0006] In an information processing method according to an aspect of the disclosure, an information processing apparatus acquires, at a predetermined timing, viewing history information of each program of each viewer and viewing prediction information including a viewing prediction of each viewer for each future program based on the viewing history information. The process involves inputting at least advertising inventory, predetermined constraints regarding advertising restrictions, and metrics regarding display guaranteed by contract, and then, at predetermined intervals, inputting viewing performance information, viewing forecast information, and constraints regarding target users for advertising based on viewing performance information, updating the ad allocation including advertisements for the contract period to satisfy those constraints using the viewing forecast information, and obtaining the updated ad allocation data output by the mathematical model. Execute this. [Effects of the Invention]

[0007] According to disclosure technology, guaranteeing display metrics in advertising, including television commercials, makes it possible to appropriately allocate advertising space. [Brief explanation of the drawing]

[0008] [Figure 1] This figure shows an example of an overview of a disclosure technology platform. [Figure 2] This figure shows an example of the configuration of an information processing system according to one embodiment. [Figure 3] This figure shows an example of the configuration of a collaborative server according to one embodiment. [Figure 4] This figure shows an example of the configuration of a mathematical model processing server according to one embodiment. [Figure 5] This figure shows an example of a mathematical model according to one embodiment. [Figure 6] This is a diagram illustrating an advertising space and an advertising playlist according to one embodiment. [Figure 7] This figure shows an example of frame data according to one embodiment. [Figure 8] This figure shows an example of material allocation data according to one embodiment. [Figure 9] This figure shows an example of constraint information according to one embodiment. [Figure 10] This figure shows an example of viewing history information and viewing prediction information according to one embodiment. [Figure 11] This figure shows an example of advertising viewing performance information and viewing prediction information according to one embodiment. [Figure 12] This figure shows an example of advertising viewing performance information and viewing prediction information according to one embodiment. [Figure 13] This figure shows an example of advertising viewing performance information and viewing prediction information according to one embodiment. [Figure 14] This flowchart shows an example of a process related to frame updating according to one embodiment. [Figure 15] This flowchart shows an example of the update process in Embodiment 1 according to one embodiment. [Figure 16] This flowchart shows an example of the update process in Embodiment 2 according to one embodiment. [Modes for carrying out the invention]

[0009] A preferred embodiment of the disclosed technology will be described with reference to the attached drawings. In each drawing, components denoted by the same reference numerals have the same or similar components.

[0010] The following describes the information processing systems in disclosure technology. In disclosure technology, "broadcast" is a concept that includes television broadcasting and radio broadcasting. "Program" includes television and radio programs broadcast during predetermined broadcast times. Advertising is the act of informing the general public, and includes advertisements broadcast via terrestrial broadcasting, etc., and in some cases, advertisements distributed via the internet. The collective term for television commercials and radio commercials is "commercial (CM)," and a commercial is a type of advertisement.

[0011] <Target CM> Advertising products that can be targeted by the disclosed technology are, for example, advertisements that guarantee the number of displays called impressions, and advertisements that may have restrictions such as targets and periods. Also, as will be described later, advertisements that guarantee reach may also be target advertisements of the disclosed technology. In addition, so-called time (net time / local time) and spot advertisements may be included in the target advertisements.

[0012] <Conventional Advertising Playlist Determination Flow> Here, an overview of the process until the advertising playlist of a conventional CM is determined will be described. (1) Contract The amount of the CM, broadcast period, exposure (GRP), time zone, and other conditions are determined. (2) Advertising Frame Acquisition The approximate date and time (program or frame ID) for sending the CM of the above contract is determined more than about one week before the CM is broadcast. At this time, for time, since the target program has been determined with the conclusion of the contract, the contract involves frame acquisition. For spots, multiple frame acquisitions are performed during the period to meet the conditions such as the contracted GRP. The frame acquisition plan may be adjusted with the advertising company and determined through agreement. Frame acquisition is performed, for example, while taking into account restriction rules (restriction conditions) such as NG for combinations of consumer finance and news programs. Here, for frame acquisition, it is necessary to perform frame acquisition that satisfies all contracts, so the workload of the operator is large. In recent years, efforts have been made to support the generation of frame acquisition plans by a system called automatic work. Here, the frame acquisition data of the advertisement linked to the contract is also referred to as "frame acquisition data". (3) Advertising Material Allocation Specific advertising materials are allocated to the frame acquisition data of each advertisement. The allocation is generally set 4 business days before on-air for the contract and for which position in the advertising frame (date and time information when the advertisement is broadcast) the advertising material is allocated. Here, the data in which advertising materials are allocated to each position of the contract is also referred to as "material allocation data". (4) In-CM Frame Editing Based on ad slot allocation and ad material assignment, an ad playlist is determined for each ad slot. Generally, the ad playlist is finalized at least one business day in advance. The ad playlist assigns ad materials to each position within the ad slot.

[0013] <Advertisement delivery procedure> As described above, the finalized conventional advertising playlist is included in the on-air data (OA data) which contains a day's worth of broadcast programs, and the finalized OA data is transmitted to the broadcasting system (broadcasting equipment) no later than the day before broadcast.

[0014] The broadcasting system generates a linear television broadcast signal by playing back and switching pre-recorded program material, advertising material, and live signals based on the transmitted OA data, and transmits them at the appropriate timings. The transmitted OA data specifies the transmission timing for each commercial slot and the composition of advertising material within each commercial slot (the advertising playlist described above). The advertising playlist is formed by including the advertising material ID and the ranking within the commercial slot.

[0015] [Embodiment] In the embodiments of the disclosure technology described below, one of the display metrics (hereinafter also referred to as "display metrics"), namely impressions (hereinafter also referred to as "Imp"), is used to maximize reach to the target audience while guaranteeing Imp, for example, by narrowing the target audience to a certain extent. In addition to impressions, display metrics also include reach, which indicates the number of users who saw the advertisement, and frequency, which indicates the number of times each user was exposed to the advertisement. Reach may include the number of times a user was exposed to the advertisement. In the embodiments, for example, in order to increase reach, a new scheme for ad placement that allows for reach control is provided. Next, an information processing system that realizes this new scheme will be described.

[0016] <Platform Overview> Figure 1 shows an example of an overview of a disclosure technology platform. According to the example platform shown in Figure 1, in the broadcasting system, the sales department performs tasks such as slot allocation and material allocation based on the advertising contract terms. For example, the sales department generates slot allocation data that includes the broadcast time slots for advertisements with guaranteed impression numbers, and generates material allocation data that includes the advertising materials to be allocated to each slot within the slot allocation data. In addition, the broadcasting preparation department of the broadcasting system generates OA data that includes the broadcast program of a predetermined amount of time (e.g., one day's worth) of programs and / or advertisements, and transmits it to the broadcasting system and the cooperating server. Note that the above processing by the broadcasting system may also be performed by having the contract terms and other information entered on a website and acquiring the entered data.

[0017] The collaborative server allows for modification of the ad slots and ad materials based on the aforementioned slot allocation data and material allocation data before the OA data is transmitted to the broadcasting system. Even after the OA data has been transmitted from the broadcasting system to the broadcasting system, the collaborative server allows for modification of the ad slots and ad materials based on the OA data shared and stored on the collaborative server. As described later, the collaborative server may also use mathematical models to allocate appropriate ad slots for ads with guaranteed impressions, using guaranteed impression counts, constraints, and audience predictions or actual audience data.

[0018] OA data transmitted from the broadcasting system is stored in the broadcasting system's data server. Within the OA data, if a recorded program is identified, the program server is notified and the recorded program is sent to the switcher. If a live broadcast studio signal is identified, the identified studio signal is sent to the switcher. If the first commercial bank, which stores advertising materials, is identified, commercials are sent from the first commercial bank to the switcher according to the advertising playlist.

[0019] In the program within the OA data, the second commercial bank is identified, and if there is any unassigned advertising material, the second commercial bank queries the linked server for commercials as needed, and sends the real-time determined commercials to the switcher as necessary. The linked server, for example, uses real-time bidding (RTB) to perform processes such as assigning commercial materials in real time.

[0020] The switcher appropriately switches between programs and commercials based on OA data and transmits a broadcast signal containing the data to be switched. The first and second commercial banks serve as commercial management servers and do not need to be configured in separate devices; they may be configured in the same device. In the disclosed technology, the processing of reviewing the time slots at predetermined timings by the coordinating server may be the main new configuration.

[0021] As described above, a disclosure platform allows for flexible changes to advertising content. Next, we will explain the information processing system that implements the aforementioned platform, focusing mainly on the review process of ad slot allocation at predetermined timings by the linked server.

[0022] <System Configuration> Figure 2 is a diagram showing an example of the configuration of an information processing system 1 according to one embodiment of the disclosure. In the example shown in Figure 2, the information processing system 1 that implements the scheme for increasing reach described above may include a cooperating server (first information processing device or first server) 10, a mathematical model processing server (second information processing device or second server) 20, a survey system 30, a broadcasting system 40, a broadcasting system 50, and one or more display devices 80.

[0023] Each device in the information processing system 1 sends and receives data to and from each other via the network N. Each server may consist of multiple processing units (which may include a database), and the number of display devices 80 can be any number. The 'n' in the nth information processing unit or server is a number used to identify each information processing unit or server, and the number can be changed as needed.

[0024] Network N consists of wireless and wired networks. Examples of networks include mobile phone networks, PHS (Personal Handy-phone System) networks, wireless LAN (Local Area Network, including communication compliant with IEEE 802.11 (so-called Wi-Fi®)), 3G (3rd Generation), LTE (Long Term Evolution), 4G (4th Generation), 5G (5th Generation), WiMAX®, infrared communication, visible light communication, Bluetooth®, wired LAN, telephone lines, power line communication networks, and networks compliant with IEEE 1394, etc.

[0025] The collaborative server 10 may be a core server that implements the scheme described above. For example, the collaborative server 10 modifies the content related to advertisements in the frame allocation data obtained from the mathematical model processing server 20 or the frame allocation data, material allocation data, etc. obtained from the broadcasting system 40 by executing a predetermined program (hereinafter also referred to as the "server program"). The collaborative server 10 may also change advertisement slots, change allocated advertisement materials, or allocate advertisement materials to unallocated positions. In the case of the disclosed technology, the collaborative server 10 performs frame allocation modification processing at predetermined timings in order to review the frame allocation during the contract period of a predetermined advertisement. The predetermined timing may be, for example, a predetermined time each day.

[0026] The mathematical model processing server 20 is a server that attempts to solve the problem of advertising slot allocation using mathematical optimization processing, for example. For example, the mathematical model processing server 20 inputs the amount of advertising inventory, the guaranteed input amount, and constraints and contract details into a mathematical model and allocates television advertising slots in a way that satisfies the constraints. The mathematical model is, as an example, software that takes the amount of advertising inventory and the guaranteed input amount as input data and allocates advertising slots in a way that satisfies the constraints and contract details. Specifically, when the mathematical model processing server 20 receives the amount of advertising inventory, the guaranteed input amount, and constraints and / or contract details set at a predetermined timing from the cooperating server 10, it executes the above software, allocates advertising slots in a way that satisfies the set constraints and / or contract details, and notifies the cooperating server 10 of the updated advertising slot allocation. The mathematical model may also be a model that uses AI-based machine learning to allocate advertising slots in a way that satisfies the constraints and / or contract.

[0027] The survey system 30 is a system for investigating the viewership ratings of each television program. The survey system 30 calculates viewership ratings from the results of investigating the television viewing habits of randomly selected survey subjects (sample households) from households that own televisions within each area. The survey system 30 also calculates viewership ratings in units as small as one minute or five minutes, and then recalculates these ratings by program broadcast time slots to determine the average viewership rating for a program, or to determine the average viewership rating for time segments such as 6:00 to 24:00.

[0028] The broadcasting system 40 generates time slot data, material allocation data, and / or OA data based on contract information for commercials, transmits the OA data to the broadcasting system 50, and transmits the time slot data, material allocation data, and / or OA data to the cooperating server 10.

[0029] The broadcasting system 50 includes the data server shown in Figure 1, acquires OA data, controls the system to generate appropriate broadcast signals according to the programs and advertising playlists set in the OA data, and transmits the generated broadcast signals.

[0030] The display device 80 is, for example, a device that receives and displays terrestrial broadcasts. The display device 80 is, for example, a television or a device having a display capable of receiving broadcast signals, and receives terrestrial (broadcast signals) including advertisements, pre-recorded programs, live broadcasts, etc., which are appropriately switched and transmitted by a switcher, and displays television programs, advertisements, etc.

[0031] <Structure> The following describes the configurations of the collaborative server 10 and the mathematical model processing server 20, which have the main functions for realizing the platform described above, among the devices of the information processing system 1. The hardware configuration of the other devices is the same as that of the collaborative server 10 and the mathematical model processing server 20.

[0032] Figure 3 shows an example of the configuration of a collaborative server 10 according to one embodiment of the disclosure. The collaborative server 10 includes one or more processors (CPUs) 110, one or more communication interfaces 120, a storage device 130, a user interface 150, and one or more communication buses 170 for interconnecting these components.

[0033] The user interface 150 is connected to a display and an input device (such as a keyboard and / or mouse, or any other pointing device), and has display and input functions.

[0034] The storage device 130 is, for example, a high-speed random access memory such as DRAM, SRAM, or other random access solid-state memory, and may also be a non-volatile memory such as one or more magnetic disk storage devices, optical disk storage devices, flash memory devices, or other non-volatile solid-state memory devices, or a non-temporary recording medium.

[0035] The storage device 130 stores data used by the information processing system 1. For example, the storage device 130 stores information such as advertising inventory, advertising materials, and advertising slots.

[0036] Another example of the storage device 130 may be one or more storage devices located remotely from the processor 110. In one embodiment, the memory 130 stores programs, modules, and data structures, or subsets thereof, executed by the processor 110.

[0037] The processor 110 executes an application program stored in the storage device 130 to form the control unit 111, which in turn comprises an acquisition unit 112, an update control unit 113, an update unit 114, a setting unit 115, and an output unit 116.

[0038] As described above, the control unit 111 performs various processes to review the ad placement data at predetermined timings, for example, in order to increase reach. For example, when a predetermined timing arrives, the control unit 111 determines whether it is necessary to change the ad placement for impression-guaranteed advertisements in order to increase reach, and if it determines that a change is necessary, it changes the ad placement using the logic described later. To perform this ad placement change, the control unit 111 has an acquisition unit 112, an update control unit 113, an update unit 114, a setting unit 115, and an output unit 116.

[0039] The acquisition unit 112 acquires indicators related to the broadcasting or distribution of television advertisements (hereinafter also referred to as "performance indicators") for contracted television advertisements that guarantee indicators related to display (display indicators). For example, the acquisition unit 112 may acquire performance indicators from the research system 30 that indicate a given television advertisement was actually broadcast or distributed. Display indicators and performance indicators include one of the following, for example, GRP, impressions, frequency, and reach, and they may be the same or different. For example, the display indicator may be impressions, and the performance indicator may be frequency.

[0040] Furthermore, the acquisition unit 112 may acquire time slot data including broadcast time slots that indicate the broadcast date and time of the television advertisement. The acquisition unit 112 may, for example, acquire time slot data from the broadcasting system 40 or the mathematical model processing server 20, or acquire time slot data that has been previously acquired and stored in the storage device 130.

[0041] The update control unit 113 controls the updating of television advertising slots based on performance indicators acquired by the acquisition unit 112. For example, the update control unit 113 includes an analysis unit that analyzes performance indicators related to the broadcasting or distribution of television advertisements at predetermined timings within the contract period of the television advertisements included in the slot allocation data, and controls whether or not to allocate slots according to the analysis results. For example, the predetermined timing may be one of the following: the middle day of the television advertisement contract period, every predetermined day after the start of the contract period, a predetermined number of days after the start of the contract period, or a predetermined number of days before the end of the contract period.

[0042] When the predetermined timing described above arrives, the analysis unit of the update control unit 113 analyzes the performance indicators of a particular television advertisement A, and analyzes whether it has been repeatedly shown to the same users, etc. For example, the analysis unit of the update control unit 113 can use the viewership ratings (household viewership ratings and individual viewership ratings) of each program obtained from the survey system 30 to analyze how much television advertisement A was shown to which target audience. As an example, the analysis unit of the update control unit 113 may compare the cumulative value of the number of users (performance indicators) based on the viewership ratings of each program on which television advertisement A was shown with a baseline value, and if the cumulative value is lower than the baseline value, it may determine that a change in the time slot is necessary.

[0043] Furthermore, if individual viewership ratings are obtained from the survey system 30, the analysis unit of the update control unit 113 may calculate the number of users based on the individual viewership ratings corresponding to the target group, and compare this cumulative value of the number of users (performance indicator) with a baseline value.

[0044] The analysis unit of the update control unit 113 determines, at a predetermined timing, whether TV advertisement A is reaching a variety of users based on its performance indicators. For example, if the analysis unit of the update control unit 113 obtains the reach of a predetermined advertisement from the research system 30, it predicts the reach for the second period (from the predetermined timing to the end of the contract period) from the actual reach for the first period (from the start of the contract period to the predetermined timing) and calculates this predicted value. The prediction of the reach for the second period can be calculated from the predicted viewership ratings of similar programs to be broadcast in the future, based on the trends (viewership ratings, etc.) of programs with a proven track record in performance indicators. The analysis unit of the update control unit 113 compares the actual reach + predicted value with a baseline value to determine whether TV advertisement A is reaching a variety of users. If the determination result is positive (actual reach + predicted value ≥ baseline value), the analysis unit of the update control unit 113 determines that updating the time slot is unnecessary, and if the determination result is negative (actual reach + predicted value < baseline value), it determines that updating the time slot is necessary. Furthermore, the analysis unit of the update control unit 113 may compare the actual reach with a reference value that takes into account the period from the start of the contract period to a predetermined timing, in order to determine whether or not a change in the allocation is necessary.

[0045] The update unit 114 updates the time slots for television advertisements from a predetermined timing onward based on the analysis results of the indicators performed by the analysis unit of the update control unit 113. For example, the update unit 114 may update the time slots for the target television advertisement so that the advertisement is scheduled during programs that have historically had high viewership ratings among the target audience.

[0046] The above processing makes it possible to ensure appropriate advertising slots when guaranteeing display metrics for advertisements, including television advertisements. For example, according to the information processing system 1 of this disclosure, during the contract period of a television advertisement, if it is determined that the performance metrics of the television advertisement have not reached a wide range of users, it becomes possible to update the advertising slots for the target television advertisement so that it can reach a wide range of users.

[0047] Below, we will describe two examples of the update logic for ad slot allocation. Example 1 is a case where ad slot allocation is performed using a mathematical model, and Example 2 is a case where ad slot allocation is performed using the main inventory.

[0048] ≪Example 1≫ In Example 1, the mathematical model processing server 20 will be described first. Figure 4 shows an example of the configuration of the mathematical model processing server 20 according to one embodiment of the disclosure. The mathematical model processing server 20 includes one or more processors (CPUs) 210, one or more communication interfaces 220, a storage device 230, a user interface 250, and one or more communication buses 270 for interconnecting these components.

[0049] The user interface 250 is connected to a display and an input device (such as a keyboard and / or mouse, or any other pointing device), and has display and input functions.

[0050] The storage device 230 is, for example, a high-speed random access memory such as DRAM, SRAM, or other random access solid-state memory, and may also be a non-volatile memory such as one or more magnetic disk storage devices, optical disk storage devices, flash memory devices, or other non-volatile solid-state memory devices, or a non-temporary recording medium.

[0051] The storage device 230 stores data and programs used by the information processing system 1. For example, the storage device 230 stores application programs for the mathematical model processing server 20 in the information processing system 1.

[0052] The processor 210 constitutes the control unit 211 by executing a program stored in the storage device 230. The control unit 211 includes an acquisition unit 212, a processing unit 213, and an output unit 214. The control unit 211 uses a mathematical model to allocate advertising space, for example, by maximizing the total remaining inventory of multiple advertising slots (e.g., overall optimization), satisfying predetermined constraints and achieving guaranteed display metrics for target attributes based on the contract.

[0053] Figure 5 shows an example of a mathematical model according to one embodiment of the disclosure. In the mathematical model shown in Figure 5, the acquisition unit 112 acquires information necessary for broadcasting and distributing advertisements. The acquisition unit 112 acquires, for example, the guaranteed number of impressions (guaranteed Imps) set for an advertisement, the amount of advertisement inventory (remaining Imps), and variables input to the mathematical model for setting guaranteed advertisements in predetermined advertisement slots, from the memory 130 or an external device (e.g., the cooperating server 10). The variables input to the mathematical model include, for example, inventory variables related to the display inventory of each advertisement, guarantee variables related to the guarantee of advertisement display, and variables related to constraints on advertisement display.

[0054] The processing unit 213 may input inventory variables relating to the display inventory of multiple advertisements, including the first advertisement, guarantee variables relating to the display guarantee of multiple advertisements, including the first advertisement, and constraints relating to the display of advertisements into a mathematical model (see Figure 5) in which an objective function is set that satisfies predetermined conditions relating to the display inventory of advertisements for terrestrial broadcasting. The processing unit 213 may then determine one or more advertising materials to be allocated to predetermined advertising slots by allocating advertising slots that satisfy the constraints and maximize the inventory.

[0055] For example, the processing unit 213 may input the following variables and use a mathematical optimization algorithm (mathematical model) to search for advertising materials so that the objective of the objective function is achieved while satisfying the constraints, reserve a slot for a given advertisement, and determine one or more advertising materials to be allocated to this advertising slot. • Objective function: Achieve guaranteed impressions, maximize remaining inventory (remaining impressions). • Constraints: Must meet pre-specified broadcasting rules, etc. (e.g., time constraints, avoidance of certain topics, etc.) • Inventory variables: Remaining inventory (remaining impressions) for each ad, order conditions for each campaign (budget, target, area, etc.) • Guaranteed variable: Guaranteed number of Imps (if a target is set, the guaranteed number of Imps per target)

[0056] The output unit 214 outputs the frame allocation data determined by the mathematical model, or the allocated advertising materials, to the linked server 10. Each time the mathematical model performs frame allocation processing, the output unit 214 outputs frame allocation data that satisfies the constraints to the linked server 10.

[0057] The linked server 10 may obtain requests for advertising playlists from the broadcasting system 50 for advertising slots in the slot allocation data output by the mathematical model and acquired. The linked server 10 generates advertising playlists in response to requests. Advertising playlists will be described later using Figure 6.

[0058] <Advertising space and ad playlists> Here, we will explain the advertising playlist using Figure 6. Figure 6 is a diagram illustrating an advertising slot and an advertising playlist according to one embodiment of the disclosure. The advertising slot shown in Figure 6 is a predetermined advertising broadcast time slot for broadcasting advertisements for goods or services, etc. Each advertising slot has at least a broadcast date and time and a broadcasting station determined. In the example shown in Figure 6, the advertising slot is an advertising slot for television commercials. However, it is not limited to this, and the advertisement broadcast using the advertising slot may also be a radio commercial.

[0059] Each advertising slot is associated with advertising slot information. This advertising slot information is general information about each advertising slot and is provided by the advertising slot provider, such as a broadcasting station. The advertising slot information includes at least the date and time the advertisement will be broadcast in the advertising slot (advertising broadcast time slot) and the broadcasting station, but may also include the type of advertising slot, the length of the advertising slot in seconds, the name of the program associated with the advertising slot (specifically, the name of the program in which the advertising slot is scheduled within or immediately before the broadcast time), and the price of the advertising slot (purchase price). Here, the type (genre) of the advertising slot is information that shows the relationship between the advertising slot and the broadcast time of the program, and specifically includes "PT (Participating Commercial: time set within the program broadcast time)" and "SB (Station Break: time between one program and the next)."

[0060] Furthermore, advertising slot information may include additional information beyond that mentioned above, such as broadcast area (regional information), broadcast period, broadcast time slot (the time of day when the advertisement is broadcast: Daypart), the genre of the program associated with the advertising slot and information about the program's cast, information about events related to the program, and the target audience described below. However, these are merely examples and the information is not limited to these.

[0061] A single advertising slot can broadcast multiple advertisements. The order in which advertisements are broadcast within a single advertising slot is called the ad position. The ad position is also known as "PIB (Position-in-Break)," and for example, if four 15-second advertisements can be broadcast within a single advertising slot (number of positions = 4), it indicates which position the advertisement will be broadcast in.

[0062] Furthermore, each ad slot is assigned an ad slot identifier (ad slot ID) that uniquely identifies that ad slot. This ad slot ID allows a single ad slot to be identified from among many others.

[0063] Furthermore, Figure 6 will be used to explain the advertising playlist. The advertising playlist is a list that includes an advertising slot ID, an advertising material ID that identifies the advertising material to be broadcast in that advertising slot, the position of the advertisement, and the duration information (length information) of the advertising material. Figure 6 shows an example of an advertising playlist with 4 positions, where the duration information and advertising material IDs are listed in order of position. For example, a guaranteed advertisement may be assigned to position 1 by the broadcasting system 40, guaranteed advertisements may be assigned to positions 2 and 4 by the mathematical model processing server 20, and a programmatic advertisement that was won in the previous auction may be assigned to position 3.

[0064] The cooperating server 10 may receive a request for an advertising playlist from the broadcasting system 50, generate an advertising playlist listing the advertisements to be broadcast in that advertising slot, and send the advertising playlist to the broadcasting system 50 as a response.

[0065] <Data Example> Next, we will describe an example of data related to the playlist mentioned above. Figure 7 is a diagram showing an example of slot allocation data according to one embodiment of the disclosure. As shown in Figure 7, the slot allocation data for each contract includes a "contract ID" that identifies the contract, the number of seconds of the advertisement, a "position ID" that identifies the position in which the advertisement will be broadcast, and a "possible time" in which the advertisement may be broadcast.

[0066] In the example shown in Figure 7, the slot allocation data for a contract with contract ID "ABC" assigns the position ID "AAA" and the broadcast time "October 1st 9:00:00~11:10:00" to a 15-second ad; the position ID "DDD" and the broadcast time "October 1st 12:00:00~14:00:00" to a 15-second ad; the position ID "EEE" and the broadcast time "October 2nd 19:00:00~20:00:00" to a 15-second ad. The slot allocation data allows for the general allocation of positions to which contracted ads will be assigned.

[0067] Figure 8 shows an example of material allocation data according to one embodiment of the disclosure. As shown in Figure 8, the material allocation data allocates advertising materials to each position allocated in the frame allocation data.

[0068] In the example shown in Figure 8, the material allocation data is as follows: For contracts with contract ID "ABC", a 15-second advertisement is assigned the position ID "AAA", the broadcast time for the advertisement is "October 1st 9:00:00~11:10:00", and the material ID for the advertisement is "A"; for contracts with position ID "DDD", the broadcast time for the advertisement is "October 1st 12:00:00~14:00:00", and the material ID for the advertisement is "A"; for contracts with position ID "EEE", the broadcast time for the advertisement is "October 2nd 19:00:00~20:00:00", and the material ID for the advertisement is "E". The material allocation data assigns the advertisement material to each position.

[0069] Now, let's assume that the acquisition unit 112 has received a change request for the material allocation data shown in Figure 8. The change request is, for example, a request to change material ID "A" to material ID "C" for contract ID "ABC" and position ID "AAA". This change request may be made, for example, by an advertiser operating from a sales site, or by an advertising company's system via API integration.

[0070] Furthermore, regarding the material allocation data shown in Figure 8, the acquisition unit 112 may receive a change request to modify the data on a contract-by-contract basis. A change request might be, for example, a request to change contract ID "ABC" to contract ID "BCD". As mentioned above, this change request may be sent from a processing unit used by the advertiser or advertising agency.

[0071] Figure 9 shows an example of constraint information according to one embodiment of the disclosure. As shown in Figure 9, constraint information sets constraint content, genre, time, and remarks for advertising materials for a specified product or service. For example, for advertising material "Advertisement A", a genre restriction is imposed, with the genres to be restricted being "news" and "anime recommended for young people", and the time restriction is set to "7:00-12:00", so that "Advertisement A" is not assigned to positions that fall under these set constraints.

[0072] Now, returning to Figure 3, we will explain the process of setting or changing television advertising slots using a mathematical model in cooperation with the linked server 10. The acquisition unit 112 of the linked server 10 acquires viewing history information for each program for each viewer, and viewing prediction information including future viewing predictions for each program for each viewer based on the viewing history information, at a predetermined timing (see, for example, Figure 10). As mentioned above, the predetermined timing is any timing, but for example, it could be midnight every day. For example, the acquisition unit 112 may acquire viewing history information and / or viewing prediction information from the survey system 30.

[0073] The update control unit 113 of the linked server 10 inputs at least the inventory related to the advertisement, predetermined constraints related to the advertisement constraints, and metrics related to the display guaranteed by contract, and at predetermined times, inputs viewing performance information, viewing forecast information, and constraints related to the users to whom the advertisement is to be displayed, identified based on the viewing performance information, to a mathematical model that allocates advertising slots to satisfy the predetermined constraints, and updates the slot allocation, including advertisements for the contract period, to satisfy the constraints using the viewing forecast information.

[0074] For example, when the acquisition unit 112 acquires viewing history information and viewing forecast information, the update control unit 113 inputs the acquired viewing history information and viewing forecast information into the mathematical model at the same time or at different times. Thus, the update control unit 113 can update the slot allocation data by inputting the updated viewing history information and viewing forecast information into the mathematical model. Alternatively, the update control unit 113 may input only the viewing forecast information into the mathematical model. For example, the update control unit 113 may identify target users based on viewing history information and allocate slots to broadcast predetermined advertisements in programs that these targets are likely to watch, in order to reach them.

[0075] The update control unit 113 may input the latest viewing performance information and viewing prediction information into the mathematical model each time a predetermined timing (for example, daily or at a predetermined time every predetermined number of days) arrives. Furthermore, the update control unit 113 may identify users (targets) who have not viewed a predetermined advertisement based on the viewing performance information, and input the constraint that these users view the predetermined advertisement into the mathematical model. For example, the update control unit 113 can set constraints in the mathematical model so that advertising slots are allocated to programs that the target is predicted to view based on the viewing prediction information. Users who have not viewed a predetermined advertisement may include users whose number of views of the predetermined advertisement is less than a threshold.

[0076] The update unit 114 updates the existing ad slot data with updated ad slot data output by the mathematical model. The update unit 114 can then perform ad slot allocation at predetermined intervals, for example, for ads with guaranteed impressions, in a way that increases reach.

[0077] Through the above process, when guaranteeing display metrics for advertisements, including television commercials, it becomes possible to appropriately allocate advertising slots. As mentioned above, even during the contract period (campaign period) of a television commercial, it becomes possible to change the slot allocation for each designated advertisement in order to increase reach.

[0078] The setting unit 115 sets a constraint condition regarding the users to be shown the ad, which is to display the first ad to a predetermined number of users who have not viewed the first ad. For example, in order to increase the reach of the first ad identified by predetermined criteria, the setting unit 115 identifies users who are presumed not to have viewed the first ad based on viewing history information, and sets a constraint condition that these users view the ad. The constraint conditions set in the setting unit 115 are input into the mathematical model by the update control unit 113.

[0079] Through the above process, it becomes possible to set conditions in the mathematical model that increase reach for each given advertisement, thereby making the ad slot allocation data output from the mathematical model more appropriate and increasing the satisfaction of advertisers and advertising agencies.

[0080] The setting unit 115 may include identifying advertisements that may not meet the guaranteed metrics based on viewing performance information and metrics guaranteed by contract, and setting the identified advertisements as the first advertisement. For example, the setting unit 115 may set a given advertisement as the first advertisement if, based on viewing performance information, the difference between the number of impressions already displayed and the number of impressions guaranteed by contract is greater than or equal to a threshold. The setting unit 115 may appropriately change the threshold depending on when the threshold determination is made within the contract period. For example, at the beginning of the contract period, the threshold may be increased by adding an offset to the default threshold, or at the end of the contract period, the threshold may be decreased by subtracting an offset from the default threshold.

[0081] The above process makes it possible to properly identify ads whose ad slots need to be changed. As a result, when reviewing ad slot data, it becomes possible to identify ads that may not achieve the guaranteed number of impressions and, by changing the ad slot data for these ads, increase the likelihood of achieving the guaranteed number of impressions while also increasing reach.

[0082] The setting unit 115 may also include overlaying the viewing performance information with the broadcast performance information of each advertisement broadcast by a designated broadcasting station to identify users who did not watch the first advertisement during broadcasts by the designated broadcasting station. For example, the setting unit 115 can grasp the viewership rating for each program for each user based on the viewing performance information, and can grasp where each advertisement was broadcast in each program using the designated broadcasting station's OA data, etc., and overlay this viewing performance information with the broadcast performance information for each program and each advertisement on the same time axis. Based on the overlaid viewing performance information and broadcast performance information, the setting unit 115 may identify users (targets) who did not watch the first advertisement during broadcasts by the designated broadcasting station.

[0083] Through the above process, it becomes possible to identify users who have not yet seen the first advertisement in broadcasts by a designated broadcasting station and who can be targeted for increased reach. This allows for scheduling the first advertisement in programs that these users are likely to watch.

[0084] The setting unit 115 may include overlaying the viewing performance information with the broadcast performance information of each advertisement broadcast on multiple broadcasting stations to identify users who did not watch the first advertisement in broadcasts by multiple broadcasting stations. For example, the setting unit 115 can grasp the viewership rating of each program for each user based on the viewing performance information, and can grasp where each advertisement was broadcast in each program using OA data from multiple broadcasting stations, etc., and overlay this viewing performance information with the broadcast performance information of each program and each advertisement on multiple broadcasting stations on the same time axis. Based on the overlaid viewing performance information and broadcast performance information, the setting unit 115 may identify users (targets) who did not watch the first advertisement in broadcasts by multiple broadcasting stations. Note that the multiple broadcasting stations may be all broadcasting stations capable of terrestrial broadcasting.

[0085] Through the above process, it becomes possible to identify users who have not yet seen the first advertisement in broadcasts by multiple broadcasting stations, and who have the potential for increased reach. This allows for the allocation of the first advertisement slot to programs that these users are most likely to watch.

[0086] The setting unit 115 may also include setting a predetermined number based on the guaranteed metrics for the first advertisement and the viewing performance information. The setting unit 115 may, for example, determine how many users should view the first advertisement by subtracting the number of reach estimated to have viewed the first advertisement based on the viewing performance information from the guaranteed number of impressions, and setting a predetermined number based on this difference. As a specific example, if the guaranteed number of impressions is 200,000 and the number of reach calculated based on the viewing rate is 100,000, the setting unit 115 may set a predetermined number (for example, 100,000 × α (0 < α < 1)) based on the difference of 100,000.

[0087] Through the above process, when setting a target as a constraint, it becomes possible to specify a concrete number of users you want to view the content. This increases the likelihood of increasing reach while still meeting impression guarantees.

[0088] <Specific examples of viewing history information and viewing forecast information> Figure 10 shows an example of viewing history information and viewing prediction information according to one embodiment of the disclosure. The viewing history information shown in Figure 10 includes information indicating whether or not each viewer selected as a sample watched each program on a predetermined broadcasting station. For example, it indicates that user A watched program A on October 1st on station AA, but did not watch program B.

[0089] The viewing prediction information shown in Figure 10 includes a prediction of whether each viewer included in the viewing performance information will watch or not watch each future program (in the example shown in Figure 10, programs from October 2nd onwards). The viewing prediction information may be generated on the survey system 30 side, or it may be generated by the cooperating server 10 based on the viewing performance information.

[0090] The update control unit 113 can identify users who have watched a program in which advertisement A is inserted, based on viewing history information, and can also identify predicted viewers who are not yet watching the program and are predicted to watch a predetermined program in the future, based on viewing prediction information. At this time, the update control unit 113 may input constraints into the mathematical model such that, for a number of predicted viewers equal to the remaining number of guaranteed Imps from the contracted guaranteed Imps, the advertisement is allocated to the program that these predicted viewers are predicted to watch. For example, the update control unit 113 identifies users who are presumed not to have watched a particular advertisement from the viewing history information shown in Figure 10, generates constraints based on viewing prediction information such that the advertisement is allocated to the program that these users are predicted to watch, and inputs these constraints into the mathematical model.

[0091] This allows us to input viewing history information, viewing forecast information, and the number of unique users that each Imp guarantee contract must achieve into the mathematical model shown in Figure 5, enabling us to allocate advertising slots that guarantee a certain degree of reach. As a result, it becomes possible to propose contracts that guarantee reach, which was difficult to achieve with television advertising.

[0092] Furthermore, based on the viewing history information shown in Figure 10, it is possible to create viewing history for any advertisement (or contract) broadcast on station AA. Figure 11 is a diagram showing an example of viewing history information and viewing prediction information according to one embodiment of the disclosure. For example, if user A watches program A broadcast on October 1st without leaving, it will be considered that user A watched advertisement 001 broadcast during program A. The update control unit 113 can measure the number of times each user (each viewer) has watched any advertisement by overlaying such program viewing history with the broadcast history of advertisements from a predetermined broadcasting station (station AA).

[0093] Based on the information shown in Figure 11, the update unit 114 may determine the number of times each advertisement has been displayed based on viewing history information, generate constraints to allocate advertising slots for programs that users who have not yet watched a given television advertisement are likely to watch in the future, and input these constraints into a mathematical model. For example, the update control unit 113 may use viewing history information to identify users who have not watched a given advertisement or who have watched it below a certain threshold, set constraints to allocate advertising slots for programs that these users are predicted to watch, and input these constraints into a mathematical model.

[0094] For example, the system of this disclosed technology allows for daily changes to ad placement, making it possible to maximize the reach of a given advertisement by using both past performance reach and future reach predictions.

[0095] Figure 12 shows an example of viewing performance information and viewing prediction information according to one embodiment of the disclosure. The viewing performance information shown in Figure 12 is performance information that takes into account each program of all broadcasting stations. Viewing performance for each program of all broadcasting stations can be obtained from the survey system 30, and past advertising broadcast performance can be obtained by analyzing broadcasts of each broadcasting station, etc.

[0096] The update control unit 113 inputs viewing performance information from all broadcasting stations into a mathematical model. It then uses past performance data, including the total user reach considering all broadcasting stations, and future viewing predictions, using data from a designated broadcasting station to set constraints. For example, the update control unit 113 may generate a constraint condition that reserves airtime for programs that users with a reach below a predetermined value across all broadcasting stations are likely to watch, and input this constraint condition into the mathematical model. This allows for the identification of users not yet reached by all broadcasting stations before maximizing reach.

[0097] Furthermore, if the viewing prediction information includes viewing probability rather than just a viewing prediction of whether or not a user has watched, the update control unit 113 may allocate program slots for a given user so that the viewing probability of that user exceeds 1. For example, the update control unit 113 identifies a user to be reached by a given advertisement (a user who has not watched), and allocates slots in multiple programs so that the viewing probability of this user exceeds 1. For example, in the example shown in Figure 12, if user C is the target, and the sum of the predicted viewing probability of program D on October 2nd and the predicted viewing probability of program G on October 3rd exceeds 1, the update control unit 113 may allocate slots so that the given advertisement is broadcast in relation to these programs D and G.

[0098] In this way, by having a predetermined number of users watch a predetermined advertisement if the sum of the predicted viewership ratings of a predetermined number of programs exceeds 1, it becomes possible to appropriately increase reach. Specifically, if a predetermined advertisement is broadcast in relation to a predetermined number of programs whose sum of predicted viewing probabilities exceeds 1, then since the probability of this user watching any of those programs is 1 or greater, the probability of them watching the predetermined advertisement is also 1 or greater, making it highly likely that the advertisement will reach the target audience.

[0099] Figure 13 shows an example of viewing performance information and viewing prediction information according to one embodiment of the disclosure. The viewing performance information shown in Figure 13 is the same as the viewing performance information shown in Figure 11, and the viewing prediction information includes information that includes the predicted viewing probability. For example, if a predetermined advertisement is definitively reserved for each user by spot and / or time slot, the update control unit 113 adds the predicted viewing probability of each reserved program to the value of the viewing performance information. The update control unit 1113 may further add the predicted viewing probability of programs with vacant slots, set a constraint that the sum of the viewing performance information and the predicted viewing information exceeds a threshold, and input this into the mathematical model.

[0100] Let's explain using a specific example shown in Figure 13. The update control unit 113 determines the number of times a given advertisement, "Advertisement 001," has been displayed to user A, specifically three times. It is already known that "Advertisement 001" is scheduled to be displayed (e.g., broadcast) in "Program E" on October 2nd and "Program G" on October 3rd, using PT / SB, etc., as a result of time and spot scheduling.

[0101] At this time, the update control unit 113 adds the predicted viewing probability of "Program E" (0.1) and the predicted viewing probability of "Program G" (0.05) to the number of impressions "User A" has made (3) for "Advertisement 001". For example, if the threshold per user is 4, the update control unit 113 determines that User A needs an additional 0.85(4-(3+0.15)) impressions (impression index).

[0102] Similarly, the update control unit 113 adds the predicted viewing probability of the program in which "Advertisement 001" is scheduled, taking into account the time and spot scheduling, to the number of times "Advertisement 001" is displayed for each user. The update control unit 113 may also generate constraints on "Advertisement 001" so that each user exceeds a threshold based on the number of displays and predicted viewing probability calculated for each user, and input these constraints into a mathematical model. For example, using the example of "User A" and "Advertisement 001" described above, the update control unit 113 generates constraints so that the predicted viewing probability of the unscheduled program slots is added so that a remaining predicted viewing probability of 0.85 can be obtained. Alternatively, the update control unit 113 may repeat the above process for each predetermined advertisement and perform scheduling for each predetermined advertisement.

[0103] The update control unit 113 may, for example, identify a target user for each predetermined advertisement who has not viewed the advertisement (or whose number of impressions is less than a threshold), and perform the above-described process for this target user.

[0104] Through the above processing, by adding the predicted viewing probability of advertisements displayed in programs that have already been allocated slots in spot or time slots to the viewing history information, it becomes possible to allocate slots more appropriately to each user, taking into account programs that they are likely to watch in the future. Furthermore, thresholds for display metrics may be set for each user. In addition, it is also possible to use the viewing history information for all channels shown in Figure 12 with the viewing history information shown in Figure 13, and the same processing as described above should be performed.

[0105] As described above in Example 1, by using viewing performance information and viewing forecast information, it is possible to construct a framework that can guarantee display metrics (e.g., reach). For example, as mentioned above, by analyzing the performance metrics at a predetermined time and if the guaranteed display metrics are unlikely to be achieved, the framework can be reviewed, allowing for the timely implementation of measures to achieve the guaranteed display metrics.

[0106] Example 2 In Example 2, the objective is the same as in Example 1, but the measures implemented are different. In Example 2, regarding the allocation of ad slots with guaranteed impressions, the ad slot that will acquire the largest number of impressions is defined as the first inventory (also referred to as the main inventory), and this first inventory is allocated to the ad slots.

[0107] In Embodiment 2, the update unit 114 includes reserving a first inventory that is expected to achieve a first indicator or higher based on guaranteed indicators (display indicators) related to display. For example, when a predetermined timing arrives, the analysis unit of the update control unit 113 determines whether the current impression performance will achieve the guaranteed impression. If the analysis unit of the update control unit 113 determines that the guaranteed impression is unlikely to be achieved, the update unit 114 may reserve the first inventory after the predetermined timing.

[0108] This allows for the allocation of initial inventory slots at a predetermined time to ensure the target is met if the guaranteed number of impressions is unlikely to be achieved.

[0109] Furthermore, the following restrictions may be imposed on the first inventory: (1) The impressions obtained by the first inventory must be at least n% of the guaranteed impressions. The constraint in (1) above is to ensure that impressions exceeding the threshold can be obtained at once. (2) The allocation of the first inventory and other inventory must have different information (for example, the basic program ID of the first inventory and the basic program ID before the change must be different). The constraint in (2) above is to allow for the allocation of time slots in different programs. (3) The first inventory will be allocated in the initial stages after the predetermined timing. The constraint in (3) above is intended to establish a path to achieving the goal as early as possible. (4) If the guaranteed impression is large and the above n% exceeds the predetermined value, consumption beyond the predetermined value of impressions shall be restricted. The constraint in (4) above is to ensure that, when n% exceeds a predetermined value, instead of seeking many impressions at once, multiple slots are allocated for a second inventory (sub-inventory) that has different attributes from the first inventory and is likely to achieve impressions above the predetermined value.

[0110] In other words, the update unit 114 may include allocating the first inventory at a predetermined time and within a predetermined period (for example, (3) constraint conditions). This allows for the rapid implementation of measures to increase impressions after the decision to change the allocation has been made.

[0111] Furthermore, the update unit 114 may also include reserving slots for at least one second inventory item from among inventory items with different attributes from the first inventory item, which is expected to acquire a second indicator or higher based on the display indicator (for example, (4) constraint conditions). This allows for the use of multiple second inventory items to reserve television advertising slots when the value of the display indicator that was expected to be acquired with the first inventory item exceeds a threshold.

[0112] As described above, according to Example 2, performance information regarding indicators can be checked during the contract period, and if the performance information has not reached a predetermined standard value, it becomes possible to implement boost measures for securing the quota.

[0113] <Operation Description> Next, we will describe each operation of the information processing system 1. Figure 14 is a flowchart showing an example of the process related to frame setting updates according to one embodiment of the disclosure. In the example shown in Figure 14, each process is executed for one television advertisement.

[0114] In step S102, the acquisition unit 112 acquires performance indicators related to the broadcast or distribution of television advertisements for contracted television advertisements that guarantee display indicators. For example, the acquisition unit 112 may acquire performance indicators for predetermined television advertisements from the research system 30.

[0115] In step S104, the analysis unit of the update control unit 113 analyzes performance indicators at a predetermined timing within the contract period of the television advertisement. For example, the analysis unit of the update control unit 113 may compare the actual reach with a baseline value that takes into account the period from the start of the contract period to the predetermined timing, and determine whether or not a change in the time slot allocation is necessary.

[0116] In step S106, the update unit 114 updates the allocation of advertising slots, including broadcast time slots, for television advertisements from a predetermined timing onward, based on the results of the performance indicator analysis. The update unit 114 may also update the allocation of the target television advertisement so that the target television advertisement is allocated during programs that have historically had high viewership ratings among the target audience.

[0117] Figure 15 is a flowchart showing an example of the update process in Embodiment 1 according to one embodiment of the disclosure. In step S202 shown in Figure 15, the acquisition unit 112 acquires viewing history information for each program for each viewer and viewing prediction information, which includes a prediction of each viewer's future viewing of each program based on the viewing history information, at a predetermined timing. The acquisition unit 112 may also acquire viewing history information, viewing prediction information, etc., from the survey system 30 as shown in Figures 10 to 13.

[0118] In step S204, the update control unit 113 inputs at least the inventory of advertisements, predetermined constraints regarding advertisement restrictions, and indicators regarding display guaranteed by contract. At predetermined times, it inputs viewing performance information, viewing forecast information, and constraints regarding users to whom advertisements are displayed, identified based on the viewing performance information, to a mathematical model that allocates advertising slots to satisfy the predetermined constraints. The update control unit updates the slot allocation, including advertisements for the contract period, to satisfy the constraints using the viewing forecast information. The constraints include, for example, identifying users who have not been able to reach the target television advertisement and allocating slots to programs that can reach these users, in order to increase reach. An example of updating advertising slots will be explained using steps S206 and S208.

[0119] In step S206, the update control unit 113 may set a constraint condition regarding the users to be displayed, which is to display the first advertisement to a predetermined number of users who have not viewed the first advertisement. This makes it possible to identify the target audience for the first advertisement.

[0120] In step S208, the update control unit 113 may overlay the viewing history information with the broadcast history information of each advertisement broadcast by the designated broadcasting station to identify users who have not viewed the first advertisement in the broadcast by the designated broadcasting station. This makes it possible to identify advertisements that may not meet the guaranteed number of views in the display metric.

[0121] Figure 16 is a flowchart showing an example of the update process in Embodiment 2 according to one embodiment of the disclosure. In step S302 shown in Figure 16, the update unit 114 may also frame the first inventory that is expected to achieve a first indicator or higher based on guaranteed indicators (display indicators) related to the display. An example of framing the first inventory will be explained using steps S304 and S306.

[0122] In step S304, the update unit 114 may reserve the first inventory at a predetermined timing and within a predetermined period.

[0123] In step S306, the update unit 114 may select one or more second inventory items from among the inventory items with different attributes from the first inventory items, which are expected to achieve a second indicator or higher based on the display indicator. For example, if the display indicator to be achieved is a large number, it becomes possible to select multiple second inventory items by dividing this display indicator and using them to achieve the target.

[0124] The embodiments described above are illustrative examples for explaining the disclosed technology and are not intended to limit the disclosed technology to these embodiments only. The disclosed technology can be modified in various ways without departing from its essence. For example, the processes of each device may be integrated as appropriate, or the processes may be transferred to the other device.

[0125] Furthermore, one of the processes described above (e.g., a mathematical model) may be implemented using a machine learning model based on artificial intelligence (AI). For example, the allocation of advertising slots and the determination of advertising materials can be implemented using reinforcement learning, in which the reward function is set so that the reward increases as the total remaining inventory of all advertising slots increases. In addition, this disclosed technology makes it possible to handle new advertisements that guarantee reach in television advertising, thereby improving usability.

[0126] [Differentiation] Although the present invention has been described based on the above embodiments, the following cases are also included in the present invention.

[0127] [Example 1] At least a portion of each configuration in the collaborative server 10 according to the above embodiment may be provided by other servers. Also, at least a portion of each process of other servers may be provided by the collaborative server 10. For example, the process related to ad slot allocation in the mathematical model processing server 20 may be implemented in the collaborative server 10. [Explanation of Symbols]

[0128] 1…Information processing system, 10…Linkage server, 20…Mathematical model processing server, 30…Survey system, 40…Broadcasting system, 50…Broadcasting system, 80…Display device, 110…Processor, 130…Memory, 111…Control unit, 112…Acquisition unit, 113…Update control unit, 114…Update unit, 115…Setting unit, 116…Output unit, 130…Storage device, 210…Processor, 211…Control unit, 212…Acquisition unit, 213…Processing unit, 214…Output unit

Claims

1. Information processing device, The system acquires viewing history information for each program for each viewer, and viewing forecast information, including future viewing predictions for each program based on the aforementioned viewing history information, at predetermined times. A mathematical model that inputs at least advertising inventory, predetermined constraints regarding advertising restrictions, and metrics regarding display guaranteed by contract, and allocates advertising slots to satisfy the predetermined constraints, is then input, at predetermined times, the viewing performance information, the viewing forecast information, and constraints regarding the target users of the advertisement identified based on the viewing performance information, and updates the slot allocation, including advertisements for the contract period, to satisfy the constraints using the viewing forecast information. To obtain the updated frame data output by the aforementioned mathematical model, An information processing method that performs the following.

2. The information processing method according to claim 1, further comprising setting the constraints for the users to be displayed to display the first advertisement to a predetermined number of users who have not viewed the first advertisement.

3. The above setting means, Based on the aforementioned viewing performance information and the metrics guaranteed by the aforementioned contract, identify advertisements that may not meet the guaranteed metrics. The information processing method according to claim 2, comprising setting a specified advertisement as the first advertisement.

4. The above setting means, The information processing method according to claim 2, further comprising overlaying the viewing history information with the broadcast history information of each advertisement broadcast by a designated broadcasting station to identify users who did not view the first advertisement in the broadcast by the designated broadcasting station.

5. The above setting means, The information processing method according to claim 2, further comprising overlaying the viewing history information with the broadcast history information of each advertisement broadcast by each of the multiple broadcasting stations, and identifying users who did not view the first advertisement in the broadcasts by the multiple broadcasting stations.

6. The above setting means, The information processing method according to claim 2, comprising setting the predetermined number based on the guaranteed indicator for the first advertisement and the viewing performance information.

7. In an information processing device, The system acquires viewing history information for each program for each viewer, and viewing forecast information, including future viewing predictions for each program based on the aforementioned viewing history information, at predetermined times. A mathematical model that inputs at least advertising inventory, predetermined constraints regarding advertising restrictions, and metrics regarding display guaranteed by contract, and allocates advertising slots to satisfy the predetermined constraints, is then input each time the predetermined timing arrives, the viewing performance information, the viewing forecast information, and constraints regarding the target users of the advertisement identified based on the viewing performance information, and updates the slot allocation, including advertisements for the contract period, to satisfy the constraints using the viewing forecast information. To obtain the updated frame data output by the aforementioned mathematical model, A program that executes something.

8. An acquisition unit that acquires viewing history information for each program for each viewer, and viewing prediction information including future viewing predictions for each program for each viewer based on the viewing history information, at a predetermined timing. An update control unit that inputs at least advertising inventory, predetermined constraints regarding advertising restrictions, and metrics regarding display guaranteed by contract, and allocates advertising slots to satisfy the predetermined constraints, and at each predetermined time the viewing performance information, the viewing forecast information, and constraints regarding users to whom the advertisement is to be displayed, identified based on the viewing performance information, and updates the slot allocation, including advertisements for the contract period, to satisfy the constraints using the viewing forecast information, An update unit that acquires the updated frame data output by the mathematical model, An information processing device equipped with the following features.

Images