System

Abstract

To provide a system.SOLUTION: A system includes: means for generating a new advertisement talent by using a prompt text for instructing generation of a new advertisement talent and a generative AI model; means for finding an advertisement talent which can recognize at least one of better service and product by comparing the generated new advertisement talent and an existing advertisement talent by using generative AI; means for broadcasting an advertisement produced by using the advertisement talent that is found, for a certain period of time; means for analyzing the emotional reaction of a user who viewed the broadcasted advertisement by using an emotion engine; and means for evaluating the effect of the broadcasted advertisement based on the analyzed emotional reaction of the user.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] Selecting advertising talent is an important factor in increasing awareness of products and services. However, it is difficult to select the best talent from among existing advertising talent, and it is also not easy to discover new talent. [Means for solving the problem]

[0005] To solve this problem, we propose a system that uses generative AI to find talent that can better recognize a service by comparing them with existing advertising talent at regular intervals. This system makes it possible to efficiently discover and utilize new talent. [Brief explanation of the drawings]

[0006] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 2 is a sequence diagram showing a flow of processing in the data processing system according to the first embodiment of the first form example. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Embodiment 1. [Figure 13] FIG. 10 is a sequence diagram showing a processing flow of a data processing system in a second embodiment of the second form example. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 of Embodiment Example 2. [Figure 15] FIG. 10 is a sequence diagram showing the flow of processing in a data processing system according to a third embodiment of the third embodiment. [Figure 16]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 3 of Embodiment 3. [Figure 17] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the first embodiment of the first form example when an emotion engine is combined. [Figure 18] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Form Example 1 when an emotion engine is combined. [Figure 19] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the second embodiment of the second form example when an emotion engine is combined. [Figure 20] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 of Form Example 2 when an emotion engine is combined. [Figure 21] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the third embodiment of the third form example when an emotion engine is combined. [Figure 22] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 3 of Form Example 3 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0007] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0008] First, the terms used in the following description will be explained.

[0009] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, the processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), or a TPU (TENSOR PROCESSING UNIT (registered trademark)).

[0010] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0011] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0012] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0013] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0014] [First embodiment]

[0015] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0016] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0017] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0018] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0019] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0020] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0021] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0022] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0023] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0024] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0025] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0026] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0027] "Example 1"

[0028] One aspect of the present invention is a system that utilizes generative AI. This system analyzes the characteristics and performance of advertising talent and generates new talent based on that. Specifically, the AI ​​learns performance data and viewer response data of existing advertising talent, and generates new talent based on that. The generated talent is compared with existing advertising talent to find talent that can better promote awareness of the service.

[0029] "Example 2"

[0030] The present invention also provides a system for comparing a new talent with existing advertising talent at regular intervals. This regular interval can be adjusted according to the duration of an advertising campaign or market trends. For example, a new talent can be generated to coincide with the launch of a new product or a specific season, and compared with existing talent.

[0031] "Example 3"

[0032] Furthermore, a system for utilizing the discovered talent is also an embodiment of the present invention. Specifically, the generated talent is used in an advertisement and its effectiveness is verified. For example, an advertisement using the generated talent is aired for a certain period of time, and the effectiveness of the talent can be evaluated by analyzing the response.

[0033] The processing flow of each embodiment will be described below.

[0034] "Example 1"

[0035] Step 1: The generative AI collects performance data of existing advertising talent and viewer response data.

[0036] Step 2: Based on the collected data, the AI ​​learns and generates new talents.

[0037] Step 3: The generated talents are compared with existing advertising talents to find talents that can better raise awareness of the service.

[0038] "Example 2"

[0039] Step 1: Set a fixed period, taking into account the duration of the advertising campaign and market trends.

[0040] Step 2: At each set time period, generate new talents and compare them with existing talents.

[0041] "Example 3"

[0042] Step 1: Hire the generated talent for your ads.

[0043] Step 2: Air the advertisement using the talent you selected for a certain period of time.

[0044] Step 3: Analyze the response to the aired advertisement and evaluate the effectiveness of the talent.

[0045] Example 1

[0046] Next, a description will be given of Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0047] Conventional methods for selecting advertising talent have difficulty finding effective talent because they do not fully utilize viewer response and performance data. It is also difficult to flexibly adjust talent according to the duration of the advertising campaign and market trends. Therefore, a system that can efficiently generate and evaluate new talent to maximize advertising effectiveness is needed.

[0048] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0049] In this invention, the server includes a means for utilizing generative AI, a means for collecting performance data and viewer reaction data of existing advertising talent, a means for preprocessing the collected data, a means for inputting the preprocessed data into an AI model for learning, a means for generating new talent using the learned AI model, a means for evaluating the performance of the generated talent, and a means for outputting the evaluation results. This makes it possible to analyze the characteristics and performance of advertising talent in detail and efficiently generate and evaluate more effective talent.

[0050] "Generative AI" is an artificial intelligence technology that generates new data and models based on existing data.

[0051] "Performance data" refers to data that shows the effectiveness of an advertising talent, such as the number of views, click rate, and purchase rate.

[0052] "Viewer response data" refers to data that records viewers' reactions to advertisements, such as comments, ratings, and number of shares.

[0053] "Preprocessing" refers to the process of converting collected data into a format suitable for an AI model by performing processes such as filling in missing values, removing outliers, and normalizing the data.

[0054] An "AI model" is an algorithm built using a machine learning framework that learns patterns from data and makes predictions or generation decisions.

[0055] A "talent" is a person or character who appears in an advertisement.

[0056] "Evaluation" refers to the process of measuring the performance of the generated talent using indicators such as number of views, click rate, and purchase rate, and comparing it with existing talent.

[0057] "Output" means providing the evaluation results in the form of a report or data.

[0058] This invention relates to a system that uses generative AI to analyze the characteristics and performance of advertising talent and generate new talent. Specific embodiments of this system are described below.

[0059] The server first collects performance data of existing advertising talent and viewer response data, using APIs to retrieve data from company databases and social media platforms, including the number of ad views, click-through rates, purchase rates, viewer comments, ratings, and number of shares.

[0060] The server then preprocesses the collected data, which includes using a data analysis library to impute missing values, remove outliers, and normalize the data. For example, data with extremely low views or abnormally high click-through rates is removed, and the data is normalized using min-max scaling.

[0061] The preprocessed data is then fed into an AI model. The server uses a deep neural network (DNN) model built with machine learning frameworks such as TENSORFLOW® or PyTorch. The model learns the characteristics of advertising talent and audience reactions, and finds patterns to generate new talent.

[0062] Once the learning is complete, the server generates a new talent. The generated talent is expected to have similar characteristics to the existing talent, but with higher performance. For example, a talent with specific attributes such as "female in her 20s, cheerful personality, sports lover" may be generated.

[0063] The performance of the generated talent is evaluated by the server. The evaluation uses indicators such as the number of views, click rate, and purchase rate, and compares it with existing talent. If the evaluation results show that the generated talent performs better than the existing talent, that talent is selected.

[0064] Finally, the server outputs the evaluation results, which are output as an Excel report and emailed to the user, who can then use the report to plan a new advertising campaign.

[0065] As a concrete example, consider a company looking for a new advertising talent for a beverage. The server collects performance data (number of views, click-through rate, purchase rate, etc.) and viewer response data (comments, ratings, number of shares, etc.) of previously used beverage advertising talent, inputs this data into an AI model, and generates new talent. If the generated talent performs better than existing talent, it will be used in the new advertising campaign.

[0066] Example prompt sentence:

[0067] "Generate new beverage advertising talent based on performance data and viewer response data of past beverage advertising talent. The generated talent must exceed existing talent in terms of views, click-through rates, or purchase rates."

[0068] In this way, the server utilizes generative AI to analyze the characteristics and performance of advertising talent, creating a system that generates new talent.

[0069] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0070] Step 1: Data collection

[0071] The server collects existing advertising talent performance data and viewer response data from the company's database and social media platforms. Specifically, it uses APIs to obtain data such as the number of views, click-through rates, conversion rates, comments, ratings, and shares. The input is the raw data obtained from the API, and the output is the collected raw data.

[0072] Step 2: Data Preprocessing

[0073] The server preprocesses the collected data. Specifically, it uses a data analysis library to fill in missing values, remove outliers, and normalize the data. For example, it removes data with extremely low view counts and abnormally high click-through rates, and normalizes the data using min-max scaling. The input is the collected raw data, and the output is the preprocessed, clean data.

[0074] Step 3: Training the AI ​​model

[0075] The server inputs the preprocessed data into the AI ​​model and begins learning. The AI ​​model used is a DNN (deep neural network) model built using machine learning frameworks such as TensorFlow and PyTorch. The model learns the characteristics of advertising talent and viewer reactions, and finds patterns to generate new talent. The input is the preprocessed data, and the output is the trained AI model.

[0076] Step 4: Generate new talent

[0077] The server uses the trained AI model to generate new talent. The generated talent is expected to have similar characteristics to existing talents, but with higher performance. For example, a talent with specific attributes such as "female in her 20s, cheerful personality, sports lover" may be generated. The input is the trained AI model, and the output is the characteristics of the newly generated talent.

[0078] Step 5: Performance evaluation

[0079] The server evaluates the performance of the generated talent. Specifically, it compares it with existing talent using metrics such as the number of views, click-through rate, and purchase rate. If the evaluation results show that the generated talent performs better than the existing talent, that talent is selected. The input is the characteristics of the generated talent, and the output is the evaluation result.

[0080] Step 6: Output the results

[0081] The server outputs the evaluation results. Specifically, it outputs the evaluation results as an Excel report and sends it to the user by email. The user then plans a new advertising campaign based on this report. The input is the evaluation results, and the output is the report provided to the user.

[0082] (Application example 1)

[0083] Next, a description will be given of Application Example 1 of Embodiment Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0084] Traditional methods for selecting advertising talent make it difficult to fully utilize existing talent performance data and viewer response data, making it difficult to maximize the effectiveness of advertising campaigns. Furthermore, selecting advertising talent takes time and money, making it difficult to respond to market fluctuations that require rapid response. This has led to the issue of a lower success rate for advertising campaigns.

[0085] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0086] In this invention, the server includes a means for utilizing generative AI, a means for comparing with existing advertising talent at regular intervals, and a means for searching for talent who can better recognize the service. This optimizes the advertising campaign, and a means for using an application installed on a smartphone. This automates the advertising talent selection process, quickly and effectively generates new talent, and increases the success rate of advertising campaigns.

[0087] "Generative AI" is an artificial intelligence technology that learns from existing data and generates new data and models.

[0088] A "certain period" is a unit of time that can be adjusted depending on the duration of the advertising campaign and market trends.

[0089] An "advertising talent" is a person or character who appears in an advertisement.

[0090] "Performance Data" means data relating to the activities and achievements of advertising talent.

[0091] "Viewer response data" refers to data regarding the reactions and feedback of viewers who have viewed advertisements.

[0092] An "application installed on a smartphone" is a software program that runs on a smartphone.

[0093] "Advertising campaign optimization measures" are methods and processes for maximizing the effectiveness of advertising.

[0094] The following system configuration will be described as an embodiment of the present invention.

[0095] The server includes a means for utilizing generative AI, a means for comparing with existing advertising talent at regular intervals, a means for finding talent who can better recognize the service, a means for optimizing advertising campaigns, and a means for using applications installed on smartphones.

[0096] System Program

[0097] The server builds a generative AI model. Specifically, it reads performance data and viewer response data of existing advertising talent and builds a neural network model using LSTM (Long Short-Term Memory) layers and Dense layers. The model is compiled using the Adam optimizer and categorical cross-entropy loss function. New talent is generated using the trained model.

[0098] Processing Description

[0099] The server does the following:

[0100] 1. Data Loading: The server loads existing advertising talent performance data and viewer response data, which may include retrieving data from databases and cloud storage.

[0101] 2. Building a model: The server builds a neural network model using LSTM layers and Dense layers, which enables processing of time series data.

[0102] 3. Compile the model: The server compiles the model using the Adam optimizer and categorical cross-entropy loss function, which allows for efficient model training.

[0103] 4. Model training: The server uses the loaded data to train a model, which can then analyze the characteristics and performance of advertising talent.

[0104] 5. New talent generation: The server generates new talent using the trained model, which allows for faster discovery of more effective advertising talent.

[0105] Specific examples

[0106] For example, when an advertising agency launches an advertising campaign for a new beverage, it can input the performance data of existing talent and viewer response data to generate the most suitable new talent, thereby increasing the success rate of the advertising campaign.

[0107] Prompt Sentence Examples

[0108] "Generate new talent using performance data and viewer response data from existing advertising talent."

[0109] In this way, the server can automate the advertising talent selection process, quickly and effectively generating new talent and increasing the success of advertising campaigns.

[0110] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0111] Step 1:

[0112] The server loads existing advertising talent performance data and viewer response data from a database or cloud storage. The input data is the advertising talent's activity history and viewer feedback data, and uses these for subsequent processing. The output is the loaded dataset.

[0113] Step 2:

[0114] The server preprocesses the loaded data, specifically imputing missing values, normalizing data, encoding categorical data, etc. The input is the dataset loaded in step 1, and the output is the preprocessed dataset.

[0115] Step 3:

[0116] The server uses the preprocessed data to build a neural network model with an LSTM layer and a Dense layer. The input is the preprocessed dataset, and the output is the built neural network model.

[0117] Step 4:

[0118] The server compiles the constructed model, specifically using the Adam optimizer and categorical cross-entropy loss function. The input is the model constructed in step 3, and the output is the compiled model.

[0119] Step 5:

[0120] The server trains the dataset with the compiled model: the input is the preprocessed dataset and the compiled model, and the output is the trained model.

[0121] Step 6:

[0122] The server generates new advertising talent using the trained model. Specifically, it uses performance data of existing talent as input to predict the characteristics of the new talent. The input is the performance data of the existing talent, and the output is the characteristics of the generated new talent.

[0123] Step 7:

[0124] The server compares the characteristics of the generated new talent with the existing talents and selects the more effective talent. The inputs are the characteristics of the generated new talent and the characteristics of the existing talents, and the output is the selected optimal talent.

[0125] Step 8:

[0126] The server generates data for utilizing the selected optimal talent in an advertising campaign and sends it to an application installed on a smartphone. The input is the data of the selected optimal talent, and the output is the data to be utilized in the advertising campaign.

[0127] Example 2

[0128] Next, a description will be given of Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0129] Traditional methods for selecting advertising talent make it difficult to quickly find the optimal talent based on market trends and the duration of an advertising campaign. Furthermore, creating new talent and comparing them with existing talent is done manually, which is inefficient, time-consuming, and costly.

[0130] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0131] In this invention, the server includes a means for a user to input a prompt sentence, a means for the server to receive the prompt sentence and send it to the generative AI model, a means for the generative AI model to generate an image and profile of a new talent, a means for the server to compare the generated talent with existing talent, and a means for the server to provide the comparison result to the user, thereby enabling the selection of the optimal talent quickly and efficiently according to the duration of the advertising campaign and market trends.

[0132] "Generative AI" refers to artificial intelligence that generates images and profiles of new advertising talent based on prompts entered by the user.

[0133] A "prompt sentence" is the text that the user enters to instruct the generative AI to generate a new talent.

[0134] The "server" is a computer system that receives prompt text from the user, sends it to the generation AI, manages and compares the information of the generated talent, and provides the results to the user.

[0135] "Advertising personality" means a person or character used to promote a product or service in an advertising campaign.

[0136] "Image" refers to the visual representation of the advertising talent generated by generative AI.

[0137] A "profile" refers to information such as the advertising talent's name, age, and career history that is generated by generative AI.

[0138] "Comparison" refers to the process of evaluating the new talent generated against existing talent to determine which is more suitable for the advertising campaign.

[0139] "User" means a person or organization that uses the system to generate new advertising talent and receive comparison results.

[0140] A "certain period of time" is a time that can be adjusted depending on the duration of the advertising campaign and market trends.

[0141] This invention relates to a system that uses generative AI to generate advertising talent and compares them with existing talent. Specific embodiments of this system are described below.

[0142] First, the user inputs a prompt through the system interface. The prompt is text that instructs the generative AI to generate a new talent. For example, the user might input a prompt such as, "I would like you to generate a youthful female talent suitable for our new summer campaign."

[0143] The server receives a prompt sentence entered by the user. The received prompt sentence is then sent to a generative AI model. For example, OpenAI's GPT-4 (registered trademark) or DALL-E is used as this generative AI model.

[0144] The generative AI model analyzes the prompt and generates a new talent image and profile based on it. The generated image is a visual representation of the advertising talent, and the profile includes information such as name, age, and career history.

[0145] The server then compares the generated new talent information with existing talent based on the duration of the advertising campaign and market trends, including the popularity and market response of the existing talent.

[0146] Finally, the server provides the comparison results to the user, who can then select a new advertising talent based on the results, for example, by hiring the new talent if it is determined that the new talent is more suitable for the campaign than the existing talent.

[0147] As a specific example, if a user inputs a prompt such as "I want you to generate a youthful female talent suitable for our new summer campaign," the server sends this prompt to a generative AI model, and the model generates an image of a youthful female talent and a profile such as "Name: Hanako Sato, Age: 25, Experience: 5 years of modeling experience." The server compares the information of this new talent with existing talents and evaluates which is more suitable for the summer campaign. The evaluation results are provided to the user, who then decides whether to hire the new talent.

[0148] This system makes it possible to quickly and efficiently select the most suitable talent depending on the duration of the advertising campaign and market trends.

[0149] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0150] Step 1:

[0151] The user enters a prompt statement.

[0152] The user inputs a prompt for generating a new advertising talent through the system interface, for example, a specific request such as "Please generate a youthful female talent suitable for the new summer campaign."

[0153] Input: User prompt text

[0154] Output: The prompt statement is sent to the system.

[0155] Step 2:

[0156] The server receives the prompt.

[0157] The server receives the prompt text entered by the user and prepares for the next process.

[0158] Input: The prompt text sent by the user

[0159] Output: The prompt statement is saved on the server.

[0160] Step 3:

[0161] The server sends a prompt to the generative AI model.

[0162] The server sends the received prompt to a generative AI model (e.g., OpenAI's GPT-4 or DALL-E), which generates an image and profile of the new talent based on the prompt.

[0163] Input: Prompt text stored on the server

[0164] Output: The prompt is sent to the generative AI model.

[0165] Step 4:

[0166] A generative AI model generates images and profiles for new talent.

[0167] The generative AI model analyzes the prompt and generates an image and profile of a new talent based on it, such as an image of a youthful female talent, along with the talent's name, age, and background information.

[0168] Input: The prompt sent to the generative AI model

[0169] Output: New talent image and profile

[0170] Step 5:

[0171] The server compares the generated talent with existing talents.

[0172] The server compares the generated new talent information with existing advertising talent based on the duration of the advertising campaign and market trends, for example, taking into account the popularity and market response of the existing talent.

[0173] Input: New talent image and profile, existing talent information

[0174] Output: Comparison results

[0175] Step 6:

[0176] The server provides the comparison results to the user.

[0177] The server provides the comparison results to the user, who can then select a new advertising talent based on the results. For example, if the generated talent is determined to be more suitable for the campaign than the existing talent, the user can hire the talent.

[0178] Input: Comparison result

[0179] Output: The comparison results are provided to the user

[0180] (Application example 2)

[0181] Next, a description will be given of Application Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0182] In conventional advertising campaigns, there are limited means to evaluate the effectiveness of existing advertising talent, making it difficult to generate and compare new talent. Furthermore, it is not possible to select the optimal talent based on market trends and the duration of the advertising campaign, making it difficult to maximize advertising effectiveness. Therefore, in order to improve the effectiveness of advertising campaigns, a system is needed that uses generative AI models to generate new talent and compares them with existing talent to select the optimal talent.

[0183] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0184] In this invention, the server includes a means for utilizing generative AI, a means for comparing with existing advertising talent at regular intervals, and a means for searching for talent that can better recognize the service. This makes it possible to generate new talent based on the duration of the advertising campaign and market trends, compare the generated talent with existing talent, evaluate which is more effective, and select the most suitable talent to reflect in the advertising campaign.

[0185] "Generative AI" is a system that uses artificial intelligence technology to generate new data and information.

[0186] A "certain period" is a unit of time that can be adjusted depending on the duration of the advertising campaign and market trends.

[0187] An "advertising talent" is a person or character who appears in an advertisement and is selected to enhance the effectiveness of the advertisement.

[0188] A "comparison measure" is a method for evaluating existing advertising talent and newly generated talent to determine which is more effective.

[0189] A "talent capable of raising awareness of a service" is a talent who has the ability to effectively raise awareness of a specific service or product among consumers through advertising.

[0190] An "advertising campaign" is an advertising activity carried out over a specific period of time with the purpose of promoting a product or increasing brand awareness.

[0191] "Market trends" refer to changes and trends in the market that affect advertising strategies, such as consumer preferences, purchasing behavior, and competitor actions.

[0192] "Generated talent" refers to advertising talent newly created using generative AI.

[0193] "Means of evaluation" refers to a method for comparing the generated talent with existing talent and measuring its effectiveness.

[0194] The "optimal talent" is the talent who can most effectively raise awareness of a service or product in an advertising campaign.

[0195] The system for implementing this invention utilizes generative AI to periodically compare existing advertising talent to find talent who can better recognize the service, and select the talent who is best suited to the advertising campaign. A specific embodiment of this system is described below.

[0196] The server generates new advertising talent using a generative AI. The generative AI model used is OpenAI's (registered trademark) GPT-3 (registered trademark). The server generates prompts based on the duration of the advertising campaign and market trends, and inputs these prompts into the generative AI model to generate new talent.

[0197] For example, if the advertising campaign period is "October 1, 2023 to December 31, 2023" and the market trend is "Fall / Winter fashion trends," the following prompt sentences will be generated:

[0198] Advertising campaign period: October 1, 2023 to December 31, 2023, Market trends: Generate new advertising talent based on autumn / winter fashion trends.

[0199] The generated talent is compared to existing advertising talent by the server, using a generative AI model to generate prompts such as the following and evaluate them:

[0200] Compare Existing Talent: Existing Popular Talent A with New Talent: Newly Generated Talent B and evaluate which is more effective.

[0201] The server evaluates the effectiveness of the generated talent and existing talent, selects the most suitable talent, and determines which talent to include in the advertising campaign based on the evaluation results.

[0202] This system generates new talent according to the duration of the advertising campaign and market trends, and compares them with existing talent to select the most suitable talent, thereby maximizing advertising effectiveness.In addition, by using a generative AI model, it is possible to quickly and efficiently generate and evaluate new talent.

[0203] In terms of specific hardware, the server uses a computer system equipped with a high-performance processor and sufficient memory. In terms of software, it uses an API (e.g., OpenAI (registered trademark) API) for running the generative AI model. Data processing and calculation are performed by the generative AI model, and algorithms are implemented to analyze and compare the characteristics and performance of the generated talent.

[0204] In this way, it is possible to select the most suitable talent for an advertising campaign and maximize the effectiveness of the advertisement.

[0205] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0206] Step 1:

[0207] The server receives data about the duration of the advertising campaign and market trends as input. This data includes the campaign start and end dates, as well as market trend information. Based on this data, the server generates prompts to be fed into the generative AI model. Specifically, the server generates prompts like the following:

[0208] Advertising campaign period: October 1, 2023 to December 31, 2023, Market trends: Generate new advertising talent based on autumn / winter fashion trends.

[0209] Inputs: duration of advertising campaign, market trends

[0210] Output: A prompt to be input to the generative AI model

[0211] Step 2:

[0212] The server sends the generated prompt to a generative AI model (e.g., OpenAI's GPT-3) to generate a new advertising talent. The generative AI model generates the characteristics and profile of the new talent based on the prompt. The generated talent information is returned to the server.

[0213] Input: prompt statement

[0214] Output: Information about the new advertising talent generated

[0215] Step 3:

[0216] The server retrieves information about existing advertising talent from a database, including the talent's name, characteristics, past performance data, etc.

[0217] Input: Query to get information about existing advertising talent

[0218] Output: Existing advertising talent information

[0219] Step 4:

[0220] The server generates a prompt to compare the new talent with the existing talents, such as:

[0221] Compare Existing Talent: Existing Popular Talent A with New Talent: Newly Generated Talent B and evaluate which is more effective.

[0222] Input: Information about the new talent created, information about the existing talent

[0223] Output: Prompt for comparison

[0224] Step 5:

[0225] The server sends the generated prompt sentence for comparison to the generative AI model for evaluation. The generative AI model compares the effects of the new talent with the existing talent based on the prompt sentence and returns the evaluation result to the server.

[0226] Input: Prompt sentence for comparison

[0227] Output: Evaluation results

[0228] Step 6:

[0229] The server selects the most suitable talent based on the evaluation results, and the information on the selected talent is reflected in the advertising campaign plan.

[0230] Input: Evaluation result

[0231] Output: Best talent information

[0232] Step 7:

[0233] The server transmits information about the selected optimal talent to the advertising campaign management system and reflects it in the advertising campaign, thereby enabling the optimal talent to be used in the advertising campaign.

[0234] Input: Optimal talent information

[0235] Output: Talent information reflected in advertising campaign

[0236] Example 3

[0237] Next, a description will be given of a third embodiment of the third embodiment. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0238] Traditional methods for selecting and evaluating advertising talent make it difficult to accurately measure talent effectiveness and select the most suitable talent. Furthermore, there was a lack of means to analyze the effectiveness of advertising campaigns in real time and respond quickly. This resulted in advertising effectiveness not being maximized and advertising costs being wasted.

[0239] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0240] In this invention, the server includes means for a user to input a prompt text, means for the server to receive the prompt text and input it into a generative AI model, means for the server to generate a talent using the generative AI model, means for the server to create images and videos of the generated talent, means for the server to create advertising materials and air advertisements, means for the server to collect advertisement response data, and means for the server to analyze the collected data and evaluate the advertising effectiveness. This makes it possible to quickly generate optimal talent using the generative AI model and analyze and evaluate the effectiveness of advertising campaigns in real time.

[0241] "Generative AI" is an artificial intelligence technology that generates new talents based on prompts entered by users.

[0242] A "prompt sentence" is text that the user inputs into the generative AI, describing the characteristics of the talent they want to generate.

[0243] The "server" is a computer system that uses generative AI to generate talent, create advertising materials, and collect and analyze advertising response data.

[0244] "Talent" refers to a person or character used in advertisements generated by generative AI.

[0245] "Advertising materials" are advertising content created using images and videos of the generated talent.

[0246] "Response data" refers to data such as click-through rates, viewing times, and conversion rates collected after an advertisement has aired.

[0247] "Advertising effectiveness" is an indicator that shows the impact and reaction of an advertisement to viewers, and is evaluated based on response data.

[0248] This invention relates to a system for generating advertising talent using generative AI and evaluating the effectiveness of the talent. Specific embodiments of this system are described below.

[0249] First, the user enters a prompt statement describing the characteristics of the celebrity they wish to create through the system interface. For example, they can enter a prompt statement such as, "Please create a celebrity who is a woman in her twenties with a bright and energetic image."

[0250] Next, the server receives the prompt sentence sent by the user and inputs it into a generative AI model. For example, OpenAI's GPT-4 (registered trademark) can be used as this generative AI model. The server uses the generative AI model to generate a talent based on the prompt sentence. This talent is generated in the form of an image or video.

[0251] The generated images and videos of the talent are processed by the server and converted into a format that can be used as advertising material. For example, image editing software or video editing software can be used. This allows the generated talent's background to be made transparent or a video tailored to a specific scene to be created.

[0252] The server then uses the generated talent to create advertising materials, which are then broadcast for a set period of time on websites, social media, television, etc. For example, a cosmetics advertisement featuring the generated talent can be broadcast for one month.

[0253] While the advertisement is being aired, the server uses tools such as web page access analysis services to collect data such as click-through rates, viewing times, and conversion rates. This data is then stored on the server as advertisement response data.

[0254] Finally, the server analyzes the collected response data and evaluates the advertising effectiveness of the generated talent. This evaluation result is reflected in the next talent generation and advertising strategy. For example, if the click-through rate is high, it can be determined that the talent's characteristics were effective.

[0255] As a concrete example, if a user inputs the prompt "Please generate a male celebrity in his 30s who looks like a trustworthy businessman," the server will use the generative AI model to generate a trustworthy businessman-like celebrity. An advertisement for business software using this celebrity will be aired for two weeks, and the data collected during that time will be analyzed. Based on the analysis results, the advertising effectiveness of the celebrity can be evaluated.

[0256] In this way, a system for generating and evaluating advertising talent using generative AI can be implemented. The flow of the identification process in the third embodiment will be described with reference to FIG.

[0257] Step 1:

[0258] The user enters a prompt statement.

[0259] The user inputs a prompt statement describing the characteristics of the talent they wish to create through the system interface. For example, they might input, "Please generate a talent who is a woman in her twenties with a bright and energetic image." The input prompt statement is then sent to the server.

[0260] Step 2:

[0261] The server receives the prompt sentence and inputs it into the generative AI model.

[0262] The server receives the prompt sent by the user and inputs it into a generative AI model (e.g., OpenAI's GPT-4®). Based on the input prompt, the generative AI model analyzes the talent's characteristics and prepares the data required for generation.

[0263] Step 3:

[0264] The server generates talents using a generative AI model.

[0265] The server uses a generative AI model to generate a talent based on the prompt. The generative AI model analyzes the content of the prompt and generates appropriate images and videos. The data of the generated talent is stored on the server.

[0266] Step 4:

[0267] The server creates images and videos of the generated talent.

[0268] The server processes the generated images and videos of the talent and formats them into a format that can be used as advertising materials. For example, image editing software is used to make the background transparent. Also, video editing software is used to create videos tailored to specific scenes. The processed images and videos are stored on the server.

[0269] Step 5:

[0270] The server creates advertising materials and broadcasts the advertisements.

[0271] The server uses the generated talent to create advertising materials. For example, it creates a cosmetics advertisement featuring the generated talent. This advertising material is broadcast for a certain period of time on websites, social media, television, etc. The broadcast schedule and media are managed by the server.

[0272] Step 6:

[0273] The server collects advertisement response data.

[0274] While the advertisement is being aired, the server uses tools such as web page access analysis services to collect data such as click-through rates, viewing times, and conversion rates. The collected response data is stored on the server.

[0275] Step 7:

[0276] The server analyzes the collected data and evaluates the effectiveness of the advertisements.

[0277] The server analyzes the collected response data and evaluates the advertising effectiveness of the generated talent. For example, it quantifies the talent's effectiveness based on click rates and conversion rates. The analysis results are reflected in the next talent generation and advertising strategy.

[0278] (Application example 3)

[0279] Next, a description will be given of Application Example 3 of Form Example 3. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0280] Traditional methods for selecting advertising talent and measuring effectiveness have been problematic in that talent selection is time-consuming and costly, and it is difficult to accurately evaluate the effectiveness of advertising. These issues are particularly significant obstacles to maximizing advertising effectiveness in today's market, where real-time analysis of advertising campaign effectiveness and rapid response are required.

[0281] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[0282] In this invention, the server includes means for utilizing generative AI, means for comparing with existing advertising talent at regular intervals, means for finding talent who can better recognize the service, means for utilizing the found talent, means for creating advertisements using the generated talent, means for broadcasting the advertisements for a certain period of time, means for collecting and analyzing responses to the advertisements during the broadcast period, and means for evaluating effectiveness and generating reports. This makes it possible to streamline the process from selecting advertising talent to measuring effectiveness and evaluate the effectiveness of advertisements in real time.

[0283] "Generative AI" is a system that uses artificial intelligence technology to generate new data and information.

[0284] "Advertising talent" refers to people or characters who appear in advertisements.

[0285] "Fixed period" refers to a time frame that can be adjusted depending on the duration of the advertising campaign and market trends.

[0286] "Means for comparison" refers to a method or system for comparing generated talent with existing talent based on evaluation criteria.

[0287] A "recognizable talent" refers to a talent who has the ability to raise awareness of a service or product through advertising.

[0288] "Search means" refers to methods and systems that use generative AI and other technologies to find suitable talent.

[0289] "Means of utilization" refers to the methods and systems for utilizing the talent identified in advertising and other promotional activities.

[0290] "Means for creating advertisements" refers to a method or system for creating advertising content using generated talent.

[0291] "Means for broadcasting advertisements" refers to the method or system for releasing the created advertisements in the media for a certain period of time.

[0292] "Means for collecting and analyzing responses" refers to methods and systems for collecting and analyzing viewer responses and data obtained during the advertisement's broadcast period.

[0293] "Means for evaluating effectiveness and generating reports" refers to methods or systems for evaluating the effectiveness of advertising based on collected and analyzed data and compiling the results into a report.

[0294] The following system configuration and processing procedure will be described as an embodiment of the present invention.

[0295] System Configuration

[0296] This system uses the following hardware and software:

[0297] Hardware: Servers, smartphones

[0298] Software: Generative AI models (e.g., OpenAI® API), advertising platform APIs, data analysis tools (e.g., Python®)

[0299] Processing steps

[0300] 1. Ways to use generative AI:

[0301] The server generates a new talent using a generative AI model. Specifically, it inputs a prompt sentence and uses the generated text as the talent.

[0302] Example: Prompt: "Generate a new advertising talent. He / she is in his / her 20s and has a relatable character."

[0303] 2. Comparison with existing advertising talent over a period of time:

[0304] The server compares the generated talent with existing advertising talent based on evaluation criteria, including audience response and advertising effectiveness.

[0305] 3. Ways to find talent that can better recognize your services:

[0306] The server uses generative AI and other technologies to find talent with the ability to increase awareness of services and products.

[0307] 4. How to utilize the talent you find:

[0308] The server uses the talent it finds for advertising and other promotional activities.

[0309] 5. How to create advertisements using generated talent:

[0310] The server uses the generated talent to create advertising content, combining advertising copy and visual materials to create attractive advertisements.

[0311] 6. Means of broadcasting advertisements for a certain period of time:

[0312] The server publishes the created advertisements on the media for a certain period of time, and manages the advertisement broadcast schedule using the advertising platform API.

[0313] 7. Methods for collecting and analyzing the response to advertisements during their broadcast period:

[0314] The server collects and analyzes viewer responses and data obtained during the advertisement broadcast period, and uses data analysis tools to evaluate viewer responses in real time.

[0315] 8. Means of evaluating effectiveness and generating reports:

[0316] The server evaluates the effectiveness of the advertisements based on the collected and analyzed data and compiles the results into a report. The evaluation results will be used to improve the advertising campaign and to select the next talent.

[0317] Specific examples

[0318] For example, if a generative AI model is used to input the prompt "Generate a new talent for advertising. He / she is in his / her 20s and has a friendly character," the model will generate a talent: "The new talent, Yamada Taro, is in his / her 20s and has a friendly character." An advertisement is created using this talent and aired for a certain period of time. Viewer responses are then collected and analyzed to evaluate the effectiveness of the advertisement, which can be used to improve the next advertising campaign.

[0319] In this way, the process from selecting advertising talent to measuring effectiveness can be made more efficient, and the effectiveness of advertising can be evaluated in real time.

[0320] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0321] Step 1:

[0322] The server generates a new talent using a generative AI model. Specifically, it receives a prompt sentence, "Please generate a new advertising talent. He / she is in his / her twenties and has a friendly character," as input and sends it to the generative AI model. The generative AI model generates the talent's text data based on this prompt sentence and outputs it to the server.

[0323] Step 2:

[0324] The server compares the generated talent with existing advertising talent. As input, it receives the text data of the generated talent and the data of the existing advertising talent. The server compares the two based on evaluation criteria (e.g., viewer response, advertising effectiveness) and determines which is more effective. As output, it obtains the comparison results.

[0325] Step 3:

[0326] The server searches for talents who can increase awareness of the service. As input, it receives the generated talent data and comparison results. Using generative AI and other technologies, the server selects talents who have the ability to increase awareness of the service or product. As output, it obtains data on the selected talents.

[0327] Step 4:

[0328] The server creates advertisements using the talent it finds. It receives data on the selected talent as input. It combines advertising copy and visual materials to create attractive advertising content. It obtains the created advertising data as output.

[0329] Step 5:

[0330] The server broadcasts the created advertisement for a certain period of time. As input, it receives advertisement data and broadcast schedule. Using the advertisement platform API, the server manages the advertisement broadcast schedule and publishes it on the media. As output, it obtains confirmation data of the start of broadcast.

[0331] Step 6:

[0332] The server collects and analyzes the response to the advertisement during the broadcast period. It receives the advertisement broadcast data as input. The server uses the advertising platform API to collect viewer response data and analyzes it using data analysis tools. It obtains the analysis result data as output.

[0333] Step 7:

[0334] The server evaluates the effectiveness and generates a report. It receives the analysis result data as input. The server evaluates the effectiveness of the advertisement based on the collected and analyzed data and compiles the results into a report. The server obtains an evaluation report as output.

[0335] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0336] "Example 1"

[0337] In one embodiment of the present invention, the generative AI includes an emotion engine that recognizes user emotions. This emotion engine analyzes emotions from user reaction data and uses the results to generate advertising talent. Specifically, it analyzes users' reactions when viewing advertisements (e.g., facial expressions, eye movements, comments on social media, etc.) to understand their emotions (happiness, surprise, interest, indifference, etc.). Then, based on this emotion data, it generates a new talent that users are likely to respond favorably to.

[0338] "Example 2"

[0339] In addition, the system analyzes users' emotional reactions to the performance of the generated advertising talent and uses the results to evaluate the advertising talent. Specifically, the emotion engine analyzes users' reactions when the generated talent appears in an advertisement, and if the reaction is good, the talent's rating is increased, and if the reaction is bad, the rating is decreased. This makes it possible to evaluate talent based on users' emotions.

[0340] "Example 3"

[0341] Furthermore, the emotion engine can adjust the parameters of emotion analysis according to the duration of the advertising campaign and market trends. For example, during certain periods, such as the Christmas season, users' emotional responses may differ from normal. In such cases, the emotion engine learns users' emotional tendencies during that period and reflects the results in the generation and evaluation of advertising talent. This enables flexible generation and evaluation of advertising talent in response to market trends.

[0342] The processing flow of each embodiment will be described below.

[0343] "Example 1"

[0344] Step 1: User watches an ad.

[0345] Step 2: The emotion engine collects user reaction data (e.g., facial expressions, eye movements, comments on social media, etc.).

[0346] Step 3: The emotion engine analyzes the user's emotions (happiness, surprise, interest, indifference, etc.) from the collected reaction data.

[0347] Step 4: Based on the analysis results, the generative AI generates new talents that users are likely to respond favorably to.

[0348] "Example 2"

[0349] Step 1: The generated talent appears in the advertisement.

[0350] Step 2: The emotion engine analyzes the user's reaction.

[0351] Step 3: If the response is positive, increase the talent's rating; if the response is negative, decrease the rating.

[0352] "Example 3"

[0353] Step 1: The sentiment engine understands the duration of the advertising campaign and market trends.

[0354] Step 2: During a specific period (e.g., Christmas season), the emotion engine learns the user's emotional tendencies during that period.

[0355] Step 3: Reflect the learning results in generating and evaluating advertising talent.

[0356] Example 1

[0357] Next, a description will be given of Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0358] Traditional methods for selecting advertising talent have made it difficult to find effective talent because they do not adequately reflect viewer reactions or advertising performance. It is also difficult to flexibly adjust talent according to the duration of the advertising campaign and market trends. This has led to the issue of not maximizing advertising effectiveness.

[0359] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0360] In this invention, the server includes means for utilizing generative AI, means for comparing with existing advertising talent at regular intervals, means for finding a talent who is more likely to recognize the service, means for utilizing the found talent, means for collecting performance data and viewer response data of the advertising talent, means for preprocessing the collected data, means for training an AI model using the preprocessed data, means for generating new talent using the trained AI model, means for displaying the generated talent and collecting user responses, and means for analyzing the collected user response data and retraining the model. This automates the advertising talent selection process and enables the generation and selection of effective talent that reflects viewer responses.

[0361] "Generative AI" is an artificial intelligence technology that generates new data and content based on existing data.

[0362] An "advertising talent" is a person or character who appears in an advertisement and whose role is to raise awareness of products and services among viewers.

[0363] "Performance data" refers to data used to measure the effectiveness of advertising, including click-through rates, viewing times, and conversion rates.

[0364] "Viewer response data" is data that records the viewer's reactions when watching an advertisement, and includes facial expressions, eye movements, comments on social media, etc.

[0365] "Preprocessing" refers to the process of converting collected data into a format suitable for analysis and learning, and includes data cleaning, normalization, and feature extraction.

[0366] "Training" is the process of teaching an AI model data so that the model can understand patterns and features in the data and make predictions or generate results from new data.

[0367] "Generation" is the process of creating new data or content based on the data that the AI ​​model has learned.

[0368] "User response" refers to changes in a user's behavior and emotions when viewing an advertisement, and is data used to evaluate the effectiveness of an advertisement.

[0369] "Retraining" is the process of retraining an AI model based on new data collected in order to improve the model's accuracy and performance.

[0370] This invention is a system for optimizing advertising talent using generative AI. Specific embodiments of this system are described below.

[0371] System configuration

[0372] This system consists of three main components: a server, a terminal, and a user.

[0373] server

[0374] The server serves as the central point for generating and evaluating advertising talent using generative AI models. The server uses the following hardware and software:

[0375] Hardware: A server machine equipped with a high-performance CPU, GPU, memory, and storage

[0376] Software: Machine learning libraries, data analysis libraries for data preprocessing, APIs for data collection

[0377] The server collects and pre-processes performance data of advertising talent and viewer response data, and uses the pre-processed data to train a generative AI model to generate new advertising talent.

[0378] Terminal

[0379] The terminal is responsible for displaying the generated new advertising talent to the user and collecting the user's reactions. The terminal uses the following hardware and software:

[0380] Hardware: A computer or smartphone with a camera, microphone, and eye-tracking device

[0381] Software: Eye tracking software, facial expression recognition software

[0382] The device records the user's reactions (facial expressions, eye movements, tone of voice, etc.) when viewing an advertisement and sends this to the server.

[0383] User

[0384] Users are responsible for watching advertisements and providing their reactions to the system, which then use the data to evaluate the advertising talent and retrain the generative AI model.

[0385] Specific examples

[0386] Here is a concrete example of how the system works:

[0387] 1. Data collection: The server collects performance data for existing advertising talents A and B. For example, data such as click-through rates, viewing times, and conversion rates is obtained using tools such as web page access analysis services. Viewer response data (facial expressions, eye movements, comments on social media, etc.) is also collected.

[0388] 2. Data Preprocessing: The server preprocesses the collected data, for example, using a data analysis library to impute missing values ​​and normalize the data.

[0389] 3. Training an AI model: The server uses the preprocessed data to train a generative AI model (e.g., GPT-4®).

[0390] 4. New talent generation: The server inputs a prompt into the trained AI model to generate a new advertising talent. For example, the prompt can be "Please generate a talent that will delight viewers."

[0391] 5. Talent Evaluation: The device displays the new talent to the user and collects their reactions, for example, by recording their facial expressions and eye movements using a webcam and eye-tracking software.

[0392] 6. Feedback collection and retraining: The server analyzes user response data collected from the device and evaluates the performance of the generative AI model. If necessary, it retrains the model.

[0393] Prompt Sentence Examples

[0394] Here are some examples of prompts:

[0395] Generate talent that your audience will enjoy.

[0396] In this way, a system is realized in which the server, terminals, and users work together to optimize advertising talent.

[0397] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0398] Step 1: Data collection

[0399] The server collects performance data of existing advertising talent and viewer response data. Specifically, it uses a web page access analysis service to obtain performance data such as click-through rates, viewing times, and conversion rates. It also collects response data such as viewers' facial expressions, eye movements, and comments on social media.

[0400] Input: Advertising talent performance data, viewer response data

[0401] Output: Collected dataset

[0402] Step 2: Data Preprocessing

[0403] The server preprocesses the collected data by using data analysis libraries to impute missing values, remove outliers, and normalize the data, as well as extracting features and converting the data into a format suitable for training AI models.

[0404] Input: Collected dataset

[0405] Output: Preprocessed dataset

[0406] Step 3: Training the AI ​​model

[0407] The server uses the preprocessed data to train a generative AI model, specifically using a machine learning library to train the AI ​​model using existing advertising talent performance data and viewer response data as input.

[0408] Input: Preprocessed dataset

[0409] Output: A trained AI model

[0410] Step 4: Generate new talent

[0411] The server generates new advertising talent using the trained AI model. Specifically, it inputs a prompt statement into the AI ​​model to generate the characteristics of the new talent. For example, the prompt statement can be "Please generate a talent that will delight viewers."

[0412] Input: Trained AI model, prompt

[0413] Output: New advertising talent generated

[0414] Step 5: Talent Assessment

[0415] The device displays the new talent to the user and collects their reactions, using a webcam and eye-tracking software to record their facial expressions and eye movements, as well as their tone of voice using a microphone.

[0416] Input: New advertising talent generated

[0417] Output: User response data

[0418] Step 6: Gather feedback and retrain

[0419] The server analyzes user response data collected from the device and evaluates the performance of the generative AI model. If necessary, it retrains the model by preprocessing the data again and retraining the AI ​​model using the preprocessed data.

[0420] Input: User response data

[0421] Output: Retrained AI model

[0422] In this way, a system is realized in which the server, terminals, and users work together to optimize advertising talent.

[0423] (Application example 1)

[0424] Next, a description will be given of Application Example 1 of Embodiment Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0425] Traditional methods for selecting advertising talent have had the problem of not being able to fully consider the emotions and reactions of viewers, making it difficult to maximize the effectiveness of advertising. Furthermore, there was a lack of a way to monitor the effectiveness of advertising campaigns in real time and quickly select the most suitable talent.

[0426] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0427] In this invention, the server includes means for utilizing generative AI, means for comparing with existing advertising talent at regular intervals, means for finding talent who can better recognize the service, means for utilizing the found talent, means for collecting viewer response data, means for analyzing viewer emotions using an emotion engine, means for generating new talent based on the analysis results, means for selecting the most suitable talent from among the generated talents, and means for monitoring the effectiveness of the advertising campaign in real time. This makes it possible to quickly select the most suitable advertising talent based on viewer emotions and reactions, and maximize the effectiveness of the advertising.

[0428] "Generative AI" is an artificial intelligence technology that generates new data and models based on existing data.

[0429] An "advertising talent" refers to a person or character who appears in an advertisement.

[0430] "Viewer response data" refers to data that indicates changes in the viewer's behavior and emotions when viewing an advertisement.

[0431] An "emotion engine" refers to software or hardware that analyzes emotions from viewers' facial expressions and behavior.

[0432] The "analysis results" are the analysis results of the viewer's emotional data obtained by the emotion engine.

[0433] "Advertising Campaign" means an advertising campaign conducted over a specific period of time.

[0434] "Monitoring" refers to the real-time monitoring of the effectiveness and progress of an advertising campaign.

[0435] "Talent selection" is the process of selecting the most suitable talent from among the multiple talents generated.

[0436] The system for carrying out this invention utilizes generative AI to select and optimize advertising talent. A specific embodiment of this system will be described below.

[0437] System configuration

[0438] The system consists of the following main components:

[0439] 1. Generative AI model: Learns performance data and viewer response data of existing advertising talent and generates new talent.

[0440] 2. Emotion engine: Collects viewer reaction data and analyzes viewer emotions.

[0441] 3. Data collection module: Collects viewer response data from advertising videos.

[0442] 4. Talent selection module: Select the most suitable talent from the generated talents.

[0443] 5. Monitoring module: Monitor the effectiveness of your advertising campaigns in real time.

[0444] Hardware and software used

[0445] Hardware: Smartphones, cameras

[0446] Software: Libraries with functions such as image processing, image analysis, and machine learning

[0447] Data processing and calculation

[0448] 1. Data collection: Using a device (smartphone or camera), the faces of viewers watching the advertising video are detected, and emotion data is collected using an emotion recognition model.

[0449] 2. Talent generation: The server generates new advertising talent using a generative AI model based on the collected emotional data.

[0450] 3. Talent selection: The server selects the most suitable talent from the multiple talents generated based on the emotion data.

[0451] 4. Monitoring: The server monitors the effectiveness of the advertising campaign in real time and reevaluates talent selection as needed.

[0452] Specific examples

[0453] For example, to collect viewers' reactions to a video advertisement for a new smartphone, data such as whether viewers smile, express surprise, post positive comments on social media, etc. Based on this data, the following prompt sentences can be input into the generative AI model:

[0454] Generate a new advertising talent based on the following data: [Happy emotion intensity: 0.8, Surprise emotion intensity: 0.7, Interest emotion intensity: 0.9]

[0455] Using this prompt, new advertising talent can be generated based on the viewer's emotional data, and the most suitable talent can be selected.

[0456] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0457] Step 1:

[0458] The device uses a camera to detect the faces of viewers watching advertising videos and collects video data in real time. The input is the viewer's video data, and the output is the face detection results. Specifically, face detection is performed and the detected face area is cut out.

[0459] Step 2:

[0460] The device inputs the face detection results into an emotion recognition model and analyzes the viewer's emotional data. The input is the face detection results, and the output is the emotional data. Specifically, the emotion recognition model is executed to estimate the viewer's emotions (happiness, surprise, interest, etc.) from their facial expressions.

[0461] Step 3:

[0462] The server inputs prompt sentences into a generative AI model based on the collected emotion data to generate new advertising talent. The input is emotion data, and the output is the generated advertising talent. Specifically, the server generates prompt sentences and inputs them into a GPT-3 (registered trademark) model to generate new talent.

[0463] Step 4:

[0464] The server selects the optimal talent from the generated advertising talents based on the emotional data. The input is the generated advertising talent and emotional data, and the output is the optimal advertising talent. Specifically, the server executes an algorithm that evaluates the performance of each talent and compares it with the emotional data to select the optimal talent.

[0465] Step 5:

[0466] The server runs an advertising campaign using the selected advertising talent and monitors its effectiveness in real time. The input is the advertising campaign implementation data, and the output is the monitoring results. Specifically, it collects the number of ad views, click rates, reactions on social media, etc., and evaluates the effectiveness of the advertising campaign.

[0467] Step 6:

[0468] The server reevaluates talent selection as necessary based on the monitoring results and optimizes the advertising campaign. The input is the monitoring results, and the output is the reevaluated advertising talent. Specifically, the server analyzes the monitoring results, and if the advertising effectiveness is declining, it executes a process to generate new talent.

[0469] Example 2

[0470] Next, a description will be given of Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0471] Conventional advertising talent selection systems have the problem that talent selection relies on subjective judgment and does not adequately reflect the user's emotional reactions. In addition, there is a lack of a means to objectively evaluate talent performance and select the most suitable talent to maximize the effectiveness of advertising campaigns.

[0472] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for utilizing generative AI, a means for comparing with existing advertising talent at regular intervals, a means for finding a talent who is more likely to recognize the service, a means for utilizing the found talent, a means for collecting users' emotional responses, a means for analyzing the collected emotional responses, and a means for evaluating the talent based on the analysis results. This enables objective talent evaluation based on users' emotional responses and selection of the optimal talent that maximizes the effectiveness of the advertising campaign.

[0473] "Generative AI" is a system that uses artificial intelligence technology to generate new advertising talent.

[0474] A "certain period" is a unit of time that can be adjusted depending on the duration of the advertising campaign and market trends.

[0475] "Advertising talent" refers to people or characters who appear in advertisements.

[0476] "User emotional response" refers to the emotional response a user shows when viewing an advertisement, and is expressed through facial expressions, tone of voice, etc.

[0477] An "emotion engine" is software or a system for analyzing a user's emotional response and quantifying their emotional state.

[0478] "Talent evaluation" is the process of numerically evaluating the effectiveness of advertising talent based on users' emotional responses and performance data.

[0479] A "prompt" is an instruction entered into the generative AI to specify the characteristics and conditions of the advertising talent to be generated.

[0480] The present invention relates to a system for generating, comparing, and evaluating advertising talent. Specific embodiments of this system are described below.

[0481] The server generates a new advertising talent using a generative AI model. For example, a general natural language processing model is used as this generative AI model. A prompt sentence is input to the generative AI model, which generates talent characteristics suitable for the advertising campaign. An example of a specific prompt sentence is, "Please generate a new talent suitable for the summer season for an advertising campaign for a new soft drink."

[0482] The server compares the generated talent with existing talent, using data on the talent's past performance and market response, such as click-through rates and viewing times from past advertising campaigns.

[0483] The device then collects the user's emotional responses while watching the advertisements by using a camera and microphone to record the user's facial expressions and tone of voice, thereby obtaining real-time emotional data.

[0484] The collected emotion data is analyzed by an emotion engine on the server. For example, a general emotion analysis API is used as the emotion engine. As a result of the analysis, the user's emotional state (joy, surprise, sadness, etc.) is quantified.

[0485] Finally, the server evaluates the talent based on the analysis results. Specifically, if the user's emotional response is positive, the talent's evaluation score is increased, and if the response is negative, the evaluation score is decreased. This evaluation score is used in the next talent selection.

[0486] For example, when running an advertising campaign for a new soft drink, the server uses a generative AI model to generate a new talent and compares it with existing talent. The emotion engine analyzes the emotional response of users when they watch the advertisement and evaluates the talent based on the results. In this way, it becomes possible to select the optimal talent based on the user's emotions.

[0487] This system enables optimal talent selection to maximize the effectiveness of advertising campaigns.

[0488] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0489] Step 1: Generating advertising talent

[0490] The server generates a new advertising talent using a generative AI model. As input, it provides the generative AI model with a prompt statement: "Generate a new talent that suits the summer season for a new soft drink advertising campaign." The generative AI model generates the talent's characteristics (e.g., youthful appearance, lively personality) based on this prompt, and obtains the new talent's data as output.

[0491] Step 2: Compare advertising talent

[0492] The server compares the generated new talent with existing talent. It uses the new talent data and the existing talent database as input. The server evaluates which talent is more effective based on performance data such as click-through rate and viewing time from past advertising campaigns, and obtains the comparison result as output.

[0493] Step 3: Collecting users' emotional responses

[0494] The device collects the user's emotional responses while watching the advertisement. As input, it uses a camera and a microphone to record the user's facial expressions and tone of voice. The device collects this data in real time and obtains the user's emotional data as output.

[0495] Step 4: Analyze emotional responses

[0496] The server analyzes the collected user emotion data using an emotion engine. The emotion engine receives inputs such as the user's facial expressions and tone of voice. The emotion engine analyzes these data, quantifies the user's emotional state (happiness, surprise, sadness, etc.), and outputs the analysis results.

[0497] Step 5: Talent Evaluation

[0498] The server evaluates the talent based on the analysis results. As input, it uses the analysis results obtained from the emotion engine and the talent's performance data. If the user's emotional response is good, the server increases the talent's evaluation score, and if the response is bad, it decreases the evaluation score. As output, it obtains the updated talent's evaluation score.

[0499] In this way, a system is realized in which the server, terminals, and users work together to generate, compare, and evaluate advertising talent.

[0500] (Application example 2)

[0501] Next, a description will be given of Application Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0502] Conventional methods for selecting advertising talent have the problem of not maximizing advertising effectiveness because they select talent without considering users' emotional responses. Furthermore, there is a lack of a means for generating new talent during an advertising campaign and comparing them with existing talent, making it difficult to select the optimal talent. Furthermore, there is no means for analyzing users' emotional responses in real time and evaluating talent based on the results, making it difficult to accurately evaluate advertising effectiveness.

[0503] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0504] In this invention, the server includes a means for utilizing generative AI, a means for periodically comparing with existing advertising talent, a means for finding talent who can better recognize the service, a means for utilizing the found talent, a means for analyzing the user's emotional response, and a means for evaluating the talent based on the analysis results. This makes it possible to select the most suitable advertising talent based on the user's emotional response and maximize the advertising effectiveness.

[0505] "Generative AI" is a technology that uses artificial intelligence to generate new advertising talent.

[0506] A "certain period" is a unit of time that can be adjusted depending on the duration of the advertising campaign and market trends.

[0507] "Existing advertising talent" refers to people or characters who appear in advertisements currently in use.

[0508] A "service-recognizing talent" is a person or character who has the ability to effectively raise awareness of a specific service or product through advertising.

[0509] The "user's emotional response" refers to the emotional response shown by the user after viewing the advertisement, and includes emotions such as joy, surprise, and sadness.

[0510] The "means for evaluating a talent based on the analysis results" is a technology that analyzes the emotional reactions of users and evaluates the effectiveness of advertising talent based on the results.

[0511] As an embodiment of the present invention, the following system is constructed: The server includes means for utilizing generative AI, means for comparing with existing advertising talent at regular intervals, means for finding talent who can better recognize the service, means for utilizing the found talent, means for analyzing the user's emotional response, and means for evaluating the talent based on the analysis results.

[0512] Hardware and software used

[0513] Hardware: Smartphone camera

[0514] Software: Libraries with functions such as image processing, image analysis, and machine learning

[0515] Data processing and calculation

[0516] The server captures the user's face using the smartphone camera and pre-processes the facial image. The pre-processed image is then input into an emotion analysis model to analyze the user's emotional response. The analysis results are fed back to a generative AI model and used to evaluate advertising talent.

[0517] Specific examples

[0518] For example, to generate a promotional talent for a new product, the following prompt sentences could be used:

[0519] "Generate advertising talent for new products."

[0520] "Generate advertising talent suitable for your summer campaign."

[0521] "Generate advertising talent for the Christmas season."

[0522] The generated advertising talent is compared with existing talent at regular intervals and evaluated based on the user's emotional response, allowing the system to select the optimal advertising talent and maximize advertising effectiveness.

[0523] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0524] Step 1:

[0525] The server captures the user's face using the smartphone's camera. The input is real-time video captured by the camera, and the output is the captured image of the face. Specifically, the camera detects the user's face and captures the image as a frame.

[0526] Step 2:

[0527] The server preprocesses the captured face image. The input is the face image acquired in step 1, and the output is the preprocessed face image. Specifically, the image is converted to grayscale, and the face area is cropped and resized.

[0528] Step 3:

[0529] The server inputs the preprocessed facial image into the emotion analysis model to analyze the user's emotional response. The input is the facial image preprocessed in step 2, and the output is the emotion analysis result. Specifically, the emotion analysis model analyzes the image and classifies emotions such as joy, surprise, and sadness.

[0530] Step 4:

[0531] The server generates a new advertising talent using a generative AI model. The input is a prompt statement (e.g., "Please generate an advertising talent for a new product."), and the output is information about the generated advertising talent. Specifically, the prompt statement is input into the generative AI model, and the model generates the characteristics of the new talent.

[0532] Step 5:

[0533] The server compares the generated advertising talent with existing advertising talent at regular intervals. The input is information about the generated advertising talent and information about the existing advertising talent, and the output is the comparison result. Specifically, it compares the talent's characteristics and performance and evaluates which is more effective.

[0534] Step 6:

[0535] The server evaluates advertising talent based on the user's emotional response. The input is the emotion analysis result obtained in step 3 and the comparison result obtained in step 5, and the output is the selection of the optimal advertising talent. Specifically, the emotion analysis result is used as the evaluation criterion to select the most effective talent.

[0536] Step 7:

[0537] The server utilizes the selected talent in the advertising campaign. The input is the information about the talent selected in step 6, and the output is the results of the advertising campaign. Specifically, the selected talent is used in the advertisement and the campaign is implemented.

[0538] Example 3

[0539] Next, a description will be given of a third embodiment of the third embodiment. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0540] Traditional methods for selecting and evaluating advertising talent lack the flexibility to maximize advertising effectiveness, making it difficult to quickly respond to market trends and changes in user sentiment. Furthermore, the process of regularly evaluating the performance of existing talent and generating new talent is inefficient, making it difficult to maximize the effectiveness of advertising campaigns.

[0541] The identification process by the identification processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means. In this invention, the server includes means for utilizing generative artificial intelligence, means for comparing with existing advertising talent at regular intervals, means for finding talent who is more likely to recognize the service, means for utilizing the found talent, means for creating advertisements using the generated talent, means for distributing advertisements, means for measuring the effectiveness of advertisements, and means for adjusting emotion analysis parameters according to the duration of the advertising campaign and market trends using an emotion analysis engine. This makes it possible to quickly respond to market trends and changes in user emotions and maximize the effectiveness of the advertising campaign.

[0542] "Generative AI" is AI that has the ability to generate new information and content based on data.

[0543] "Advertising talent" refers to people or characters who appear in advertisements.

[0544] A "certain period" refers to a specific time frame that is set depending on the duration of an advertising campaign or market trends.

[0545] "Comparative means" refers to methods or techniques for evaluating and comparing the performance and characteristics of existing advertising talent with newly generated talent.

[0546] "Means of discovery" refers to methods and technologies for using generative artificial intelligence to find more effective advertising talent.

[0547] "Means of utilization" refers to the methods and techniques used to incorporate the identified advertising talent into actual advertising campaigns and maximize their effectiveness.

[0548] "Means for creating advertisements" refers to the methods and techniques for producing advertising content using generated talent.

[0549] "Means of delivering advertising" refers to the methods and technologies used to deliver created advertisements to target audiences.

[0550] "Means of measuring advertising effectiveness" refers to methods and technologies for collecting and analyzing data such as click-through rates, viewing times, and engagement rates to evaluate the success of advertising campaigns.

[0551] "Emotion analysis engine" refers to technology that analyzes users' emotional responses and adjusts the parameters of advertising campaigns based on that data.

[0552] This invention is a system that utilizes generative artificial intelligence to generate advertising talent and maximize the effectiveness of advertising campaigns using the talent. Specific embodiments of this system are described below.

[0553] System Overview

[0554] This system works in collaboration with three parties: a server, a terminal, and a user. The server uses generative artificial intelligence and an emotion analysis engine to generate advertising talent, create and distribute advertisements, and measure their effectiveness. The terminal is a device through which users view advertisements and is responsible for collecting users' emotional responses.

[0555] Hardware and software used

[0556] Generative artificial intelligence: For example, using generative AI models such as OpenAI's GPT-4.

[0557] Sentiment analysis engine: Software for analyzing users' emotional responses.

[0558] Video editing software: Create your ads.

[0559] Social media platforms: To distribute advertisements.

[0560] Data processing and calculation

[0561] 1. Creating a prompt for talent generation

[0562] The server creates a prompt to be input to the generative AI model, which specifically describes the characteristics and attributes of the talent to be generated.

[0563] Example: "Generate a female talent in her twenties. She has a bright, energetic, and friendly personality."

[0564] 2. Talent Generation

[0565] The server inputs prompt text into the generative AI model to generate new talents.

[0566] Example: The server sends prompts to a generative AI model and receives data on the generated talent.

[0567] 3. Ad Creation

[0568] The server uses the generated talent to create advertisements, which can be in the form of video ads or banner ads.

[0569] Example: A server uses video editing software to create a video advertisement that includes the generated talent.

[0570] 4. Advertisement Delivery

[0571] The server distributes the created advertisements, which can be distributed to social media sites, websites, etc.

[0572] Example: A server uses a social media platform to deliver video ads.

[0573] 5. Measuring effectiveness

[0574] The server measures the effectiveness of the ads, specifically collecting data such as click-through rates, viewing times, and engagement rates.

[0575] Example: The server uses the social media platform's analytics tools to collect information such as click-through rates and viewing times for ads, and stores this information in a database.

[0576] 6. Tuning your emotional engine

[0577] The server uses the emotion engine to adjust the parameters of the emotion analysis according to the duration of the advertising campaign and market trends.

[0578] Example: The server collects user emotion data during the Christmas season and adjusts the parameters of the emotion engine.

[0579] Specific examples

[0580] Example prompts for generative AI models

[0581] "Generate a male talent in his 30s suitable for the Christmas season. He has a warm, family-oriented character."

[0582] "For Valentine's Day, please create a female talent in her 20s who will inspire romantic feelings. She has a romantic and charming character."

[0583] In this way, the server is a system that generates and evaluates advertising talent by utilizing a generative AI model and a sentiment analysis engine. This enables rapid response to market trends and changes in user sentiment, maximizing the effectiveness of advertising campaigns. The flow of the identification process in Example 3 will be explained using Figure 21.

[0584] Step 1: Create a prompt statement

[0585] The server creates a prompt to be input to the generative AI model, which specifically describes the characteristics and attributes of the talent to be generated.

[0586] Input: Information about the objectives and target audience of your ad campaign.

[0587] Data processing: Generate appropriate prompts for the generative AI model based on the objectives of the advertising campaign and information about the target audience.

[0588] Output: The prompt sentence to be input to the generative AI model.

[0589] Specific operation: The server generates the prompt sentence, "Please generate a female talent in her twenties. She has a bright, energetic, and friendly character."

[0590] Step 2: Talent Generation

[0591] The server inputs prompt text into the generative AI model to generate new talents.

[0592] Input: The prompt statement created in step 1.

[0593] Data Calculation: The generative AI model analyzes the prompt and generates a talent with the specified characteristics and attributes.

[0594] Output: Generated talent data.

[0595] Specific operation: The server sends a prompt to the generative AI model and receives data on the generated talent.

[0596] Step 3: Create your ad

[0597] The server uses the generated talent to create advertisements, which can be in the form of video ads or banner ads.

[0598] Input: Talent data generated in step 2.

[0599] Data processing: Creating advertising content based on the generated talent data.

[0600] Output: The finished ad content.

[0601] What happens: The server uses video editing software to create a video advertisement that includes the generated talent.

[0602] Step 4: Serving Ads

[0603] The server distributes the created advertisements, which can be distributed to social media sites, websites, etc.

[0604] Input: The ad content created in step 3.

[0605] Data processing: Converting advertising content into a format suitable for the distribution platform.

[0606] Output: The ad that was served.

[0607] Specific operation: The server uses a social networking platform to deliver video ads.

[0608] Step 5: Measuring advertising effectiveness

[0609] The server measures the effectiveness of the ads, specifically collecting data such as click-through rates, viewing times, and engagement rates.

[0610] Input: Performance data of the ads served.

[0611] Data calculations: Analyzing collected performance data and evaluating the effectiveness of advertising.

[0612] Output: A report on the effectiveness of your ads.

[0613] What it does: The server uses the social media platform's analytics tools to collect ad click-through rates and viewing times, and stores them in a database.

[0614] Step 6: Tuning your Emotional Engine

[0615] The server uses the emotion engine to adjust the parameters of the emotion analysis according to the duration of the advertising campaign and market trends.

[0616] Input: User sentiment data and market trend data.

[0617] Data calculation: Analyze sentiment data and market trend data and adjust the parameters of the sentiment engine.

[0618] Output: Tuned emotion engine parameters.

[0619] Specific operation: The server collects user emotion data during the Christmas season and adjusts the parameters of the emotion engine.

[0620] Step 7: Flexible talent generation and assessment

[0621] The server generates and evaluates advertising talent flexibly according to market trends.

[0622] Input: Market trend data and advertising effectiveness data.

[0623] Data calculation: Analyze market trend data and advertising effectiveness data to generate and evaluate new talent.

[0624] Output: The newly generated talent and its evaluation results.

[0625] Specific operation: The server generates a talent that stimulates romantic feelings in preparation for Valentine's Day, and distributes the advertisement again to evaluate its effectiveness and measure its effectiveness.

[0626] (Application example 3)

[0627] Next, a description will be given of Application Example 3 of Form Example 3. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0628] Conventional methods for selecting and evaluating advertising talent make it difficult to fully reflect users' emotional reactions and market trends, making it difficult to maximize advertising effectiveness. Furthermore, flexible responses to specific periods and market fluctuations are required, but achieving this has been difficult.

[0629] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[0630] In this invention, the server includes means for utilizing generative AI, means for comparing with existing advertising talent at regular intervals, means for finding talent who can better recognize the service, means for utilizing the found talent, means for analyzing users' emotional responses using an emotion engine, means for adjusting emotion analysis parameters depending on the duration of the advertising campaign and market trends, and means for evaluating the effectiveness of advertising using the generated talent. This enables flexible generation and evaluation of advertising talent that reflects users' emotional responses and market trends.

[0631] "Generative AI" is a system that uses artificial intelligence technology to generate new data and content.

[0632] An "advertising talent" is a person or character used in an advertisement, whose role is to raise awareness of a product or service among viewers and consumers.

[0633] An "emotion engine" is software or hardware that analyzes a user's emotional response and adjusts the system's behavior based on the results.

[0634] "User emotional response" refers to a user's emotional reaction or feedback to an advertisement or content, including emotions such as joy, sadness, and anger.

[0635] An "advertising campaign" is an advertising activity carried out over a specific period of time, which is carried out in a planned manner to increase awareness of a product or service.

[0636] "Market trends" refer to changes in consumer behavior, trends, competitor actions, etc. in a particular market.

[0637] "Emotion analysis parameters" are settings and reference values ​​used when analyzing a user's emotional response, and affect the accuracy and results of the analysis.

[0638] The "effectiveness of advertising" refers to the impact and response that an advertisement has on viewers and consumers, and refers to concrete results such as increased product awareness and sales.

[0639] The system for implementing this invention is composed of a server including a generative AI, an emotion engine, and an advertising effectiveness evaluation module, and a terminal that collects user emotional responses. Specific embodiments of the system are described below.

[0640] The server generates advertising talent using generative AI. The generative AI analyzes the characteristics and performance of advertising talent and generates new talent based on that. The generated talent is compared with existing advertising talent at regular intervals, and the talent that can best raise awareness of the service is selected.

[0641] The emotion engine is used to analyze users' emotional responses. Users' emotional responses are collected via devices such as smartphones and tablets. The emotion engine adjusts the emotion analysis parameters according to the duration of the advertising campaign and market trends. For example, the "joy" parameter is set high during the Christmas season.

[0642] The advertisement using the generated talent is aired for a certain period of time, and its effectiveness is evaluated by an advertisement effectiveness evaluation module, which analyzes the response to the advertisement and evaluates the effectiveness of the generated talent.

[0643] For example, to generate a talent that emphasizes the emotion of "joy" during the Christmas season, enter the following prompt into the generation AI:

[0644] Generate a talent with the following emotional parameters: {Happy: 0.8, Sad: 0.1, Anger: 0.1}

[0645] By inputting this prompt into a generative AI, it is possible to generate talents that are perfect for the Christmas season.

[0646] The hardware used includes servers, smartphones, tablets, and other devices, and the software is Python (registered trademark), OpenAI (registered trademark) API, emotion analysis software, and an advertising effectiveness evaluation module. The server runs the generative AI and emotion engine, while the device collects users' emotional responses.

[0647] In this way, it becomes possible to flexibly generate and evaluate advertising talent that reflects users' emotional responses and market trends.

[0648] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0649] Step 1:

[0650] The server generates advertising talent using generative AI. Specifically, the server analyzes the characteristics and performance of advertising talent and creates prompts to generate new talent based on that analysis. For example, to generate a talent that emphasizes the emotion of "joy" during the Christmas season, the server generates the following prompt:

[0651] Generate a talent with the following emotional parameters: {Happy: 0.8, Sad: 0.1, Anger: 0.1}

[0652] Input: Advertising talent characteristics data, emotion parameters

[0653] Output: Generated talent prompt

[0654] Step 2:

[0655] The server inputs a prompt into the generative AI to generate a new advertising talent, which then generates an image and profile of the talent based on the prompt.

[0656] Input: prompt statement

[0657] Output: Generated advertising talent images and profiles

[0658] Step 3:

[0659] The server compares the generated advertising talent with existing advertising talent at regular intervals, specifically, the server compares performance data of the existing advertising talent with performance data of the generated talent, and selects a more effective talent.

[0660] Input: Existing advertising talent performance data, Generated talent performance data

[0661] Output: Selected advertising talent

[0662] Step 4:

[0663] The device collects the user's emotional response. Specifically, it collects emotional response data (e.g., facial expressions and voice) when the user views an advertisement via a smartphone or tablet.

[0664] Input: User viewing data

[0665] Output: User's emotional response data

[0666] Step 5:

[0667] The server analyzes the user's emotional response using an emotion engine that analyzes the collected emotional response data to identify the user's emotional state.

[0668] Input: User emotional response data

[0669] Output: Parsed user's emotional state

[0670] Step 6:

[0671] The server adjusts the emotion analysis parameters according to the duration of the advertising campaign and market trends, for example, by setting the "happiness" parameter higher during a specific period.

[0672] Input: Advertising campaign duration, market trend data

[0673] Output: Adjusted sentiment analysis parameters

[0674] Step 7:

[0675] The server broadcasts the advertisement using the generated talent for a certain period of time and evaluates its effectiveness. The advertisement effectiveness evaluation module collects response data for the advertisement and evaluates the effectiveness of the generated talent.

[0676] Input: Generated talent, advertising response data

[0677] Output: Advertising effectiveness evaluation results

[0678] In this way, it becomes possible to flexibly generate and evaluate advertising talent that reflects users' emotional responses and market trends.

[0679] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0680] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0681] Another example of generative AI is Gemini (registered trademark) (Internet search engine). <url: https: gemini.google.com ?hl="ja">) are mentioned.

[0682] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0683] [Second embodiment]

[0684] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0685] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0686] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0687] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0688] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0689] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0690] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0691] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0692] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0693] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0694] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0695] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0696] "Example 1"

[0697] One aspect of the present invention is a system that utilizes generative AI. This system analyzes the characteristics and performance of advertising talent and generates new talent based on that. Specifically, the AI ​​learns performance data and viewer response data of existing advertising talent, and generates new talent based on that. The generated talent is compared with existing advertising talent to find talent that can better promote awareness of the service.

[0698] "Example 2"

[0699] The present invention also provides a system for comparing a new talent with existing advertising talent at regular intervals. This regular interval can be adjusted according to the duration of an advertising campaign or market trends. For example, a new talent can be generated to coincide with the launch of a new product or a specific season, and compared with existing talent.

[0700] "Example 3"

[0701] Furthermore, a system for utilizing the discovered talent is also an embodiment of the present invention. Specifically, the generated talent is used in an advertisement and its effectiveness is verified. For example, an advertisement using the generated talent is aired for a certain period of time, and the effectiveness of the talent can be evaluated by analyzing the response.

[0702] The processing flow of each embodiment will be described below.

[0703] "Example 1"

[0704] Step 1: The generative AI collects performance data of existing advertising talent and viewer response data.

[0705] Step 2: Based on the collected data, the AI ​​learns and generates new talents.

[0706] Step 3: The generated talents are compared with existing advertising talents to find talents that can better raise awareness of the service.

[0707] "Example 2"

[0708] Step 1: Set a fixed period, taking into account the duration of the advertising campaign and market trends.

[0709] Step 2: At each set time period, generate new talents and compare them with existing talents.

[0710] "Example 3"

[0711] Step 1: Hire the generated talent for your ads.

[0712] Step 2: Air the advertisement using the talent you selected for a certain period of time.

[0713] Step 3: Analyze the response to the aired advertisement and evaluate the effectiveness of the talent.

[0714] Example 1

[0715] Next, a description will be given of Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0716] Conventional methods for selecting advertising talent have difficulty finding effective talent because they do not fully utilize viewer response and performance data. It is also difficult to flexibly adjust talent according to the duration of the advertising campaign and market trends. Therefore, a system that can efficiently generate and evaluate new talent to maximize advertising effectiveness is needed.

[0717] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0718] In this invention, the server includes a means for utilizing generative AI, a means for collecting performance data and viewer reaction data of existing advertising talent, a means for preprocessing the collected data, a means for inputting the preprocessed data into an AI model for learning, a means for generating new talent using the learned AI model, a means for evaluating the performance of the generated talent, and a means for outputting the evaluation results. This makes it possible to analyze the characteristics and performance of advertising talent in detail and efficiently generate and evaluate more effective talent.

[0719] "Generative AI" is an artificial intelligence technology that generates new data and models based on existing data.

[0720] "Performance data" refers to data that shows the effectiveness of an advertising talent, such as the number of views, click rate, and purchase rate.

[0721] "Viewer response data" refers to data that records viewers' reactions to advertisements, such as comments, ratings, and number of shares.

[0722] "Preprocessing" refers to the process of converting collected data into a format suitable for an AI model by performing processes such as filling in missing values, removing outliers, and normalizing the data.

[0723] An "AI model" is an algorithm built using a machine learning framework that learns patterns from data and makes predictions or generates results.

[0724] A "talent" is a person or character who appears in an advertisement.

[0725] "Evaluation" refers to the process of measuring the performance of the generated talent using indicators such as number of views, click rate, and purchase rate, and comparing it with existing talent.

[0726] "Output" means providing the evaluation results in the form of a report or data.

[0727] This invention relates to a system that uses generative AI to analyze the characteristics and performance of advertising talent and generate new talent. Specific embodiments of this system are described below.

[0728] The server first collects performance data of existing advertising talent and viewer response data, using APIs to retrieve data from company databases and social media platforms, including the number of ad views, click-through rates, purchase rates, viewer comments, ratings, and number of shares.

[0729] The server then preprocesses the collected data, which includes using a data analysis library to impute missing values, remove outliers, and normalize the data. For example, data with extremely low views or abnormally high click-through rates is removed, and the data is normalized using min-max scaling.

[0730] The pre-processed data is then fed into an AI model. The server uses a deep neural network (DNN) model built using a machine learning framework. The model learns the characteristics of advertising talent and audience reactions, and finds patterns to generate new talent.

[0731] Once the learning is complete, the server generates a new talent. The generated talent is expected to have similar characteristics to the existing talent, but with higher performance. For example, a talent with specific attributes such as "female in her 20s, cheerful personality, sports lover" may be generated.

[0732] The performance of the generated talent is evaluated by the server. The evaluation uses indicators such as the number of views, click rate, and purchase rate, and compares it with existing talent. If the evaluation results show that the generated talent performs better than the existing talent, that talent is selected.

[0733] Finally, the server outputs the evaluation results, which are output as an Excel report and emailed to the user, who can then use the report to plan a new advertising campaign.

[0734] As a concrete example, consider a company looking for a new advertising talent for a beverage. The server collects performance data (number of views, click-through rate, purchase rate, etc.) and viewer response data (comments, ratings, number of shares, etc.) of previously used beverage advertising talent, inputs this data into an AI model, and generates new talent. If the generated talent performs better than existing talent, it will be used in the new advertising campaign.

[0735] Example prompt sentence:

[0736] "Generate new beverage advertising talent based on performance data and viewer response data of past beverage advertising talent. The generated talent must exceed existing talent in terms of views, click-through rates, or purchase rates."

[0737] In this way, the server utilizes generative AI to analyze the characteristics and performance of advertising talent, creating a system that generates new talent.

[0738] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0739] Step 1: Data collection

[0740] The server collects existing advertising talent performance data and viewer response data from the company's database and social media platforms. Specifically, it uses APIs to obtain data such as the number of views, click-through rates, conversion rates, comments, ratings, and shares. The input is the raw data obtained from the API, and the output is the collected raw data.

[0741] Step 2: Data Preprocessing

[0742] The server preprocesses the collected data. Specifically, it uses a data analysis library to fill in missing values, remove outliers, and normalize the data. For example, it removes data with extremely low view counts and abnormally high click-through rates, and normalizes the data using min-max scaling. The input is the collected raw data, and the output is the preprocessed, clean data.

[0743] Step 3: Training the AI ​​model

[0744] The server inputs the preprocessed data into the AI ​​model and begins learning. The AI ​​model used is a DNN (deep neural network) model built using a machine learning framework. The model learns the characteristics of advertising talent and viewer reactions, and finds patterns to generate new talent. The input is the preprocessed data, and the output is the trained AI model.

[0745] Step 4: Generate new talent

[0746] The server uses the trained AI model to generate new talent. The generated talent is expected to have similar characteristics to existing talents, but with higher performance. For example, a talent with specific attributes such as "female in her 20s, cheerful personality, sports lover" may be generated. The input is the trained AI model, and the output is the characteristics of the newly generated talent.

[0747] Step 5: Performance evaluation

[0748] The server evaluates the performance of the generated talent. Specifically, it compares it with existing talent using metrics such as the number of views, click-through rate, and purchase rate. If the evaluation results show that the generated talent performs better than the existing talent, that talent is selected. The input is the characteristics of the generated talent, and the output is the evaluation result.

[0749] Step 6: Output the results

[0750] The server outputs the evaluation results. Specifically, it outputs the evaluation results as an Excel report and sends it to the user by email. The user then plans a new advertising campaign based on this report. The input is the evaluation results, and the output is the report provided to the user.

[0751] (Application example 1)

[0752] Next, a description will be given of Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0753] Traditional methods for selecting advertising talent make it difficult to fully utilize existing talent performance data and viewer response data, making it difficult to maximize the effectiveness of advertising campaigns. Furthermore, selecting advertising talent takes time and money, making it difficult to respond to market fluctuations that require rapid response. This has led to the issue of a lower success rate for advertising campaigns.

[0754] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0755] In this invention, the server includes a means for utilizing generative AI, a means for comparing with existing advertising talent at regular intervals, and a means for searching for talent who can better recognize the service. This optimizes the advertising campaign, and a means for using an application installed on a smartphone. This automates the advertising talent selection process, quickly and effectively generates new talent, and increases the success rate of advertising campaigns.

[0756] "Generative AI" is an artificial intelligence technology that learns from existing data and generates new data and models.

[0757] A "certain period" is a unit of time that can be adjusted depending on the duration of the advertising campaign and market trends.

[0758] An "advertising talent" is a person or character who appears in an advertisement.

[0759] "Performance Data" means data relating to the activities and achievements of advertising talent.

[0760] "Viewer response data" refers to data regarding the reactions and feedback of viewers who have viewed advertisements.

[0761] An "application installed on a smartphone" is a software program that runs on a smartphone.

[0762] "Advertising campaign optimization measures" are methods and processes for maximizing the effectiveness of advertising.

[0763] The following system configuration will be described as an embodiment of the present invention.

[0764] The server includes a means for utilizing generative AI, a means for comparing with existing advertising talent at regular intervals, a means for finding talent who can better recognize the service, a means for optimizing advertising campaigns, and a means for using applications installed on smartphones.

[0765] System Program

[0766] The server builds a generative AI model. Specifically, it reads performance data and viewer response data of existing advertising talent and builds a neural network model using LSTM (Long Short-Term Memory) layers and Dense layers. The model is compiled using the Adam optimizer and categorical cross-entropy loss function. New talent is generated using the trained model.

[0767] Processing Description

[0768] The server does the following:

[0769] 1. Data Loading: The server loads existing advertising talent performance data and viewer response data, which may include retrieving data from databases and cloud storage.

[0770] 2. Building a model: The server builds a neural network model using LSTM layers and Dense layers, which enables processing of time series data.

[0771] 3. Compile the model: The server compiles the model using the Adam optimizer and categorical cross-entropy loss function, which allows for efficient model training.

[0772] 4. Model training: The server uses the loaded data to train a model, which can then analyze the characteristics and performance of advertising talent.

[0773] 5. New talent generation: The server generates new talent using the trained model, which allows for faster discovery of more effective advertising talent.

[0774] Specific examples

[0775] For example, when an advertising agency launches an advertising campaign for a new beverage, it can input the performance data of existing talent and viewer response data to generate the most suitable new talent, thereby increasing the success rate of the advertising campaign.

[0776] Prompt Sentence Examples

[0777] "Generate new talent using performance data and viewer response data from existing advertising talent."

[0778] In this way, the server can automate the advertising talent selection process, quickly and effectively generating new talent and increasing the success of advertising campaigns.

[0779] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0780] Step 1:

[0781] The server loads existing advertising talent performance data and viewer response data from a database or cloud storage. The input data is the advertising talent's activity history and viewer feedback data, and uses these for subsequent processing. The output is the loaded dataset.

[0782] Step 2:

[0783] The server preprocesses the loaded data, specifically imputing missing values, normalizing data, encoding categorical data, etc. The input is the dataset loaded in step 1, and the output is the preprocessed dataset.

[0784] Step 3:

[0785] The server uses the preprocessed data to build a neural network model with an LSTM layer and a Dense layer. The input is the preprocessed dataset, and the output is the built neural network model.

[0786] Step 4:

[0787] The server compiles the constructed model, specifically using the Adam optimizer and categorical cross-entropy loss function. The input is the model constructed in step 3, and the output is the compiled model.

[0788] Step 5:

[0789] The server trains the dataset with the compiled model: the input is the preprocessed dataset and the compiled model, and the output is the trained model.

[0790] Step 6:

[0791] The server generates new advertising talent using the trained model. Specifically, it uses performance data of existing talent as input to predict the characteristics of the new talent. The input is the performance data of the existing talent, and the output is the characteristics of the generated new talent.

[0792] Step 7:

[0793] The server compares the characteristics of the generated new talent with the existing talents and selects the more effective talent. The inputs are the characteristics of the generated new talent and the characteristics of the existing talents, and the output is the selected optimal talent.

[0794] Step 8:

[0795] The server generates data for utilizing the selected optimal talent in an advertising campaign and sends it to an application installed on a smartphone. The input is the data of the selected optimal talent, and the output is the data to be utilized in the advertising campaign.

[0796] Example 2

[0797] Next, a description will be given of Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0798] Traditional methods for selecting advertising talent make it difficult to quickly find the optimal talent based on market trends and the duration of an advertising campaign. Furthermore, creating new talent and comparing them with existing talent is done manually, which is inefficient, time-consuming, and costly.

[0799] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0800] In this invention, the server includes a means for a user to input a prompt sentence, a means for the server to receive the prompt sentence and send it to the generative AI model, a means for the generative AI model to generate an image and profile of a new talent, a means for the server to compare the generated talent with existing talent, and a means for the server to provide the comparison result to the user, thereby enabling the selection of the optimal talent quickly and efficiently according to the duration of the advertising campaign and market trends.

[0801] "Generative AI" refers to artificial intelligence that generates images and profiles of new advertising talent based on prompts entered by the user.

[0802] A "prompt sentence" is the text that the user enters to instruct the generative AI to generate a new talent.

[0803] The "server" is a computer system that receives prompt text from the user, sends it to the generation AI, manages and compares the information of the generated talent, and provides the results to the user.

[0804] "Advertising personality" means a person or character used to promote a product or service in an advertising campaign.

[0805] "Image" refers to the visual representation of the advertising talent generated by generative AI.

[0806] A "profile" refers to information such as the advertising talent's name, age, and career history that is generated by generative AI.

[0807] "Comparison" refers to the process of evaluating the new talent generated against existing talent to determine which is more suitable for the advertising campaign.

[0808] "User" means a person or organization that uses the system to generate new advertising talent and receive comparison results.

[0809] A "certain period of time" is a time that can be adjusted depending on the duration of the advertising campaign and market trends.

[0810] This invention relates to a system that uses generative AI to generate advertising talent and compares them with existing talent. Specific embodiments of this system are described below.

[0811] First, the user inputs a prompt through the system interface. The prompt is text that instructs the generative AI to generate a new talent. For example, the user might input a prompt such as, "I would like you to generate a youthful female talent suitable for our new summer campaign."

[0812] The server receives a prompt sentence entered by the user. The received prompt sentence is then sent to a generative AI model. For example, OpenAI's GPT-4 (registered trademark) or DALL-E is used as this generative AI model.

[0813] The generative AI model analyzes the prompt and generates a new talent image and profile based on it. The generated image is a visual representation of the advertising talent, and the profile includes information such as name, age, and career history.

[0814] The server then compares the generated new talent information with existing talent based on the duration of the advertising campaign and market trends, including the popularity and market response of the existing talent.

[0815] Finally, the server provides the comparison results to the user, who can then select a new advertising talent based on the results, for example, by hiring the new talent if it is determined that the new talent is more suitable for the campaign than the existing talent.

[0816] As a specific example, if a user inputs a prompt such as "I want you to generate a youthful female talent suitable for our new summer campaign," the server sends this prompt to a generative AI model, and the model generates an image of a youthful female talent and a profile such as "Name: Hanako Sato, Age: 25, Experience: 5 years of modeling experience." The server compares the information of this new talent with existing talents and evaluates which is more suitable for the summer campaign. The evaluation results are provided to the user, who then decides whether to hire the new talent.

[0817] This system makes it possible to quickly and efficiently select the most suitable talent depending on the duration of the advertising campaign and market trends.

[0818] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0819] Step 1:

[0820] The user enters a prompt statement.

[0821] The user inputs a prompt for generating a new advertising talent through the system interface, for example, a specific request such as "Please generate a youthful female talent suitable for the new summer campaign."

[0822] Input: User prompt text

[0823] Output: The prompt statement is sent to the system.

[0824] Step 2:

[0825] The server receives the prompt.

[0826] The server receives the prompt text entered by the user and prepares for the next process.

[0827] Input: The prompt text sent by the user

[0828] Output: The prompt statement is saved on the server.

[0829] Step 3:

[0830] The server sends a prompt to the generative AI model.

[0831] The server sends the received prompt to the generative AI model, which then generates an image and profile of the new talent based on the prompt.

[0832] Input: Prompt text stored on the server

[0833] Output: The prompt is sent to the generative AI model.

[0834] Step 4:

[0835] A generative AI model generates images and profiles for new talent.

[0836] The generative AI model analyzes the prompt and generates an image and profile of a new talent based on it, such as an image of a youthful female talent, along with the talent's name, age, and background information.

[0837] Input: The prompt sent to the generative AI model

[0838] Output: New talent image and profile

[0839] Step 5:

[0840] The server compares the generated talent with existing talents.

[0841] The server compares the generated new talent information with existing advertising talent based on the duration of the advertising campaign and market trends, for example, taking into account the popularity and market response of the existing talent.

[0842] Input: New talent image and profile, existing talent information

[0843] Output: Comparison results

[0844] Step 6:

[0845] The server provides the comparison results to the user.

[0846] The server provides the comparison results to the user, who can then select a new advertising talent based on the results. For example, if the generated talent is determined to be more suitable for the campaign than the existing talent, the user can hire the talent.

[0847] Input: Comparison result

[0848] Output: The comparison results are provided to the user

[0849] (Application example 2)

[0850] Next, a description will be given of Application Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0851] In conventional advertising campaigns, there are limited means to evaluate the effectiveness of existing advertising talent, making it difficult to generate and compare new talent. Furthermore, it is not possible to select the optimal talent based on market trends and the duration of the advertising campaign, making it difficult to maximize advertising effectiveness. Therefore, in order to improve the effectiveness of advertising campaigns, a system is needed that uses generative AI models to generate new talent and compares them with existing talent to select the optimal talent.

[0852] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0853] In this invention, the server includes a means for utilizing generative AI, a means for comparing with existing advertising talent at regular intervals, and a means for searching for talent that can better recognize the service. This makes it possible to generate new talent based on the duration of the advertising campaign and market trends, compare the generated talent with existing talent, evaluate which is more effective, and select the most suitable talent to reflect in the advertising campaign.

[0854] "Generative AI" is a system that uses artificial intelligence technology to generate new data and information.

[0855] A "certain period" is a unit of time that can be adjusted depending on the duration of the advertising campaign and market trends.

[0856] An "advertising talent" is a person or character who appears in an advertisement and is selected to enhance the effectiveness of the advertisement.

[0857] A "comparison measure" is a method for evaluating existing advertising talent and newly generated talent to determine which is more effective.

[0858] A "talent capable of raising awareness of a service" is a talent who has the ability to effectively raise awareness of a specific service or product among consumers through advertising.

[0859] An "advertising campaign" is an advertising activity carried out over a specific period of time with the purpose of promoting a product or increasing brand awareness.

[0860] "Market trends" refer to changes and trends in the market that affect advertising strategies, such as consumer preferences, purchasing behavior, and competitor actions.

[0861] "Generated talent" refers to advertising talent newly created using generative AI.

[0862] "Means of evaluation" refers to a method for comparing the generated talent with existing talent and measuring its effectiveness.

[0863] The "optimal talent" is the talent who can most effectively raise awareness of a service or product in an advertising campaign.

[0864] The system for implementing this invention utilizes generative AI to periodically compare existing advertising talent to find talent who can better recognize the service, and select the talent who is best suited to the advertising campaign. A specific embodiment of this system is described below.

[0865] The server generates new advertising talent using generative AI. Generative AI models such as OpenAI's GPT-3 are used as generative AI. The server generates prompts based on the duration of the advertising campaign and market trends, and inputs these prompts into the generative AI model to generate new talent.

[0866] For example, if the advertising campaign period is "October 1, 2023 to December 31, 2023" and the market trend is "Fall / Winter fashion trends," the following prompt sentences will be generated:

[0867] Advertising campaign period: October 1, 2023 to December 31, 2023, Market trends: Generate new advertising talent based on autumn / winter fashion trends.

[0868] The generated talent is compared to existing advertising talent by the server, using a generative AI model to generate prompts such as the following and evaluate them:

[0869] Compare Existing Talent: Existing Popular Talent A with New Talent: Newly Generated Talent B and evaluate which is more effective.

[0870] The server evaluates the effectiveness of the generated talent and existing talent, selects the most suitable talent, and determines which talent to include in the advertising campaign based on the evaluation results.

[0871] This system generates new talent according to the duration of the advertising campaign and market trends, and compares them with existing talent to select the most suitable talent, thereby maximizing advertising effectiveness.In addition, by using a generative AI model, it is possible to quickly and efficiently generate and evaluate new talent.

[0872] The server uses a computer system equipped with a high-performance processor and ample memory for its hardware. The software uses an API for running the generative AI model. The generative AI model processes and calculates data, and algorithms are implemented to analyze and compare the characteristics and performance of the generated talent.

[0873] In this way, it is possible to select the most suitable talent for an advertising campaign and maximize the effectiveness of the advertisement.

[0874] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0875] Step 1:

[0876] The server receives data about the duration of the advertising campaign and market trends as input. This data includes the campaign start and end dates, as well as market trend information. Based on this data, the server generates prompts to be fed into the generative AI model. Specifically, the server generates prompts like the following:

[0877] Advertising campaign period: October 1, 2023 to December 31, 2023, Market trends: Generate new advertising talent based on autumn / winter fashion trends.

[0878] Inputs: duration of advertising campaign, market trends

[0879] Output: A prompt to be input to the generative AI model

[0880] Step 2:

[0881] The server sends the generated prompt text to the generative AI model to generate a new advertising talent. The generative AI model generates the characteristics and profile of the new talent based on the prompt text. The generated talent information is returned to the server.

[0882] Input: prompt statement

[0883] Output: Information about the new advertising talent generated

[0884] Step 3:

[0885] The server retrieves information about existing advertising talent from a database, including the talent's name, characteristics, past performance data, etc.

[0886] Input: Query to get information about existing advertising talent

[0887] Output: Existing advertising talent information

[0888] Step 4:

[0889] The server generates a prompt to compare the new talent with the existing talents, such as:

[0890] Compare Existing Talent: Existing Popular Talent A with New Talent: Newly Generated Talent B and evaluate which is more effective.

[0891] Input: Information about the new talent created, information about the existing talent

[0892] Output: Prompt for comparison

[0893] Step 5:

[0894] The server sends the generated prompt sentence for comparison to the generative AI model for evaluation. The generative AI model compares the effects of the new talent with the existing talent based on the prompt sentence and returns the evaluation result to the server.

[0895] Input: Prompt sentence for comparison

[0896] Output: Evaluation results

[0897] Step 6:

[0898] The server selects the most suitable talent based on the evaluation results, and the information on the selected talent is reflected in the advertising campaign plan.

[0899] Input: Evaluation result

[0900] Output: Best talent information

[0901] Step 7:

[0902] The server transmits information about the selected optimal talent to the advertising campaign management system and reflects it in the advertising campaign, thereby enabling the optimal talent to be used in the advertising campaign.

[0903] Input: Optimal talent information

[0904] Output: Talent information reflected in advertising campaign

[0905] Example 3

[0906] Next, a description will be given of Example 3 of Form Example 3. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0907] Traditional methods for selecting and evaluating advertising talent make it difficult to accurately measure talent effectiveness and select the most suitable talent. Furthermore, there was a lack of means to analyze the effectiveness of advertising campaigns in real time and respond quickly. This resulted in advertising effectiveness not being maximized and advertising costs being wasted.

[0908] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0909] In this invention, the server includes means for a user to input a prompt text, means for the server to receive the prompt text and input it into a generative AI model, means for the server to generate a talent using the generative AI model, means for the server to create images and videos of the generated talent, means for the server to create advertising materials and air advertisements, means for the server to collect advertisement response data, and means for the server to analyze the collected data and evaluate the advertising effectiveness. This makes it possible to quickly generate optimal talent using the generative AI model and analyze and evaluate the effectiveness of advertising campaigns in real time.

[0910] "Generative AI" is an artificial intelligence technology that generates new talents based on prompts entered by users.

[0911] A "prompt sentence" is text that the user inputs into the generative AI, describing the characteristics of the talent they want to generate.

[0912] The "server" is a computer system that uses generative AI to generate talent, create advertising materials, and collect and analyze advertising response data.

[0913] "Talent" refers to a person or character used in advertisements generated by generative AI.

[0914] "Advertising materials" are advertising content created using images and videos of the generated talent.

[0915] "Response data" refers to data such as click-through rates, viewing times, and conversion rates collected after an advertisement has aired.

[0916] "Advertising effectiveness" is an indicator that shows the impact and reaction of an advertisement to viewers, and is evaluated based on response data.

[0917] This invention relates to a system for generating advertising talent using generative AI and evaluating the effectiveness of the talent. Specific embodiments of this system are described below.

[0918] First, the user enters a prompt statement describing the characteristics of the celebrity they wish to create through the system interface. For example, they can enter a prompt statement such as, "Please create a celebrity who is a woman in her twenties with a bright and energetic image."

[0919] Next, the server receives the prompt sentence sent by the user and inputs it into a generative AI model. For example, OpenAI's GPT-4 (registered trademark) can be used as this generative AI model. The server uses the generative AI model to generate a talent based on the prompt sentence. This talent is generated in the form of an image or video.

[0920] The generated images and videos of the talent are processed by the server and converted into a format that can be used as advertising material. For example, image editing software or video editing software can be used. This allows the generated talent's background to be made transparent or a video tailored to a specific scene to be created.

[0921] The server then uses the generated talent to create advertising materials, which are then broadcast for a set period of time on websites, social media, television, etc. For example, a cosmetics advertisement featuring the generated talent can be broadcast for one month.

[0922] While the advertisement is being aired, the server uses tools such as web page access analysis services to collect data such as click-through rates, viewing times, and conversion rates. This data is then stored on the server as advertisement response data.

[0923] Finally, the server analyzes the collected response data and evaluates the advertising effectiveness of the generated talent. This evaluation result is reflected in the next talent generation and advertising strategy. For example, if the click-through rate is high, it can be determined that the talent's characteristics were effective.

[0924] As a concrete example, if a user inputs the prompt "Please generate a male celebrity in his 30s who looks like a trustworthy businessman," the server will use the generative AI model to generate a trustworthy businessman-like celebrity. An advertisement for business software using this celebrity will be aired for two weeks, and the data collected during that time will be analyzed. Based on the analysis results, the advertising effectiveness of the celebrity can be evaluated.

[0925] In this way, a system for generating and evaluating advertising talent using generative AI can be implemented. The flow of the identification process in the third embodiment will be described with reference to FIG.

[0926] Step 1:

[0927] The user enters a prompt statement.

[0928] The user inputs a prompt statement describing the characteristics of the talent they wish to create through the system interface. For example, they might input, "Please generate a talent who is a woman in her twenties with a bright and energetic image." The input prompt statement is then sent to the server.

[0929] Step 2:

[0930] The server receives the prompt sentence and inputs it into the generative AI model.

[0931] The server receives the prompt sent by the user and inputs it into the generative AI model. Based on the input prompt, the generative AI model analyzes the talent's characteristics and prepares the data necessary for generation.

[0932] Step 3:

[0933] The server generates talents using a generative AI model.

[0934] The server uses a generative AI model to generate a talent based on the prompt. The generative AI model analyzes the content of the prompt and generates appropriate images and videos. The data of the generated talent is stored on the server.

[0935] Step 4:

[0936] The server creates images and videos of the generated talent.

[0937] The server processes the generated images and videos of the talent and formats them into a format that can be used as advertising materials. For example, image editing software is used to make the background transparent. Also, video editing software is used to create videos tailored to specific scenes. The processed images and videos are stored on the server.

[0938] Step 5:

[0939] The server creates advertising materials and broadcasts the advertisements.

[0940] The server uses the generated talent to create advertising materials. For example, it creates a cosmetics advertisement featuring the generated talent. This advertising material is broadcast for a certain period of time on websites, social media, television, etc. The broadcast schedule and media are managed by the server.

[0941] Step 6:

[0942] The server collects advertisement response data.

[0943] While the advertisement is being aired, the server uses tools such as web page access analysis services to collect data such as click-through rates, viewing times, and conversion rates. The collected response data is stored on the server.

[0944] Step 7:

[0945] The server analyzes the collected data and evaluates the effectiveness of the advertisements.

[0946] The server analyzes the collected response data and evaluates the advertising effectiveness of the generated talent. For example, it quantifies the talent's effectiveness based on click rates and conversion rates. The analysis results are reflected in the next talent generation and advertising strategy.

[0947] (Application example 3)

[0948] Next, a description will be given of Application Example 3 of Form Example 3. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0949] Traditional methods for selecting advertising talent and measuring effectiveness have been problematic in that talent selection is time-consuming and costly, and it is difficult to accurately evaluate the effectiveness of advertising. These issues are particularly significant obstacles to maximizing advertising effectiveness in today's market, where real-time analysis of advertising campaign effectiveness and rapid response are required.

[0950] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[0951] In this invention, the server includes means for utilizing generative AI, means for comparing with existing advertising talent at regular intervals, means for finding talent who can better recognize the service, means for utilizing the found talent, means for creating advertisements using the generated talent, means for broadcasting the advertisements for a certain period of time, means for collecting and analyzing responses to the advertisements during the broadcast period, and means for evaluating effectiveness and generating reports. This makes it possible to streamline the process from selecting advertising talent to measuring effectiveness and evaluate the effectiveness of advertisements in real time.

[0952] "Generative AI" is a system that uses artificial intelligence technology to generate new data and information.

[0953] "Advertising talent" refers to people or characters who appear in advertisements.

[0954] "Fixed period" refers to a time frame that can be adjusted depending on the duration of the advertising campaign and market trends.

[0955] "Means for comparison" refers to a method or system for comparing generated talent with existing talent based on evaluation criteria.

[0956] A "recognizable talent" refers to a talent who has the ability to raise awareness of a service or product through advertising.

[0957] "Search means" refers to methods and systems that use generative AI and other technologies to find suitable talent.

[0958] "Means of utilization" refers to the methods and systems for utilizing the talent identified in advertising and other promotional activities.

[0959] "Means for creating advertisements" refers to a method or system for creating advertising content using generated talent.

[0960] "Means for broadcasting advertisements" refers to the method or system for releasing the created advertisements in the media for a certain period of time.

[0961] "Means for collecting and analyzing responses" refers to methods and systems for collecting and analyzing viewer responses and data obtained during the advertisement's broadcast period.

[0962] "Means for evaluating effectiveness and generating reports" refers to methods or systems for evaluating the effectiveness of advertising based on collected and analyzed data and compiling the results into a report.

[0963] The following system configuration and processing procedure will be described as an embodiment of the present invention.

[0964] System Configuration

[0965] This system uses the following hardware and software:

[0966] Hardware: Servers, smartphones

[0967] Software: Generative AI models (e.g., OpenAI® API), advertising platform APIs, data analysis tools (e.g., Python®)

[0968] Processing steps

[0969] 1. Ways to use generative AI:

[0970] The server generates a new talent using a generative AI model. Specifically, it inputs a prompt sentence and uses the generated text as the talent.

[0971] Example: Prompt: "Generate a new advertising talent. He / she is in his / her 20s and has a relatable character."

[0972] 2. Comparison with existing advertising talent over a period of time:

[0973] The server compares the generated talent with existing advertising talent based on evaluation criteria, including audience response and advertising effectiveness.

[0974] 3. Ways to find talent that can better recognize your services:

[0975] The server uses generative AI and other technologies to find talent with the ability to increase awareness of services and products.

[0976] 4. How to utilize the talent you find:

[0977] The server uses the talent it finds for advertising and other promotional activities.

[0978] 5. How to create advertisements using generated talent:

[0979] The server uses the generated talent to create advertising content, combining advertising copy and visual materials to create attractive advertisements.

[0980] 6. Means of broadcasting advertisements for a certain period of time:

[0981] The server publishes the created advertisements on the media for a certain period of time, and manages the advertisement broadcast schedule using the advertising platform API.

[0982] 7. Methods for collecting and analyzing the response to advertisements during their broadcast period:

[0983] The server collects and analyzes viewer responses and data obtained during the advertisement broadcast period, and uses data analysis tools to evaluate viewer responses in real time.

[0984] 8. Means of evaluating effectiveness and generating reports:

[0985] The server evaluates the effectiveness of the advertisements based on the collected and analyzed data and compiles the results into a report. The evaluation results will be used to improve the advertising campaign and to select the next talent.

[0986] Specific examples

[0987] For example, if a generative AI model is used to input the prompt "Generate a new talent for advertising. He / she is in his / her 20s and has a friendly character," the model will generate a talent: "The new talent, Yamada Taro, is in his / her 20s and has a friendly character." An advertisement is created using this talent and aired for a certain period of time. Viewer responses are then collected and analyzed to evaluate the effectiveness of the advertisement, which can be used to improve the next advertising campaign.

[0988] In this way, the process from selecting advertising talent to measuring effectiveness can be made more efficient, and the effectiveness of advertising can be evaluated in real time.

[0989] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0990] Step 1:

[0991] The server generates a new talent using a generative AI model. Specifically, it receives a prompt sentence, "Please generate a new advertising talent. He / she is in his / her twenties and has a friendly character," as input and sends it to the generative AI model. The generative AI model generates the talent's text data based on this prompt sentence and outputs it to the server.

[0992] Step 2:

[0993] The server compares the generated talent with existing advertising talent. As input, it receives the text data of the generated talent and the data of the existing advertising talent. The server compares the two based on evaluation criteria (e.g., viewer response, advertising effectiveness) and determines which is more effective. As output, it obtains the comparison results.

[0994] Step 3:

[0995] The server searches for talents who can increase awareness of the service. As input, it receives the generated talent data and comparison results. Using generative AI and other technologies, the server selects talents who have the ability to increase awareness of the service or product. As output, it obtains data on the selected talents.

[0996] Step 4:

[0997] The server creates advertisements using the talent it finds. It receives data on the selected talent as input. It combines advertising copy and visual materials to create attractive advertising content. It obtains the created advertising data as output.

[0998] Step 5:

[0999] The server broadcasts the created advertisement for a certain period of time. As input, it receives advertisement data and broadcast schedule. Using the advertisement platform API, the server manages the advertisement broadcast schedule and publishes it on the media. As output, it obtains confirmation data of the start of broadcast.

[1000] Step 6:

[1001] The server collects and analyzes the response to the advertisement during the broadcast period. It receives the advertisement broadcast data as input. The server uses the advertising platform API to collect viewer response data and analyzes it using data analysis tools. It obtains the analysis result data as output.

[1002] Step 7:

[1003] The server evaluates the effectiveness and generates a report. It receives the analysis result data as input. The server evaluates the effectiveness of the advertisement based on the collected and analyzed data and compiles the results into a report. The server obtains an evaluation report as output.

[1004] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1005] "Example 1"

[1006] In one embodiment of the present invention, the generative AI includes an emotion engine that recognizes user emotions. This emotion engine analyzes emotions from user reaction data and uses the results to generate advertising talent. Specifically, it analyzes users' reactions when viewing advertisements (e.g., facial expressions, eye movements, comments on social media, etc.) to understand their emotions (happiness, surprise, interest, indifference, etc.). Then, based on this emotion data, it generates a new talent that users are likely to respond favorably to.

[1007] "Example 2"

[1008] In addition, the system analyzes users' emotional reactions to the performance of the generated advertising talent and uses the results to evaluate the advertising talent. Specifically, the emotion engine analyzes users' reactions when the generated talent appears in an advertisement, and if the reaction is good, the talent's rating is increased, and if the reaction is bad, the rating is decreased. This makes it possible to evaluate talent based on users' emotions.

[1009] "Example 3"

[1010] Furthermore, the emotion engine can adjust the parameters of emotion analysis according to the duration of the advertising campaign and market trends. For example, during certain periods, such as the Christmas season, users' emotional responses may differ from normal. In such cases, the emotion engine learns users' emotional tendencies during that period and reflects the results in the generation and evaluation of advertising talent. This enables flexible generation and evaluation of advertising talent in response to market trends.

[1011] The processing flow of each embodiment will be described below.

[1012] "Example 1"

[1013] Step 1: User watches an ad.

[1014] Step 2: The emotion engine collects user reaction data (e.g., facial expressions, eye movements, comments on social media, etc.).

[1015] Step 3: The emotion engine analyzes the user's emotions (happiness, surprise, interest, indifference, etc.) from the collected reaction data.

[1016] Step 4: Based on the analysis results, the generative AI generates new talents that users are likely to respond favorably to.

[1017] "Example 2"

[1018] Step 1: The generated talent appears in the advertisement.

[1019] Step 2: The emotion engine analyzes the user's reaction.

[1020] Step 3: If the response is positive, increase the talent's rating; if the response is negative, decrease the rating.

[1021] "Example 3"

[1022] Step 1: The sentiment engine understands the duration of the advertising campaign and market trends.

[1023] Step 2: During a specific period (e.g., Christmas season), the emotion engine learns the user's emotional tendencies during that period.

[1024] Step 3: Reflect the learning results in generating and evaluating advertising talent.

[1025] Example 1

[1026] Next, a description will be given of Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1027] Traditional methods for selecting advertising talent have made it difficult to find effective talent because they do not adequately reflect viewer reactions or advertising performance. It is also difficult to flexibly adjust talent according to the duration of the advertising campaign and market trends. This has led to the issue of not maximizing advertising effectiveness.

[1028] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1029] In this invention, the server includes means for utilizing generative AI, means for comparing with existing advertising talent at regular intervals, means for finding a talent who is more likely to recognize the service, means for utilizing the found talent, means for collecting performance data and viewer response data of the advertising talent, means for preprocessing the collected data, means for training an AI model using the preprocessed data, means for generating new talent using the trained AI model, means for displaying the generated talent and collecting user responses, and means for analyzing the collected user response data and retraining the model. This automates the advertising talent selection process and enables the generation and selection of effective talent that reflects viewer responses.

[1030] "Generative AI" is an artificial intelligence technology that generates new data and content based on existing data.

[1031] An "advertising talent" is a person or character who appears in an advertisement and whose role is to raise awareness of products and services among viewers.

[1032] "Performance data" refers to data used to measure the effectiveness of advertising, including click-through rates, viewing times, and conversion rates.

[1033] "Viewer response data" is data that records the viewer's reactions when watching an advertisement, and includes facial expressions, eye movements, comments on social media, etc.

[1034] "Preprocessing" refers to the process of converting collected data into a format suitable for analysis and learning, and includes data cleaning, normalization, and feature extraction.

[1035] "Training" is the process of teaching an AI model data so that the model can understand patterns and features in the data and make predictions or generate results from new data.

[1036] "Generation" is the process of creating new data or content based on the data that the AI ​​model has learned.

[1037] "User response" refers to changes in a user's behavior and emotions when viewing an advertisement, and is data used to evaluate the effectiveness of an advertisement.

[1038] "Retraining" is the process of retraining an AI model based on new data collected in order to improve the model's accuracy and performance.

[1039] This invention is a system for optimizing advertising talent using generative AI. Specific embodiments of this system are described below.

[1040] System configuration

[1041] This system consists of three main components: a server, a terminal, and a user.

[1042] server

[1043] The server serves as the central point for generating and evaluating advertising talent using generative AI models. The server uses the following hardware and software:

[1044] Hardware: A server machine equipped with a high-performance CPU, GPU, memory, and storage

[1045] Software: Machine learning libraries, data analysis libraries for data preprocessing, APIs for data collection

[1046] The server collects and pre-processes performance data of advertising talent and viewer response data, and uses the pre-processed data to train a generative AI model to generate new advertising talent.

[1047] Terminal

[1048] The terminal is responsible for displaying the generated new advertising talent to the user and collecting the user's reactions. The terminal uses the following hardware and software:

[1049] Hardware: A computer or smartphone with a camera, microphone, and eye-tracking device

[1050] Software: Eye tracking software, facial expression recognition software

[1051] The device records the user's reactions (facial expressions, eye movements, tone of voice, etc.) when viewing an advertisement and sends this to the server.

[1052] User

[1053] Users are responsible for watching advertisements and providing their reactions to the system, which then use the data to evaluate the advertising talent and retrain the generative AI model.

[1054] Specific examples

[1055] Here is a concrete example of how the system works:

[1056] 1. Data collection: The server collects performance data for existing advertising talents A and B. For example, data such as click-through rates, viewing times, and conversion rates is obtained using tools such as web page access analysis services. Viewer response data (facial expressions, eye movements, comments on social media, etc.) is also collected.

[1057] 2. Data Preprocessing: The server preprocesses the collected data, for example, using a data analysis library to impute missing values ​​and normalize the data.

[1058] 3. Training the AI ​​model: The server uses the preprocessed data to train the generative AI model.

[1059] 4. New talent generation: The server inputs a prompt into the trained AI model to generate a new advertising talent. For example, the prompt can be "Please generate a talent that will delight viewers."

[1060] 5. Talent Evaluation: The device displays the new talent to the user and collects their reactions, for example, by recording their facial expressions and eye movements using a webcam and eye-tracking software.

[1061] 6. Feedback collection and retraining: The server analyzes user response data collected from the device and evaluates the performance of the generative AI model. If necessary, it retrains the model.

[1062] Prompt Sentence Examples

[1063] Here are some examples of prompts:

[1064] Generate talent that your audience will enjoy.

[1065] In this way, a system is realized in which the server, terminals, and users work together to optimize advertising talent.

[1066] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1067] Step 1: Data collection

[1068] The server collects performance data of existing advertising talent and viewer response data. Specifically, it uses tools such as web page access analysis services to obtain performance data such as click-through rates, viewing times, and conversion rates. It also collects response data such as viewer facial expressions, eye movements, and comments on social media.

[1069] Input: Advertising talent performance data, viewer response data

[1070] Output: Collected dataset

[1071] Step 2: Data Preprocessing

[1072] The server preprocesses the collected data by using data analysis libraries to impute missing values, remove outliers, and normalize the data, as well as extracting features and converting the data into a format suitable for training AI models.

[1073] Input: Collected dataset

[1074] Output: Preprocessed dataset

[1075] Step 3: Training the AI ​​model

[1076] The server uses the preprocessed data to train a generative AI model, specifically using a machine learning library to train the AI ​​model using existing advertising talent performance data and viewer response data as input.

[1077] Input: Preprocessed dataset

[1078] Output: A trained AI model

[1079] Step 4: Generate new talent

[1080] The server generates new advertising talent using the trained AI model. Specifically, it inputs a prompt statement into the AI ​​model to generate the characteristics of the new talent. For example, the prompt statement can be "Please generate a talent that will delight viewers."

[1081] Input: Trained AI model, prompt

[1082] Output: New advertising talent generated

[1083] Step 5: Talent Evaluation

[1084] The device displays the new talent to the user and collects their reactions, using a webcam and eye-tracking software to record their facial expressions and eye movements, as well as their tone of voice using a microphone.

[1085] Input: New advertising talent generated

[1086] Output: User response data

[1087] Step 6: Gather feedback and retrain

[1088] The server analyzes user response data collected from the device and evaluates the performance of the generative AI model. If necessary, it retrains the model by preprocessing the data again and retraining the AI ​​model using the preprocessed data.

[1089] Input: User response data

[1090] Output: Retrained AI model

[1091] In this way, a system is realized in which the server, terminals, and users work together to optimize advertising talent.

[1092] (Application example 1)

[1093] Next, a description will be given of Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1094] Traditional methods for selecting advertising talent have had the problem of not being able to fully consider the emotions and reactions of viewers, making it difficult to maximize the effectiveness of advertising. Furthermore, there was a lack of a way to monitor the effectiveness of advertising campaigns in real time and quickly select the most suitable talent.

[1095] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1096] In this invention, the server includes means for utilizing generative AI, means for comparing with existing advertising talent at regular intervals, means for finding talent who can better recognize the service, means for utilizing the found talent, means for collecting viewer response data, means for analyzing viewer emotions using an emotion engine, means for generating new talent based on the analysis results, means for selecting the most suitable talent from among the generated talents, and means for monitoring the effectiveness of the advertising campaign in real time. This makes it possible to quickly select the most suitable advertising talent based on viewer emotions and reactions, and maximize the effectiveness of the advertising.

[1097] "Generative AI" is an artificial intelligence technology that generates new data and models based on existing data.

[1098] An "advertising talent" refers to a person or character who appears in an advertisement.

[1099] "Viewer response data" refers to data that indicates changes in the viewer's behavior and emotions when viewing an advertisement.

[1100] An "emotion engine" refers to software or hardware that analyzes emotions from viewers' facial expressions and behavior.

[1101] The "analysis results" are the analysis results of the viewer's emotional data obtained by the emotion engine.

[1102] "Advertising Campaign" means an advertising campaign conducted over a specific period of time.

[1103] "Monitoring" refers to the real-time monitoring of the effectiveness and progress of an advertising campaign.

[1104] "Talent selection" is the process of selecting the most suitable talent from among the multiple talents generated.

[1105] The system for carrying out this invention utilizes generative AI to select and optimize advertising talent. A specific embodiment of this system will be described below.

[1106] System configuration

[1107] The system consists of the following main components:

[1108] 1. Generative AI model: Learns performance data and viewer response data of existing advertising talent and generates new talent.

[1109] 2. Emotion engine: Collects viewer reaction data and analyzes viewer emotions.

[1110] 3. Data collection module: Collects viewer response data from advertising videos.

[1111] 4. Talent selection module: Select the most suitable talent from the generated talents.

[1112] 5. Monitoring module: Monitor the effectiveness of your advertising campaigns in real time.

[1113] Hardware and software used

[1114] Hardware: Smartphones, cameras

[1115] Software: Libraries with functions such as image processing, image analysis, and machine learning

[1116] Data processing and calculation

[1117] 1. Data collection: Using a device (smartphone or camera), the faces of viewers watching the advertising video are detected, and emotion data is collected using an emotion recognition model.

[1118] 2. Talent generation: The server generates new advertising talent using a generative AI model based on the collected emotional data.

[1119] 3. Talent selection: The server selects the most suitable talent from the multiple talents generated based on the emotion data.

[1120] 4. Monitoring: The server monitors the effectiveness of the advertising campaign in real time and reevaluates talent selection as needed.

[1121] Specific examples

[1122] For example, to collect viewers' reactions to a video advertisement for a new smartphone, data such as whether viewers smile, express surprise, post positive comments on social media, etc. Based on this data, the following prompt sentences can be input into the generative AI model:

[1123] "Generate a talent with the following emotional parameters: [Happy Emotion Intensity: 0.8, Surprise Emotion Intensity: 0.7, Interest Emotion Intensity: 0.9]

[1124] Using this prompt, new advertising talent can be generated based on the viewer's emotional data, and the most suitable talent can be selected.

[1125] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1126] Step 1:

[1127] The device uses a camera to detect the faces of viewers watching advertising videos and collects video data in real time. The input is the viewer's video data, and the output is the face detection results. Specifically, face detection is performed using OpenCV, and the detected face area is cut out.

[1128] Step 2:

[1129] The device inputs the face detection results into an emotion recognition model and analyzes the viewer's emotional data. The input is the face detection results, and the output is emotion data. Specifically, the emotion recognition model is run using Keras to estimate emotions (happiness, surprise, interest, etc.) from the viewer's facial expressions.

[1130] Step 3:

[1131] The server inputs prompt sentences into a generative AI model based on the collected emotion data to generate new advertising talent. The input is emotion data, and the output is the generated advertising talent. Specifically, the prompt sentences are generated using Transformers and input into the GPT-3 model to generate new talent.

[1132] Step 4:

[1133] The server selects the optimal talent from the generated advertising talents based on the emotional data. The input is the generated advertising talent and emotional data, and the output is the optimal advertising talent. Specifically, the server executes an algorithm that evaluates the performance of each talent and compares it with the emotional data to select the optimal talent.

[1134] Step 5:

[1135] The server runs an advertising campaign using the selected advertising talent and monitors its effectiveness in real time. The input is the advertising campaign implementation data, and the output is the monitoring results. Specifically, it collects the number of ad views, click rates, reactions on social media, etc., and evaluates the effectiveness of the advertising campaign.

[1136] Step 6:

[1137] The server reevaluates talent selection as necessary based on the monitoring results and optimizes the advertising campaign. The input is the monitoring results, and the output is the reevaluated advertising talent. Specifically, the server analyzes the monitoring results, and if the advertising effectiveness is declining, it executes a process to generate new talent.

[1138] Example 2

[1139] Next, a description will be given of Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1140] Conventional advertising talent selection systems have the problem that talent selection relies on subjective judgment and does not adequately reflect the user's emotional reactions. In addition, there is a lack of a means to objectively evaluate talent performance and select the most suitable talent to maximize the effectiveness of advertising campaigns.

[1141] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for utilizing generative AI, a means for comparing with existing advertising talent at regular intervals, a means for finding a talent who is more likely to recognize the service, a means for utilizing the found talent, a means for collecting users' emotional responses, a means for analyzing the collected emotional responses, and a means for evaluating the talent based on the analysis results. This enables objective talent evaluation based on users' emotional responses and selection of the optimal talent that maximizes the effectiveness of the advertising campaign.

[1142] "Generative AI" is a system that uses artificial intelligence technology to generate new advertising talent.

[1143] A "certain period" is a unit of time that can be adjusted depending on the duration of the advertising campaign and market trends.

[1144] "Advertising talent" refers to people or characters who appear in advertisements.

[1145] "User emotional response" refers to the emotional response a user shows when viewing an advertisement, and is expressed through facial expressions, tone of voice, etc.

[1146] An "emotion engine" is software or a system for analyzing a user's emotional response and quantifying their emotional state.

[1147] "Talent evaluation" is the process of numerically evaluating the effectiveness of advertising talent based on users' emotional responses and performance data.

[1148] A "prompt" is an instruction entered into the generative AI to specify the characteristics and conditions of the advertising talent to be generated.

[1149] The present invention relates to a system for generating, comparing, and evaluating advertising talent. Specific embodiments of this system are described below.

[1150] The server generates a new advertising talent using a generative AI model. For example, a general natural language processing model is used as this generative AI model. A prompt sentence is input to the generative AI model, which generates talent characteristics suitable for the advertising campaign. An example of a specific prompt sentence is, "Please generate a new talent suitable for the summer season for an advertising campaign for a new soft drink."

[1151] The server compares the generated talent with existing talent, using data on the talent's past performance and market response, such as click-through rates and viewing times from past advertising campaigns.

[1152] The device then collects the user's emotional responses while watching the advertisements by using a camera and microphone to record the user's facial expressions and tone of voice, thereby obtaining real-time emotional data.

[1153] The collected emotion data is analyzed by an emotion engine on the server. For example, a general emotion analysis API is used as the emotion engine. As a result of the analysis, the user's emotional state (joy, surprise, sadness, etc.) is quantified.

[1154] Finally, the server evaluates the talent based on the analysis results. Specifically, if the user's emotional response is positive, the talent's evaluation score is increased, and if the response is negative, the evaluation score is decreased. This evaluation score is used in the next talent selection.

[1155] For example, when running an advertising campaign for a new soft drink, the server uses a generative AI model to generate a new talent and compares it with existing talent. The emotion engine analyzes the emotional response of users when they watch the advertisement and evaluates the talent based on the results. In this way, it becomes possible to select the optimal talent based on the user's emotions.

[1156] This system enables optimal talent selection to maximize the effectiveness of advertising campaigns.

[1157] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1158] Step 1: Generating advertising talent

[1159] The server generates a new advertising talent using a generative AI model. As input, it provides the generative AI model with a prompt statement: "Generate a new talent that suits the summer season for a new soft drink advertising campaign." The generative AI model generates the talent's characteristics (e.g., youthful appearance, lively personality) based on this prompt, and obtains the new talent's data as output.

[1160] Step 2: Compare advertising talent

[1161] The server compares the generated new talent with existing talent. It uses the new talent data and the existing talent database as input. The server evaluates which talent is more effective based on performance data such as click-through rate and viewing time from past advertising campaigns, and obtains the comparison result as output.

[1162] Step 3: Collecting users' emotional responses

[1163] The device collects the user's emotional responses while watching the advertisement. As input, it uses a camera and a microphone to record the user's facial expressions and tone of voice. The device collects this data in real time and obtains the user's emotional data as output.

[1164] Step 4: Analyze emotional responses

[1165] The server analyzes the collected user emotion data using an emotion engine. The emotion engine receives inputs such as the user's facial expressions and tone of voice. The emotion engine analyzes these data, quantifies the user's emotional state (happiness, surprise, sadness, etc.), and outputs the analysis results.

[1166] Step 5: Talent Evaluation

[1167] The server evaluates the talent based on the analysis results. As input, it uses the analysis results obtained from the emotion engine and the talent's performance data. If the user's emotional response is good, the server increases the talent's evaluation score, and if the response is bad, it decreases the evaluation score. As output, it obtains the updated talent's evaluation score.

[1168] In this way, a system is realized in which the server, terminals, and users work together to generate, compare, and evaluate advertising talent.

[1169] (Application example 2)

[1170] Next, a description will be given of Application Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1171] Conventional methods for selecting advertising talent have the problem of not maximizing advertising effectiveness because they select talent without considering users' emotional responses. Furthermore, there is a lack of a means for generating new talent during an advertising campaign and comparing them with existing talent, making it difficult to select the optimal talent. Furthermore, there is no means for analyzing users' emotional responses in real time and evaluating talent based on the results, making it difficult to accurately evaluate advertising effectiveness.

[1172] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1173] In this invention, the server includes a means for utilizing generative AI, a means for periodically comparing with existing advertising talent, a means for finding talent who can better recognize the service, a means for utilizing the found talent, a means for analyzing the user's emotional response, and a means for evaluating the talent based on the analysis results. This makes it possible to select the most suitable advertising talent based on the user's emotional response and maximize the advertising effectiveness.

[1174] "Generative AI" is a technology that uses artificial intelligence to generate new advertising talent.

[1175] A "certain period" is a unit of time that can be adjusted depending on the duration of the advertising campaign and market trends.

[1176] "Existing advertising talent" refers to people or characters who appear in advertisements currently in use.

[1177] A "service-recognizing talent" is a person or character who has the ability to effectively raise awareness of a specific service or product through advertising.

[1178] The "user's emotional response" refers to the emotional response shown by the user after viewing the advertisement, and includes emotions such as joy, surprise, and sadness.

[1179] The "means for evaluating a talent based on the analysis results" is a technology that analyzes the emotional reactions of users and evaluates the effectiveness of advertising talent based on the results.

[1180] As an embodiment of the present invention, the following system is constructed: The server includes means for utilizing generative AI, means for comparing with existing advertising talent at regular intervals, means for finding talent who can better recognize the service, means for utilizing the found talent, means for analyzing the user's emotional response, and means for evaluating the talent based on the analysis results.

[1181] Hardware and software used

[1182] Hardware: Smartphone camera

[1183] Software: Libraries with functions such as image processing, image analysis, and machine learning

[1184] Data processing and calculation

[1185] The server captures the user's face using the smartphone camera and pre-processes the facial image. The pre-processed image is then input into an emotion analysis model to analyze the user's emotional response. The analysis results are fed back to a generative AI model (such as GPT-3) and used to evaluate advertising talent.

[1186] Specific examples

[1187] For example, to generate a promotional talent for a new product, the following prompt sentences could be used:

[1188] "Generate advertising talent for new products."

[1189] "Generate advertising talent suitable for your summer campaign."

[1190] "Generate advertising talent for the Christmas season."

[1191] The generated advertising talent is compared with existing talent at regular intervals and evaluated based on the user's emotional response, allowing the system to select the optimal advertising talent and maximize advertising effectiveness.

[1192] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1193] Step 1:

[1194] The server captures the user's face using the smartphone's camera. The input is real-time video captured by the camera, and the output is the captured image of the face. Specifically, the camera detects the user's face and captures the image as a frame.

[1195] Step 2:

[1196] The server preprocesses the captured face image using OpenCV. The input is the face image acquired in step 1, and the output is the preprocessed face image. Specifically, the image is converted to grayscale, and the face area is cropped and resized.

[1197] Step 3:

[1198] The server inputs the preprocessed facial image into the emotion analysis model to analyze the user's emotional response. The input is the facial image preprocessed in step 2, and the output is the emotion analysis result. Specifically, the emotion analysis model analyzes the image and classifies emotions such as joy, surprise, and sadness.

[1199] Step 4:

[1200] The server generates a new advertising talent using a generative AI model (such as GPT-3). The input is a prompt sentence (e.g., "Please generate an advertising talent for a new product."), and the output is information about the generated advertising talent. Specifically, the prompt sentence is input into the generative AI model, and the model generates the characteristics of the new talent.

[1201] Step 5:

[1202] The server compares the generated advertising talent with existing advertising talent at regular intervals. The input is information about the generated advertising talent and information about the existing advertising talent, and the output is the comparison result. Specifically, it compares the talent's characteristics and performance and evaluates which is more effective.

[1203] Step 6:

[1204] The server evaluates advertising talent based on the user's emotional response. The input is the emotion analysis result obtained in step 3 and the comparison result obtained in step 5, and the output is the selection of the optimal advertising talent. Specifically, the emotion analysis result is used as the evaluation criterion to select the most effective talent.

[1205] Step 7:

[1206] The server utilizes the selected talent in the advertising campaign. The input is the information about the talent selected in step 6, and the output is the results of the advertising campaign. Specifically, the selected talent is used in the advertisement and the campaign is implemented.

[1207] Example 3

[1208] Next, a description will be given of Example 3 of Form Example 3. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1209] Traditional methods for selecting and evaluating advertising talent lack the flexibility to maximize advertising effectiveness, making it difficult to quickly respond to market trends and changes in user sentiment. Furthermore, the process of regularly evaluating the performance of existing talent and generating new talent is inefficient, making it difficult to maximize the effectiveness of advertising campaigns.

[1210] The identification process by the identification processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means. In this invention, the server includes means for utilizing generative artificial intelligence, means for comparing with existing advertising talent at regular intervals, means for finding talent who is more likely to recognize the service, means for utilizing the found talent, means for creating advertisements using the generated talent, means for distributing advertisements, means for measuring the effectiveness of advertisements, and means for adjusting emotion analysis parameters according to the duration of the advertising campaign and market trends using an emotion analysis engine. This makes it possible to quickly respond to market trends and changes in user emotions and maximize the effectiveness of the advertising campaign.

[1211] "Generative AI" is AI that has the ability to generate new information and content based on data.

[1212] "Advertising talent" refers to people or characters who appear in advertisements.

[1213] A "certain period" refers to a specific time frame that is set depending on the duration of an advertising campaign or market trends.

[1214] "Comparative means" refers to methods or techniques for evaluating and comparing the performance and characteristics of existing advertising talent with newly generated talent.

[1215] "Means of discovery" refers to methods and technologies for using generative artificial intelligence to find more effective advertising talent.

[1216] "Means of utilization" refers to the methods and techniques used to incorporate the identified advertising talent into actual advertising campaigns and maximize their effectiveness.

[1217] "Means for creating advertisements" refers to the methods and techniques for producing advertising content using generated talent.

[1218] "Means of delivering advertising" refers to the methods and technologies used to deliver created advertisements to target audiences.

[1219] "Means of measuring advertising effectiveness" refers to methods and technologies for collecting and analyzing data such as click-through rates, viewing times, and engagement rates to evaluate the success of advertising campaigns.

[1220] "Emotion analysis engine" refers to technology that analyzes users' emotional responses and adjusts the parameters of advertising campaigns based on that data.

[1221] This invention is a system that utilizes generative artificial intelligence to generate advertising talent and maximize the effectiveness of advertising campaigns using the talent. Specific embodiments of this system are described below.

[1222] System Overview

[1223] This system works in collaboration with three parties: a server, a terminal, and a user. The server uses generative artificial intelligence and an emotion analysis engine to generate advertising talent, create and distribute advertisements, and measure their effectiveness. The terminal is a device through which users view advertisements and is responsible for collecting users' emotional responses.

[1224] Hardware and software used

[1225] Generative artificial intelligence: For example, using generative AI models such as OpenAI's GPT-4.

[1226] Sentiment analysis engine: Software for analyzing users' emotional responses.

[1227] Video editing software: Create your ads.

[1228] Social media platforms: To distribute advertisements.

[1229] Data processing and calculation

[1230] 1. Creating a prompt for talent generation

[1231] The server creates a prompt to be input to the generative AI model, which specifically describes the characteristics and attributes of the talent to be generated.

[1232] Example: "Generate a female talent in her twenties. She has a bright, energetic, and friendly personality."

[1233] 2. Talent Generation

[1234] The server inputs prompt text into the generative AI model to generate new talents.

[1235] Example: The server sends prompts to a generative AI model and receives data on the generated talent.

[1236] 3. Ad Creation

[1237] The server uses the generated talent to create advertisements, which can be in the form of video ads or banner ads.

[1238] Example: A server uses video editing software to create a video advertisement that includes the generated talent.

[1239] 4. Advertisement Delivery

[1240] The server distributes the created advertisements, which can be distributed to social media sites, websites, etc.

[1241] Example: A server uses a social media platform to deliver video ads.

[1242] 5. Measuring effectiveness

[1243] The server measures the effectiveness of the ads, specifically collecting data such as click-through rates, viewing times, and engagement rates.

[1244] Example: The server uses the social media platform's analytics tools to collect information such as click-through rates and viewing times for ads, and stores this information in a database.

[1245] 6. Tuning your emotional engine

[1246] The server uses the emotion engine to adjust the parameters of the emotion analysis according to the duration of the advertising campaign and market trends.

[1247] Example: The server collects user emotion data during the Christmas season and adjusts the parameters of the emotion engine.

[1248] Specific examples

[1249] Example prompts for generative AI models

[1250] "Generate a male talent in his 30s suitable for the Christmas season. He has a warm, family-oriented character."

[1251] "For Valentine's Day, please create a female talent in her 20s who will inspire romantic feelings. She has a romantic and charming character."

[1252] In this way, the server is a system that generates and evaluates advertising talent by utilizing a generative AI model and a sentiment analysis engine. This enables rapid response to market trends and changes in user sentiment, maximizing the effectiveness of advertising campaigns. The flow of the identification process in Example 3 will be explained using Figure 21.

[1253] Step 1: Create a prompt statement

[1254] The server creates a prompt to be input to the generative AI model, which specifically describes the characteristics and attributes of the talent to be generated.

[1255] Input: Information about the objectives and target audience of your ad campaign.

[1256] Data processing: Generate appropriate prompts for the generative AI model based on the objectives of the advertising campaign and information about the target audience.

[1257] Output: The prompt sentence to be input to the generative AI model.

[1258] Specific operation: The server generates the prompt sentence, "Please generate a female talent in her twenties. She has a bright, energetic, and friendly character."

[1259] Step 2: Talent Generation

[1260] The server inputs prompt text into the generative AI model to generate new talents.

[1261] Input: The prompt statement created in step 1.

[1262] Data Calculation: The generative AI model analyzes the prompt and generates a talent with the specified characteristics and attributes.

[1263] Output: Generated talent data.

[1264] Specific operation: The server sends a prompt to the generative AI model and receives data on the generated talent.

[1265] Step 3: Create your ad

[1266] The server uses the generated talent to create advertisements, which can be in the form of video ads or banner ads.

[1267] Input: Talent data generated in step 2.

[1268] Data processing: Creating advertising content based on the generated talent data.

[1269] Output: The finished ad content.

[1270] What happens: The server uses video editing software to create a video advertisement that includes the generated talent.

[1271] Step 4: Serving Ads

[1272] The server distributes the created advertisements, which can be distributed to social media sites, websites, etc.

[1273] Input: The ad content created in step 3.

[1274] Data processing: Converting advertising content into a format suitable for the distribution platform.

[1275] Output: The ad that was served.

[1276] Specific operation: The server uses a social networking platform to deliver video ads.

[1277] Step 5: Measuring advertising effectiveness

[1278] The server measures the effectiveness of the ads, specifically collecting data such as click-through rates, viewing times, and engagement rates.

[1279] Input: Performance data of the ads served.

[1280] Data calculations: Analyzing collected performance data and evaluating the effectiveness of advertising.

[1281] Output: A report on the effectiveness of your ads.

[1282] What it does: The server uses the social media platform's analytics tools to collect ad click-through rates and viewing times, and stores them in a database.

[1283] Step 6: Tuning your Emotional Engine

[1284] The server uses the emotion engine to adjust the parameters of the emotion analysis according to the duration of the advertising campaign and market trends.

[1285] Input: User sentiment data and market trend data.

[1286] Data calculation: Analyze sentiment data and market trend data and adjust the parameters of the sentiment engine.

[1287] Output: Tuned emotion engine parameters.

[1288] Specific operation: The server collects user emotion data during the Christmas season and adjusts the parameters of the emotion engine.

[1289] Step 7: Flexible talent generation and assessment

[1290] The server generates and evaluates advertising talent flexibly according to market trends.

[1291] Input: Market trend data and advertising effectiveness data.

[1292] Data calculation: Analyze market trend data and advertising effectiveness data to generate and evaluate new talent.

[1293] Output: The newly generated talent and its evaluation results.

[1294] Specific operation: The server generates a talent that stimulates romantic feelings in preparation for Valentine's Day, and distributes the advertisement again to evaluate its effectiveness and measure its effectiveness.

[1295] (Application example 3)

[1296] Next, a description will be given of Application Example 3 of Form Example 3. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1297] Conventional methods for selecting and evaluating advertising talent make it difficult to fully reflect users' emotional reactions and market trends, making it difficult to maximize advertising effectiveness. Furthermore, flexible responses to specific periods and market fluctuations are required, but achieving this has been difficult.

[1298] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[1299] In this invention, the server includes means for utilizing generative AI, means for comparing with existing advertising talent at regular intervals, means for finding talent who can better recognize the service, means for utilizing the found talent, means for analyzing users' emotional responses using an emotion engine, means for adjusting emotion analysis parameters depending on the duration of the advertising campaign and market trends, and means for evaluating the effectiveness of advertising using the generated talent. This enables flexible generation and evaluation of advertising talent that reflects users' emotional responses and market trends.

[1300] "Generative AI" is a system that uses artificial intelligence technology to generate new data and content.

[1301] An "advertising talent" is a person or character used in an advertisement, whose role is to raise awareness of a product or service among viewers and consumers.

[1302] An "emotion engine" is software or hardware that analyzes a user's emotional response and adjusts the system's behavior based on the results.

[1303] "User emotional response" refers to a user's emotional reaction or feedback to an advertisement or content, including emotions such as joy, sadness, and anger.

[1304] An "advertising campaign" is an advertising activity carried out over a specific period of time, which is carried out in a planned manner to increase awareness of a product or service.

[1305] "Market trends" refer to changes in consumer behavior, trends, competitor actions, etc. in a particular market.

[1306] "Emotion analysis parameters" are settings and reference values ​​used when analyzing a user's emotional response, and affect the accuracy and results of the analysis.

[1307] The "effectiveness of advertising" refers to the impact and response that an advertisement has on viewers and consumers, and refers to concrete results such as increased product awareness and sales.

[1308] The system for implementing this invention is composed of a server including a generative AI, an emotion engine, and an advertising effectiveness evaluation module, and a terminal that collects user emotional responses. Specific embodiments of the system are described below.

[1309] The server generates advertising talent using generative AI. The generative AI analyzes the characteristics and performance of advertising talent and generates new talent based on that. The generated talent is compared with existing advertising talent at regular intervals, and the talent that can best raise awareness of the service is selected.

[1310] The emotion engine is used to analyze users' emotional responses. Users' emotional responses are collected via devices such as smartphones and tablets. The emotion engine adjusts the emotion analysis parameters according to the duration of the advertising campaign and market trends. For example, the "joy" parameter is set high during the Christmas season.

[1311] The advertisement using the generated talent is aired for a certain period of time, and its effectiveness is evaluated by an advertisement effectiveness evaluation module, which analyzes the response to the advertisement and evaluates the effectiveness of the generated talent.

[1312] For example, to generate a talent that emphasizes the emotion of "joy" during the Christmas season, enter the following prompt into the generation AI:

[1313] Generate a talent with the following emotional parameters: {Happy: 0.8, Sad: 0.1, Anger: 0.1}

[1314] By inputting this prompt into a generative AI, it is possible to generate talents that are perfect for the Christmas season.

[1315] The hardware used includes servers, smartphones, tablets, and other devices, while the software is Python, OpenAI API, emotion analysis software, and an advertising effectiveness evaluation module. The server runs the generative AI and emotion engine, while the device collects users' emotional responses.

[1316] In this way, it becomes possible to flexibly generate and evaluate advertising talent that reflects users' emotional responses and market trends.

[1317] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[1318] Step 1:

[1319] The server generates advertising talent using generative AI. Specifically, the server analyzes the characteristics and performance of advertising talent and creates prompts to generate new talent based on that analysis. For example, to generate a talent that emphasizes the emotion of "joy" during the Christmas season, the server generates the following prompt:

[1320] Generate a talent with the following emotional parameters: {Happy: 0.8, Sad: 0.1, Anger: 0.1}

[1321] Input: Advertising talent characteristics data, emotion parameters

[1322] Output: Generated talent prompt

[1323] Step 2:

[1324] The server inputs a prompt into the generative AI to generate a new advertising talent, which then generates an image and profile of the talent based on the prompt.

[1325] Input: prompt statement

[1326] Output: Generated advertising talent images and profiles

[1327] Step 3:

[1328] The server compares the generated advertising talent with existing advertising talent at regular intervals, specifically, the server compares performance data of the existing advertising talent with performance data of the generated talent, and selects a more effective talent.

[1329] Input: Existing advertising talent performance data, Generated talent performance data

[1330] Output: Selected advertising talent

[1331] Step 4:

[1332] The device collects the user's emotional response. Specifically, it collects emotional response data (e.g., facial expressions and voice) when the user views an advertisement via a smartphone or tablet.

[1333] Input: User viewing data

[1334] Output: User's emotional response data

[1335] Step 5:

[1336] The server analyzes the user's emotional response using an emotion engine that analyzes the collected emotional response data to identify the user's emotional state.

[1337] Input: User emotional response data

[1338] Output: Parsed user's emotional state

[1339] Step 6:

[1340] The server adjusts the emotion analysis parameters according to the duration of the advertising campaign and market trends, for example, by setting the "happiness" parameter higher during a specific period.

[1341] Input: Advertising campaign duration, market trend data

[1342] Output: Adjusted sentiment analysis parameters

[1343] Step 7:

[1344] The server broadcasts the advertisement using the generated talent for a certain period of time and evaluates its effectiveness. The advertisement effectiveness evaluation module collects response data for the advertisement and evaluates the effectiveness of the generated talent.

[1345] Input: Generated talent, advertising response data

[1346] Output: Advertising effectiveness evaluation results

[1347] In this way, it becomes possible to flexibly generate and evaluate advertising talent that reflects users' emotional responses and market trends.

[1348] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1349] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1350] Another example of generative AI is Gemini (internet search engine). <url: https: gemini.google.com ?hl="ja">) are mentioned.

[1351] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[1352] [Third embodiment]

[1353] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1354] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[1355] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1356] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[1357] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1358] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1359] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1360] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1361] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1362] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1363] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1364] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[1365] "Example 1"

[1366] One aspect of the present invention is a system that utilizes generative AI. This system analyzes the characteristics and performance of advertising talent and generates new talent based on that. Specifically, the AI ​​learns performance data and viewer response data of existing advertising talent, and generates new talent based on that. The generated talent is compared with existing advertising talent to find talent that can better promote awareness of the service.

[1367] "Example 2"

[1368] The present invention also provides a system for comparing a new talent with existing advertising talent at regular intervals. This regular interval can be adjusted according to the duration of an advertising campaign or market trends. For example, a new talent can be generated to coincide with the launch of a new product or a specific season, and compared with existing talent.

[1369] "Example 3"

[1370] Furthermore, a system for utilizing the discovered talent is also an embodiment of the present invention. Specifically, the generated talent is used in an advertisement and its effectiveness is verified. For example, an advertisement using the generated talent is aired for a certain period of time, and the effectiveness of the talent can be evaluated by analyzing the response.

[1371] The processing flow of each embodiment will be described below.

[1372] "Example 1"

[1373] Step 1: The generative AI collects performance data of existing advertising talent and viewer response data.

[1374] Step 2: Based on the collected data, the AI ​​learns and generates new talents.

[1375] Step 3: The generated talents are compared with existing advertising talents to find talents that can better raise awareness of the service.

[1376] "Example 2"

[1377] Step 1: Set a fixed period, taking into account the duration of the advertising campaign and market trends.

[1378] Step 2: At each set time period, generate new talents and compare them with existing talents.

[1379] "Example 3"

[1380] Step 1: Hire the generated talent for your ads.

[1381] Step 2: Air the advertisement using the talent you selected for a certain period of time.

[1382] Step 3: Analyze the response to the aired advertisement and evaluate the effectiveness of the talent.

[1383] Example 1

[1384] Next, a description will be given of Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1385] Conventional methods for selecting advertising talent have difficulty finding effective talent because they do not fully utilize viewer response and performance data. It is also difficult to flexibly adjust talent according to the duration of the advertising campaign and market trends. Therefore, a system that can efficiently generate and evaluate new talent to maximize advertising effectiveness is needed.

[1386] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1387] In this invention, the server includes a means for utilizing generative AI, a means for collecting performance data and viewer reaction data of existing advertising talent, a means for preprocessing the collected data, a means for inputting the preprocessed data into an AI model for learning, a means for generating new talent using the learned AI model, a means for evaluating the performance of the generated talent, and a means for outputting the evaluation results. This makes it possible to analyze the characteristics and performance of advertising talent in detail and efficiently generate and evaluate more effective talent.

[1388] "Generative AI" is an artificial intelligence technology that generates new data and models based on existing data.

[1389] "Performance data" refers to data that shows the effectiveness of an advertising talent, such as the number of views, click rate, and purchase rate.

[1390] "Viewer response data" refers to data that records viewers' reactions to advertisements, such as comments, ratings, and number of shares.

[1391] "Preprocessing" refers to the process of converting collected data into a format suitable for an AI model by performing processes such as filling in missing values, removing outliers, and normalizing the data.

[1392] An "AI model" is an algorithm built using a machine learning framework that learns patterns from data and makes predictions or generation decisions.

[1393] A "talent" is a person or character who appears in an advertisement.

[1394] "Evaluation" refers to the process of measuring the performance of the generated talent using indicators such as number of views, click rate, and purchase rate, and comparing it with existing talent.

[1395] "Output" means providing the evaluation results in the form of a report or data.

[1396] This invention relates to a system that uses generative AI to analyze the characteristics and performance of advertising talent and generate new talent. Specific embodiments of this system are described below.

[1397] The server first collects performance data of existing advertising talent and viewer response data, using APIs to retrieve data from company databases and social media platforms, including the number of ad views, click-through rates, purchase rates, viewer comments, ratings, and number of shares.

[1398] The server then preprocesses the collected data, which includes using a data analysis library to impute missing values, remove outliers, and normalize the data. For example, data with extremely low views or abnormally high click-through rates is removed, and the data is normalized using min-max scaling.

[1399] The pre-processed data is then fed into an AI model. The server uses a deep neural network (DNN) model built using a machine learning framework. The model learns the characteristics of advertising talent and audience reactions, and finds patterns to generate new talent.

[1400] Once the learning is complete, the server generates a new talent. The generated talent is expected to have similar characteristics to the existing talent, but with higher performance. For example, a talent with specific attributes such as "female in her 20s, cheerful personality, sports lover" may be generated.

[1401] The performance of the generated talent is evaluated by the server. The evaluation uses indicators such as the number of views, click rate, and purchase rate, and compares it with existing talent. If the evaluation results show that the generated talent performs better than the existing talent, that talent is selected.

[1402] Finally, the server outputs the evaluation results, which are output as an Excel report and emailed to the user, who can then use the report to plan a new advertising campaign.

[1403] As a concrete example, consider a company looking for a new advertising talent for a beverage. The server collects performance data (number of views, click-through rate, purchase rate, etc.) and viewer response data (comments, ratings, number of shares, etc.) of previously used beverage advertising talent, inputs this data into an AI model, and generates new talent. If the generated talent performs better than existing talent, it will be used in the new advertising campaign.

[1404] Example prompt sentence:

[1405] "Generate new beverage advertising talent based on performance data and viewer response data of past beverage advertising talent. The generated talent must exceed existing talent in terms of views, click-through rates, or purchase rates."

[1406] In this way, the server utilizes generative AI to analyze the characteristics and performance of advertising talent, creating a system that generates new talent.

[1407] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1408] Step 1: Data collection

[1409] The server collects existing advertising talent performance data and viewer response data from the company's database and social media platforms. Specifically, it uses APIs to obtain data such as the number of views, click-through rates, conversion rates, comments, ratings, and shares. The input is the raw data obtained from the API, and the output is the collected raw data.

[1410] Step 2: Data Preprocessing

[1411] The server preprocesses the collected data. Specifically, it uses a data analysis library to fill in missing values, remove outliers, and normalize the data. For example, it removes data with extremely low view counts and abnormally high click-through rates, and normalizes the data using min-max scaling. The input is the collected raw data, and the output is the preprocessed, clean data.

[1412] Step 3: Training the AI ​​model

[1413] The server inputs the preprocessed data into the AI ​​model and begins learning. The AI ​​model used is a DNN (deep neural network) model built using a machine learning framework. The model learns the characteristics of advertising talent and viewer reactions, and finds patterns to generate new talent. The input is the preprocessed data, and the output is the trained AI model.

[1414] Step 4: Generate new talent

[1415] The server uses the trained AI model to generate new talent. The generated talent is expected to have similar characteristics to existing talents, but with higher performance. For example, a talent with specific attributes such as "female in her 20s, cheerful personality, sports lover" may be generated. The input is the trained AI model, and the output is the characteristics of the newly generated talent.

[1416] Step 5: Performance evaluation

[1417] The server evaluates the performance of the generated talent. Specifically, it compares it with existing talent using metrics such as the number of views, click-through rate, and purchase rate. If the evaluation results show that the generated talent performs better than the existing talent, that talent is selected. The input is the characteristics of the generated talent, and the output is the evaluation result.

[1418] Step 6: Output the results

[1419] The server outputs the evaluation results. Specifically, it outputs the evaluation results as an Excel report and sends it to the user by email. The user then plans a new advertising campaign based on this report. The input is the evaluation results, and the output is the report provided to the user.

[1420] (Application example 1)

[1421] Next, a description will be given of Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1422] Traditional methods for selecting advertising talent make it difficult to fully utilize existing talent performance data and viewer response data, making it difficult to maximize the effectiveness of advertising campaigns. Furthermore, selecting advertising talent takes time and money, making it difficult to respond to market fluctuations that require rapid response. This has led to the issue of a lower success rate for advertising campaigns.

[1423] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1424] In this invention, the server includes a means for utilizing generative AI, a means for comparing with existing advertising talent at regular intervals, and a means for searching for talent who can better recognize the service. This optimizes the advertising campaign, and a means for using an application installed on a smartphone. This automates the advertising talent selection process, quickly and effectively generates new talent, and increases the success rate of advertising campaigns.

[1425] "Generative AI" is an artificial intelligence technology that learns from existing data and generates new data and models.

[1426] A "certain period" is a unit of time that can be adjusted depending on the duration of the advertising campaign and market trends.

[1427] An "advertising talent" is a person or character who appears in an advertisement.

[1428] "Performance Data" means data relating to the activities and achievements of advertising talent.

[1429] "Viewer response data" refers to data regarding the reactions and feedback of viewers who have viewed advertisements.

[1430] An "application installed on a smartphone" is a software program that runs on a smartphone.

[1431] "Advertising campaign optimization measures" are methods and processes for maximizing the effectiveness of advertising.

[1432] The following system configuration will be described as an embodiment of the present invention.

[1433] The server includes a means for utilizing generative AI, a means for comparing with existing advertising talent at regular intervals, a means for finding talent who can better recognize the service, a means for optimizing advertising campaigns, and a means for using applications installed on smartphones.

[1434] System Program

[1435] The server builds a generative AI model. Specifically, it reads performance data and viewer response data of existing advertising talent and builds a neural network model using LSTM (Long Short-Term Memory) layers and Dense layers. The model is compiled using the Adam optimizer and categorical cross-entropy loss function. New talent is generated using the trained model.

[1436] Processing Description

[1437] The server does the following:

[1438] 1. Data Loading: The server loads existing advertising talent performance data and viewer response data, which may include retrieving data from databases and cloud storage.

[1439] 2. Building a model: The server builds a neural network model using LSTM layers and Dense layers, which enables processing of time series data.

[1440] 3. Compile the model: The server compiles the model using the Adam optimizer and categorical cross-entropy loss function, which allows for efficient model training.

[1441] 4. Model training: The server uses the loaded data to train a model, which can then analyze the characteristics and performance of advertising talent.

[1442] 5. New talent generation: The server generates new talent using the trained model, which allows for faster discovery of more effective advertising talent.

[1443] Specific examples

[1444] For example, when an advertising agency launches an advertising campaign for a new beverage, it can input the performance data of existing talent and viewer response data to generate the most suitable new talent, thereby increasing the success rate of the advertising campaign.

[1445] Prompt Sentence Examples

[1446] "Generate new talent using performance data and viewer response data from existing advertising talent."

[1447] In this way, the server can automate the advertising talent selection process, quickly and effectively generating new talent and increasing the success of advertising campaigns.

[1448] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1449] Step 1:

[1450] The server loads existing advertising talent performance data and viewer response data from a database or cloud storage. The input data is the advertising talent's activity history and viewer feedback data, and uses these for subsequent processing. The output is the loaded dataset.

[1451] Step 2:

[1452] The server preprocesses the loaded data, specifically imputing missing values, normalizing data, encoding categorical data, etc. The input is the dataset loaded in step 1, and the output is the preprocessed dataset.

[1453] Step 3:

[1454] The server uses the preprocessed data to build a neural network model with an LSTM layer and a Dense layer. The input is the preprocessed dataset, and the output is the built neural network model.

[1455] Step 4:

[1456] The server compiles the constructed model, specifically using the Adam optimizer and categorical cross-entropy loss function. The input is the model constructed in step 3, and the output is the compiled model.

[1457] Step 5:

[1458] The server trains the dataset with the compiled model: the input is the preprocessed dataset and the compiled model, and the output is the trained model.

[1459] Step 6:

[1460] The server generates new advertising talent using the trained model. Specifically, it uses performance data of existing talent as input to predict the characteristics of the new talent. The input is the performance data of the existing talent, and the output is the characteristics of the generated new talent.

[1461] Step 7:

[1462] The server compares the characteristics of the generated new talent with the existing talents and selects the more effective talent. The input is the characteristics of the generated new talent and the characteristics of the existing talents, and the output is the selected optimal talent.

[1463] Step 8:

[1464] The server generates data for utilizing the selected optimal talent in an advertising campaign and sends it to an application installed on a smartphone. The input is the data for the selected optimal talent, and the output is the data for utilization in an advertising campaign.

[1465] Example 2

[1466] Next, a description will be given of Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1467] Traditional methods for selecting advertising talent make it difficult to quickly find the optimal talent based on market trends and the duration of an advertising campaign. Furthermore, creating new talent and comparing them with existing talent is done manually, which is inefficient, time-consuming, and costly.

[1468] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1469] In this invention, the server includes a means for a user to input a prompt sentence, a means for the server to receive the prompt sentence and send it to the generative AI model, a means for the generative AI model to generate an image and profile of a new talent, a means for the server to compare the generated talent with existing talent, and a means for the server to provide the comparison result to the user, thereby enabling the selection of the optimal talent quickly and efficiently according to the duration of the advertising campaign and market trends.

[1470] "Generative AI" refers to artificial intelligence that generates images and profiles of new advertising talent based on prompts entered by the user.

[1471] A "prompt sentence" is the text that the user enters to instruct the generative AI to generate a new talent.

[1472] The "server" is a computer system that receives prompt text from the user, sends it to the generation AI, manages and compares the information of the generated talent, and provides the results to the user.

[1473] "Advertising personality" means a person or character used to promote a product or service in an advertising campaign.

[1474] "Image" refers to the visual representation of the advertising talent generated by generative AI.

[1475] A "profile" refers to information such as the advertising talent's name, age, and career history that is generated by generative AI.

[1476] "Comparison" refers to the process of evaluating the new talent generated against existing talent to determine which is more suitable for the advertising campaign.

[1477] "User" means a person or organization that uses the system to generate new advertising talent and receive comparison results.

[1478] A "certain period of time" is a time that can be adjusted depending on the duration of the advertising campaign and market trends.

[1479] This invention relates to a system that uses generative AI to generate advertising talent and compares them with existing talent. Specific embodiments of this system are described below.

[1480] First, the user inputs a prompt through the system interface. The prompt is text that instructs the generative AI to generate a new talent. For example, the user might input a prompt such as, "I would like you to generate a youthful female talent suitable for our new summer campaign."

[1481] The server receives a prompt sentence entered by the user. The received prompt sentence is then sent to a generative AI model. For example, OpenAI's GPT-4 (registered trademark) or DALL-E is used as this generative AI model.

[1482] The generative AI model analyzes the prompt and generates a new talent image and profile based on it. The generated image is a visual representation of the advertising talent, and the profile includes information such as name, age, and career history.

[1483] The server then compares the generated new talent information with existing talent based on the duration of the advertising campaign and market trends, including the popularity and market response of the existing talent.

[1484] Finally, the server provides the comparison results to the user, who can then select a new advertising talent based on the results, for example, by hiring the new talent if it is determined that the new talent is more suitable for the campaign than the existing talent.

[1485] As a specific example, if a user inputs a prompt such as "I want you to generate a youthful female talent suitable for our new summer campaign," the server sends this prompt to a generative AI model, and the model generates an image of a youthful female talent and a profile such as "Name: Hanako Sato, Age: 25, Experience: 5 years of modeling experience." The server compares the information of this new talent with existing talents and evaluates which is more suitable for the summer campaign. The evaluation results are provided to the user, who then decides whether to hire the new talent.

[1486] This system makes it possible to quickly and efficiently select the most suitable talent depending on the duration of the advertising campaign and market trends.

[1487] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1488] Step 1:

[1489] The user enters a prompt statement.

[1490] The user inputs a prompt for generating a new advertising talent through the system interface, for example, a specific request such as "Please generate a youthful female talent suitable for the new summer campaign."

[1491] Input: User prompt text

[1492] Output: The prompt statement is sent to the system.

[1493] Step 2:

[1494] The server receives the prompt.

[1495] The server receives the prompt text entered by the user and prepares for the next process.

[1496] Input: The prompt text sent by the user

[1497] Output: The prompt statement is saved on the server.

[1498] Step 3:

[1499] The server sends a prompt to the generative AI model.

[1500] The server sends the received prompt to the generative AI model, which then generates an image and profile of the new talent based on the prompt.

[1501] Input: Prompt text stored on the server

[1502] Output: The prompt is sent to the generative AI model.

[1503] Step 4:

[1504] A generative AI model generates images and profiles for new talent.

[1505] The generative AI model analyzes the prompt and generates an image and profile of a new talent based on it, such as an image of a youthful female talent, along with the talent's name, age, and background information.

[1506] Input: The prompt sent to the generative AI model

[1507] Output: New talent image and profile

[1508] Step 5:

[1509] The server compares the generated talent with existing talents.

[1510] The server compares the generated new talent information with existing advertising talent based on the duration of the advertising campaign and market trends, for example, taking into account the popularity and market response of the existing talent.

[1511] Input: New talent image and profile, existing talent information

[1512] Output: Comparison results

[1513] Step 6:

[1514] The server provides the comparison results to the user.

[1515] The server provides the comparison results to the user, who can then select a new advertising talent based on the results. For example, if the generated talent is determined to be more suitable for the campaign than the existing talent, the user can hire the talent.

[1516] Input: Comparison result

[1517] Output: The comparison results are provided to the user

[1518] (Application example 2)

[1519] Next, a description will be given of Application Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1520] In conventional advertising campaigns, there are limited means to evaluate the effectiveness of existing advertising talent, making it difficult to generate and compare new talent. Furthermore, it is not possible to select the optimal talent based on market trends and the duration of the advertising campaign, making it difficult to maximize advertising effectiveness. Therefore, in order to improve the effectiveness of advertising campaigns, a system is needed that uses generative AI models to generate new talent and compares them with existing talent to select the optimal talent.

[1521] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1522] In this invention, the server includes a means for utilizing generative AI, a means for comparing with existing advertising talent at regular intervals, and a means for searching for talent that can better recognize the service. This makes it possible to generate new talent based on the duration of the advertising campaign and market trends, compare the generated talent with existing talent, evaluate which is more effective, and select the most suitable talent to reflect in the advertising campaign.

[1523] "Generative AI" is a system that uses artificial intelligence technology to generate new data and information.

[1524] A "certain period" is a unit of time that can be adjusted depending on the duration of the advertising campaign and market trends.

[1525] An "advertising talent" is a person or character who appears in an advertisement and is selected to enhance the effectiveness of the advertisement.

[1526] A "comparison measure" is a method for evaluating existing advertising talent and newly generated talent to determine which is more effective.

[1527] A "talent capable of raising awareness of a service" is a talent who has the ability to effectively raise awareness of a specific service or product among consumers through advertising.

[1528] An "advertising campaign" is an advertising activity carried out over a specific period of time with the purpose of promoting a product or increasing brand awareness.

[1529] "Market trends" refer to changes and trends in the market that affect advertising strategies, such as consumer preferences, purchasing behavior, and competitor actions.

[1530] "Generated talent" refers to advertising talent newly created using generative AI.

[1531] "Means of evaluation" refers to a method for comparing the generated talent with existing talent and measuring its effectiveness.

[1532] The "optimal talent" is the talent who can most effectively raise awareness of a service or product in an advertising campaign.

[1533] The system for implementing this invention utilizes generative AI to periodically compare existing advertising talent to find talent who can better recognize the service, and select the talent who is best suited to the advertising campaign. A specific embodiment of this system is described below.

[1534] The server generates new advertising talent using generative AI. Generative AI models such as OpenAI's GPT-3 are used as generative AI. The server generates prompts based on the duration of the advertising campaign and market trends, and inputs these prompts into the generative AI model to generate new talent.

[1535] For example, if the advertising campaign period is "October 1, 2023 to December 31, 2023" and the market trend is "Fall / Winter fashion trends," the following prompt sentences will be generated:

[1536] Advertising campaign period: October 1, 2023 to December 31, 2023, Market trends: Generate new advertising talent based on autumn / winter fashion trends.

[1537] The generated talent is compared to existing advertising talent by the server, using a generative AI model to generate prompts such as the following and evaluate them:

[1538] Compare Existing Talent: Existing Popular Talent A with New Talent: Newly Generated Talent B and evaluate which is more effective.

[1539] The server evaluates the effectiveness of the generated talent and existing talent, selects the most suitable talent, and determines which talent to include in the advertising campaign based on the evaluation results.

[1540] This system generates new talent according to the duration of the advertising campaign and market trends, and compares them with existing talent to select the most suitable talent, thereby maximizing advertising effectiveness.In addition, by using a generative AI model, it is possible to quickly and efficiently generate and evaluate new talent.

[1541] In terms of specific hardware, the server uses a computer system equipped with a high-performance processor and sufficient memory. In terms of software, it uses an API (e.g., OpenAI (registered trademark) API) for running the generative AI model. Data processing and calculation are performed by the generative AI model, and algorithms are implemented to analyze and compare the characteristics and performance of the generated talent.

[1542] In this way, it is possible to select the most suitable talent for an advertising campaign and maximize the effectiveness of the advertisement.

[1543] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1544] Step 1:

[1545] The server receives data about the duration of the advertising campaign and market trends as input. This data includes the campaign start and end dates, as well as market trend information. Based on this data, the server generates prompts to be fed into the generative AI model. Specifically, the server generates prompts like the following:

[1546] Advertising campaign period: October 1, 2023 to December 31, 2023, Market trends: Generate new advertising talent based on autumn / winter fashion trends.

[1547] Inputs: duration of advertising campaign, market trends

[1548] Output: A prompt to be input to the generative AI model

[1549] Step 2:

[1550] The server sends the generated prompt to a generative AI model (e.g., OpenAI's GPT-3) to generate a new advertising talent. The generative AI model generates the characteristics and profile of the new talent based on the prompt. The generated talent information is returned to the server.

[1551] Input: prompt statement

[1552] Output: Information about the new advertising talent generated

[1553] Step 3:

[1554] The server retrieves information about existing advertising talent from a database, including the talent's name, characteristics, past performance data, etc.

[1555] Input: Query to get information about existing advertising talent

[1556] Output: Existing advertising talent information

[1557] Step 4:

[1558] The server generates a prompt to compare the new talent with the existing talents, such as:

[1559] Compare Existing Talent: Existing Popular Talent A with New Talent: Newly Generated Talent B and evaluate which is more effective.

[1560] Input: Information about the new talent created, information about the existing talent

[1561] Output: Prompt for comparison

[1562] Step 5:

[1563] The server sends the generated prompt sentence for comparison to the generative AI model for evaluation. The generative AI model compares the effects of the new talent with the existing talent based on the prompt sentence and returns the evaluation result to the server.

[1564] Input: Prompt sentence for comparison

[1565] Output: Evaluation results

[1566] Step 6:

[1567] The server selects the most suitable talent based on the evaluation results, and the information on the selected talent is reflected in the advertising campaign plan.

[1568] Input: Evaluation result

[1569] Output: Information on the best talent

[1570] Step 7:

[1571] The server transmits information about the selected optimal talent to the advertising campaign management system and reflects it in the advertising campaign, thereby enabling the optimal talent to be used in the advertising campaign.

[1572] Input: Optimal talent information

[1573] Output: Talent information reflected in advertising campaign

[1574] Example 3

[1575] Next, a third embodiment of the third embodiment will be described. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1576] Traditional methods for selecting and evaluating advertising talent make it difficult to accurately measure talent effectiveness and select the most suitable talent. Furthermore, there was a lack of means to analyze the effectiveness of advertising campaigns in real time and respond quickly. This resulted in advertising effectiveness not being maximized and advertising costs being wasted.

[1577] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1578] In this invention, the server includes means for a user to input a prompt text, means for the server to receive the prompt text and input it into a generative AI model, means for the server to generate a talent using the generative AI model, means for the server to create images and videos of the generated talent, means for the server to create advertising materials and air advertisements, means for the server to collect advertisement response data, and means for the server to analyze the collected data and evaluate the advertising effectiveness. This makes it possible to quickly generate optimal talent using the generative AI model and analyze and evaluate the effectiveness of advertising campaigns in real time.

[1579] "Generative AI" is an artificial intelligence technology that generates new talents based on prompts entered by users.

[1580] A "prompt sentence" is text that the user inputs into the generative AI, describing the characteristics of the talent they want to generate.

[1581] The "server" is a computer system that uses generative AI to generate talent, create advertising materials, and collect and analyze advertising response data.

[1582] "Talent" refers to a person or character used in advertisements generated by generative AI.

[1583] "Advertising materials" are advertising content created using images and videos of the generated talent.

[1584] "Response data" refers to data such as click-through rates, viewing times, and conversion rates collected after an advertisement has aired.

[1585] "Advertising effectiveness" is an indicator that shows the impact and reaction of an advertisement to viewers, and is evaluated based on response data.

[1586] This invention relates to a system for generating advertising talent using generative AI and evaluating the effectiveness of the talent. Specific embodiments of this system are described below.

[1587] First, the user enters a prompt statement describing the characteristics of the celebrity they wish to create through the system interface. For example, they can enter a prompt statement such as, "Please create a celebrity who is a woman in her twenties with a bright and energetic image."

[1588] Next, the server receives the prompt sentence sent by the user and inputs it into a generative AI model. For example, OpenAI's GPT-4 (registered trademark) can be used as this generative AI model. The server uses the generative AI model to generate a talent based on the prompt sentence. This talent is generated in the form of an image or video.

[1589] The generated images and videos of the talent are processed by the server and converted into a format that can be used as advertising material. For example, image editing software or video editing software can be used. This allows the generated talent's background to be made transparent or a video tailored to a specific scene to be created.

[1590] The server then uses the generated talent to create advertising materials, which are then broadcast for a set period of time on websites, social media, television, etc. For example, a cosmetics advertisement featuring the generated talent can be broadcast for one month.

[1591] While the advertisement is being aired, the server uses tools such as web page access analysis services to collect data such as click-through rates, viewing times, and conversion rates. This data is then stored on the server as advertisement response data.

[1592] Finally, the server analyzes the collected response data and evaluates the advertising effectiveness of the generated talent. This evaluation result is reflected in the next talent generation and advertising strategy. For example, if the click-through rate is high, it can be determined that the talent's characteristics were effective.

[1593] As a concrete example, if a user inputs the prompt "Please generate a male celebrity in his 30s who looks like a trustworthy businessman," the server will use the generative AI model to generate a trustworthy businessman-like celebrity. An advertisement for business software using this celebrity will be aired for two weeks, and the data collected during that time will be analyzed. Based on the analysis results, the advertising effectiveness of the celebrity can be evaluated.

[1594] In this way, a system for generating and evaluating advertising talent using generative AI can be implemented. The flow of the identification process in the third embodiment will be described with reference to FIG.

[1595] Step 1:

[1596] The user enters a prompt statement.

[1597] The user inputs a prompt statement describing the characteristics of the talent they wish to create through the system interface. For example, they might input, "Please generate a talent who is a woman in her twenties with a bright and energetic image." The input prompt statement is then sent to the server.

[1598] Step 2:

[1599] The server receives the prompt sentence and inputs it into the generative AI model.

[1600] The server receives the prompt sent by the user and inputs it into a generative AI model (e.g., OpenAI's GPT-4®). Based on the input prompt, the generative AI model analyzes the talent's characteristics and prepares the data required for generation.

[1601] Step 3:

[1602] The server generates talents using a generative AI model.

[1603] The server uses a generative AI model to generate a talent based on the prompt. The generative AI model analyzes the content of the prompt and generates appropriate images and videos. The data of the generated talent is stored on the server.

[1604] Step 4:

[1605] The server creates images and videos of the generated talent.

[1606] The server processes the generated images and videos of the talent and formats them into a format that can be used as advertising materials. For example, image editing software is used to make the background transparent. Also, video editing software is used to create videos tailored to specific scenes. The processed images and videos are stored on the server.

[1607] Step 5:

[1608] The server creates advertising materials and broadcasts the advertisements.

[1609] The server uses the generated talent to create advertising materials. For example, it creates a cosmetics advertisement featuring the generated talent. This advertising material is broadcast for a certain period of time on websites, social media, television, etc. The broadcast schedule and media are managed by the server.

[1610] Step 6:

[1611] The server collects advertisement response data.

[1612] While the advertisement is being aired, the server uses tools such as web page access analysis services to collect data such as click-through rates, viewing times, and conversion rates. The collected response data is stored on the server.

[1613] Step 7:

[1614] The server analyzes the collected data and evaluates the effectiveness of the advertisements.

[1615] The server analyzes the collected response data and evaluates the advertising effectiveness of the generated talent. For example, it quantifies the talent's effectiveness based on click rates and conversion rates. The analysis results are reflected in the next talent generation and advertising strategy.

[1616] (Application example 3)

[1617] Next, a description will be given of Application Example 3 of Form Example 3. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1618] Traditional methods for selecting advertising talent and measuring effectiveness have been problematic in that talent selection is time-consuming and costly, and it is difficult to accurately evaluate the effectiveness of advertising. These issues are particularly significant obstacles to maximizing advertising effectiveness in today's market, where real-time analysis of advertising campaign effectiveness and rapid response are required.

[1619] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[1620] In this invention, the server includes means for utilizing generative AI, means for comparing with existing advertising talent at regular intervals, means for finding talent who can better recognize the service, means for utilizing the found talent, means for creating advertisements using the generated talent, means for broadcasting the advertisements for a certain period of time, means for collecting and analyzing responses to the advertisements during the broadcast period, and means for evaluating effectiveness and generating reports. This makes it possible to streamline the process from selecting advertising talent to measuring effectiveness and evaluate the effectiveness of advertisements in real time.

[1621] "Generative AI" is a system that uses artificial intelligence technology to generate new data and information.

[1622] "Advertising talent" refers to people or characters who appear in advertisements.

[1623] "Fixed period" refers to a time frame that can be adjusted depending on the duration of the advertising campaign and market trends.

[1624] "Means for comparison" refers to a method or system for comparing generated talent with existing talent based on evaluation criteria.

[1625] A "recognizable talent" refers to a talent who has the ability to raise awareness of a service or product through advertising.

[1626] "Search means" refers to methods and systems that use generative AI and other technologies to find suitable talent.

[1627] "Means of utilization" refers to the methods and systems for utilizing the talent identified in advertising and other promotional activities.

[1628] "Means for creating advertisements" refers to a method or system for creating advertising content using generated talent.

[1629] "Means for broadcasting advertisements" refers to the method or system for releasing the created advertisements in the media for a certain period of time.

[1630] "Means for collecting and analyzing responses" refers to methods and systems for collecting and analyzing viewer responses and data obtained during the advertisement's broadcast period.

[1631] "Means for evaluating effectiveness and generating reports" refers to methods or systems for evaluating the effectiveness of advertising based on collected and analyzed data and compiling the results into a report.

[1632] The following system configuration and processing procedure will be described as an embodiment of the present invention.

[1633] System Configuration

[1634] This system uses the following hardware and software:

[1635] Hardware: Servers, smartphones

[1636] Software: Generative AI models (e.g., OpenAI® API), advertising platform APIs, data analysis tools (e.g., Python®)

[1637] Processing steps

[1638] 1. Ways to use generative AI:

[1639] The server generates a new talent using a generative AI model. Specifically, it inputs a prompt sentence and uses the generated text as the talent.

[1640] Example: Prompt: "Generate a new advertising talent. He / she is in his / her 20s and has a relatable character."

[1641] 2. Comparison with existing advertising talent over a period of time:

[1642] The server compares the generated talent with existing advertising talent based on evaluation criteria, including audience response and advertising effectiveness.

[1643] 3. Ways to find talent that can better recognize your services:

[1644] The server uses generative AI and other technologies to find talent with the ability to increase awareness of services and products.

[1645] 4. How to utilize the talent you find:

[1646] The server uses the talent it finds for advertising and other promotional activities.

[1647] 5. How to create advertisements using generated talent:

[1648] The server uses the generated talent to create advertising content, combining advertising copy and visual materials to create attractive advertisements.

[1649] 6. Means of broadcasting advertisements for a certain period of time:

[1650] The server publishes the created advertisements on the media for a certain period of time, and manages the advertisement broadcast schedule using the advertising platform API.

[1651] 7. Methods for collecting and analyzing the response to advertisements during their broadcast period:

[1652] The server collects and analyzes viewer responses and data obtained during the advertisement broadcast period, and uses data analysis tools to evaluate viewer responses in real time.

[1653] 8. Means of evaluating effectiveness and generating reports:

[1654] The server evaluates the effectiveness of the advertisements based on the collected and analyzed data and compiles the results into a report. The evaluation results will be used to improve the advertising campaign and to select the next talent.

[1655] Specific examples

[1656] For example, if a generative AI model is used to input the prompt "Generate a new talent for advertising. He / she is in his / her 20s and has a friendly character," the model will generate a talent: "The new talent, Yamada Taro, is in his / her 20s and has a friendly character." An advertisement is created using this talent and aired for a certain period of time. Viewer responses are then collected and analyzed to evaluate the effectiveness of the advertisement, which can be used to improve the next advertising campaign.

[1657] In this way, the process from selecting advertising talent to measuring effectiveness can be made more efficient, and the effectiveness of advertising can be evaluated in real time.

[1658] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[1659] Step 1:

[1660] The server generates a new talent using a generative AI model. Specifically, it receives a prompt sentence, "Please generate a new advertising talent. He / she is in his / her twenties and has a friendly character," as input and sends it to the generative AI model. The generative AI model generates the talent's text data based on this prompt sentence and outputs it to the server.

[1661] Step 2:

[1662] The server compares the generated talent with existing advertising talent. As input, it receives the text data of the generated talent and the data of the existing advertising talent. The server compares the two based on evaluation criteria (e.g., viewer response, advertising effectiveness) and determines which is more effective. As output, it obtains the comparison results.

[1663] Step 3:

[1664] The server searches for talents who can increase awareness of the service. As input, it receives the generated talent data and comparison results. Using generative AI and other technologies, the server selects talents who have the ability to increase awareness of the service or product. As output, it obtains data on the selected talents.

[1665] Step 4:

[1666] The server creates advertisements using the talent it finds. It receives data on the selected talent as input. It combines advertising copy and visual materials to create attractive advertising content. It obtains the created advertising data as output.

[1667] Step 5:

[1668] The server broadcasts the created advertisement for a certain period of time. As input, it receives advertisement data and broadcast schedule. Using the advertisement platform API, the server manages the advertisement broadcast schedule and publishes it on the media. As output, it obtains confirmation data of the start of broadcast.

[1669] Step 6:

[1670] The server collects and analyzes the response to the advertisement during the broadcast period. It receives the advertisement broadcast data as input. The server uses the advertising platform API to collect viewer response data and analyzes it using data analysis tools. It obtains the analysis result data as output.

[1671] Step 7:

[1672] The server evaluates the effectiveness and generates a report. It receives the analysis result data as input. The server evaluates the effectiveness of the advertisement based on the collected and analyzed data and compiles the results into a report. The server obtains an evaluation report as output.

[1673] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1674] "Example 1"

[1675] In one embodiment of the present invention, the generative AI includes an emotion engine that recognizes user emotions. This emotion engine analyzes emotions from user reaction data and uses the results to generate advertising talent. Specifically, it analyzes users' reactions when viewing advertisements (e.g., facial expressions, eye movements, comments on social media, etc.) to understand their emotions (happiness, surprise, interest, indifference, etc.). Then, based on this emotion data, it generates a new talent that users are likely to respond favorably to.

[1676] "Example 2"

[1677] In addition, the system analyzes users' emotional reactions to the performance of the generated advertising talent and uses the results to evaluate the advertising talent. Specifically, the emotion engine analyzes users' reactions when the generated talent appears in an advertisement, and if the reaction is good, the talent's rating is increased, and if the reaction is bad, the rating is decreased. This makes it possible to evaluate talent based on users' emotions.

[1678] "Example 3"

[1679] Furthermore, the emotion engine can adjust the parameters of emotion analysis according to the duration of the advertising campaign and market trends. For example, during certain periods, such as the Christmas season, users' emotional responses may differ from normal. In such cases, the emotion engine learns users' emotional tendencies during that period and reflects the results in the generation and evaluation of advertising talent. This enables flexible generation and evaluation of advertising talent in response to market trends.

[1680] The processing flow of each embodiment will be described below.

[1681] "Example 1"

[1682] Step 1: User watches an ad.

[1683] Step 2: The emotion engine collects user reaction data (e.g., facial expressions, eye movements, comments on social media, etc.).

[1684] Step 3: The emotion engine analyzes the user's emotions (happiness, surprise, interest, indifference, etc.) from the collected reaction data.

[1685] Step 4: Based on the analysis results, the generative AI generates new talents that users are likely to respond favorably to.

[1686] "Example 2"

[1687] Step 1: The generated talent appears in the advertisement.

[1688] Step 2: The emotion engine analyzes the user's reaction.

[1689] Step 3: If the response is positive, increase the talent's rating; if the response is negative, decrease the rating.

[1690] "Example 3"

[1691] Step 1: The sentiment engine understands the duration of the advertising campaign and market trends.

[1692] Step 2: During a specific period (e.g., Christmas season), the emotion engine learns the user's emotional tendencies during that period.

[1693] Step 3: Reflect the learning results in generating and evaluating advertising talent.

[1694] Example 1

[1695] Next, a description will be given of Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1696] Traditional methods for selecting advertising talent have made it difficult to find effective talent because they do not adequately reflect viewer reactions or advertising performance. It is also difficult to flexibly adjust talent according to the duration of the advertising campaign and market trends. This has led to the issue of not maximizing advertising effectiveness.

[1697] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1698] In this invention, the server includes means for utilizing generative AI, means for comparing with existing advertising talent at regular intervals, means for finding a talent who is more likely to recognize the service, means for utilizing the found talent, means for collecting performance data and viewer response data of the advertising talent, means for preprocessing the collected data, means for training an AI model using the preprocessed data, means for generating new talent using the trained AI model, means for displaying the generated talent and collecting user responses, and means for analyzing the collected user response data and retraining the model. This automates the advertising talent selection process and enables the generation and selection of effective talent that reflects viewer responses.

[1699] "Generative AI" is an artificial intelligence technology that generates new data and content based on existing data.

[1700] An "advertising talent" is a person or character who appears in an advertisement and whose role is to raise awareness of products and services among viewers.

[1701] "Performance data" refers to data used to measure the effectiveness of advertising, including click-through rates, viewing times, and conversion rates.

[1702] "Viewer response data" is data that records the viewer's reactions when watching an advertisement, and includes facial expressions, eye movements, comments on social media, etc.

[1703] "Preprocessing" refers to the process of converting collected data into a format suitable for analysis and learning, and includes data cleaning, normalization, and feature extraction.

[1704] "Training" is the process of teaching an AI model data so that the model can understand patterns and features in the data and make predictions or generate results from new data.

[1705] "Generation" is the process of creating new data or content based on the data that the AI ​​model has learned.

[1706] "User response" refers to changes in a user's behavior and emotions when viewing an advertisement, and is data used to evaluate the effectiveness of an advertisement.

[1707] "Retraining" is the process of retraining an AI model based on new data collected in order to improve the model's accuracy and performance.

[1708] This invention is a system for optimizing advertising talent using generative AI. Specific embodiments of this system are described below.

[1709] System configuration

[1710] This system consists of three main components: a server, a terminal, and a user.

[1711] server

[1712] The server serves as the central point for generating and evaluating advertising talent using generative AI models. The server uses the following hardware and software:

[1713] Hardware: A server machine equipped with a high-performance CPU, GPU, memory, and storage

[1714] Software: Machine learning libraries, data analysis libraries for data preprocessing, APIs for data collection

[1715] The server collects and pre-processes performance data of advertising talent and viewer response data, and uses the pre-processed data to train a generative AI model to generate new advertising talent.

[1716] Terminal

[1717] The terminal is responsible for displaying the generated new advertising talent to the user and collecting the user's reactions. The terminal uses the following hardware and software:

[1718] Hardware: A computer or smartphone with a camera, microphone, and eye-tracking device

[1719] Software: Eye tracking software, facial expression recognition software

[1720] The device records the user's reactions (facial expressions, eye movements, tone of voice, etc.) when viewing an advertisement and sends this to the server.

[1721] User

[1722] Users are responsible for watching advertisements and providing their reactions to the system, which then use the data to evaluate the advertising talent and retrain the generative AI model.

[1723] Specific examples

[1724] Here is a concrete example of how the system works:

[1725] 1. Data collection: The server collects performance data for existing advertising talents A and B. For example, it obtains data such as click-through rate, viewing time, and conversion rate. It also collects viewer response data (facial expressions, eye movements, comments on social media, etc.).

[1726] 2. Data Preprocessing: The server preprocesses the collected data, for example, using a data analysis library to impute missing values ​​and normalize the data.

[1727] 3. Training the AI ​​model: The server uses the preprocessed data to train a generative AI model (e.g., GPT-4).

[1728] 4. New talent generation: The server inputs a prompt into the trained AI model to generate a new advertising talent. For example, the prompt can be "Please generate a talent that will delight viewers."

[1729] 5. Talent Evaluation: The device displays the new talent to the user and collects their reactions, for example, by recording their facial expressions and eye movements using a webcam and eye-tracking software.

[1730] 6. Feedback collection and retraining: The server analyzes user response data collected from the device and evaluates the performance of the generative AI model. If necessary, it retrains the model.

[1731] Prompt Sentence Examples

[1732] Here are some examples of prompts:

[1733] Generate talent that your audience will enjoy.

[1734] In this way, a system is realized in which the server, terminals, and users work together to optimize advertising talent.

[1735] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1736] Step 1: Data collection

[1737] The server collects performance data of existing advertising talent and viewer response data, specifically, click-through rates, viewing times, conversion rates, etc. It also collects response data such as viewers' facial expressions, eye movements, and comments on social media.

[1738] Input: Advertising talent performance data, viewer response data

[1739] Output: Collected dataset

[1740] Step 2: Data Preprocessing

[1741] The server preprocesses the collected data by using data analysis libraries to impute missing values, remove outliers, and normalize the data, as well as extracting features and converting the data into a format suitable for training AI models.

[1742] Input: Collected dataset

[1743] Output: Preprocessed dataset

[1744] Step 3: Training the AI ​​model

[1745] The server uses the preprocessed data to train a generative AI model, specifically using a machine learning library to train the AI ​​model using existing advertising talent performance data and viewer response data as input.

[1746] Input: Preprocessed dataset

[1747] Output: A trained AI model

[1748] Step 4: Generate new talent

[1749] The server generates new advertising talent using the trained AI model. Specifically, it inputs a prompt statement into the AI ​​model to generate the characteristics of the new talent. For example, the prompt statement can be "Please generate a talent that will delight viewers."

[1750] Input: Trained AI model, prompt

[1751] Output: New advertising talent generated

[1752] Step 5: Talent Assessment

[1753] The device displays the new talent to the user and collects their reactions, using a webcam and eye-tracking software to record their facial expressions and eye movements, as well as their tone of voice using a microphone.

[1754] Input: New advertising talent generated

[1755] Output: User response data

[1756] Step 6: Gather feedback and retrain

[1757] The server analyzes user response data collected from the device and evaluates the performance of the generative AI model. If necessary, it retrains the model by preprocessing the data again and retraining the AI ​​model using the preprocessed data.

[1758] Input: User response data

[1759] Output: Retrained AI model

[1760] In this way, a system is realized in which the server, terminals, and users work together to optimize advertising talent.

[1761] (Application example 1)

[1762] Next, a description will be given of Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1763] Traditional methods for selecting advertising talent have had the problem of not being able to fully consider the emotions and reactions of viewers, making it difficult to maximize the effectiveness of advertising. Furthermore, there was a lack of a way to monitor the effectiveness of advertising campaigns in real time and quickly select the most suitable talent.

[1764] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1765] In this invention, the server includes means for utilizing generative AI, means for comparing with existing advertising talent at regular intervals, means for finding talent who can better recognize the service, means for utilizing the found talent, means for collecting viewer response data, means for analyzing viewer emotions using an emotion engine, means for generating new talent based on the analysis results, means for selecting the most suitable talent from among the generated talents, and means for monitoring the effectiveness of the advertising campaign in real time. This makes it possible to quickly select the most suitable advertising talent based on viewer emotions and reactions, and maximize the effectiveness of the advertising.

[1766] "Generative AI" is an artificial intelligence technology that generates new data and models based on existing data.

[1767] An "advertising talent" refers to a person or character who appears in an advertisement.

[1768] "Viewer response data" refers to data that indicates changes in the viewer's behavior and emotions when viewing an advertisement.

[1769] An "emotion engine" refers to software or hardware that analyzes emotions from viewers' facial expressions and behavior.

[1770] The "analysis results" are the analysis results of the viewer's emotional data obtained by the emotion engine.

[1771] "Advertising Campaign" means an advertising campaign conducted over a specific period of time. ...

Claims

1. A means for generating new advertising talent using a prompt sentence instructing the generation of a new advertising talent and the generative AI model; A means for comparing the new advertising talent generated using a generating AI with existing advertising talent to find an advertising talent that is more recognizable for at least one of a service and a product; a means for broadcasting an advertisement produced using the found advertising talent for a certain period of time; A means for analyzing an emotional response of a user who viewed the broadcasted advertisement using an emotion engine; means for evaluating the effectiveness of the aired advertisements based on the analyzed emotional responses of the users; A system including:

2. The system of claim 1 , wherein the generating AI generates the new advertising talent based on an evaluation result of the effectiveness of the aired advertisement.

3. The system of claim 1 , further comprising means for adjusting parameters for analyzing said emotional response depending on at least one of the timing of said period and market trends.

Images