Information processing method, program, and information processing device

Abstract

By at least performing post-processing on the drafted ad placement, the likelihood of creating a design that meets the user's needs can be improved. [Solution] An information processing method comprising: an information processing device acquiring one or more spot contracts; outputting to another information processing device input information used in a mathematical model for creating advertising slots for each spot contract, and information on available advertising slots for a predetermined period; obtaining a proposed advertising slot layout from the other information processing device that has executed the mathematical model, based on the input information and the available slot information; changing the input information if there is a change instruction for the proposed layout; outputting the changed input information to the other information processing device; and obtaining the updated proposed advertising slot layout from the other information processing device that has executed the mathematical model again.

Description

Technical Field

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

Background Art

[0002] Conventionally, a technology is known that includes: an information acquisition unit having an ad slot information acquisition unit that acquires predetermined information related to ad slots provided by an ad slot provider, and a contract information acquisition unit that acquires predetermined information related to a contract for any ad slot concluded between the ad slot provider and an ad slot user; a matrix generation unit that generates a matrix composed of a plurality of cells in which ad slot items and contract items are two-dimensionally represented as row elements and column elements; a score setting unit that sets an arbitrary score for each cell of the matrix; a deviation degree calculation unit that calculates a degree of deviation from the average value of the scores set for all the cells; and an ad slot setting unit that allocates ad slots to the ad slot user based on the degree of deviation (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] In the prior art, various technologies have been proposed for processes that automatically generate plans for ad slots, but the actual situation is that no attention has been paid to inputting data to an apparatus that performs automatic plan generation or acquiring the plan generation results.

[0005] Accordingly, the disclosed technology has an object to provide an information processing method, a program, and an information processing apparatus that improve the possibility of generating a plan desired by a user by performing at least post-processing on the plan generation results for ad slot plan generation.

Means for Solving the Problems

[0006] An information processing method in one aspect of the disclosure involves an information processing device acquiring one or more spot contracts, outputting to another information processing device input information used in a mathematical model for creating advertising slots for each spot contract, and information on available advertising slots for a predetermined period, obtaining a proposed advertising slot layout from the other information processing device that has executed the mathematical model, modifying the input information if there is a change instruction for the proposed layout, outputting the modified input information to the other information processing device, and obtaining an updated proposed advertising slot layout from the other information processing device that has executed the mathematical model again. [Effects of the Invention]

[0007] According to the disclosure technology, by at least post-processing the drafting results of the ad slot design, the likelihood of creating a design that meets the user's needs is improved. [Brief explanation of the drawing]

[0008] [Figure 1] This diagram illustrates an example of the system configuration of an information processing system according to an embodiment. [Figure 2] This diagram illustrates an example of the configuration of an intermediate server according to the embodiment. [Figure 3] This figure shows an example of contract information according to the embodiment. [Figure 4] This figure shows an example of available frame information according to the embodiment. [Figure 5] This diagram shows an overview of the creation of the frame design in the embodiment. [Figure 6] This figure shows an example of the division process according to the embodiment. [Figure 7] This is a diagram showing an example of a pattern according to the embodiment. [Figure 8] This is a sequence diagram showing an example of the frame design creation process according to the 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 configuration.

[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, as well as programs distributed in response to user requests. 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] The advertising slots are, for example, advertising slots for television commercials. However, they are not limited to this; advertisements broadcast using these advertising slots may also be radio commercials.

[0012] The advertising slot information may include, for example, information provided by the provider's equipment, such as a broadcasting station. The advertising slot information may also include, for example, advertising slot identification information (advertising slot ID) that uniquely identifies the advertising slot, the day and time when the advertisement is broadcast in the advertising slot (advertising broadcast time slot), the broadcasting station, 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 incorporated within or immediately before the broadcast time), the viewership rating, and / or the price of the advertising slot (purchase price).

[0013] Furthermore, advertising slot information may include, for example, the start time of the advertising slot, the start time of the program, the end time of the program, and / or program information (the genre of the program (news program, information program, etc.)). The type of advertising slot refers to information that shows the relationship between the advertising slot and the broadcast time of the program, and specifically includes "PT (Participating Commercial: a time set within the program's broadcast time)" and "SB (Station Break: the time between one program and the next)."

[0014] The advertising slot information may include other information besides that mentioned above, such as broadcast area (regional information), broadcast period, broadcast time (daytime when the advertisement is broadcast), 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.

[0015] <System Configuration and Overview> Referring to Figure 1, an example of the system configuration of the information processing system 1 according to this embodiment will be described. The information processing system 1 shown in Figure 1 may include a broadcast server 100, an intermediate server 200, an execution server 300, a display device 400, and an AI (Artificial Intelligence) server 500.

[0016] The advertising server 100 is an information processing device that, for example, the sales department uses to perform or assist in tasks such as scheduling ad slots and allocating materials based on the terms and conditions of advertising contracts. For example, the advertising server 100 outputs advertising contract information to the intermediate server 200 so that the execution server 300, which performs the planning using computer processing, can create the advertisement.

[0017] Further, the sales and broadcasting server 100 creates scheduling data (for example, advertisement slot information) including the transmission time slot of the advertisement that is the subject of the contract described above, or creates material allocation data including advertisement materials allocated to each schedule in the scheduling data. The scheduling data and / or the material allocation data created in the sales and broadcasting server 100 are transmitted to the intermediate server 200.

[0018] The intermediate server (cooperation server) 200 is an information processing device that exchanges data between the sales and broadcasting server 100 and the execution server 300, and may be a server provided on the cloud. The intermediate server 200 acquires contract information and advertisement slot information from the sales and broadcasting server 100, and performs preprocessing as necessary so that the information can be processed in the execution server 300. The preprocessing includes, for example, a process of converting each piece of information in the acquired contract information and / or advertisement slot information into the format of input information for the mathematical optimization algorithm (mathematical model) processed by the execution server 300. As a specific example, the intermediate server 200 performs a process of extractable schedulable slot information in the advertisement slot information and each piece of information in the contract information for input information, and / or converting them into a format.

[0019] Further, the intermediate server 200 acquires the draft schedule created by the execution server 300, and outputs this draft schedule to the sales and broadcasting server 100 or to the display device 400. When a modification of the draft schedule is requested by the display device 400 or the sales and broadcasting server 100, the intermediate server 200 modifies the input information in response to the modification request, and outputs the modified input information to the execution server 300. For example, the intermediate server 200 may perform the update processing of the draft schedule until the draft schedule satisfies a predetermined condition.

[0020] Upon acquiring the input information and the free slot information from the intermediate server 200, the execution server 300 uses a mathematical optimization algorithm (mathematical model) based on the input information and the free slot information to perform scheduling and search for various contracts for the free slots so that the objective by the objective function is achieved while satisfying the constraints, and performs appropriate scheduling. • Objective function: Maximize the unused GRP (Gross Rating Point) for the specified number of episodes (Σ variable related to the number of seconds of empty time slot (e.g., number of seconds empty / 15) × viewership rating). The goal is to allocate slots for each contract while satisfying the constraints, and maximize the GRP (Gross Rate Points) of the remaining available slots. • Constraints: Must meet pre-specified broadcasting rules (e.g., time constraints, avoidance of certain topics, etc.) The allocation information for each contract determined by the mathematical model is output to the intermediate server 200. Based on the allocation information output by the intermediate server 200 to the broadcast server 100, the advertising material indicated by the advertising material ID is included in the broadcast wave and transmitted. For example, based on the broadcast wave received by the television equipment, the broadcast program and broadcast advertisement are displayed on the screen, and the broadcast advertisement is displayed on the television equipment.

[0021] Furthermore, the execution server 300 may execute each process in the algorithm for creating the framework when a predetermined timing arrives, based on input information already acquired (e.g., the latest input information). For example, the execution server 300 may execute each process of this algorithm and create the framework when a predetermined time of day arrives (e.g., midnight).

[0022] The display device 400 is a device that displays the frame proposals obtained from the intermediate server 200, and is capable of accepting changes to the frame proposals and confirmed advertising slots. The display device 400 may be a device used by employees belonging to the department related to broadcasting, and may be a device equipped with at least a display and processor, such as a personal computer, tablet terminal, or smartphone.

[0023] The AI ​​(Artificial Intelligence) server 500 processes requests from the intermediate server 200, including setting rules for creating frameworks, creating initial solutions for frameworks, and creating frameworks to be executed by the execution server 300, using AI. The AI ​​may be a generative AI using a large language model (LLM), or an AI using a machine learning model.

[0024] The AI ​​server 500 may, for example, acquire contract information and slot information from the intermediate server 200, generate an initial solution for the slot plan using AI processing, and output it to the intermediate server 200. In this case, the intermediate server 200 outputs the initial solution for the slot plan to the execution server 300, and the execution server 300 can simulate the allocation of advertising slots based on the initial solution for the slot plan, thereby reducing processing time.

[0025] The AI ​​server 500 may use AI to generate constraint rules (constraint information) regarding constraint conditions that result in a more appropriate solution by learning from past advertising slots (program genre, time slot, assigned advertising material), and output them to the intermediate server 200. In this case, the intermediate server 200 outputs the constraint information obtained by the AI ​​server 500 to the execution server 300, causing the execution server 300 to execute the slot selection that satisfies the constraint information.

[0026] The AI ​​server 500 may, for example, perform the processing that the execution server 300 would normally do, using AI. As mentioned above, the AI ​​processing may use a machine learning model or a generative AI based on LLM.

[0027] The network to which each device within Information Processing System 1 is connected 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. The various configurations in the disclosed technology will be described below.

[0028] <Server Configuration> Figure 2 illustrates an example of the configuration of the intermediate server 200 according to this embodiment. As shown in Figure 2, the intermediate server 200 includes one or more processors (CPUs) 210, a storage device 230, a user interface 250, one or more communication interfaces 260, and one or more communication buses 270 for interconnecting these components.

[0029] The communication interface 260 obtains broadcasting contract information and / or advertising slot information from the broadcasting server 100 via the network N, and outputs contract information and / or available slot information to the execution server 300. The communication interface 260 also obtains slot proposals from the execution server 300 and outputs slot proposals to the broadcasting server 100 or the display device 400.

[0030] 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.

[0031] The storage device 230 may be, for example, a high-speed random access memory such as DRAM, SRAM, or other random access solid-state memory, or it may 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 it may be a non-temporary recording medium that can be read by a computer.

[0032] The storage device 230 stores data used by the information processing system 1. For example, the storage device 230 stores data related to each advertising material, contract information related to each advertisement contract, advertising slot information, etc. In addition, television advertising materials received by the broadcasting station from advertisers may be stored in an external system such as an CM bank (database).

[0033] Another example of the storage device 230 may be one or more storage devices located remotely from the processor 210. In one embodiment, the storage device 230 stores programs, modules, and data structures, or subsets thereof, executed by the processor 210.

[0034] The processor 210 executes a program stored in the storage device 230 to configure a control unit 211 having the functions of the intermediate server 200 described above. The control unit 211 includes, for example, a first acquisition unit 212, an output unit 213, a second acquisition unit 214, a modification unit 215, a presentation unit 216, and a proposal unit 217.

[0035] <<Framework creation process>> The control unit 211 controls the creation of a slot plan for a predetermined advertising slot. The first acquisition unit 212 acquires one or more spot contracts. For example, the first acquisition unit 212 acquires contract information for one or more spot contracts transmitted by the broadcast server 100. The contract information includes, for example, the broadcast contract ID, contract period, number of seconds, designated episode number, period ratio, target, contract GRP, image pattern, degree of dispersion, product genre, slot allocation data, importance, and at least one of the NG program genres. The designated episode number is data indicating which week's weekly report (viewership rating) average of the previous four weeks should be used as the basis.

[0036] The output unit 213 outputs to another information processing device (e.g., execution server 300) the input information used in the mathematical model for designing advertising slots for each spot contract, and the information on available advertising slots for a predetermined period. The output unit 213 outputs the input information, for example, which has been converted (e.g., format converted) as necessary from the contract information. The input information includes, for example, the contract period, the number of seconds, the specified number of episodes, the period ratio, the contract GRP, the image pattern, the degree of dispersion, the product genre, the available slot key (slot ID), the importance level, and the NG genre, at least one of these. The available slot key is information indicating the ID of an available slot that satisfies the image, NG genre, etc., and the control unit 211 may set the available slot key as a preprocessing step. The available slot information may be, for example, the advertising slot indicated by the available slot key, with the viewership rating indicated by the specified number of episodes added to it.

[0037] The output unit 213 may extract one or more data from the contract information acquired by the first acquisition unit 212, and output input information to the execution server 300 in which the data format and other aspects have been converted for the extracted one or more data. Furthermore, if the first acquisition unit 212 acquires contract information from multiple companies, the output unit 213 may output input information in which the contract information acquired from each company has been converted into information in a unified format used for the mathematical model. Here, conversion may include extracting data necessary for the mathematical model from the data included in the contract information.

[0038] The second acquisition unit 214 acquires proposed advertising slots created based on input information and available slot information from other information processing devices (e.g., execution server 300) that have executed the mathematical model. For example, the execution server 300 acquires input information and available slot information, inputs them into the mathematical model, and executes it. Through the execution of the mathematical model, each contract included in the input information is allocated to the available advertising slots indicated by the available slot information so as to satisfy the constraints. The second acquisition unit 214 outputs the proposed slots created by the mathematical model to the display device 400 and the broadcasting server 100.

[0039] The modification unit 215 modifies the input information if there are modification instructions for the created ad slot plan. For example, the modification unit 215 may automatically relax the constraints included in the contract information according to the created ad slot plan. For example, the modification unit 215 may make modification instructions by relaxing or deleting the period ratio in the input information (±5% → ±10%).

[0040] The output unit 213 outputs the modified input information from the modification unit 215 to another information processing device (e.g., the execution server 300). The output unit 213 may output the modified input information to the execution server 300 each time the input information is modified by the modification unit 215.

[0041] The second acquisition unit 214 may acquire the updated ad slot layout from another information processing device (e.g., execution server 300) that has run the mathematical model again. The second acquisition unit 214 outputs the updated ad slot layout to the display device 400 or the broadcast server 100.

[0042] Through the above processing, by at least performing post-processing on the drafted advertising space, the likelihood of creating a draft that meets the user's needs can be improved. For example, if there are change instructions for a draft advertising space created using a mathematical model, modifying the input information will allow for more flexible creation of draft advertising space.

[0043] In addition to the processing described above, the modification unit 215 may also include mitigating the distribution of advertising slots in the input information for each spot contract. For example, the modification unit 215 may change the degree of distribution of contracts included in the contract information for contracts where advertising slots were not allocated, or for contracts where slots were allocated but have a low priority, in a direction that does not distribute them. The distribution information includes the degree of distribution or priority.

[0044] The contract stipulates that if the degree of dispersion (1-3: higher numbers indicate a higher degree of dispersion) is 3 within the contract period, the modification section 215 may reduce the degree of dispersion (change it to 2 or 1) in order to increase the likelihood of advertising slots being allocated to this contract, or to allocate slots to the entire contract. The modification section 215 may also automatically reduce the degree of dispersion (by decrementing it by one from the current degree of dispersion).

[0045] By performing the above processes, the degree of dispersion of the ad slots that are reserved is reduced, thereby increasing the likelihood of securing an ad slot and making it possible to increase the number of ad slots that are reserved.

[0046] In addition to or instead of any of the above-described processes, the modification unit 215 may also include deleting spot contracts from the input information whose priority is lower than a predetermined value. For example, if a contract for which an advertising slot has not been allocated has a high priority, the modification unit 215 may be modified to delete the contract information of spot contracts with a lower priority than that priority from the input information. A higher priority indicates that the contract will be allocated an available slot earlier. The modification unit 215 may also perform the priority change automatically.

[0047] Through the above process, it becomes possible to allocate slots to contracts with higher priority, using the priority levels included in the contract.

[0048] The modification unit 215 may, in addition to or instead of any of the above-described processes, include dividing a predetermined period included in the contract. For example, the modification unit 215 may include adjusting the input information for each divided period for contracts that were not allocated or for contracts with low priority that were allocated. For example, if the contract period is 1 to 30 days, the modification unit 215 may divide it into the first half (1 to 15 days) and the second half (16 to 30 days), distributing the contract GRP between the first and second halves, or it may use different patterns for the first and second halves. The modification unit 215 may also automatically divide the predetermined period based on a pre-set division method.

[0049] In this case, the output unit 213 may also include outputting the adjusted input information to another information processing device (e.g., execution server 300) for each divided period of the predetermined contract. For example, the output unit 213 may output to the execution server 300 in a manner that satisfies the constraints for each divided period of the predetermined contract.

[0050] By dividing the contract period as described above, it becomes possible to create advertising space proposals with greater flexibility.

[0051] In addition to or instead of any of the above-described processes, the modification unit 215 may also determine that there is a change instruction and modify the input information if the frame proposal obtained from another information processing device (e.g., execution server 300) does not satisfy predetermined conditions for a suitable frame proposal. For example, the predetermined conditions may include, for frame proposals created by a mathematical model, that frame allocation was successful for all contracts, that the number of contracts that have not been allocated is less than or equal to a predetermined value, or that advertising slots have been allocated for contracts with an importance higher than a predetermined value. If the predetermined conditions are not met, the modification unit 215 may determine that there is a change instruction and automatically modify the input information. Furthermore, this modification process may be performed recursively until the predetermined conditions are met. If there is a change instruction, the modification unit 215 may execute at least one of the above-described modification processes.

[0052] Through the above process, the modification unit 215 can recursively perform modifications to the input information and recreate the framework based on predetermined conditions for the framework created by the mathematical model.

[0053] ≪Using a GUI (Graphical User Interface)≫ The presentation unit 216 presents the user with a framework created using a mathematical model. For example, the presentation unit 216 outputs the framework to the display device 400, and the display device 400 displays the framework and input information (including the conditions that satisfy the framework). In some cases, a change instruction may be given to the input information that forms the basis of the framework displayed using the display device 400. For example, there are cases where the framework created by the execution server 300 is better modified by a human eye.

[0054] At this time, the user checks the frame proposal displayed on the display device 400 and gives change instructions such as removing the framed advertisement frame or moving it to another advertisement frame, or relaxes the items included in the input conditions. The presentation unit 216 receives change instructions from the user to whom the frame proposal has been presented, and the modification unit 215 modifies the input information according to these change instructions. For example, the modification unit 215 may relax the constraints of a given advertisement or change the parameters of some of the input information in the contract information according to the change instructions.

[0055] Through the above process, the created framework proposal is presented, and if necessary, the user can provide instructions for changes. The presentation unit 216 may also present the initial framework proposal compared with the framework proposal approved by the user, or it may present the initially entered input information compared with the input information of the framework proposal approved by the user. This allows the user to understand the input information used when a good framework proposal is created.

[0056] In addition to the processing described above, the presentation unit 216 may accept the user's designation of one or more spot contracts. The modification unit 215 may confirm the ad slots allocated to the one or more spot contracts for which the designation has been accepted. The modification unit 215 may also modify the available slot information and the input information so as to remove the ad slots of confirmed spot contracts from the available slot information and remove confirmed spot contracts from the input information. In this case, the output unit 213 may output the modified input information and available slot information to the execution server 300 and cause the execution server 300 to create the ad slot plan again.

[0057] Through the above process, by changing the input information of unconfirmed contracts other than those that have been finalized, it becomes possible to finalize contracts for which appropriate ad slots have been allocated, and to re-allocate ad slots for other contracts. User arrangements may be assisted or replaced by the AI ​​agent function of AI Server 500. Specifically, the AI ​​may reflect the user's natural language-based image of the proposed ad slots in the appropriate contract input information changes, or the AI ​​may autonomously make corrections based on pre-set evaluation condition prompts for the proposed ad slots.

[0058] ≪Using AI≫ The proposal unit 217 inputs input information and available slot information into a learning model that has been trained using the content of each spot contract and the advertising slots allocated for each spot contract as training data. For example, the AI ​​server 500 has this learning model and trains using the input information of each spot contract that has actually been allocated and the information of each advertising slot (period, date and time, viewership rating, etc.) as training data. As a result, this learning model outputs slot proposal information in which each spot contract has been allocated to a predetermined advertising slot by inputting the input information of each spot contract. The AI ​​server 500 may be a server that is linked to an external data source, and it may also utilize an external general-purpose AI model to realize AI processing functions.

[0059] For example, before the input information and available slot information are output to the mathematical model, the proposal unit 217 may output the input information and available slot information to the AI ​​server 500 to obtain the slot plan information for the advertising slots reserved in each spot contract. The proposal unit 217 may also consider the slot plan information obtained from the AI ​​server 500 as the initial solution.

[0060] The output unit 213 may also output the design created by the AI ​​server 500's learning model as the initial solution to another information processing device (e.g., execution server 300). This allows the execution server 300 to perform a process to search for the optimal solution based on the initial solution, thereby reducing processing time and processing load.

[0061] The mathematical model of the execution server 300 may be a learning model that has learned using the contents of each spot contract and the advertising slots allocated to each spot contract as training data, or it may be a generative AI that designs advertising slots based on the contents of each spot contract. For example, the mathematical model of the execution server 300 may be a model that is used with AI.

[0062] If the mathematical model is a pre-trained model, it may be the same model used by the AI ​​server 500 mentioned above. If the mathematical model is a model that uses generative AI, the generative AI can learn the history of previously allocated advertising slot information (advertising slot information that satisfies the constraints) so that when each spot contract is acquired, it can output appropriate slot suggestions.

[0063] The AI ​​server 500 may use machine learning to generate constraint rules to guarantee good image patterns (for example, the character "ヨ" described later). By training it with advertising slots allocated in past contracts, it may generate constraint rules to guarantee good image patterns for each advertising genre. Alternatively, the AI ​​server 500 may create constraint rules for each advertiser or advertising agency by inputting information on past slot allocations for each advertiser or advertising agency.

[0064] Furthermore, the AI ​​server 500 may use reinforcement learning to allow the AI ​​agent to relax constraints and perform recursive processing. The AI ​​server 500 may also use unused GRPs for the objective function of the mathematical model of the execution server 300 as a reward, and create a framework that maximizes this reward while relaxing constraints.

[0065] Recursive processing As described above, the control unit 211 may repeat the creation of a draft framework until the predetermined conditions are met, but it may also repeat the draft framework creation process recursively as follows. (1) Optimization using a mathematical model under initial conditions (2) If a solution cannot be obtained (infeasible), adjust the conditions (relax, reduce the contracts being used) and repeat mathematical optimization until a solution is obtained. Condition adjustments involve inputting high-priority contracts first to fix the allocation, or mitigating the degree of dispersion for lower-priority contracts, before re-inputting the input information for each contract into the mathematical model. (3) Use the remaining slots to input into the mathematical model how to accommodate as many low-priority contracts as possible that could not be included in (2).

[0066] <Data Example> Figure 3 shows an example of contract information according to one embodiment of the disclosure. In the example shown in Figure 3, the contract information for each contract includes a contract ID, contract period, duration in seconds, designated issue number, image, frame ID, constraint information, etc. "Contract" indicates the type of advertising contract, "Period" indicates the duration of the contract, "Duration" indicates the duration of the advertisement in this contract, and "Designated Issue Number" indicates which week's weekly report (viewership rating) average of the previous four weeks will be used as the basis.

[0067] "Pattern" indicates the distance between allocated slots. For example, in a list of programs for which a contract has been allocated slots, it shows an indicator of whether the allocated slots are distributed evenly when a predetermined contract is applied to an already allocated slot. "Slot ID" indicates an available advertising slot, and "Restriction" indicates restriction information to prevent the contract from being allocated slots in PT or SB.

[0068] Figure 4 shows an example of available slot information according to one embodiment of the disclosure. In the example shown in Figure 4, the available slot information includes slot ID, time slot, duration in seconds, program genre, episode number × viewership rating, etc. "Slot ID" is the slot ID shown in Figure 3 and may be used as a primary key. "Time slot" indicates the start and end times of the advertisement, "Duration in seconds" indicates the number of seconds that can be broadcast as this advertisement slot, "Program genre" indicates the genre of the program broadcast before, after, or after this advertisement slot, and "Episode number × viewership rating" indicates an index used as an indicator for the objective function of the mathematical model, for example, episode number × 5.1%.

[0069] <Specific example> Figure 5 shows an overview of the framework creation process in one embodiment of the disclosure. As shown in Figure 5, the execution server 300 obtains contract information for contracts A, B, C, etc., and available slot information from the intermediate server 200. The intermediate server 200 generates each contract information and available slot information as preprocessing for input into the mathematical model of the execution server 300.

[0070] The intermediate server 200 generates input information, including broadcast date and time information, designated episode number, contract GRP information, picture pattern, distribution degree, slot ID, genre, NG program genre, and priority, based on the contract information obtained from the broadcast server 100. The following is an example of input information. Broadcast information: April 1st to April 15th, 15 seconds each. Designated issue number: 26, 24 Contract GRP: 100 Pattern: Yo Degree of dispersion: 3 Frame ID: 20250401_··· Genre: Consumer finance Forbidden genres: Not suitable for minors Priority:2

[0071] The intermediate server 200 generates available time slot information based on contract information and viewership rating information published by companies that survey viewership ratings. For example, the intermediate server 200 generates available time slot information such as broadcast time, duration in seconds, program genre, and episode number × viewership rating for each time slot ID included in the input information. The following is an example of available time slot information. Frame ID: 20250401_··· Broadcast time: 4 / 1 19:00~19:54 Duration: 120 seconds Program genre: News Episode number × viewership rating: Episode 26 × 5.1%

[0072] The execution server 300 obtains the above-mentioned input information and available frame information as input parameters for the mathematical model and executes the mathematical model. By executing the mathematical model, the execution server 300 creates a frame proposal that satisfies the objective function. The frame proposal includes a frame proposal in which each contract included in the input information has been framed. If the execution server 300 is unable to frame contracts exceeding a predetermined value among the contracts included in the input information, it may output the frame allocation failure result to the intermediate server 200.

[0073] The intermediate server 200 outputs the time slot proposal obtained from the execution server 300 to the display device 400 for display. The display device 400 may, for example, display a table showing where a given contract has been allocated time slots for a one-week program schedule. This allows users such as advertisers, advertising agencies, or broadcasting departments to determine whether appropriate time slots have been allocated for a given contract.

[0074] If there is a change request for the framework created by the execution server 300, the intermediate server 200 will loosen the degree of distribution or lower the priority for the specified contract according to the content of the change request. As described above, the intermediate server 200 changes the input information conditions and outputs the modified input information to the execution server 300, causing it to re-execute the mathematical model.

[0075] The mathematical model may be re-executed by the intermediate server 200, modifying the input information until predetermined termination conditions for the framework are met. These termination conditions include, for example, that all contracts included in the input information are framed, and that the evaluation value of each framed contract is equal to or greater than a predetermined value.

[0076] (Division of contract period) Figure 6 is a diagram illustrating an example of division processing according to an embodiment of the disclosure. In the example shown in Figure 6, the contract period from 4 / 1 to 4 / 14 is divided into 5 segments each lasting 3 days. At this time, the intermediate server 200 may divide the GRP into 5 segments based on the contract GRP and the reservable slots, as a constraint condition, so that each divided period satisfies the minimum number of slots. Further, the intermediate server 200 may set a constraint condition such that at least one slot is reserved on the start day and the last day.

[0077] For example, in the example shown in Figure 6, the intermediate server 200 may set a constraint condition to reserve the minimum number of slots (e.g., 2 slots) in each period of days 1 to 3, days 4 to 6, days 7 to 9, days 10 to 12, and days 13 to 14.

[0078] Figure 7 is a diagram illustrating an example of a pattern according to an embodiment of the disclosure. In the example shown in Figure 7, the character "ヨ" is presented, indicating that a slot is reserved in the upward-pointing arrow frame. The slot reservation shown in Figure 7 is an example of a good slot reservation. Note that patterns include "one", "inverted L", "U-shape", "all days", and the like. The pattern may be specified in a contract, for example.

[0079] The execution server 300 executes various slot reservation patterns such that the variance (or standard deviation) of the X coordinate (day of the week) and / or Y coordinate (OA time) when patterned in the contract is equal to or greater than a level value defined for each pattern type.

[0080] In the execution server 300, it is preferable that a solution to the mathematical model can be obtained. When relaxing constraint conditions, the X coordinate (day of the week) may be eliminated to determine whether the variance of the Y coordinate (OA time) is equal to or greater than the level value, or the leveling period unit may be divided to determine whether the variance is equal to or greater than the level value in each period.

[0081] The intermediate server 200 may change the relaxation method for constraint conditions according to the purpose of slot reservation. For example, the intermediate server 200 may process at least one of the following relaxation methods for the following purposes. Purpose A: Maximize contract accommodation while relaxing partial conditions (exploration of compromise points) • Relaxable conditions: Period ratio, degree of dispersion • If the mathematical model does not produce a solution (i.e., it is not executable), the execution server 300 is instructed to relax constraints such as the period ratio and degree of dispersion for contracts of lower importance and to try to find a solution again. • If execution fails, instruct the execution server 300 to remove the restrictions from contracts of lower importance and then seek a solution again. • Set constraints with a range (e.g., ±5%) for the period ratio and degree of dispersion, which are conditions that can be relaxed.

[0082] Objective B: To optimize the overall framework for all contracts to be included. • For contracts spanning multiple months, the specified monthly GRP will be used (if not specified, it will be automatically prorated based on the number of days). • Allows for framing that may partially violate the constraints. etc.

[0083] <Action Processing> Figure 8 is a sequence diagram showing an example of the frame draft creation process according to the embodiment. The example shown in Figure 8 illustrates a process in which data is exchanged between each device, a frame draft is created by the execution server 300, and the frame draft is modified.

[0084] In step S102, the first acquisition unit 212 of the intermediate server 200 acquires contract information for each broadcasting contract from the broadcasting server 100. The contract information includes, for example, the broadcasting contract ID, contract period, number of seconds, designated episode number, period ratio, target, contract GRP, picture pattern, degree of dispersion, product genre, time slot data, importance, and at least one of the NG program genres.

[0085] In step S104, the output unit 213 of the intermediate server 200 outputs input information based on contract information and available slot information to the execution server 300. The input information is information processed by the control unit 211 and includes information that has been formatted so that it can be processed by the execution server 300, based on predetermined information from the information included in the contract information. The available slot information is information processed by the control unit 211 and includes information that adds viewership ratings to available slots.

[0086] In step S106, the execution server 300 obtains input information and available slot information from the intermediate server 200. Based on the input information and available slot information, it uses a mathematical model to search for various contracts for the available slots, ensuring that the objective is achieved while satisfying the constraints, and then selects an appropriate contract.

[0087] In step S108, the second acquisition unit 214 of the intermediate server 200 acquires information on the framework proposals for each contract from the execution server 300.

[0088] In step S110, the presentation unit 216 of the intermediate server 200 outputs the frame plan information obtained from the execution server 300 to the broadcast server 100.

[0089] In step S112, the presentation unit 216 of the intermediate server 200 outputs the frame plan information acquired from the execution server 300 to the display device 400.

[0090] In step S114, the display device 400 indicates that it has received a change request from the user who has reviewed the frame design.

[0091] In step S116, the modification unit 215 of the intermediate server 200 receives a change request from the display device 400 and modifies the input information, such as constraint information, according to its contents. Note that if the intermediate server 200 automatically relaxed the constraint information or otherwise performed the modification process for input information before steps S110 and S112, the modification unit 215 will perform the modification process for input information.

[0092] In step S118, the output unit 213 of the intermediate server 200 outputs the modified input information from the modification unit 215 to the execution server 300. The output unit 213 may output the modified input information to the execution server 300 each time the input information is modified by the modification unit 215.

[0093] In step S120, when the execution server 300 obtains the modified input information from the intermediate server 200, it uses the mathematical model again to search for various contracts for the available slots based on the available slot information and the modified input information, and then selects an appropriate slot.

[0094] In step S122, the second acquisition unit 214 of the intermediate server 200 acquires information on the updated frame plan, which has been reframed, from the execution server 300.

[0095] In step S124, the presentation unit 216 of the intermediate server 200 outputs the updated frame plan information obtained from the execution server 300 to the broadcast server 100.

[0096] In step S112, the display unit 216 of the intermediate server 200 outputs the updated frame plan information obtained from the execution server 300 to the display device 400.

[0097] Steps S114 to S126 are executed each time the input information (e.g., constraint information) is changed. Also, as described above, the intermediate server 200 may be modified to automatically relax the constraint information if the execution server 300 is unable to frame the model, and output the input information including the changed constraint information to the execution server 300 to recursively execute the mathematical model.

[0098] Through the above process, by applying post-processing to the draft of the ad space, it becomes possible to create the ad space that the user desires.

[0099] The next steps, in the S200 series, describe the process of creating an initial solution to be input to the execution server 300. In step S202, the first acquisition unit 212 of the intermediate server 200 acquires contract information for each broadcasting contract from the broadcasting server 100. The processing in step S202 is the same as the processing in step S102.

[0100] In step S204, the proposal unit 217 of the intermediate server 200 inputs the input information and available slot information into a learning model that has been trained using the content of each spot contract and the advertising slots allocated for each spot contract as training data. For example, the AI ​​server 500 has this learning model and trains using the input information of each spot contract that has actually been allocated and the information of each advertising slot (period, date and time, viewership rating, etc.) as training data. As a result, this learning model outputs slot proposal information in which each spot contract is allocated to a predetermined advertising slot by inputting the input information of each spot contract.

[0101] In step S206, the AI ​​server 500 has this learning model and has trained it using the input information of each spot contract that has actually been allocated and the information of each advertising slot (period, date and time, viewership rating, etc.) as training data. By inputting the input information of each spot contract into this learning model, it outputs slot allocation information in which each spot contract has been allocated to a predetermined advertising slot. This slot allocation information is created as the initial solution.

[0102] Furthermore, the AI ​​server 500 may use reinforcement learning to perform constraint relaxation and recursive processing by the AI ​​agent. The AI ​​server 500 may also create a framework that maximizes the reward by using unused GRPs for the objective function of the mathematical model of the execution server 300 as a reward, while relaxing the constraints.

[0103] In step S208, the proposal unit 217 of the intermediate server 200 obtains initial solution framework information from the AI ​​server 500.

[0104] In step S210, the output unit 213 of the intermediate server 200 outputs input information based on contract information, available slot information, and initial solution framework information to the execution server 300. In subsequent processing, from step S106 onward, the mathematical model is executed based on the initial solution framework so that the objective function achieves its goal.

[0105] By performing the above steps, it becomes possible to search for the optimal solution based on the initial solution, thereby reducing processing time and processing load.

[0106] The above embodiments are illustrative examples for explaining the present invention and are not intended to limit the present invention to those embodiments alone. Furthermore, the present invention can be modified in various ways without departing from its essence. Moreover, those skilled in the art can adopt embodiments in which each of the elements described above is replaced with equivalent ones, and such embodiments are also included within the scope of the present invention.

[0107] [Differentiation] Although the present invention has been described based on the above embodiments, the following cases are also included in the present invention. The execution server 300 in the above embodiment may be executed by an AI server 500. The AI ​​server 500 may cause a generation AI using an LLM (Large Language Model) to store past frame assignments, input a prompt to perform frame assignment based on input information and available frame information, and create frame proposal information based on this prompt. [Explanation of Symbols]

[0108] 1... Information processing system, 100... Broadcast server, Intermediate servers... 200, 210... Processor, 211... Control unit, 212... First acquisition unit, 213... Output unit, 214... Second acquisition unit, 215... Modification unit, 216... Presentation unit, 217... Proposal unit, 220... Communication interface, 230... Storage device, 300... Execution server, 400... Display device, 500... AI server.

Claims

1. An information processing system comprising one or more processors, The one or more processors described above To obtain one or more spot contracts, Using input information that includes at least the contract period, duration in seconds, and viewership data for each spot contract, and information on available advertising slots for a predetermined period, a mathematical model is executed to allocate each spot contract included in the input information to the available slots indicated by the available slot information in a manner that satisfies predetermined constraints. By executing the mathematical model, each spot contract obtains a proposed advertising slot allocation based on the input information and the available slot information, If there is an instruction to change the aforementioned frame plan, the prescribed constraints, or at least one of the contract period and the viewership data included in the input information, Re-execute the mathematical model using the modified constraints or input information. By executing the aforementioned mathematical model, the updated ad slot layouts are obtained. An information processing system that performs [this action].

2. The aforementioned changes are, The information processing system according to claim 1, further comprising mitigating the distribution state of advertising slots with respect to the distribution information of advertising slots included in the input information for each spot contract.

3. The aforementioned changes are, The information processing system according to claim 1, further comprising deleting spot contracts from the input information that have a priority lower than a predetermined value for each of the spot contracts.

4. The aforementioned changes are, To divide the aforementioned contract period, This includes adjusting the input information for each divided period, Running the aforementioned mathematical model again means The information processing system according to claim 1, comprising running the mathematical model again using the adjusted input information for each divided period.

5. The aforementioned changes are, An information processing system according to any one of claims 1 to 4, which includes determining that there is a change instruction if the aforementioned frame proposal does not satisfy predetermined conditions for a suitable frame proposal, and changing the input information.

6. To present the aforementioned framework proposal to the user, The information processing system according to claim 1, comprising the actions of obtaining the change instruction from the user to whom the aforementioned framework proposal has been presented.

7. The above presentation means that Accepting the designation of one or more spot contracts from the aforementioned user, The information processing system according to claim 6, comprising determining the advertising slots allocated to the one or more spot contracts that have been accepted.

8. The input information and the available slot information are input into a learning model that has been trained using the details of each spot contract and the advertising slots allocated to each spot contract as training data. The process involves obtaining ad slot proposals generated by the aforementioned learning model, Executing the aforementioned mathematical model means The information processing system according to claim 1, comprising executing the mathematical model using the framework generated by the learning model as the initial solution.

9. The information processing system according to claim 1, wherein the mathematical model is a learning model that has learned the contents of each spot contract and the advertising slots allocated for each spot contract as training data, or a generative AI that designs advertising slots based on the contents of each spot contract.

10. One or more processors included in the information processing system To obtain one or more spot contracts, Using input information that includes at least the contract period, duration in seconds, and viewership data for each spot contract, and information on available advertising slots for a predetermined period, a mathematical model is executed to allocate each spot contract included in the input information to the available slots indicated by the available slot information in a manner that satisfies predetermined constraints. By executing the mathematical model, each spot contract obtains a proposed advertising slot allocation based on the input information and the available slot information, If there is an instruction to change the aforementioned frame plan, the prescribed constraints, or at least one of the contract period and the viewership data included in the input information, Re-execute the mathematical model using the modified constraints or input information. By executing the aforementioned mathematical model, the updated ad slot layouts are obtained. A program that executes the command.

11. One or more processors included in the information processing system To obtain one or more spot contracts, Using input information that includes at least the contract period, duration in seconds, and viewership data for each spot contract, and information on available advertising slots for a predetermined period, a mathematical model is executed to allocate each spot contract included in the input information to the available slots indicated by the available slot information in a manner that satisfies predetermined constraints. By executing the mathematical model, each spot contract obtains a proposed advertising slot allocation based on the input information and the available slot information, If there is an instruction to change the aforementioned frame plan, the prescribed constraints, or at least one of the contract period and the viewership data included in the input information, Re-execute the mathematical model using the modified constraints or input information. By executing the aforementioned mathematical model, the updated ad slot layouts are obtained. An information processing method that performs the following.

Images