System

Abstract

A system is provided.SOLUTION: A system comprising: means for generating a video for an advertisement using a generative model; means for analyzing big data and generating a scenario of the advertisement; means for operating the generated video and measuring an effect thereof; and means for optimizing operation of the advertisement based on a result of the measurement.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] As the video advertising market expands, there is a demand for efficient ad production and operation. However, producing video ads is costly and time-consuming, and it is difficult to guarantee quality. Furthermore, optimizing ad operation to maximize effectiveness is difficult. These issues pose a major problem for corporate clients and digital marketing departments, resulting in unnecessary resource waste. To solve these issues, there is a need to utilize generative models and big data to consistently improve efficiency from production to operation. [Means for solving the problem]

[0005] The present invention provides a system that includes a means for generating videos for advertising using a generative model and a means for generating advertising scenarios by analyzing big data. It also includes a means for managing the generated videos and measuring their effectiveness, and a means for optimizing advertising management based on the measurement results. The system further includes a means for modifying the videos for advertising using a generative model and a means for improving the advertising scenarios based on big data analysis, thereby achieving higher quality video advertising production. The system also includes a means for receiving advertising requirements from a user, providing the user with a video generated based on the received requirements, receiving feedback, and a means for improving and regenerating the video based on the feedback, thereby achieving efficient and effective advertising management.

[0006] A "generative model" refers to an algorithm or program that uses AI to generate new data from specific input data.

[0007] An "advertising scenario" refers to the design of the message, storyline, and scene flow used in an advertisement.

[0008] "Big data" refers to large amounts of diverse data, and useful information and patterns can be extracted by analyzing that data.

[0009] "Video operation" refers to the activity of actually distributing the generated video advertisements on the platform and measuring their effectiveness.

[0010] "Measuring effectiveness" refers to collecting and analyzing performance data such as the number of times an ad is displayed, click-through rate, and conversion rate.

[0011] "Optimizing operations" refers to adjusting the delivery method and content of advertising based on the results of effectiveness measurements, and making improvements to maximize effectiveness.

[0012] "Feedback" refers to opinions, evaluations, and information requesting improvements provided by users or customers. [Brief explanation of the drawings]

[0013] [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. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0014] 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.

[0015] First, the terms used in the following description will be explained.

[0016] 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, a 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), and an APU (Accelerated Processing Unit).

[0017] 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.

[0018] 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.

[0019] 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.

[0020] 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."

[0021] [First embodiment]

[0022] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0023] 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.

[0024] 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).

[0025] 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.

[0026] 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.

[0027] 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.

[0028] 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.

[0029] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0030] 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.

[0031] 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.

[0032] 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.

[0033] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. 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."

[0034] The present invention provides a system for efficiently implementing the entire process from video advertising production to operation, and specific embodiments thereof are described below. This system integrates generative models, big data analysis, feedback processing, and optimization functions to effectively produce and operate video advertising.

[0035] System configuration

[0036] The system consists of the following main components:

[0037] 1. Server

[0038] Generative Model

[0039] Big Data Analysis Unit

[0040] Effectiveness Measurement Unit

[0041] Optimization Unit

[0042] 2. Terminal

[0043] User Interface

[0044] Dashboard

[0045] 3. Users

[0046] Advertisement requirements provided

[0047] Providing Feedback

[0048] Program processing flow

[0049] Video ad requirements definition

[0050] User:

[0051] Users first log in to the portal site and enter the requirements for a new advertising campaign, including target demographics, product information, advertising objectives, etc. After entering the requirements, users can provide detailed information by clicking the "Submit" button to send them to the server.

[0052] server:

[0053] The server analyzes the requirements data received from the user and performs preprocessing before passing it to the Generative Model. This preprocessing converts the requirements data into an appropriate format so that the Generative Model can operate efficiently.

[0054] Scenario generation and video generation

[0055] server:

[0056] The server passes the preprocessed data to a generative model to generate an advertising scenario. Based on the generated scenario, the server collects additional materials such as images, audio, and text, and uses the generative model to automatically generate a video advertisement. The generated video is then encoded for initial review, and a preview link is generated.

[0057] User:

[0058] The user can review the generated video by checking the provided preview link. If feedback is required, the user can enter comments or corrections on the platform and send them back to the server.

[0059] Improving video quality and preparing for operation

[0060] server:

[0061] The server receives feedback from users and analyzes the big data to improve the advertising scenario. It uses patterns obtained from the big data to reflect improvements to the scenario. Based on the improved scenario, it encodes the regenerated video and prepares it for deployment. Once the deployment is ready, the server sends a notification to the user.

[0062] Device:

[0063] The terminal notifies the user that the operation preparation is complete. The user confirms the contents and clicks the "Start operation" button to start the advertisement operation.

[0064] Efficacy monitoring and optimization

[0065] server:

[0066] As soon as the server starts running the ad, it collects performance data such as the number of impressions, click-through rates, and number of conversions in real time. The collected data is analyzed by an effectiveness measurement unit, and if performance is poor, the ad is regenerated and optimized using a generative model. The server also adjusts the timing and targeting of ad displays as appropriate.

[0067] Examples:

[0068] Suppose a company creates an advertising video for a new smartphone product and starts running it. If the click-through rate on the first day is low, the server uses a generative model to improve the video and generates a new scenario based on big data analysis. Then, it changes the timing of the ad display to the optimal time and runs it again.

[0069] Measurement and reporting

[0070] server:

[0071] Once the ad run period is over, the server aggregates all data and measures its effectiveness. A detailed report summarizing the analysis results is generated and provided to the user. The report includes detailed results for each performance indicator and insights for future ad runs.

[0072] User:

[0073] Users can receive the provided report, check the results of their advertising operations, and plan their next advertising strategy based on the report contents.

[0074] The above is a specific description of the embodiment of the present invention. The present system enables users to efficiently create and operate high-quality, effective video advertisements.

[0075] The processing flow will be explained below.

[0076] Step 1:

[0077] User:

[0078] Users log in to the portal site and enter the requirements for a new advertising campaign, providing detailed information such as target demographic, product information, and advertising objectives (e.g., brand awareness, increased conversions). They then confirm the information they entered and click the "Submit" button.

[0079] Step 2:

[0080] server:

[0081] The server analyzes the requirements data received from the user and performs preprocessing before passing it to the generative model. Preprocessing converts the requirements data into an appropriate format so that the generative model can operate efficiently.

[0082] Step 3:

[0083] server:

[0084] The server passes the preprocessed data to a generative model to generate advertising scenarios. The generative model automatically generates effective advertising scenarios based on the input data.

[0085] Step 4:

[0086] server:

[0087] Based on the generated scenario, the server collects materials such as images, audio, and text. Using these materials, the generative model automatically creates a video ad. The generated video is encoded for early review and a preview link is generated.

[0088] Step 5:

[0089] User:

[0090] The user can review the generated video content by checking the provided preview link, and provide feedback and input corrections or additional requirements as needed.

[0091] Step 6:

[0092] server:

[0093] The server receives feedback from users and analyzes the big data to improve the advertising scenario. It utilizes patterns obtained from the big data to improve the scenario.

[0094] Step 7:

[0095] server:

[0096] Based on the improved scenario, the generative model is used to regenerate the video ad. The regenerated video is then encoded and made ready for deployment. Once ready, the server sends a notification to the user.

[0097] Step 8:

[0098] Device:

[0099] The device notifies the user that it is ready for operation. The user confirms the information and clicks the "Start operation" button to start advertising operations.

[0100] Step 9:

[0101] server:

[0102] As soon as the ad starts running, the server collects performance data in real time, such as the number of impressions, click-through rates, and conversions, which are then analyzed by the effectiveness measurement unit.

[0103] Step 10:

[0104] server:

[0105] If performance is poor, the server uses the generative model to regenerate and optimize the ads, adjusting the timing and targeting of the ads accordingly.

[0106] Step 11:

[0107] server:

[0108] At the end of the advertising period, the server aggregates all data and measures the effectiveness. Based on the analysis results, a detailed report is generated and provided to the user.

[0109] Step 12:

[0110] User:

[0111] Users can receive the provided report, check the results of their advertising operations, and plan their next advertising strategy based on the report contents.

[0112] The above are the specific processing steps for carrying out the present invention. This system allows users to efficiently create and operate high-quality, effective video advertisements.

[0113] Example 1

[0114] Next, a description will be given of 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."

[0115] Traditional advertising production and management processes are time-consuming, resource-intensive, and inefficient. In particular, they lack flexibility and make it difficult to measure and optimize advertising effectiveness in real time. Another issue is the difficulty of quickly incorporating user feedback and quickly improving advertising content.

[0116] 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.

[0117] In this invention, the server includes means for generating videos for advertisements using a generative AI model, means for analyzing large-scale data to generate advertising scenarios, means for operating the generated videos and measuring their effectiveness, means for optimizing the operation of advertisements based on the measurement results, means for receiving and preprocessing advertising requirements from users, means for receiving user feedback on the generated videos, and means for improving and regenerating the videos based on the feedback. This streamlines the process of advertisement production and operation, enables real-time measurement and optimization of effectiveness, and enables rapid improvement that reflects user feedback.

[0118] A "generative AI model" is an artificial intelligence model that automatically generates advertising videos and scenarios based on specified prompt text.

[0119] "Large-scale data" refers to large amounts of data, such as advertising operations and user feedback, and is a collection of information obtained by analyzing this data.

[0120] "Scenario generation" is the process of designing the structure, content, and storyboard of an advertising video.

[0121] "Effectiveness measurement" is the process of collecting data such as the number of impressions, click rates, and number of conversions obtained during advertising operations and evaluating the performance of the advertisements.

[0122] "Optimization" refers to the process of adjusting advertising operations based on the results of effectiveness measurement to improve performance.

[0123] "User feedback" refers to comments and correction instructions given by the user to the generated advertising video.

[0124] "Regeneration" is the process of recreating advertising videos and scenarios based on feedback and effectiveness measurement results.

[0125] "Preprocessing" is the initial process of analyzing the advertising requirement data received from the user and converting it into a format that is easy for the generative AI model to process.

[0126] The present invention is a system for efficiently implementing the entire process from video advertising production to operation, and specific embodiments thereof are described below. This system integrates generative AI models, large-scale data analysis, feedback processing, and optimization functions to effectively produce and operate video advertising.

[0127] System configuration

[0128] The system consists of the following main components:

[0129] 1. Server

[0130] Generative AI Models

[0131] Large-Scale Data Analysis Unit

[0132] Effectiveness Measurement Unit

[0133] Optimization Unit

[0134] 2. Terminal

[0135] User Interface

[0136] Dashboard

[0137] 3. Users

[0138] Advertisement requirements provided

[0139] Providing Feedback

[0140] Video ad requirements definition

[0141] User:

[0142] Users log in to the portal site and enter the requirements for a new advertising campaign. These requirements include target demographics, product information, advertising objectives, etc., and users can provide detailed information. For example, they can enter a requirement such as "Raising awareness of a new headphone for music lovers aged 25-35." After entering the information, users click the "Submit" button to send it to the server.

[0143] server:

[0144] The server analyzes the requirements data received from the user and performs preprocessing before passing it to the generative AI model. This preprocessing converts the requirements data into an appropriate format, allowing the generative AI model to operate efficiently.

[0145] Scenario generation and video generation

[0146] server:

[0147] The server then passes the preprocessed data to a generative AI model to generate an advertising scenario. For example, it uses a prompt such as, "Generate a scenario that will resonate with music lovers aged 25-35." Based on the generated scenario, the server collects additional materials such as images, audio, and text, and uses the generative model to automatically generate a video ad. The generated video is then encoded for initial review, and a preview link is generated.

[0148] User:

[0149] The user opens the provided preview link to check the content of the generated video, and if necessary, enters comments or corrections on the platform and sends it back to the server.

[0150] Improving video quality and preparing for operation

[0151] server:

[0152] The server receives user feedback, analyzes large-scale data, and refines the advertising scenario. It uses patterns extracted from the large-scale data to improve the scenario. Based on the refined scenario, the video is regenerated and ready for deployment.

[0153] Device:

[0154] The device will notify the user that it is ready for operation. The user can confirm the details and click the "Start operation" button to start advertising operations.

[0155] Efficacy monitoring and optimization

[0156] server:

[0157] As soon as the server starts running the ad, it collects performance data such as the number of impressions, click-through rates, and number of conversions in real time. The collected data is analyzed by an effectiveness measurement unit, and if performance is poor, the ad is regenerated and optimized using a generative model. The timing and targeting of ad displays are also adjusted as appropriate.

[0158] Examples:

[0159] For example, if a company is running an ad for a new smartphone product and the click-through rate on the first day is low, the server will use the generative model to generate new scenarios, improve the video, optimize the timing of the landing page, and adjust the display time of the ad.

[0160] Measurement and reporting

[0161] server:

[0162] Once the ad run period is over, the server aggregates all data and measures its effectiveness. A detailed report summarizing the analysis results is generated and provided to the user. The report includes detailed results for each performance indicator and insights for future ad runs.

[0163] User:

[0164] Users can receive the provided report, check the results of their advertising operations, and plan their next advertising strategy based on the report contents.

[0165] This system streamlines the video ad production and operation process, enabling real-time measurement and optimization of effectiveness. It also allows for quick reflection of user feedback, making it possible to constantly improve ad content to keep it up-to-date and optimal.

[0166] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0167] Step 1: Enter user requirements

[0168] User:

[0169] Input: Users log in to the portal site and enter the requirements for a new advertising campaign, providing detailed information in text format, such as the target demographic (e.g., "music lovers aged 25-35"), product information, and advertising objectives (e.g., "increase awareness of a new product").

[0170] Output: User-entered requirements data.

[0171] Action: The user clicks the "Submit" button to send the entered requirements data to the server.

[0172] Step 2: Preprocessing requirements data

[0173] server:

[0174] Input: Requirement data submitted by the user.

[0175] Output: Preprocessed requirements data (structured data).

[0176] How it works: The server parses the received requirements data and preprocesses it to be passed to the generative AI model. Specifically, it structures the text data and splits it into necessary fields.

[0177] Step 3: Scenario generation

[0178] server:

[0179] Input: Preprocessed requirements data.

[0180] Output: The generated advertising scenario.

[0181] How it works: The server inputs the preprocessed data into a generative AI model and generates advertising scenarios using prompts, such as "Generate a scenario that will resonate with music lovers aged 25-35."

[0182] Step 4: Gathering materials and creating a video

[0183] server:

[0184] Input: The generated ad scenario.

[0185] Output: The generated video ad file.

[0186] How it works: Based on the generated scenario, the server collects the necessary images, audio, and text materials, and automatically generates a video ad using a generative AI model. The generated video file is encoded and a preview link is generated.

[0187] Step 5: Video preview and feedback

[0188] User:

[0189] Input: The video generated using the preview link.

[0190] Output: User feedback (comments and suggestions for corrections).

[0191] How it works: The user opens the provided preview link, checks the generated video, and optionally enters comments and corrections on the platform, which then sends the video back to the server.

[0192] Step 6: Receive and analyze feedback

[0193] server:

[0194] Input: User feedback.

[0195] Output: The scenario with the improvements.

[0196] How it works: The server receives feedback from users and analyzes it using a large-scale data analysis unit. Based on the analysis results, it extracts areas for improvement in the scenario and generates a new scenario using a generative AI model.

[0197] Step 7: Scenario refinement and video regeneration

[0198] server:

[0199] Input: The scenario with the improvements.

[0200] Output: Regenerated video ad files.

[0201] Operation: The server gathers additional material and regenerates the video based on the improved scenario. The regenerated video file is then re-encoded and ready for production.

[0202] Step 8: Notification of operational readiness

[0203] server:

[0204] Input: The regenerated video ad file.

[0205] Output: Operational readiness notification.

[0206] Action: The server notifies the user that it is ready for operation.

[0207] Step 9: Go live

[0208] Device:

[0209] Input: Operational readiness notification.

[0210] Output: Advertising videos are launched.

[0211] Operation: The device displays a notification to the user that preparation for operation is complete. The user clicks the "Start operation" button to start ad operation.

[0212] Step 10: Collecting performance measurement data

[0213] server:

[0214] Input: Ad video performance data (number of impressions, click-through rate, number of conversions, etc.).

[0215] Output: Performance measurement data.

[0216] How it works: The server starts running ads and collects real-time performance data such as impressions, click-through rates, and conversions.

[0217] Step 11: Effectiveness analysis and optimization

[0218] server:

[0219] Input: Performance measurement data.

[0220] Output: Improved and optimized ad video.

[0221] How it works: The server analyzes the collected data using an effectiveness measurement unit, and if performance is poor, it uses a generative model to regenerate ads, adjusting the timing and targeting of ads accordingly.

[0222] Step 12: Measurement and reporting

[0223] server:

[0224] Input: Aggregated performance measurement data.

[0225] Output: Detailed performance measurement report.

[0226] How it works: After the ad campaign period ends, the server aggregates all data, measures the effectiveness, and generates a detailed report summarizing the analysis results, which is then provided to the user.

[0227] Step 13: Review reports and develop strategies

[0228] User:

[0229] Input: Detailed performance measurement report.

[0230] Output: Next advertising strategy.

[0231] How it works: Users can receive the provided report, check the results of their advertising campaigns, and plan their next advertising strategy based on the report.

[0232] (Application example 1)

[0233] Next, a description will be given of Application 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."

[0234] Traditional video ad production systems have the problem that the processes of ad creation, operation, effectiveness measurement, and optimization are fragmented, making efficient operation difficult. Furthermore, there is a lack of a system that can optimize ads in real time and immediately reflect user feedback. As a result, advertising effectiveness is reduced and user satisfaction is low.

[0235] 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.

[0236] In this invention, the server includes a means for generating videos for advertisements using a generative model, a means for analyzing big data to generate advertising scenarios, a means for operating the generated videos and measuring their effectiveness, a means for optimizing the operation of advertisements based on the measurement results, a means for displaying the optimized video advertisements in real time using smart glasses, and a means for regenerating and optimizing advertisements based on collected data. This enables the entire process from the production of video advertisements to their operation and feedback to be integrated and operated efficiently and effectively.

[0237] A "generative model" is an algorithm that automatically generates scenarios and videos for advertising.

[0238] "Big data" refers to large amounts of data, a group of data that is used to generate and optimize advertising scenarios.

[0239] An "advertising scenario" is a plan that defines the content and progression of a video advertisement.

[0240] "Operation" refers to the process of actually delivering the generated video advertisements and displaying them to users.

[0241] "Effectiveness measurement" is the process of evaluating the results of advertising and collecting and analyzing performance data.

[0242] "Optimization" refers to making adjustments based on collected data to improve the performance of your ads.

[0243] "Smart glasses" are a type of wearable device that visually displays information to the user while they are wearing them.

[0244] "Real-time" means that processing and reaction are immediate and based on the current time zone.

[0245] "Feedback" means reviews and correction instructions provided by users.

[0246] An "interface" is a point of contact or means by which a user interacts with a system.

[0247] "Data" refers to a set of information or numbers that a system processes.

[0248] "Regeneration" is the process of creating a new version of an ad based on previously generated content.

[0249] The present invention provides a system for efficiently implementing the entire process from video advertising production to operation, and is characterized by its ability to display optimized video advertising in real time using smart glasses. Specific embodiments are described below. The system integrates generative models, big data analysis, feedback processing, and optimization functions to effectively produce and operate video advertising.

[0250] System configuration

[0251] The system consists of the following main components:

[0252] 1. Server:

[0253] Generative model: An algorithm that automatically generates scenarios and videos for advertising.

[0254] Big Data Analysis Unit: A data set used to analyze large amounts of data and measure and optimize advertising effectiveness.

[0255] Effectiveness Measurement Unit: Measures the effectiveness of advertising and analyzes the collected data.

[0256] Optimization unit: Optimizes your ads based on the data collected to improve their performance.

[0257] 2. Terminal:

[0258] Smart glasses: Glasses that visually display information while worn by the user (e.g., Google Glass, Vuzix Blade).

[0259] User Interface: An interface for entering ad requirements, previewing, and providing feedback.

[0260] Dashboard: A management screen for displaying advertising results and effectiveness measurement results.

[0261] 3. User:

[0262] Provide Ad Requirements: Enter your requirements for a new ad campaign.

[0263] Providing feedback: Review the content of the generated video and provide feedback.

[0264] System Operation

[0265] 1. Video ad requirements definition:

[0266] Users input their advertising requirements through the smart glasses interface, including target demographic, product information, and advertising objectives, and then the requirements are sent to the server.

[0267] 2. Scenario and video generation:

[0268] The server analyzes the received requirements data and passes it to the generative model. The generative model generates a scenario, collects the necessary materials (images, audio, text), and automatically generates a video advertisement. The generated video is encoded for initial review and streamed to the smart glasses.

[0269] 3. Viewing videos and providing feedback:

[0270] Users can view the generated video through smart glasses and provide feedback using voice commands or touch gestures, which is then sent to the server, which regenerates and optimizes the video as needed.

[0271] 4. Effect monitoring and optimization:

[0272] The server monitors the effectiveness of the ad (number of views, viewing time, user response) in real time. If the effectiveness is low, it will be regenerated and optimized to improve the ad performance.

[0273] 5. Report Generation:

[0274] The server aggregates all data, measures effectiveness, and generates detailed reports to provide users with insights to help them plan their next advertising strategy.

[0275] Specific examples

[0276] For example, consider a company that is creating and running an advertising video for a new smartphone product. The target demographic is technology-conscious people in their 20s and 30s, and the goal of the ad is to stimulate purchasing desire and increase brand awareness. Based on this requirement, the following prompt sentences are input into the generative AI model:

[0277] Prompt statement:

[0278] The target demographic for a new smartphone is tech-savvy people in their 20s and 30s. The advertising objective is to stimulate purchase intent and increase brand awareness. Based on this requirement, generate the following scenario and creative. The ad should include high-performance camera features, a long battery life, and a seamless user experience.

[0279] As described above, this system allows users to view high-quality, effective video advertisements in real time using smart glasses, and allows for immediate feedback.

[0280] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0281] Step 1: User Input

[0282] Through the smart glasses' user interface, users input requirements for a new advertising campaign, including target demographic, product information, and advertising objectives. The input is made on the interface and transmitted to the server. The input data also specifies specific elements and conditions that should be included in the advertisement.

[0283] Input: target demographic, product information, advertising objectives

[0284] Output: Requirements data sent to the server

[0285] Step 2: Requirement Data Analysis

[0286] The server analyzes the requirements data received from the user, and the analysis results are converted into a format that can be passed to the generative AI model. This preprocessing allows the generative model to operate efficiently.

[0287] Input: Requirement data

[0288] Output: Data converted into a format to be passed to the generative model

[0289] Step 3: Scenario generation

[0290] The server passes the preprocessed data to the generative model to generate an advertising scenario. The generative model constructs a scenario based on the prompt sentence and determines the story of the advertising video.

[0291] Input: Preprocessed data, generative AI model

[0292] Output: Generated scenario

[0293] Step 4: Gathering materials and creating a video

[0294] The server collects materials such as images, audio, and text based on the generated scenario, then uses the generative model to automatically generate advertising videos, which are then encoded and prepared for streaming to the smart glasses.

[0295] Input: Generated scenario, material data

[0296] Output: Generated video ad

[0297] Step 5: Advertising and Feedback

[0298] The user watches the generated video advertisement through the smart glasses. If feedback is required, the user can provide it using voice commands or touch gestures. The feedback is then sent to the server.

[0299] Input: Generated video ad

[0300] Output: User feedback

[0301] Step 6: Feedback analysis and video regeneration

[0302] The server analyzes the feedback received from the user, refines the content of the video advertisement based on the feedback, and regenerates the video, which is then re-encoded and streamed to the smart glasses.

[0303] Input: User feedback

[0304] Output: Regenerated video ad

[0305] Step 7: Measure your results

[0306] The server starts running the ad and simultaneously collects viewing data (number of views, viewing time, user response) in real time. The collected data is analyzed by the effectiveness measurement unit and the ad is optimized as necessary.

[0307] Input: Viewing data

[0308] Output: Effectiveness measurement report, optimization instructions

[0309] Step 8: Generate reports

[0310] After the ad run period ends, the server aggregates all data and generates a detailed performance measurement report, which is provided to the user and contains insights for planning their next advertising strategy.

[0311] Input: Aggregated operational data

[0312] Output: Performance measurement report

[0313] Through the above process, the present invention makes it possible to efficiently and effectively produce, operate, collect feedback, and optimize advertising videos.

[0314] 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.

[0315] The present invention is a system for efficiently and effectively implementing the entire process from the production to the operation of video advertisements, and in particular, by combining an emotion engine that recognizes user emotions, it achieves a high level of personalization. Below, we will explain the form for implementing the present invention and the specific processing content thereof.

[0316] System configuration

[0317] The system consists of the following main components:

[0318] 1. Server

[0319] Generative Model

[0320] Big Data Analysis Unit

[0321] Effectiveness Measurement Unit

[0322] Optimization Unit

[0323] Emotion Engine

[0324] 2. Terminal

[0325] User Interface

[0326] Emotion Recognition Interface

[0327] Dashboard

[0328] 3. Users

[0329] Advertisement requirements provided

[0330] Providing Feedback

[0331] Program processing flow

[0332] Video ad requirements definition

[0333] User:

[0334] First, users log in to the portal site and enter the requirements for a new advertising campaign, providing detailed information such as target demographic, product information, and advertising objectives (e.g., brand awareness, increased conversions). They then confirm the information they entered and click the "Submit" button.

[0335] server:

[0336] The server analyzes the requirements data received from the user and performs preprocessing before passing it to the Generative Model. This preprocessing converts the requirements data into an appropriate format so that the Generative Model can operate efficiently.

[0337] Scenario generation and video generation

[0338] server:

[0339] The server passes the preprocessed data to a generative model to generate advertising scenarios. The generative model automatically generates effective advertising scenarios based on the input data.

[0340] server:

[0341] Based on the generated scenario, the server collects materials such as images, audio, and text. Using these materials, the generative model automatically creates a video ad. The generated video is encoded for early review and a preview link is generated.

[0342] User:

[0343] The user can review the generated video content by checking the provided preview link, and provide feedback and input corrections or additional requirements as needed.

[0344] Improving video quality and preparing for operation

[0345] server:

[0346] The server receives feedback from users and analyzes the big data to improve the advertising scenario. It utilizes patterns obtained from the big data to improve the scenario.

[0347] server:

[0348] Based on the improved scenario, the generative model is used to regenerate the video ad. The regenerated video is then encoded and made ready for deployment. Once ready, the server sends a notification to the user.

[0349] Device:

[0350] The device notifies the user that it is ready for operation. The user confirms the information and clicks the "Start operation" button to start advertising operations.

[0351] Effect monitoring, optimization, and emotion recognition

[0352] server:

[0353] As soon as the ad starts running, the server collects performance data in real time, such as the number of impressions, click-through rates, and conversions, which are then analyzed by the effectiveness measurement unit.

[0354] server:

[0355] The emotion engine recognizes emotions in real time from facial expressions and voice when a user views an ad, and dynamically adjusts the content of the ad based on that data. For example, if a user has a happy expression while viewing an ad, the engine can add more engaging elements to match that positive emotion.

[0356] server:

[0357] If performance is poor, the server uses the generative model to regenerate and optimize the ads, adjusting the timing and targeting of the ads accordingly.

[0358] Examples:

[0359] Suppose a company creates and starts using an advertising video for a new smartphone product. If many users who watch the ad feel uninterested, the server uses an emotion engine to analyze the emotion data, improves the advertising scenario, and regenerates it to make it more interesting.

[0360] Measurement and reporting

[0361] server:

[0362] Once the ad campaign period ends, the server aggregates all data and measures its effectiveness. Based on the analysis results, a detailed report is generated and provided to the user. The report includes detailed results for each performance indicator and insights for future ad campaigns.

[0363] User:

[0364] Users can receive the provided report, check the results of their advertising operations, and plan their next advertising strategy based on the report contents.

[0365] The above is a specific description of an embodiment of the present invention. This system allows users to efficiently create and manage high-quality, effective video advertisements. Furthermore, by utilizing the emotion engine, it is possible to achieve a high level of personalization of advertisements and provide each user with an optimal advertising experience.

[0366] The processing flow will be explained below.

[0367] Step 1:

[0368] User:

[0369] Users log in to the portal site and enter the requirements for a new advertising campaign, including detailed information such as target demographic, product information, and advertising objectives (e.g., brand awareness, increased conversions). They confirm the information they entered and click the "Submit" button.

[0370] Step 2:

[0371] server:

[0372] The server analyzes the requirements data received from the user and performs preprocessing before passing it to the Generative Model. This preprocessing converts the provided data into an appropriate format so that the Generative Model can operate efficiently.

[0373] Step 3:

[0374] server:

[0375] The server passes the preprocessed data to a generative model to generate advertising scenarios. The generative model automatically generates effective advertising scenarios based on the input data.

[0376] Step 4:

[0377] server:

[0378] Based on the generated scenario, the server collects materials such as images, audio, and text. Using these materials, the generative model automatically creates a video ad. The generated video is encoded for early review and a preview link is generated.

[0379] Step 5:

[0380] User:

[0381] The user can review the generated video content by checking the provided preview link, and provide feedback and input corrections or additional requirements as needed.

[0382] Step 6:

[0383] server:

[0384] The server receives feedback from users and analyzes the big data to improve the advertising scenario. It utilizes patterns obtained from the big data to improve the scenario.

[0385] Step 7:

[0386] server:

[0387] Based on the improved scenario, the generative model is used to regenerate the video ad. The regenerated video is then encoded and made ready for deployment. Once ready, the server sends a notification to the user.

[0388] Step 8:

[0389] Device:

[0390] The device notifies the user that it is ready for operation. The user confirms the information and clicks the "Start operation" button to start advertising operations.

[0391] Step 9:

[0392] server:

[0393] As soon as the ad starts running, performance data such as the number of impressions, click-through rate, and number of conversions is collected in real time, and the collected data is analyzed by the effectiveness measurement unit.

[0394] Step 10:

[0395] server:

[0396] The emotion engine recognizes emotions in real time from facial expressions and voice when a user views an advertisement, and dynamically adjusts the content of the advertisement based on that data. For example, if a user is not interested in an advertisement, the scenario will be improved by reflecting that emotional data.

[0397] Step 11:

[0398] server:

[0399] If performance is poor, the server uses the generative model to regenerate and optimize the ads, adjusting the timing and targeting of the ads accordingly.

[0400] Step 12:

[0401] server:

[0402] Once the advertising period ends, the server aggregates all data and measures the effectiveness. Based on the analysis results, a detailed report is generated and provided to the user.

[0403] Step 13:

[0404] User:

[0405] Users can receive the provided report, check the results of their advertising operations, and plan their next advertising strategy based on the report contents.

[0406] The above are the specific processing steps for implementing the present invention in combination with an emotion engine. This system allows users to efficiently create and manage high-quality, effective video advertisements, and by utilizing the emotion engine, highly personalized advertisements can be achieved.

[0407] Example 2

[0408] Next, a description will be given of 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."

[0409] Conventional advertising video production and operation systems lacked personalization, making it difficult to optimize by incorporating each user's emotions and real-time feedback. Furthermore, the process of measuring effectiveness and analyzing feedback was complicated, making it difficult to respond quickly to improve advertising effectiveness. This resulted in problems such as a lack of improvement in user experience and reduced advertising effectiveness.

[0410] 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.

[0411] In this invention, the server includes means for generating videos for advertisements using a generative model, means for analyzing big data to generate advertising scenarios, means for operating the generated videos and measuring their effectiveness, means for recognizing emotions from a user's facial expressions and voice and dynamically adjusting advertising content based on the emotion data, means for analyzing the content of feedback and regenerating videos based on the analysis results, and means for aggregating the results of measuring the effectiveness of advertising operation, generating detailed reports, and providing them to users. This makes it possible to highly personalize advertisements and optimize them in real time.

[0412] A "generative model" is a machine learning model that uses algorithms to automatically generate content.

[0413] "Big data" is a collection of large amounts of diverse data, the analysis of which makes it possible to discover new insights and patterns.

[0414] An "advertising scenario" is a detailed design document for planning the composition and development of an advertising video.

[0415] "Emotion recognition" is a technology that analyzes non-verbal information such as a user's facial expressions and voice to recognize their emotional state.

[0416] "Dynamic adjustment" is the process of instantly changing advertising content based on real-time data.

[0417] "Feedback" refers to the evaluations and opinions provided by users, and the system is improved and adjusted based on them.

[0418] "Effectiveness measurement" is the process of collecting and analyzing data to verify whether advertising is having the intended effect.

[0419] A "report" is a document that summarizes the results of effectiveness measurement and analysis, and is used as a guideline for the next strategy planning and improvement.

[0420] "Advertising Requirements" are detailed instructions that set out the conditions and goals necessary to run an advertising campaign.

[0421] "Operational optimization" refers to adjustments and improvement activities carried out to maximize advertising operational performance.

[0422] "Performance Data" means data used to evaluate the results of advertising operations, such as the number of times an advertisement is displayed, click-through rate, and number of conversions.

[0423] The present invention provides a video advertising production and management system that combines an emotion engine that recognizes user emotions to achieve advanced personalization. The following describes an embodiment of the system.

[0424] System configuration

[0425] This system consists of a server, a terminal, and a user. The specific hardware and software used are as follows:

[0426] 1. Server:

[0427] Generative model: Uses OpenAI's GPT model.

[0428] Big Data Analytics Unit: Uses AWS SageMaker and Google Cloud BigQuery.

[0429] Performance Measurement Unit: A Python script for collecting and analyzing performance data.

[0430] Optimization unit: A script that adjusts ad operations.

[0431] Emotion Engine: Emotion recognition software for recognizing the user's facial expressions and voice.

[0432] 2. Terminal:

[0433] User interface: Built with HTML / CSS and JavaScript, accepts input from the user.

[0434] Emotion recognition interface: Recognizes the user's emotions using a camera and microphone.

[0435] Dashboard: An interface that displays operational data and effectiveness measurement results.

[0436] 3. User:

[0437] Provide your advertising requirements: Enter your advertising campaign details.

[0438] Providing feedback: Provide feedback such as your ratings and opinions on the provided advertising video.

[0439] System operation explanation

[0440] Video ad requirements definition

[0441] User: Logs in to the portal site and enters details such as the target demographic, product information, and advertising objectives of the advertising campaign. For example, the user might enter, "The target demographic is young people in their 20s, and the purpose is to raise awareness of a new smartphone product."

[0442] Scenario generation and video generation

[0443] Server: Analyzes the requirements data sent by the user and passes it to the generative model. A Python script is used for this preprocessing. The generative model (e.g., GPT-3) generates effective advertising scenarios.

[0444] Server: Based on the generated scenario, images, audio, and text materials are collected and a video ad is automatically generated using software such as Adobe Premiere Pro or Final Cut Pro. The generated video is then encoded and a preview link is generated.

[0445] User: Check the generated video via the preview link and provide feedback if necessary. For example, "I'd like the user's voice to be more prominent in the final scene."

[0446] Improving video quality and preparing for operation

[0447] Server: Receives user feedback, analyzes the data using the big data analysis unit, refines the scenario based on the analysis results, and regenerates the video ad using the generative model.

[0448] Server: Encodes the regenerated video and notifies the user that it is ready for production.

[0449] Terminal: The device will notify the user that it is ready for operation, and the user can click the "Start operation" button to start advertising operations.

[0450] Effect monitoring, optimization, and emotion recognition

[0451] Server: Once the ad is launched, real-time performance data such as impressions, click-through rates, and conversions are collected.

[0452] Server: The emotion engine recognizes emotions from facial expressions and voice when users view advertisements, and dynamically adjusts the content of advertisements based on that data.

[0453] Server: Uses generative models to regenerate and optimize ads when performance is poor, and also dynamically adjusts ad timing and targeting.

[0454] Measurement and reporting

[0455] Server: After the advertising period ends, all data is aggregated and effectiveness is measured. A detailed report is generated based on the analysis results and provided to the user.

[0456] Users: They can review the provided reports and plan their next advertising strategy. For example, they can decide to target women in their early 20s in their next campaign.

[0457] Example prompts to input to the generative AI model

[0458] "The target demographic is young people in their 20s, and the goal is to raise awareness of a new smartphone product. Please generate an effective advertising scenario for this target demographic."

[0459] This system allows users to efficiently create and operate high-quality, effective video ads. In particular, by utilizing the emotion engine, it is possible to provide the optimal advertising experience for each user.

[0460] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0461] Step 1:

[0462] User: Enter your advertising requirements

[0463] Users log in to the portal site and enter the target demographic, product information, and purpose of the ad campaign. For example, they enter specific information such as "The target demographic is young people in their 20s, and the purpose is to raise awareness of a new smartphone product," and click the submit button. The entered data is sent to the server.

[0464] Input: Advertising requirement data

[0465] Output: Ad requirements data sent to the server

[0466] Step 2:

[0467] Server: Preprocessing requirements data

[0468] The server parses the ad requirement data received from users and converts it into a format suitable for the generative model. This preprocessing is done using Python scripts, such as cleaning and formatting the text data.

[0469] Input: Ad requirements data sent by the user

[0470] Output: Data converted into a format suitable for feeding into a generative model

[0471] Step 3:

[0472] Server: Automatic scenario generation

[0473] The server passes the preprocessed data to a generative model, which automatically generates advertising scenarios. Here, a generative AI model (e.g., GPT-3) is used. The generated scenarios are stored on the server.

[0474] Input: Data converted to a format suitable for feeding into a generative model

[0475] Output: Generated advertising scenario

[0476] Step 4:

[0477] Server: Collects advertising materials and generates videos

[0478] The server collects the necessary image, audio, and text materials based on the generated scenario. Using the collected materials, the video ad is automatically generated using software such as Adobe Premiere Pro or Final Cut Pro. The generated video is then encoded and a preview link is generated.

[0479] Input: Generated advertising scenario

[0480] Output: Generated video ad with preview link

[0481] Step 5:

[0482] Users: Video reviews and feedback

[0483] Users can watch the generated video via the provided preview link, review the video, and provide feedback on corrections or additional requirements as needed. For example, they could provide feedback such as "I would like the product name to be more prominent in the final scene."

[0484] Input: Preview link and user feedback

[0485] Output: Feedback data

[0486] Step 6:

[0487] Server: Feedback analysis and scenario refinement

[0488] The server analyzes the feedback received from the user and uses a big data analysis unit to analyze the data. Based on the analysis results, the advertising scenario is improved, for example, by modifying certain parts of the scenario based on the feedback.

[0489] Input: Feedback data

[0490] Output: Improved advertising scenario

[0491] Step 7:

[0492] Server: Regenerate videos and prepare for operation

[0493] The server regenerates the video ad using the generative model based on the improved scenario, re-encodes the video, and notifies the server that it is ready for production.

[0494] Input: Improved advertising scenario

[0495] Output: Regenerated video ads and notifications

[0496] Step 8:

[0497] Terminal: Notification of start of operation

[0498] The device will notify the user that it is ready for operation. The user will receive the notification, check the contents, and click the "Start operation" button to start advertising operations.

[0499] Input: Notification of operational readiness

[0500] Output: Ads go live

[0501] Step 9:

[0502] Server: Performance data collection and analysis

[0503] As soon as the ad starts running, the server collects performance data in real time, such as the number of impressions, click-through rates, and conversions, which are then analyzed by the effectiveness measurement unit.

[0504] Input: Ad performance data

[0505] Output: Parsed performance report

[0506] Step 10:

[0507] Server: Emotion data analysis and dynamic ad adjustment

[0508] The server's emotion engine recognizes emotions in real time from facial expressions and voice as the user views the advertisement. The content of the advertisement is dynamically adjusted based on this emotional data. For example, if the user shows interest, the content will be changed to emphasize that element.

[0509] Input: Emotion data

[0510] Output: Dynamically adjusted ad content

[0511] Step 11:

[0512] Server: Ad regeneration and optimization

[0513] The server uses the generative model to regenerate and optimize ads when performance is poor, and dynamically adjusts ad timing and targeting settings.

[0514] Input: Performance and emotion data

[0515] Output: Regenerated optimized video ad

[0516] Step 12:

[0517] Server: Performance measurement and report generation

[0518] Once the advertising period ends, the server aggregates all data and measures the effectiveness. Based on the results of the analysis, a detailed report is generated and provided to the user.

[0519] Input: All performance data

[0520] Output: Performance measurement results and reports

[0521] Step 13:

[0522] User: Check the report and consider the next strategy

[0523] Users can review the provided reports and evaluate the results of their advertising operations. Based on this evaluation, they can consider their next advertising strategy and come up with further improvements.

[0524] Input: Performance measurement report

[0525] Output: Next advertising strategy plan

[0526] (Application example 2)

[0527] Next, a description will be given of Application 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."

[0528] In the production and provision of modern advertising videos and entertainment content, it is difficult to consider the different emotions and interests of each user, and generic advertising may not be effective for all users. Furthermore, in order to maximize passenger satisfaction, entertainment systems inside autonomous vehicles are required to provide content tailored to passenger emotions in real time. However, current technology does not provide a method to achieve this.

[0529] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for generating videos for advertising using a generative model, means for generating advertising scenarios by analyzing big data, means for operating the generated videos and measuring their effectiveness, means for analyzing user emotions using an emotion recognition engine and dynamically optimizing the video content, and means for providing advertising and entertainment content based on emotion recognition. This makes it possible to provide highly personalized advertising and entertainment experiences according to the emotional state of each passenger.

[0530] A "generative model" is an algorithm that uses artificial intelligence technology to automatically create advertising videos.

[0531] "Big data" refers to a large amount of data collected from a wide variety of sources.

[0532] An "advertising scenario" is a plan or script that determines the content and structure of an advertising video.

[0533] "Operation" refers to the process of actually distributing the generated advertising video and observing its effectiveness.

[0534] "Effectiveness measurement" is an analytical method for evaluating the performance of advertising videos.

[0535] "Optimization" is the process of improving advertising videos and their delivery methods to maximize the desired effect.

[0536] An "emotion recognition engine" is a technology for analyzing emotions from a user's facial expressions and voice.

[0537] "Dynamic optimization" refers to changing the content of the advertising video in real time according to the user's state.

[0538] "Entertainment content" refers to various media materials that provide entertainment and interest to users.

[0539] "User emotional feedback" refers to the emotional response of users when they watch advertising videos or entertainment content.

[0540] The present invention provides a system for providing highly personalized advertising and entertainment experiences within an autonomous vehicle that are tailored to the emotional state of each passenger. Specific embodiments for implementing the present invention will be described below.

[0541] System configuration

[0542] This system consists of three main components: a server, a terminal, and a user.

[0543] 1. Server

[0544] Generative model: An artificial intelligence model for automatically generating advertising videos and entertainment content.

[0545] Big Data Analysis Unit: A unit that analyzes the performance of advertising scenarios and content.

[0546] Effectiveness measurement unit: A unit for measuring the effectiveness of ads and content in real time.

[0547] Optimization unit: A unit for optimizing the delivery method and content of ads and content.

[0548] Emotion engine: An engine for recognizing emotions from the user's facial expressions and voice.

[0549] 2. Terminal

[0550] In-car camera: A camera for capturing facial images of passengers.

[0551] User Interface: A display for displaying advertisements and content.

[0552] Emotion recognition interface: An interface for sending facial images captured by the in-car camera to a server.

[0553] Dashboard: A tool for monitoring the delivery and effectiveness of ads and content.

[0554] 3. Users

[0555] Passengers: Users who watch advertising and entertainment content in the vehicle.

[0556] Processing flow explanation

[0557] 1. Passenger Emotion Recognition

[0558] The device's onboard camera captures the passenger's face and transmits the image in real time to an emotion-recognition interface.

[0559] The emotion engine on the server analyzes the received images and predicts the passenger's emotional state using image processing libraries such as OpenCV and dlib, as well as deep learning models using Keras.

[0560] 2. Dynamic generation and optimization of advertising and entertainment content

[0561] The server's generative model generates advertising videos and entertainment content based on the emotion data obtained from the emotion engine.

[0562] The big data analysis unit creates content scenarios, and the effectiveness measurement unit monitors performance in real time.

[0563] The optimization unit adjusts content to maximize its effectiveness based on emotional state and collected performance data.

[0564] 3. Delivery of optimal content

[0565] The server sends the generated and optimized advertising and entertainment content to the terminal's user interface for display to the passenger.

[0566] Specific examples

[0567] For example, if a passenger makes a "disinterested" facial expression while in the car, the server's emotion engine will recognize that emotion and use a generative model to regenerate advertising content to attract more interest. An optimization unit then monitors and measures the effectiveness of the content used and makes further adjustments if necessary.

[0568] Prompt Sentence Examples

[0569] Design a system that recognizes emotions in real time from facial expressions that indicate the user's emotions when viewing an advertisement, and retrieves and displays the most appropriate video advertisement from the server based on that user's emotional state.

[0570] This enables the system to provide optimal advertising and entertainment experiences for each passenger.

[0571] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0572] Step 1:

[0573] The device's onboard camera captures the passenger's face and transmits the image in real time to an emotion-recognition interface.

[0574] Input: Passenger face image captured by in-car camera

[0575] Processing: Send the facial image to the emotion recognition interface

[0576] Output: Facial image sent to emotion recognition interface

[0577] Step 2:

[0578] The server's emotion engine analyzes the received images and predicts the passenger's emotional state.

[0579] Input: Facial image sent from emotion recognition interface

[0580] Processing: Detect faces using image processing libraries (OpenCV, dlib) and predict emotions using deep learning models (Keras)

[0581] Output: Predicted emotional state (e.g., uninterested, amused)

[0582] Step 3:

[0583] The server's generative model generates advertising videos and entertainment content based on the emotion data obtained from the emotion engine.

[0584] Input: Emotional state data from the emotion engine

[0585] Processing: Generative models use emotional state data to generate appropriate advertising and content scenarios, and collect and edit materials.

[0586] Output: Generated advertising video or entertainment content

[0587] Step 4:

[0588] The server's big data analysis unit creates content scenarios, and the effectiveness measurement unit monitors performance in real time.

[0589] Input: Generated ad video or content, various performance data (number of impressions, click-through rate, emotional feedback)

[0590] Processing: Analyzing performance data, building scenarios and making adjustments

[0591] Output: Improved scenarios and performance measurement data

[0592] Step 5:

[0593] The server's optimization unit adjusts content to maximize its effectiveness based on emotional state and collected performance data.

[0594] Input: Improved scenario, various performance data

[0595] Processing: Adjusting how and what content is delivered based on the optimization algorithms used by the optimization unit

[0596] Output: Optimized advertising videos or entertainment content

[0597] Step 6:

[0598] The server sends the generated and optimized advertising and entertainment content to the terminal's user interface for display to the passenger.

[0599] Input: Optimized ad video or entertainment content

[0600] Processing: Encode the content and send it to the device's user interface

[0601] Output: Advertising videos or entertainment content displayed on the device's display

[0602] Step 7:

[0603] The server's effectiveness measurement unit monitors the delivery status and effectiveness of advertisements and content.

[0604] Input: Various performance data (number of impressions, click-through rate, number of conversions, emotional feedback)

[0605] Processing: Analyzes the collected performance data and generates reports

[0606] Output: Reports provided to users, insights for future advertising operations

[0607] 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.

[0608] 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> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) 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.

[0609] 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.

[0610] [Second embodiment]

[0611] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0612] 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.

[0613] 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).

[0614] 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.

[0615] 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.

[0616] 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).

[0617] 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.

[0618] 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.

[0619] 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.

[0620] 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.

[0621] 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.

[0622] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0623] The present invention provides a system for efficiently implementing the entire process from video advertising production to operation, and specific embodiments thereof are described below. This system integrates generative models, big data analysis, feedback processing, and optimization functions to effectively produce and operate video advertising.

[0624] System configuration

[0625] The system consists of the following main components:

[0626] 1. Server

[0627] Generative Model

[0628] Big Data Analysis Unit

[0629] Effectiveness Measurement Unit

[0630] Optimization Unit

[0631] 2. Terminal

[0632] User Interface

[0633] Dashboard

[0634] 3. Users

[0635] Advertisement requirements provided

[0636] Providing Feedback

[0637] Program processing flow

[0638] Video ad requirements definition

[0639] User:

[0640] Users first log in to the portal site and enter the requirements for a new advertising campaign, including target demographics, product information, advertising objectives, etc. After entering the requirements, users can provide detailed information by clicking the "Submit" button to send them to the server.

[0641] server:

[0642] The server analyzes the requirements data received from the user and performs preprocessing before passing it to the Generative Model. This preprocessing converts the requirements data into an appropriate format so that the Generative Model can operate efficiently.

[0643] Scenario generation and video generation

[0644] server:

[0645] The server passes the preprocessed data to a generative model to generate an advertising scenario. Based on the generated scenario, the server collects additional materials such as images, audio, and text, and uses the generative model to automatically generate a video advertisement. The generated video is then encoded for initial review, and a preview link is generated.

[0646] User:

[0647] The user can review the generated video by checking the provided preview link. If feedback is required, the user can enter comments or corrections on the platform and send them back to the server.

[0648] Improving video quality and preparing for operation

[0649] server:

[0650] The server receives feedback from users and analyzes the big data to improve the advertising scenario. It uses patterns obtained from the big data to reflect improvements to the scenario. Based on the improved scenario, it encodes the regenerated video and prepares it for deployment. Once the deployment is ready, the server sends a notification to the user.

[0651] Device:

[0652] The terminal notifies the user that the operation preparation is complete. The user confirms the contents and clicks the "Start operation" button to start the advertisement operation.

[0653] Efficacy monitoring and optimization

[0654] server:

[0655] As soon as the server starts running the ad, it collects performance data such as the number of impressions, click-through rates, and number of conversions in real time. The collected data is analyzed by an effectiveness measurement unit, and if performance is poor, the ad is regenerated and optimized using a generative model. The server also adjusts the timing and targeting of ad displays as appropriate.

[0656] Examples:

[0657] Suppose a company creates an advertising video for a new smartphone product and starts running it. If the click-through rate on the first day is low, the server uses a generative model to improve the video and generates a new scenario based on big data analysis. Then, it changes the timing of the ad display to the optimal time and runs it again.

[0658] Measurement and reporting

[0659] server:

[0660] Once the ad run period is over, the server aggregates all data and measures its effectiveness. A detailed report summarizing the analysis results is generated and provided to the user. The report includes detailed results for each performance indicator and insights for future ad runs.

[0661] User:

[0662] Users can receive the provided report, check the results of their advertising operations, and plan their next advertising strategy based on the report contents.

[0663] The above is a specific description of the embodiment of the present invention. The present system enables users to efficiently create and operate high-quality, effective video advertisements.

[0664] The processing flow will be explained below.

[0665] Step 1:

[0666] User:

[0667] Users log in to the portal site and enter the requirements for a new advertising campaign, providing detailed information such as target demographic, product information, and advertising objectives (e.g., brand awareness, increased conversions). They then confirm the information they entered and click the "Submit" button.

[0668] Step 2:

[0669] server:

[0670] The server analyzes the requirements data received from the user and performs preprocessing before passing it to the generative model. Preprocessing converts the requirements data into an appropriate format so that the generative model can operate efficiently.

[0671] Step 3:

[0672] server:

[0673] The server passes the preprocessed data to a generative model to generate advertising scenarios. The generative model automatically generates effective advertising scenarios based on the input data.

[0674] Step 4:

[0675] server:

[0676] Based on the generated scenario, the server collects materials such as images, audio, and text. Using these materials, the generative model automatically creates a video ad. The generated video is encoded for early review and a preview link is generated.

[0677] Step 5:

[0678] User:

[0679] The user can review the generated video content by checking the provided preview link, and provide feedback and input corrections or additional requirements as needed.

[0680] Step 6:

[0681] server:

[0682] The server receives feedback from users and analyzes the big data to improve the advertising scenario. It utilizes patterns obtained from the big data to improve the scenario.

[0683] Step 7:

[0684] server:

[0685] Based on the improved scenario, the generative model is used to regenerate the video ad. The regenerated video is then encoded and made ready for deployment. Once ready, the server sends a notification to the user.

[0686] Step 8:

[0687] Device:

[0688] The device notifies the user that it is ready for operation. The user confirms the information and clicks the "Start operation" button to start advertising operations.

[0689] Step 9:

[0690] server:

[0691] As soon as the ad starts running, the server collects performance data in real time, such as the number of impressions, click-through rates, and conversions, which are then analyzed by the effectiveness measurement unit.

[0692] Step 10:

[0693] server:

[0694] If performance is poor, the server uses the generative model to regenerate and optimize the ads, adjusting the timing and targeting of the ads accordingly.

[0695] Step 11:

[0696] server:

[0697] At the end of the advertising period, the server aggregates all data and measures the effectiveness. Based on the analysis results, a detailed report is generated and provided to the user.

[0698] Step 12:

[0699] User:

[0700] Users can receive the provided report, check the results of their advertising operations, and plan their next advertising strategy based on the report contents.

[0701] The above are the specific processing steps for carrying out the present invention. This system allows users to efficiently create and operate high-quality, effective video advertisements.

[0702] Example 1

[0703] Next, a description will be given of 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."

[0704] Traditional advertising production and management processes are time-consuming, resource-intensive, and inefficient. In particular, they lack flexibility and make it difficult to measure and optimize advertising effectiveness in real time. Another issue is the difficulty of quickly incorporating user feedback and quickly improving advertising content.

[0705] 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.

[0706] In this invention, the server includes means for generating videos for advertisements using a generative AI model, means for analyzing large-scale data to generate advertising scenarios, means for operating the generated videos and measuring their effectiveness, means for optimizing the operation of advertisements based on the measurement results, means for receiving and preprocessing advertising requirements from users, means for receiving user feedback on the generated videos, and means for improving and regenerating the videos based on the feedback. This streamlines the process of advertisement production and operation, enables real-time measurement and optimization of effectiveness, and enables rapid improvement that reflects user feedback.

[0707] A "generative AI model" is an artificial intelligence model that automatically generates advertising videos and scenarios based on specified prompt text.

[0708] "Large-scale data" refers to large amounts of data, such as advertising operations and user feedback, and is a collection of information obtained by analyzing this data.

[0709] "Scenario generation" is the process of designing the structure, content, and storyboard of an advertising video.

[0710] "Effectiveness measurement" is the process of collecting data such as the number of impressions, click rates, and number of conversions obtained during advertising operations and evaluating the performance of the advertisements.

[0711] "Optimization" refers to the process of adjusting advertising operations based on the results of effectiveness measurement to improve performance.

[0712] "User feedback" refers to comments and correction instructions given by the user to the generated advertising video.

[0713] "Regeneration" is the process of recreating advertising videos and scenarios based on feedback and effectiveness measurement results.

[0714] "Preprocessing" is the initial process of analyzing the advertising requirement data received from the user and converting it into a format that is easy for the generative AI model to process.

[0715] The present invention is a system for efficiently implementing the entire process from video advertising production to operation, and specific embodiments thereof are described below. This system integrates generative AI models, large-scale data analysis, feedback processing, and optimization functions to effectively produce and operate video advertising.

[0716] System configuration

[0717] The system consists of the following main components:

[0718] 1. Server

[0719] Generative AI Models

[0720] Large-Scale Data Analysis Unit

[0721] Effectiveness Measurement Unit

[0722] Optimization Unit

[0723] 2. Terminal

[0724] User Interface

[0725] Dashboard

[0726] 3. Users

[0727] Advertisement requirements provided

[0728] Providing Feedback

[0729] Video ad requirements definition

[0730] User:

[0731] Users log in to the portal site and enter the requirements for a new advertising campaign. These requirements include target demographics, product information, advertising objectives, etc., and users can provide detailed information. For example, they can enter a requirement such as "Raising awareness of a new headphone for music lovers aged 25-35." After entering the information, users click the "Submit" button to send it to the server.

[0732] server:

[0733] The server analyzes the requirements data received from the user and performs preprocessing before passing it to the generative AI model. This preprocessing converts the requirements data into an appropriate format, allowing the generative AI model to operate efficiently.

[0734] Scenario generation and video generation

[0735] server:

[0736] The server then passes the preprocessed data to a generative AI model to generate an advertising scenario. For example, it uses a prompt such as, "Generate a scenario that will resonate with music lovers aged 25-35." Based on the generated scenario, the server collects additional materials such as images, audio, and text, and uses the generative model to automatically generate a video ad. The generated video is then encoded for initial review, and a preview link is generated.

[0737] User:

[0738] The user opens the provided preview link to check the content of the generated video, and if necessary, enters comments or corrections on the platform and sends it back to the server.

[0739] Improving video quality and preparing for operation

[0740] server:

[0741] The server receives user feedback, analyzes large-scale data, and refines the advertising scenario. It uses patterns extracted from the large-scale data to improve the scenario. Based on the refined scenario, the video is regenerated and ready for deployment.

[0742] Device:

[0743] The device will notify the user that it is ready for operation. The user can confirm the details and click the "Start operation" button to start advertising operations.

[0744] Efficacy monitoring and optimization

[0745] server:

[0746] As soon as the server starts running the ad, it collects performance data such as the number of impressions, click-through rates, and number of conversions in real time. The collected data is analyzed by an effectiveness measurement unit, and if performance is poor, the ad is regenerated and optimized using a generative model. The timing and targeting of ad displays are also adjusted as appropriate.

[0747] Examples:

[0748] For example, if a company is running an ad for a new smartphone product and the click-through rate on the first day is low, the server will use the generative model to generate new scenarios, improve the video, optimize the timing of the landing page, and adjust the display time of the ad.

[0749] Measurement and reporting

[0750] server:

[0751] Once the ad run period is over, the server aggregates all data and measures its effectiveness. A detailed report summarizing the analysis results is generated and provided to the user. The report includes detailed results for each performance indicator and insights for future ad runs.

[0752] User:

[0753] Users can receive the provided report, check the results of their advertising operations, and plan their next advertising strategy based on the report contents.

[0754] This system streamlines the video ad production and operation process, enabling real-time measurement and optimization of effectiveness. It also allows for quick reflection of user feedback, making it possible to constantly improve ad content to keep it up-to-date and optimal.

[0755] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0756] Step 1: Enter user requirements

[0757] User:

[0758] Input: Users log in to the portal site and enter the requirements for a new advertising campaign, providing detailed information in text format, such as the target demographic (e.g., "music lovers aged 25-35"), product information, and advertising objectives (e.g., "increase awareness of a new product").

[0759] Output: User-entered requirements data.

[0760] Action: The user clicks the "Submit" button to send the entered requirements data to the server.

[0761] Step 2: Preprocessing requirements data

[0762] server:

[0763] Input: Requirement data submitted by the user.

[0764] Output: Preprocessed requirements data (structured data).

[0765] How it works: The server parses the received requirements data and preprocesses it to be passed to the generative AI model. Specifically, it structures the text data and splits it into necessary fields.

[0766] Step 3: Scenario generation

[0767] server:

[0768] Input: Preprocessed requirements data.

[0769] Output: The generated advertising scenario.

[0770] How it works: The server inputs the preprocessed data into a generative AI model and generates advertising scenarios using prompts, such as "Generate a scenario that will resonate with music lovers aged 25-35."

[0771] Step 4: Gathering materials and creating a video

[0772] server:

[0773] Input: The generated ad scenario.

[0774] Output: The generated video ad file.

[0775] How it works: Based on the generated scenario, the server collects the necessary images, audio, and text materials, and automatically generates a video ad using a generative AI model. The generated video file is encoded and a preview link is generated.

[0776] Step 5: Video preview and feedback

[0777] User:

[0778] Input: The video generated using the preview link.

[0779] Output: User feedback (comments and suggestions for corrections).

[0780] How it works: The user opens the provided preview link, checks the generated video, and optionally enters comments and corrections on the platform, which then sends the video back to the server.

[0781] Step 6: Receive and analyze feedback

[0782] server:

[0783] Input: User feedback.

[0784] Output: The scenario with the improvements.

[0785] How it works: The server receives feedback from users and analyzes it using a large-scale data analysis unit. Based on the analysis results, it extracts areas for improvement in the scenario and generates a new scenario using a generative AI model.

[0786] Step 7: Scenario refinement and video regeneration

[0787] server:

[0788] Input: The scenario with the improvements.

[0789] Output: Regenerated video ad files.

[0790] Operation: The server gathers additional material and regenerates the video based on the improved scenario. The regenerated video file is then re-encoded and ready for production.

[0791] Step 8: Notification of operational readiness

[0792] server:

[0793] Input: The regenerated video ad file.

[0794] Output: Operational readiness notification.

[0795] Action: The server notifies the user that it is ready for operation.

[0796] Step 9: Go live

[0797] Device:

[0798] Input: Operational readiness notification.

[0799] Output: Advertising videos are launched.

[0800] Operation: The device displays a notification to the user that preparation for operation is complete. The user clicks the "Start operation" button to start ad operation.

[0801] Step 10: Collecting performance measurement data

[0802] server:

[0803] Input: Ad video performance data (number of impressions, click-through rate, number of conversions, etc.).

[0804] Output: Performance measurement data.

[0805] How it works: The server starts running ads and collects real-time performance data such as impressions, click-through rates, and conversions.

[0806] Step 11: Effectiveness analysis and optimization

[0807] server:

[0808] Input: Performance measurement data.

[0809] Output: Improved and optimized ad video.

[0810] How it works: The server analyzes the collected data using an effectiveness measurement unit, and if performance is poor, it uses a generative model to regenerate ads, adjusting the timing and targeting of ads accordingly.

[0811] Step 12: Measurement and reporting

[0812] server:

[0813] Input: Aggregated performance measurement data.

[0814] Output: Detailed performance measurement report.

[0815] How it works: After the ad campaign period ends, the server aggregates all data, measures the effectiveness, and generates a detailed report summarizing the analysis results, which is then provided to the user.

[0816] Step 13: Review reports and develop strategies

[0817] User:

[0818] Input: Detailed performance measurement report.

[0819] Output: Next advertising strategy.

[0820] How it works: Users can receive the provided report, check the results of their advertising campaigns, and plan their next advertising strategy based on the report.

[0821] (Application example 1)

[0822] Next, a description will be given of Application 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."

[0823] Traditional video ad production systems have the problem that the processes of ad creation, operation, effectiveness measurement, and optimization are fragmented, making efficient operation difficult. Furthermore, there is a lack of a system that can optimize ads in real time and immediately reflect user feedback. As a result, advertising effectiveness is reduced and user satisfaction is low.

[0824] 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.

[0825] In this invention, the server includes a means for generating videos for advertisements using a generative model, a means for analyzing big data to generate advertising scenarios, a means for operating the generated videos and measuring their effectiveness, a means for optimizing the operation of advertisements based on the measurement results, a means for displaying the optimized video advertisements in real time using smart glasses, and a means for regenerating and optimizing advertisements based on collected data. This enables the entire process from the production of video advertisements to their operation and feedback to be integrated and operated efficiently and effectively.

[0826] A "generative model" is an algorithm that automatically generates scenarios and videos for advertising.

[0827] "Big data" refers to large amounts of data, a group of data that is used to generate and optimize advertising scenarios.

[0828] An "advertising scenario" is a plan that defines the content and progression of a video advertisement.

[0829] "Operation" refers to the process of actually delivering the generated video advertisements and displaying them to users.

[0830] "Effectiveness measurement" is the process of evaluating the results of advertising and collecting and analyzing performance data.

[0831] "Optimization" refers to making adjustments based on collected data to improve the performance of your ads.

[0832] "Smart glasses" are a type of wearable device that visually displays information to the user while they are wearing them.

[0833] "Real-time" means that processing and reaction are immediate and based on the current time zone.

[0834] "Feedback" means reviews and correction instructions provided by users.

[0835] An "interface" is a point of contact or means by which a user interacts with a system.

[0836] "Data" refers to a set of information or numbers that a system processes.

[0837] "Regeneration" is the process of creating a new version of an ad based on previously generated content.

[0838] The present invention provides a system for efficiently implementing the entire process from video advertising production to operation, and is characterized by its ability to display optimized video advertising in real time using smart glasses. Specific embodiments are described below. The system integrates generative models, big data analysis, feedback processing, and optimization functions to effectively produce and operate video advertising.

[0839] System configuration

[0840] The system consists of the following main components:

[0841] 1. Server:

[0842] Generative model: An algorithm that automatically generates scenarios and videos for advertising.

[0843] Big Data Analysis Unit: A data set used to analyze large amounts of data and measure and optimize advertising effectiveness.

[0844] Effectiveness Measurement Unit: Measures the effectiveness of advertising and analyzes the collected data.

[0845] Optimization unit: Optimizes your ads based on the data collected to improve their performance.

[0846] 2. Terminal:

[0847] Smart glasses: Glasses that visually display information while worn by the user (e.g., Google Glass, Vuzix Blade).

[0848] User Interface: An interface for entering ad requirements, previewing, and providing feedback.

[0849] Dashboard: A management screen for displaying advertising results and effectiveness measurement results.

[0850] 3. User:

[0851] Provide Ad Requirements: Enter your requirements for a new ad campaign.

[0852] Providing feedback: Review the content of the generated video and provide feedback.

[0853] System Operation

[0854] 1. Video ad requirements definition:

[0855] Users input their advertising requirements through the smart glasses interface, including target demographic, product information, and advertising objectives, and then the requirements are sent to the server.

[0856] 2. Scenario and video generation:

[0857] The server analyzes the received requirements data and passes it to the generative model. The generative model generates a scenario, collects the necessary materials (images, audio, text), and automatically generates a video advertisement. The generated video is encoded for initial review and streamed to the smart glasses.

[0858] 3. Viewing videos and providing feedback:

[0859] Users can view the generated video through smart glasses and provide feedback using voice commands or touch gestures, which is then sent to the server, which regenerates and optimizes the video as needed.

[0860] 4. Effect monitoring and optimization:

[0861] The server monitors the effectiveness of the ad (number of views, viewing time, user response) in real time. If the effectiveness is low, it will be regenerated and optimized to improve the ad performance.

[0862] 5. Report Generation:

[0863] The server aggregates all data, measures effectiveness, and generates detailed reports to provide users with insights to help them plan their next advertising strategy.

[0864] Specific examples

[0865] For example, consider a company that is creating and running an advertising video for a new smartphone product. The target demographic is technology-conscious people in their 20s and 30s, and the goal of the ad is to stimulate purchasing desire and increase brand awareness. Based on this requirement, the following prompt sentences are input into the generative AI model:

[0866] Prompt statement:

[0867] The target demographic for a new smartphone is tech-savvy people in their 20s and 30s. The advertising objective is to stimulate purchase intent and increase brand awareness. Based on this requirement, generate the following scenario and creative. The ad should include high-performance camera features, a long battery life, and a seamless user experience.

[0868] As described above, this system allows users to view high-quality, effective video advertisements in real time using smart glasses, and allows for immediate feedback.

[0869] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0870] Step 1: User Input

[0871] Through the smart glasses' user interface, users input requirements for a new advertising campaign, including target demographic, product information, and advertising objectives. The input is made on the interface and transmitted to the server. The input data also specifies specific elements and conditions that should be included in the advertisement.

[0872] Input: target demographic, product information, advertising objectives

[0873] Output: Requirements data sent to the server

[0874] Step 2: Requirement Data Analysis

[0875] The server analyzes the requirements data received from the user, and the analysis results are converted into a format that can be passed to the generative AI model. This preprocessing allows the generative model to operate efficiently.

[0876] Input: Requirement data

[0877] Output: Data converted into a format to be passed to the generative model

[0878] Step 3: Scenario generation

[0879] The server passes the preprocessed data to the generative model to generate an advertising scenario. The generative model constructs a scenario based on the prompt sentence and determines the story of the advertising video.

[0880] Input: Preprocessed data, generative AI model

[0881] Output: Generated scenario

[0882] Step 4: Gathering materials and creating a video

[0883] The server collects materials such as images, audio, and text based on the generated scenario, then uses the generative model to automatically generate advertising videos, which are then encoded and prepared for streaming to the smart glasses.

[0884] Input: Generated scenario, material data

[0885] Output: Generated video ad

[0886] Step 5: Advertising and Feedback

[0887] The user watches the generated video advertisement through the smart glasses. If feedback is required, the user can provide it using voice commands or touch gestures. The feedback is then sent to the server.

[0888] Input: Generated video ad

[0889] Output: User feedback

[0890] Step 6: Feedback analysis and video regeneration

[0891] The server analyzes the feedback received from the user, refines the content of the video advertisement based on the feedback, and regenerates the video, which is then re-encoded and streamed to the smart glasses.

[0892] Input: User feedback

[0893] Output: Regenerated video ad

[0894] Step 7: Measure your results

[0895] The server starts running the ad and simultaneously collects viewing data (number of views, viewing time, user response) in real time. The collected data is analyzed by the effectiveness measurement unit and the ad is optimized as necessary.

[0896] Input: Viewing data

[0897] Output: Effectiveness measurement report, optimization instructions

[0898] Step 8: Generate reports

[0899] After the ad run period ends, the server aggregates all data and generates a detailed performance measurement report, which is provided to the user and contains insights for planning their next advertising strategy.

[0900] Input: Aggregated operational data

[0901] Output: Performance measurement report

[0902] Through the above process, the present invention makes it possible to efficiently and effectively produce, operate, collect feedback, and optimize advertising videos.

[0903] 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.

[0904] The present invention is a system for efficiently and effectively implementing the entire process from the production to the operation of video advertisements, and in particular, by combining an emotion engine that recognizes user emotions, it achieves a high level of personalization. Below, we will explain the form for implementing the present invention and the specific processing content thereof.

[0905] System configuration

[0906] The system consists of the following main components:

[0907] 1. Server

[0908] Generative Model

[0909] Big Data Analysis Unit

[0910] Effectiveness Measurement Unit

[0911] Optimization Unit

[0912] Emotion Engine

[0913] 2. Terminal

[0914] User Interface

[0915] Emotion Recognition Interface

[0916] Dashboard

[0917] 3. Users

[0918] Advertisement requirements provided

[0919] Providing Feedback

[0920] Program processing flow

[0921] Video ad requirements definition

[0922] User:

[0923] First, users log in to the portal site and enter the requirements for a new advertising campaign, providing detailed information such as target demographic, product information, and advertising objectives (e.g., brand awareness, increased conversions). They then confirm the information they entered and click the "Submit" button.

[0924] server:

[0925] The server analyzes the requirements data received from the user and performs preprocessing before passing it to the Generative Model. This preprocessing converts the requirements data into an appropriate format so that the Generative Model can operate efficiently.

[0926] Scenario generation and video generation

[0927] server:

[0928] The server passes the preprocessed data to a generative model to generate advertising scenarios. The generative model automatically generates effective advertising scenarios based on the input data.

[0929] server:

[0930] Based on the generated scenario, the server collects materials such as images, audio, and text. Using these materials, the generative model automatically creates a video ad. The generated video is encoded for early review and a preview link is generated.

[0931] User:

[0932] The user can review the generated video content by checking the provided preview link, and provide feedback and input corrections or additional requirements as needed.

[0933] Improving video quality and preparing for operation

[0934] server:

[0935] The server receives feedback from users and analyzes the big data to improve the advertising scenario. It utilizes patterns obtained from the big data to improve the scenario.

[0936] server:

[0937] Based on the improved scenario, the generative model is used to regenerate the video ad. The regenerated video is then encoded and made ready for deployment. Once ready, the server sends a notification to the user.

[0938] Device:

[0939] The device notifies the user that it is ready for operation. The user confirms the information and clicks the "Start operation" button to start advertising operations.

[0940] Effect monitoring, optimization, and emotion recognition

[0941] server:

[0942] As soon as the ad starts running, the server collects performance data in real time, such as the number of impressions, click-through rates, and conversions, which are then analyzed by the effectiveness measurement unit.

[0943] server:

[0944] The emotion engine recognizes emotions in real time from facial expressions and voice when a user views an ad, and dynamically adjusts the content of the ad based on that data. For example, if a user has a happy expression while viewing an ad, the engine can add more engaging elements to match that positive emotion.

[0945] server:

[0946] If performance is poor, the server uses the generative model to regenerate and optimize the ads, adjusting the timing and targeting of the ads accordingly.

[0947] Examples:

[0948] Suppose a company creates and starts using an advertising video for a new smartphone product. If many users who watch the ad feel uninterested, the server uses an emotion engine to analyze the emotion data, improves the advertising scenario, and regenerates it to make it more interesting.

[0949] Measurement and reporting

[0950] server:

[0951] Once the ad campaign period ends, the server aggregates all data and measures its effectiveness. Based on the analysis results, a detailed report is generated and provided to the user. The report includes detailed results for each performance indicator and insights for future ad campaigns.

[0952] User:

[0953] Users can receive the provided report, check the results of their advertising operations, and plan their next advertising strategy based on the report contents.

[0954] The above is a specific description of an embodiment of the present invention. This system allows users to efficiently create and manage high-quality, effective video advertisements. Furthermore, by utilizing the emotion engine, it is possible to achieve a high level of personalization of advertisements and provide each user with an optimal advertising experience.

[0955] The processing flow will be explained below.

[0956] Step 1:

[0957] User:

[0958] Users log in to the portal site and enter the requirements for a new advertising campaign, including detailed information such as target demographic, product information, and advertising objectives (e.g., brand awareness, increased conversions). They confirm the information they entered and click the "Submit" button.

[0959] Step 2:

[0960] server:

[0961] The server analyzes the requirements data received from the user and performs preprocessing before passing it to the Generative Model. This preprocessing converts the provided data into an appropriate format so that the Generative Model can operate efficiently.

[0962] Step 3:

[0963] server:

[0964] The server passes the preprocessed data to a generative model to generate advertising scenarios. The generative model automatically generates effective advertising scenarios based on the input data.

[0965] Step 4:

[0966] server:

[0967] Based on the generated scenario, the server collects materials such as images, audio, and text. Using these materials, the generative model automatically creates a video ad. The generated video is encoded for early review and a preview link is generated.

[0968] Step 5:

[0969] User:

[0970] The user can review the generated video content by checking the provided preview link, and provide feedback and input corrections or additional requirements as needed.

[0971] Step 6:

[0972] server:

[0973] The server receives feedback from users and analyzes the big data to improve the advertising scenario. It utilizes patterns obtained from the big data to improve the scenario.

[0974] Step 7:

[0975] server:

[0976] Based on the improved scenario, the generative model is used to regenerate the video ad. The regenerated video is then encoded and made ready for deployment. Once ready, the server sends a notification to the user.

[0977] Step 8:

[0978] Device:

[0979] The device notifies the user that it is ready for operation. The user confirms the information and clicks the "Start operation" button to start advertising operations.

[0980] Step 9:

[0981] server:

[0982] As soon as the ad starts running, performance data such as the number of impressions, click-through rate, and number of conversions is collected in real time, and the collected data is analyzed by the effectiveness measurement unit.

[0983] Step 10:

[0984] server:

[0985] The emotion engine recognizes emotions in real time from facial expressions and voice when a user views an advertisement, and dynamically adjusts the content of the advertisement based on that data. For example, if a user is not interested in an advertisement, the scenario will be improved by reflecting that emotional data.

[0986] Step 11:

[0987] server:

[0988] If performance is poor, the server uses the generative model to regenerate and optimize the ads, adjusting the timing and targeting of the ads accordingly.

[0989] Step 12:

[0990] server:

[0991] Once the advertising period ends, the server aggregates all data and measures the effectiveness. Based on the analysis results, a detailed report is generated and provided to the user.

[0992] Step 13:

[0993] User:

[0994] Users can receive the provided report, check the results of their advertising operations, and plan their next advertising strategy based on the report contents.

[0995] The above are the specific processing steps for implementing the present invention in combination with an emotion engine. This system allows users to efficiently create and manage high-quality, effective video advertisements, and by utilizing the emotion engine, highly personalized advertisements can be achieved.

[0996] Example 2

[0997] Next, a description will be given of 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."

[0998] Conventional advertising video production and operation systems lacked personalization, making it difficult to optimize by incorporating each user's emotions and real-time feedback. Furthermore, the process of measuring effectiveness and analyzing feedback was complicated, making it difficult to respond quickly to improve advertising effectiveness. This resulted in problems such as a lack of improvement in user experience and reduced advertising effectiveness.

[0999] 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.

[1000] In this invention, the server includes means for generating videos for advertisements using a generative model, means for analyzing big data to generate advertising scenarios, means for operating the generated videos and measuring their effectiveness, means for recognizing emotions from a user's facial expressions and voice and dynamically adjusting advertising content based on the emotion data, means for analyzing the content of feedback and regenerating videos based on the analysis results, and means for aggregating the results of measuring the effectiveness of advertising operation, generating detailed reports, and providing them to users. This makes it possible to highly personalize advertisements and optimize them in real time.

[1001] A "generative model" is a machine learning model that uses algorithms to automatically generate content.

[1002] "Big data" is a collection of large amounts of diverse data, the analysis of which makes it possible to discover new insights and patterns.

[1003] An "advertising scenario" is a detailed design document for planning the composition and development of an advertising video.

[1004] "Emotion recognition" is a technology that analyzes non-verbal information such as a user's facial expressions and voice to recognize their emotional state.

[1005] "Dynamic adjustment" is the process of instantly changing advertising content based on real-time data.

[1006] "Feedback" refers to the evaluations and opinions provided by users, and the system is improved and adjusted based on them.

[1007] "Effectiveness measurement" is the process of collecting and analyzing data to verify whether advertising is having the intended effect.

[1008] A "report" is a document that summarizes the results of effectiveness measurement and analysis, and is used as a guideline for the next strategy planning and improvement.

[1009] "Advertising Requirements" are detailed instructions that set out the conditions and goals necessary to run an advertising campaign.

[1010] "Operational optimization" refers to adjustments and improvement activities carried out to maximize advertising operational performance.

[1011] "Performance Data" means data used to evaluate the results of advertising operations, such as the number of times an advertisement is displayed, click-through rate, and number of conversions.

[1012] The present invention provides a video advertising production and management system that combines an emotion engine that recognizes user emotions to achieve advanced personalization. The following describes an embodiment of the system.

[1013] System configuration

[1014] This system consists of a server, a terminal, and a user. The specific hardware and software used are as follows:

[1015] 1. Server:

[1016] Generative model: Uses OpenAI's GPT model.

[1017] Big Data Analytics Unit: Uses AWS SageMaker and Google Cloud BigQuery.

[1018] Performance Measurement Unit: A Python script for collecting and analyzing performance data.

[1019] Optimization unit: A script that adjusts ad operations.

[1020] Emotion Engine: Emotion recognition software for recognizing the user's facial expressions and voice.

[1021] 2. Terminal:

[1022] User interface: Built with HTML / CSS and JavaScript, accepts input from the user.

[1023] Emotion recognition interface: Recognizes the user's emotions using a camera and microphone.

[1024] Dashboard: An interface that displays operational data and effectiveness measurement results.

[1025] 3. User:

[1026] Provide your advertising requirements: Enter your advertising campaign details.

[1027] Providing feedback: Provide feedback such as your ratings and opinions on the provided advertising video.

[1028] System operation explanation

[1029] Video ad requirements definition

[1030] User: Logs in to the portal site and enters details such as the target demographic, product information, and advertising objectives of the advertising campaign. For example, the user might enter, "The target demographic is young people in their 20s, and the purpose is to raise awareness of a new smartphone product."

[1031] Scenario generation and video generation

[1032] Server: Analyzes the requirements data sent by the user and passes it to the generative model. A Python script is used for this preprocessing. The generative model (e.g., GPT-3) generates effective advertising scenarios.

[1033] Server: Based on the generated scenario, images, audio, and text materials are collected and a video ad is automatically generated using software such as Adobe Premiere Pro or Final Cut Pro. The generated video is then encoded and a preview link is generated.

[1034] User: Check the generated video via the preview link and provide feedback if necessary. For example, "I'd like the user's voice to be more prominent in the final scene."

[1035] Improving video quality and preparing for operation

[1036] Server: Receives user feedback, analyzes the data using the big data analysis unit, refines the scenario based on the analysis results, and regenerates the video ad using the generative model.

[1037] Server: Encodes the regenerated video and notifies the user that it is ready for production.

[1038] Terminal: The device will notify the user that it is ready for operation, and the user can click the "Start operation" button to start advertising operations.

[1039] Effect monitoring, optimization, and emotion recognition

[1040] Server: Once the ad is launched, real-time performance data such as impressions, click-through rates, and conversions are collected.

[1041] Server: The emotion engine recognizes emotions from facial expressions and voice when users view advertisements, and dynamically adjusts the content of advertisements based on that data.

[1042] Server: Uses generative models to regenerate and optimize ads when performance is poor, and also dynamically adjusts ad timing and targeting.

[1043] Measurement and reporting

[1044] Server: After the advertising period ends, all data is aggregated and effectiveness is measured. A detailed report is generated based on the analysis results and provided to the user.

[1045] Users: They can review the provided reports and plan their next advertising strategy. For example, they can decide to target women in their early 20s in their next campaign.

[1046] Example prompts to input to the generative AI model

[1047] "The target demographic is young people in their 20s, and the goal is to raise awareness of a new smartphone product. Please generate an effective advertising scenario for this target demographic."

[1048] This system allows users to efficiently create and operate high-quality, effective video ads. In particular, by utilizing the emotion engine, it is possible to provide the optimal advertising experience for each user.

[1049] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1050] Step 1:

[1051] User: Enter your advertising requirements

[1052] Users log in to the portal site and enter the target demographic, product information, and purpose of the ad campaign. For example, they enter specific information such as "The target demographic is young people in their 20s, and the purpose is to raise awareness of a new smartphone product," and click the submit button. The entered data is sent to the server.

[1053] Input: Advertising requirement data

[1054] Output: Ad requirements data sent to the server

[1055] Step 2:

[1056] Server: Preprocessing requirements data

[1057] The server parses the ad requirement data received from users and converts it into a format suitable for the generative model. This preprocessing is done using Python scripts, such as cleaning and formatting the text data.

[1058] Input: Ad requirements data sent by the user

[1059] Output: Data converted into a format suitable for feeding into a generative model

[1060] Step 3:

[1061] Server: Automatic scenario generation

[1062] The server passes the preprocessed data to a generative model, which automatically generates advertising scenarios. Here, a generative AI model (e.g., GPT-3) is used. The generated scenarios are stored on the server.

[1063] Input: Data converted to a format suitable for feeding into a generative model

[1064] Output: Generated advertising scenario

[1065] Step 4:

[1066] Server: Collects advertising materials and generates videos

[1067] The server collects the necessary image, audio, and text materials based on the generated scenario. Using the collected materials, the video ad is automatically generated using software such as Adobe Premiere Pro or Final Cut Pro. The generated video is then encoded and a preview link is generated.

[1068] Input: Generated advertising scenario

[1069] Output: Generated video ad with preview link

[1070] Step 5:

[1071] Users: Video reviews and feedback

[1072] Users can watch the generated video via the provided preview link, review the video, and provide feedback on corrections or additional requirements as needed. For example, they could provide feedback such as "I would like the product name to be more prominent in the final scene."

[1073] Input: Preview link and user feedback

[1074] Output: Feedback data

[1075] Step 6:

[1076] Server: Feedback analysis and scenario refinement

[1077] The server analyzes the feedback received from the user and uses a big data analysis unit to analyze the data. Based on the analysis results, the advertising scenario is improved, for example, by modifying certain parts of the scenario based on the feedback.

[1078] Input: Feedback data

[1079] Output: Improved advertising scenario

[1080] Step 7:

[1081] Server: Regenerate videos and prepare for operation

[1082] The server regenerates the video ad using the generative model based on the improved scenario, re-encodes the video, and notifies the server that it is ready for production.

[1083] Input: Improved advertising scenario

[1084] Output: Regenerated video ads and notifications

[1085] Step 8:

[1086] Terminal: Notification of start of operation

[1087] The device will notify the user that it is ready for operation. The user will receive the notification, check the contents, and click the "Start operation" button to start advertising operations.

[1088] Input: Notification of operational readiness

[1089] Output: Ads go live

[1090] Step 9:

[1091] Server: Performance data collection and analysis

[1092] As soon as the ad starts running, the server collects performance data in real time, such as the number of impressions, click-through rates, and conversions, which are then analyzed by the effectiveness measurement unit.

[1093] Input: Ad performance data

[1094] Output: Parsed performance report

[1095] Step 10:

[1096] Server: Emotion data analysis and dynamic ad adjustment

[1097] The server's emotion engine recognizes emotions in real time from facial expressions and voice as the user views the advertisement. The content of the advertisement is dynamically adjusted based on this emotional data. For example, if the user shows interest, the content will be changed to emphasize that element.

[1098] Input: Emotion data

[1099] Output: Dynamically adjusted ad content

[1100] Step 11:

[1101] Server: Ad regeneration and optimization

[1102] The server uses the generative model to regenerate and optimize ads when performance is poor, and dynamically adjusts ad timing and targeting settings.

[1103] Input: Performance and emotion data

[1104] Output: Regenerated optimized video ad

[1105] Step 12:

[1106] Server: Performance measurement and report generation

[1107] Once the advertising period ends, the server aggregates all data and measures the effectiveness. Based on the results of the analysis, a detailed report is generated and provided to the user.

[1108] Input: All performance data

[1109] Output: Performance measurement results and reports

[1110] Step 13:

[1111] User: Check the report and consider the next strategy

[1112] Users can review the provided reports and evaluate the results of their advertising operations. Based on this evaluation, they can consider their next advertising strategy and come up with further improvements.

[1113] Input: Performance measurement report

[1114] Output: Next advertising strategy plan

[1115] (Application example 2)

[1116] Next, a description will be given of Application 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."

[1117] In the production and provision of modern advertising videos and entertainment content, it is difficult to consider the different emotions and interests of each user, and generic advertising may not be effective for all users. Furthermore, in order to maximize passenger satisfaction, entertainment systems inside autonomous vehicles are required to provide content tailored to passenger emotions in real time. However, current technology does not provide a method to achieve this.

[1118] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for generating videos for advertising using a generative model, means for generating advertising scenarios by analyzing big data, means for operating the generated videos and measuring their effectiveness, means for analyzing user emotions using an emotion recognition engine and dynamically optimizing the video content, and means for providing advertising and entertainment content based on emotion recognition. This makes it possible to provide highly personalized advertising and entertainment experiences according to the emotional state of each passenger.

[1119] A "generative model" is an algorithm that uses artificial intelligence technology to automatically create advertising videos.

[1120] "Big data" refers to a large amount of data collected from a wide variety of sources.

[1121] An "advertising scenario" is a plan or script that determines the content and structure of an advertising video.

[1122] "Operation" refers to the process of actually distributing the generated advertising video and observing its effectiveness.

[1123] "Effectiveness measurement" is an analytical method for evaluating the performance of advertising videos.

[1124] "Optimization" is the process of improving advertising videos and their delivery methods to maximize the desired effect.

[1125] An "emotion recognition engine" is a technology for analyzing emotions from a user's facial expressions and voice.

[1126] "Dynamic optimization" refers to changing the content of the advertising video in real time according to the user's state.

[1127] "Entertainment content" refers to various media materials that provide entertainment and interest to users.

[1128] "User emotional feedback" refers to the emotional response of users when they watch advertising videos or entertainment content.

[1129] The present invention provides a system for providing highly personalized advertising and entertainment experiences within an autonomous vehicle that are tailored to the emotional state of each passenger. Specific embodiments for implementing the present invention will be described below.

[1130] System configuration

[1131] This system consists of three main components: a server, a terminal, and a user.

[1132] 1. Server

[1133] Generative model: An artificial intelligence model for automatically generating advertising videos and entertainment content.

[1134] Big Data Analysis Unit: A unit that analyzes the performance of advertising scenarios and content.

[1135] Effectiveness measurement unit: A unit for measuring the effectiveness of ads and content in real time.

[1136] Optimization unit: A unit for optimizing the delivery method and content of ads and content.

[1137] Emotion engine: An engine for recognizing emotions from the user's facial expressions and voice.

[1138] 2. Terminal

[1139] In-car camera: A camera for capturing facial images of passengers.

[1140] User Interface: A display for displaying advertisements and content.

[1141] Emotion recognition interface: An interface for sending facial images captured by the in-car camera to a server.

[1142] Dashboard: A tool for monitoring the delivery and effectiveness of ads and content.

[1143] 3. Users

[1144] Passengers: Users who watch advertising and entertainment content in the vehicle.

[1145] Processing flow explanation

[1146] 1. Passenger Emotion Recognition

[1147] The device's onboard camera captures the passenger's face and transmits the image in real time to an emotion-recognition interface.

[1148] The emotion engine on the server analyzes the received images and predicts the passenger's emotional state using image processing libraries such as OpenCV and dlib, as well as deep learning models using Keras.

[1149] 2. Dynamic generation and optimization of advertising and entertainment content

[1150] The server's generative model generates advertising videos and entertainment content based on the emotion data obtained from the emotion engine.

[1151] The big data analysis unit creates content scenarios, and the effectiveness measurement unit monitors performance in real time.

[1152] The optimization unit adjusts content to maximize its effectiveness based on emotional state and collected performance data.

[1153] 3. Delivery of optimal content

[1154] The server sends the generated and optimized advertising and entertainment content to the terminal's user interface for display to the passenger.

[1155] Specific examples

[1156] For example, if a passenger makes a "disinterested" facial expression while in the car, the server's emotion engine will recognize that emotion and use a generative model to regenerate advertising content to attract more interest. An optimization unit then monitors and measures the effectiveness of the content used and makes further adjustments if necessary.

[1157] Prompt Sentence Examples

[1158] Design a system that recognizes emotions in real time from facial expressions that indicate the user's emotions when viewing an advertisement, and retrieves and displays the most appropriate video advertisement from the server based on that user's emotional state.

[1159] This enables the system to provide optimal advertising and entertainment experiences for each passenger.

[1160] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1161] Step 1:

[1162] The device's onboard camera captures the passenger's face and transmits the image in real time to an emotion-recognition interface.

[1163] Input: Passenger face image captured by in-car camera

[1164] Processing: Send the facial image to the emotion recognition interface

[1165] Output: Facial image sent to emotion recognition interface

[1166] Step 2:

[1167] The server's emotion engine analyzes the received images and predicts the passenger's emotional state.

[1168] Input: Facial image sent from emotion recognition interface

[1169] Processing: Detect faces using image processing libraries (OpenCV, dlib) and predict emotions using deep learning models (Keras)

[1170] Output: Predicted emotional state (e.g., uninterested, amused)

[1171] Step 3:

[1172] The server's generative model generates advertising videos and entertainment content based on the emotion data obtained from the emotion engine.

[1173] Input: Emotional state data from the emotion engine

[1174] Processing: Generative models use emotional state data to generate appropriate advertising and content scenarios, and collect and edit materials.

[1175] Output: Generated advertising video or entertainment content

[1176] Step 4:

[1177] The server's big data analysis unit creates content scenarios, and the effectiveness measurement unit monitors performance in real time.

[1178] Input: Generated ad video or content, various performance data (number of impressions, click-through rate, emotional feedback)

[1179] Processing: Analyzing performance data, building scenarios and making adjustments

[1180] Output: Improved scenarios and performance measurement data

[1181] Step 5:

[1182] The server's optimization unit adjusts content to maximize its effectiveness based on emotional state and collected performance data.

[1183] Input: Improved scenario, various performance data

[1184] Processing: Adjusting how and what content is delivered based on the optimization algorithms used by the optimization unit

[1185] Output: Optimized advertising videos or entertainment content

[1186] Step 6:

[1187] The server sends the generated and optimized advertising and entertainment content to the terminal's user interface for display to the passenger.

[1188] Input: Optimized ad video or entertainment content

[1189] Processing: Encode the content and send it to the device's user interface

[1190] Output: Advertising videos or entertainment content displayed on the device's display

[1191] Step 7:

[1192] The server's effectiveness measurement unit monitors the delivery status and effectiveness of advertisements and content.

[1193] Input: Various performance data (number of impressions, click-through rate, number of conversions, emotional feedback)

[1194] Processing: Analyzes the collected performance data and generates reports

[1195] Output: Reports provided to users, insights for future advertising operations

[1196] 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.

[1197] 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> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) 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.

[1198] 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.

[1199] [Third embodiment]

[1200] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1201] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[1202] 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).

[1203] 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.

[1204] 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.

[1205] 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).

[1206] 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.

[1207] 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.

[1208] 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.

[1209] 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.

[1210] 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.

[1211] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[1212] The present invention provides a system for efficiently implementing the entire process from video advertising production to operation, and specific embodiments thereof are described below. This system integrates generative models, big data analysis, feedback processing, and optimization functions to effectively produce and operate video advertising.

[1213] System configuration

[1214] The system consists of the following main components:

[1215] 1. Server

[1216] Generative Model

[1217] Big Data Analysis Unit

[1218] Effectiveness Measurement Unit

[1219] Optimization Unit

[1220] 2. Terminal

[1221] User Interface

[1222] Dashboard

[1223] 3. Users

[1224] Advertisement requirements provided

[1225] Providing Feedback

[1226] Program processing flow

[1227] Video ad requirements definition

[1228] User:

[1229] Users first log in to the portal site and enter the requirements for a new advertising campaign, including target demographics, product information, advertising objectives, etc. After entering the requirements, users can provide detailed information by clicking the "Submit" button to send them to the server.

[1230] server:

[1231] The server analyzes the requirements data received from the user and performs preprocessing before passing it to the Generative Model. This preprocessing converts the requirements data into an appropriate format so that the Generative Model can operate efficiently.

[1232] Scenario generation and video generation

[1233] server:

[1234] The server passes the preprocessed data to a generative model to generate an advertising scenario. Based on the generated scenario, the server collects additional materials such as images, audio, and text, and uses the generative model to automatically generate a video advertisement. The generated video is then encoded for initial review, and a preview link is generated.

[1235] User:

[1236] The user can review the generated video by checking the provided preview link. If feedback is required, the user can enter comments or corrections on the platform and send them back to the server.

[1237] Improving video quality and preparing for operation

[1238] server:

[1239] The server receives feedback from users and analyzes the big data to improve the advertising scenario. It uses patterns obtained from the big data to reflect improvements to the scenario. Based on the improved scenario, it encodes the regenerated video and prepares it for deployment. Once the deployment is ready, the server sends a notification to the user.

[1240] Device:

[1241] The terminal notifies the user that the operation preparation is complete. The user confirms the contents and clicks the "Start operation" button to start the advertisement operation.

[1242] Efficacy monitoring and optimization

[1243] server:

[1244] As soon as the server starts running the ad, it collects performance data such as the number of impressions, click-through rates, and number of conversions in real time. The collected data is analyzed by an effectiveness measurement unit, and if performance is poor, the ad is regenerated and optimized using a generative model. The server also adjusts the timing and targeting of ad displays as appropriate.

[1245] Examples:

[1246] Suppose a company creates an advertising video for a new smartphone product and starts running it. If the click-through rate on the first day is low, the server uses a generative model to improve the video and generates a new scenario based on big data analysis. Then, it changes the timing of the ad display to the optimal time and runs it again.

[1247] Measurement and reporting

[1248] server:

[1249] Once the ad run period is over, the server aggregates all data and measures its effectiveness. A detailed report summarizing the analysis results is generated and provided to the user. The report includes detailed results for each performance indicator and insights for future ad runs.

[1250] User:

[1251] Users can receive the provided report, check the results of their advertising operations, and plan their next advertising strategy based on the report contents.

[1252] The above is a specific description of the embodiment of the present invention. The present system enables users to efficiently create and operate high-quality, effective video advertisements.

[1253] The processing flow will be explained below.

[1254] Step 1:

[1255] User:

[1256] Users log in to the portal site and enter the requirements for a new advertising campaign, providing detailed information such as target demographic, product information, and advertising objectives (e.g., brand awareness, increased conversions). They then confirm the information they entered and click the "Submit" button.

[1257] Step 2:

[1258] server:

[1259] The server analyzes the requirements data received from the user and performs preprocessing before passing it to the generative model. Preprocessing converts the requirements data into an appropriate format so that the generative model can operate efficiently.

[1260] Step 3:

[1261] server:

[1262] The server passes the preprocessed data to a generative model to generate advertising scenarios. The generative model automatically generates effective advertising scenarios based on the input data.

[1263] Step 4:

[1264] server:

[1265] Based on the generated scenario, the server collects materials such as images, audio, and text. Using these materials, the generative model automatically creates a video ad. The generated video is encoded for early review and a preview link is generated.

[1266] Step 5:

[1267] User:

[1268] The user can review the generated video content by checking the provided preview link, and provide feedback and input corrections or additional requirements as needed.

[1269] Step 6:

[1270] server:

[1271] The server receives feedback from users and analyzes the big data to improve the advertising scenario. It utilizes patterns obtained from the big data to improve the scenario.

[1272] Step 7:

[1273] server:

[1274] Based on the improved scenario, the generative model is used to regenerate the video ad. The regenerated video is then encoded and made ready for deployment. Once ready, the server sends a notification to the user.

[1275] Step 8:

[1276] Device:

[1277] The device notifies the user that it is ready for operation. The user confirms the information and clicks the "Start operation" button to start advertising operations.

[1278] Step 9:

[1279] server:

[1280] As soon as the ad starts running, the server collects performance data in real time, such as the number of impressions, click-through rates, and conversions, which are then analyzed by the effectiveness measurement unit.

[1281] Step 10:

[1282] server:

[1283] If performance is poor, the server uses the generative model to regenerate and optimize the ads, adjusting the timing and targeting of the ads accordingly.

[1284] Step 11:

[1285] server:

[1286] At the end of the advertising period, the server aggregates all data and measures the effectiveness. Based on the analysis results, a detailed report is generated and provided to the user.

[1287] Step 12:

[1288] User:

[1289] Users can receive the provided report, check the results of their advertising operations, and plan their next advertising strategy based on the report contents.

[1290] The above are the specific processing steps for carrying out the present invention. This system allows users to efficiently create and operate high-quality, effective video advertisements.

[1291] Example 1

[1292] Next, a description will be given of 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."

[1293] Traditional advertising production and management processes are time-consuming, resource-intensive, and inefficient. In particular, they lack flexibility and make it difficult to measure and optimize advertising effectiveness in real time. Another issue is the difficulty of quickly incorporating user feedback and quickly improving advertising content.

[1294] 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.

[1295] In this invention, the server includes means for generating videos for advertisements using a generative AI model, means for analyzing large-scale data to generate advertising scenarios, means for operating the generated videos and measuring their effectiveness, means for optimizing the operation of advertisements based on the measurement results, means for receiving and preprocessing advertising requirements from users, means for receiving user feedback on the generated videos, and means for improving and regenerating the videos based on the feedback. This streamlines the process of advertisement production and operation, enables real-time measurement and optimization of effectiveness, and enables rapid improvement that reflects user feedback.

[1296] A "generative AI model" is an artificial intelligence model that automatically generates advertising videos and scenarios based on specified prompt text.

[1297] "Large-scale data" refers to large amounts of data, such as advertising operations and user feedback, and is a collection of information obtained by analyzing this data.

[1298] "Scenario generation" is the process of designing the structure, content, and storyboard of an advertising video.

[1299] "Effectiveness measurement" is the process of collecting data such as the number of impressions, click rates, and number of conversions obtained during advertising operations and evaluating the performance of the advertisements.

[1300] "Optimization" refers to the process of adjusting advertising operations based on the results of effectiveness measurement to improve performance.

[1301] "User feedback" refers to comments and correction instructions given by the user to the generated advertising video.

[1302] "Regeneration" is the process of recreating advertising videos and scenarios based on feedback and effectiveness measurement results.

[1303] "Preprocessing" is the initial process of analyzing the advertising requirement data received from the user and converting it into a format that is easy for the generative AI model to process.

[1304] The present invention is a system for efficiently implementing the entire process from video advertising production to operation, and specific embodiments thereof are described below. This system integrates generative AI models, large-scale data analysis, feedback processing, and optimization functions to effectively produce and operate video advertising.

[1305] System configuration

[1306] The system consists of the following main components:

[1307] 1. Server

[1308] Generative AI Models

[1309] Large-Scale Data Analysis Unit

[1310] Effectiveness Measurement Unit

[1311] Optimization Unit

[1312] 2. Terminal

[1313] User Interface

[1314] Dashboard

[1315] 3. Users

[1316] Advertisement requirements provided

[1317] Providing Feedback

[1318] Video ad requirements definition

[1319] User:

[1320] Users log in to the portal site and enter the requirements for a new advertising campaign. These requirements include target demographics, product information, advertising objectives, etc., and users can provide detailed information. For example, they can enter a requirement such as "Raising awareness of a new headphone for music lovers aged 25-35." After entering the information, users click the "Submit" button to send it to the server.

[1321] server:

[1322] The server analyzes the requirements data received from the user and performs preprocessing before passing it to the generative AI model. This preprocessing converts the requirements data into an appropriate format, allowing the generative AI model to operate efficiently.

[1323] Scenario generation and video generation

[1324] server:

[1325] The server then passes the preprocessed data to a generative AI model to generate an advertising scenario. For example, it uses a prompt such as, "Generate a scenario that will resonate with music lovers aged 25-35." Based on the generated scenario, the server collects additional materials such as images, audio, and text, and uses the generative model to automatically generate a video ad. The generated video is then encoded for initial review, and a preview link is generated.

[1326] User:

[1327] The user opens the provided preview link to check the content of the generated video, and if necessary, enters comments or corrections on the platform and sends it back to the server.

[1328] Improving video quality and preparing for operation

[1329] server:

[1330] The server receives user feedback, analyzes large-scale data, and refines the advertising scenario. It uses patterns extracted from the large-scale data to improve the scenario. Based on the refined scenario, the video is regenerated and ready for deployment.

[1331] Device:

[1332] The device will notify the user that it is ready for operation. The user can confirm the details and click the "Start operation" button to start advertising operations.

[1333] Efficacy monitoring and optimization

[1334] server:

[1335] As soon as the server starts running the ad, it collects performance data such as the number of impressions, click-through rates, and number of conversions in real time. The collected data is analyzed by an effectiveness measurement unit, and if performance is poor, the ad is regenerated and optimized using a generative model. The timing and targeting of ad displays are also adjusted as appropriate.

[1336] Examples:

[1337] For example, if a company is running an ad for a new smartphone product and the click-through rate on the first day is low, the server will use the generative model to generate new scenarios, improve the video, optimize the timing of the landing page, and adjust the display time of the ad.

[1338] Measurement and reporting

[1339] server:

[1340] Once the ad run period is over, the server aggregates all data and measures its effectiveness. A detailed report summarizing the analysis results is generated and provided to the user. The report includes detailed results for each performance indicator and insights for future ad runs.

[1341] User:

[1342] Users can receive the provided report, check the results of their advertising operations, and plan their next advertising strategy based on the report contents.

[1343] This system streamlines the video ad production and operation process, enabling real-time measurement and optimization of effectiveness. It also allows for quick reflection of user feedback, making it possible to constantly improve ad content to keep it up-to-date and optimal.

[1344] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1345] Step 1: Enter user requirements

[1346] User:

[1347] Input: Users log in to the portal site and enter the requirements for a new advertising campaign, providing detailed information in text format, such as the target demographic (e.g., "music lovers aged 25-35"), product information, and advertising objectives (e.g., "increase awareness of a new product").

[1348] Output: User-entered requirements data.

[1349] Action: The user clicks the "Submit" button to send the entered requirements data to the server.

[1350] Step 2: Preprocessing requirements data

[1351] server:

[1352] Input: Requirement data submitted by the user.

[1353] Output: Preprocessed requirements data (structured data).

[1354] How it works: The server parses the received requirements data and preprocesses it to be passed to the generative AI model. Specifically, it structures the text data and splits it into necessary fields.

[1355] Step 3: Scenario generation

[1356] server:

[1357] Input: Preprocessed requirements data.

[1358] Output: The generated advertising scenario.

[1359] How it works: The server inputs the preprocessed data into a generative AI model and generates advertising scenarios using prompts, such as "Generate a scenario that will resonate with music lovers aged 25-35."

[1360] Step 4: Gathering materials and creating a video

[1361] server:

[1362] Input: The generated ad scenario.

[1363] Output: The generated video ad file.

[1364] How it works: Based on the generated scenario, the server collects the necessary images, audio, and text materials, and automatically generates a video ad using a generative AI model. The generated video file is encoded and a preview link is generated.

[1365] Step 5: Video preview and feedback

[1366] User:

[1367] Input: The video generated using the preview link.

[1368] Output: User feedback (comments and suggestions for corrections).

[1369] How it works: The user opens the provided preview link, checks the generated video, and optionally enters comments and corrections on the platform, which then sends the video back to the server.

[1370] Step 6: Receive and analyze feedback

[1371] server:

[1372] Input: User feedback.

[1373] Output: The scenario with the improvements.

[1374] How it works: The server receives feedback from users and analyzes it using a large-scale data analysis unit. Based on the analysis results, it extracts areas for improvement in the scenario and generates a new scenario using a generative AI model.

[1375] Step 7: Scenario refinement and video regeneration

[1376] server:

[1377] Input: The scenario with the improvements.

[1378] Output: Regenerated video ad files.

[1379] Operation: The server gathers additional material and regenerates the video based on the improved scenario. The regenerated video file is then re-encoded and ready for production.

[1380] Step 8: Notification of operational readiness

[1381] server:

[1382] Input: The regenerated video ad file.

[1383] Output: Operational readiness notification.

[1384] Action: The server notifies the user that it is ready for operation.

[1385] Step 9: Go live

[1386] Device:

[1387] Input: Operational readiness notification.

[1388] Output: Advertising videos are launched.

[1389] Operation: The device displays a notification to the user that preparation for operation is complete. The user clicks the "Start operation" button to start ad operation.

[1390] Step 10: Collecting performance measurement data

[1391] server:

[1392] Input: Ad video performance data (number of impressions, click-through rate, number of conversions, etc.).

[1393] Output: Performance measurement data.

[1394] How it works: The server starts running ads and collects real-time performance data such as impressions, click-through rates, and conversions.

[1395] Step 11: Effectiveness analysis and optimization

[1396] server:

[1397] Input: Performance measurement data.

[1398] Output: Improved and optimized ad video.

[1399] How it works: The server analyzes the collected data using an effectiveness measurement unit, and if performance is poor, it uses a generative model to regenerate ads, adjusting the timing and targeting of ads accordingly.

[1400] Step 12: Measurement and reporting

[1401] server:

[1402] Input: Aggregated performance measurement data.

[1403] Output: Detailed performance measurement report.

[1404] How it works: After the ad campaign period ends, the server aggregates all data, measures the effectiveness, and generates a detailed report summarizing the analysis results, which is then provided to the user.

[1405] Step 13: Review reports and develop strategies

[1406] User:

[1407] Input: Detailed performance measurement report.

[1408] Output: Next advertising strategy.

[1409] How it works: Users can receive the provided report, check the results of their advertising campaigns, and plan their next advertising strategy based on the report.

[1410] (Application example 1)

[1411] Next, a description will be given of Application 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."

[1412] Traditional video ad production systems have the problem that the processes of ad creation, operation, effectiveness measurement, and optimization are fragmented, making efficient operation difficult. Furthermore, there is a lack of a system that can optimize ads in real time and immediately reflect user feedback. As a result, advertising effectiveness is reduced and user satisfaction is low.

[1413] 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.

[1414] In this invention, the server includes a means for generating videos for advertisements using a generative model, a means for analyzing big data to generate advertising scenarios, a means for operating the generated videos and measuring their effectiveness, a means for optimizing the operation of advertisements based on the measurement results, a means for displaying the optimized video advertisements in real time using smart glasses, and a means for regenerating and optimizing advertisements based on collected data. This enables the entire process from the production of video advertisements to their operation and feedback to be integrated and operated efficiently and effectively.

[1415] A "generative model" is an algorithm that automatically generates scenarios and videos for advertising.

[1416] "Big data" refers to large amounts of data, a group of data that is used to generate and optimize advertising scenarios.

[1417] An "advertising scenario" is a plan that defines the content and progression of a video advertisement.

[1418] "Operation" refers to the process of actually delivering the generated video advertisements and displaying them to users.

[1419] "Effectiveness measurement" is the process of evaluating the results of advertising and collecting and analyzing performance data.

[1420] "Optimization" refers to making adjustments based on collected data to improve the performance of your ads.

[1421] "Smart glasses" are a type of wearable device that visually displays information to the user while they are wearing them.

[1422] "Real-time" means that processing and reaction are immediate and based on the current time zone.

[1423] "Feedback" means reviews and correction instructions provided by users.

[1424] An "interface" is a point of contact or means by which a user interacts with a system.

[1425] "Data" refers to a set of information or numbers that a system processes.

[1426] "Regeneration" is the process of creating a new version of an ad based on previously generated content.

[1427] The present invention provides a system for efficiently implementing the entire process from video advertising production to operation, and is characterized by its ability to display optimized video advertising in real time using smart glasses. Specific embodiments are described below. The system integrates generative models, big data analysis, feedback processing, and optimization functions to effectively produce and operate video advertising.

[1428] System configuration

[1429] The system consists of the following main components:

[1430] 1. Server:

[1431] Generative model: An algorithm that automatically generates scenarios and videos for advertising.

[1432] Big Data Analysis Unit: A data set used to analyze large amounts of data and measure and optimize advertising effectiveness.

[1433] Effectiveness Measurement Unit: Measures the effectiveness of advertising and analyzes the collected data.

[1434] Optimization unit: Optimizes your ads based on the data collected to improve their performance.

[1435] 2. Terminal:

[1436] Smart glasses: Glasses that visually display information while worn by the user (e.g., Google Glass, Vuzix Blade).

[1437] User Interface: An interface for entering ad requirements, previewing, and providing feedback.

[1438] Dashboard: A management screen for displaying advertising results and effectiveness measurement results.

[1439] 3. User:

[1440] Provide Ad Requirements: Enter your requirements for a new ad campaign.

[1441] Providing feedback: Review the content of the generated video and provide feedback.

[1442] System Operation

[1443] 1. Video ad requirements definition:

[1444] Users input their advertising requirements through the smart glasses interface, including target demographic, product information, and advertising objectives, and then the requirements are sent to the server.

[1445] 2. Scenario and video generation:

[1446] The server analyzes the received requirements data and passes it to the generative model. The generative model generates a scenario, collects the necessary materials (images, audio, text), and automatically generates a video advertisement. The generated video is encoded for initial review and streamed to the smart glasses.

[1447] 3. Viewing videos and providing feedback:

[1448] Users can view the generated video through smart glasses and provide feedback using voice commands or touch gestures, which is then sent to the server, which regenerates and optimizes the video as needed.

[1449] 4. Effect monitoring and optimization:

[1450] The server monitors the effectiveness of the ad (number of views, viewing time, user response) in real time. If the effectiveness is low, it will be regenerated and optimized to improve the ad performance.

[1451] 5. Report Generation:

[1452] The server aggregates all data, measures effectiveness, and generates detailed reports to provide users with insights to help them plan their next advertising strategy.

[1453] Specific examples

[1454] For example, consider a company that is creating and running an advertising video for a new smartphone product. The target demographic is technology-conscious people in their 20s and 30s, and the goal of the ad is to stimulate purchasing desire and increase brand awareness. Based on this requirement, the following prompt sentences are input into the generative AI model:

[1455] Prompt statement:

[1456] The target demographic for a new smartphone is tech-savvy people in their 20s and 30s. The advertising objective is to stimulate purchase intent and increase brand awareness. Based on this requirement, generate the following scenario and creative. The ad should include high-performance camera features, a long battery life, and a seamless user experience.

[1457] As described above, this system allows users to view high-quality, effective video advertisements in real time using smart glasses, and allows for immediate feedback.

[1458] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1459] Step 1: User Input

[1460] Through the smart glasses' user interface, users input requirements for a new advertising campaign, including target demographic, product information, and advertising objectives. The input is made on the interface and transmitted to the server. The input data also specifies specific elements and conditions that should be included in the advertisement.

[1461] Input: target demographic, product information, advertising objectives

[1462] Output: Requirements data sent to the server

[1463] Step 2: Requirement Data Analysis

[1464] The server analyzes the requirements data received from the user, and the analysis results are converted into a format that can be passed to the generative AI model. This preprocessing allows the generative model to operate efficiently.

[1465] Input: Requirement data

[1466] Output: Data converted into a format to be passed to the generative model

[1467] Step 3: Scenario generation

[1468] The server passes the preprocessed data to the generative model to generate an advertising scenario. The generative model constructs a scenario based on the prompt sentence and determines the story of the advertising video.

[1469] Input: Preprocessed data, generative AI model

[1470] Output: Generated scenario

[1471] Step 4: Gathering materials and creating a video

[1472] The server collects materials such as images, audio, and text based on the generated scenario, then uses the generative model to automatically generate advertising videos, which are then encoded and prepared for streaming to the smart glasses.

[1473] Input: Generated scenario, material data

[1474] Output: Generated video ad

[1475] Step 5: Advertising and Feedback

[1476] The user watches the generated video advertisement through the smart glasses. If feedback is required, the user can provide it using voice commands or touch gestures. The feedback is then sent to the server.

[1477] Input: Generated video ad

[1478] Output: User feedback

[1479] Step 6: Feedback analysis and video regeneration

[1480] The server analyzes the feedback received from the user, refines the content of the video advertisement based on the feedback, and regenerates the video, which is then re-encoded and streamed to the smart glasses.

[1481] Input: User feedback

[1482] Output: Regenerated video ad

[1483] Step 7: Measure your results

[1484] The server starts running the ad and simultaneously collects viewing data (number of views, viewing time, user response) in real time. The collected data is analyzed by the effectiveness measurement unit and the ad is optimized as necessary.

[1485] Input: Viewing data

[1486] Output: Effectiveness measurement report, optimization instructions

[1487] Step 8: Generate reports

[1488] After the ad run period ends, the server aggregates all data and generates a detailed performance measurement report, which is provided to the user and contains insights for planning their next advertising strategy.

[1489] Input: Aggregated operational data

[1490] Output: Performance measurement report

[1491] Through the above process, the present invention makes it possible to efficiently and effectively produce, operate, collect feedback, and optimize advertising videos.

[1492] 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.

[1493] The present invention is a system for efficiently and effectively implementing the entire process from the production to the operation of video advertisements, and in particular, by combining an emotion engine that recognizes user emotions, it achieves a high level of personalization. Below, we will explain the form for implementing the present invention and the specific processing content thereof.

[1494] System configuration

[1495] The system consists of the following main components:

[1496] 1. Server

[1497] Generative Model

[1498] Big Data Analysis Unit

[1499] Effectiveness Measurement Unit

[1500] Optimization Unit

[1501] Emotion Engine

[1502] 2. Terminal

[1503] User Interface

[1504] Emotion Recognition Interface

[1505] Dashboard

[1506] 3. Users

[1507] Advertisement requirements provided

[1508] Providing Feedback

[1509] Program processing flow

[1510] Video ad requirements definition

[1511] User:

[1512] First, users log in to the portal site and enter the requirements for a new advertising campaign, providing detailed information such as target demographic, product information, and advertising objectives (e.g., brand awareness, increased conversions). They then confirm the information they entered and click the "Submit" button.

[1513] server:

[1514] The server analyzes the requirements data received from the user and performs preprocessing before passing it to the Generative Model. This preprocessing converts the requirements data into an appropriate format so that the Generative Model can operate efficiently.

[1515] Scenario generation and video generation

[1516] server:

[1517] The server passes the preprocessed data to a generative model to generate advertising scenarios. The generative model automatically generates effective advertising scenarios based on the input data.

[1518] server:

[1519] Based on the generated scenario, the server collects materials such as images, audio, and text. Using these materials, the generative model automatically creates a video ad. The generated video is encoded for early review and a preview link is generated.

[1520] User:

[1521] The user can review the generated video content by checking the provided preview link, and provide feedback and input corrections or additional requirements as needed.

[1522] Improving video quality and preparing for operation

[1523] server:

[1524] The server receives feedback from users and analyzes the big data to improve the advertising scenario. It utilizes patterns obtained from the big data to improve the scenario.

[1525] server:

[1526] Based on the improved scenario, the generative model is used to regenerate the video ad. The regenerated video is then encoded and made ready for deployment. Once ready, the server sends a notification to the user.

[1527] Device:

[1528] The device notifies the user that it is ready for operation. The user confirms the information and clicks the "Start operation" button to start advertising operations.

[1529] Effect monitoring, optimization, and emotion recognition

[1530] server:

[1531] As soon as the ad starts running, the server collects performance data in real time, such as the number of impressions, click-through rates, and conversions, which are then analyzed by the effectiveness measurement unit.

[1532] server:

[1533] The emotion engine recognizes emotions in real time from facial expressions and voice when a user views an ad, and dynamically adjusts the content of the ad based on that data. For example, if a user has a happy expression while viewing an ad, the engine can add more engaging elements to match that positive emotion.

[1534] server:

[1535] If performance is poor, the server uses the generative model to regenerate and optimize the ads, adjusting the timing and targeting of the ads accordingly.

[1536] Examples:

[1537] Suppose a company creates and starts using an advertising video for a new smartphone product. If many users who watch the ad feel uninterested, the server uses an emotion engine to analyze the emotion data, improves the advertising scenario, and regenerates it to make it more interesting.

[1538] Measurement and reporting

[1539] server:

[1540] Once the ad campaign period ends, the server aggregates all data and measures its effectiveness. Based on the analysis results, a detailed report is generated and provided to the user. The report includes detailed results for each performance indicator and insights for future ad campaigns.

[1541] User:

[1542] Users can receive the provided report, check the results of their advertising operations, and plan their next advertising strategy based on the report contents.

[1543] The above is a specific description of an embodiment of the present invention. This system allows users to efficiently create and manage high-quality, effective video advertisements. Furthermore, by utilizing the emotion engine, it is possible to achieve a high level of personalization of advertisements and provide each user with an optimal advertising experience.

[1544] The processing flow will be explained below.

[1545] Step 1:

[1546] User:

[1547] Users log in to the portal site and enter the requirements for a new advertising campaign, including detailed information such as target demographic, product information, and advertising objectives (e.g., brand awareness, increased conversions). They confirm the information they entered and click the "Submit" button.

[1548] Step 2:

[1549] server:

[1550] The server analyzes the requirements data received from the user and performs preprocessing before passing it to the Generative Model. This preprocessing converts the provided data into an appropriate format so that the Generative Model can operate efficiently.

[1551] Step 3:

[1552] server:

[1553] The server passes the preprocessed data to a generative model to generate advertising scenarios. The generative model automatically generates effective advertising scenarios based on the input data.

[1554] Step 4:

[1555] server:

[1556] Based on the generated scenario, the server collects materials such as images, audio, and text. Using these materials, the generative model automatically creates a video ad. The generated video is encoded for early review and a preview link is generated.

[1557] Step 5:

[1558] User:

[1559] The user can review the generated video content by checking the provided preview link, and provide feedback and input corrections or additional requirements as needed.

[1560] Step 6:

[1561] server:

[1562] The server receives feedback from users and analyzes the big data to improve the advertising scenario. It utilizes patterns obtained from the big data to improve the scenario.

[1563] Step 7:

[1564] server:

[1565] Based on the improved scenario, the generative model is used to regenerate the video ad. The regenerated video is then encoded and made ready for deployment. Once ready, the server sends a notification to the user.

[1566] Step 8:

[1567] Device:

[1568] The device notifies the user that it is ready for operation. The user confirms the information and clicks the "Start operation" button to start advertising operations.

[1569] Step 9:

[1570] server:

[1571] As soon as the ad starts running, performance data such as the number of impressions, click-through rate, and number of conversions is collected in real time, and the collected data is analyzed by the effectiveness measurement unit.

[1572] Step 10:

[1573] server:

[1574] The emotion engine recognizes emotions in real time from facial expressions and voice when a user views an advertisement, and dynamically adjusts the content of the advertisement based on that data. For example, if a user is not interested in an advertisement, the scenario will be improved by reflecting that emotional data.

[1575] Step 11:

[1576] server:

[1577] If performance is poor, the server uses the generative model to regenerate and optimize the ads, adjusting the timing and targeting of the ads accordingly.

[1578] Step 12:

[1579] server:

[1580] Once the advertising period ends, the server aggregates all data and measures the effectiveness. Based on the analysis results, a detailed report is generated and provided to the user.

[1581] Step 13:

[1582] User:

[1583] Users can receive the provided report, check the results of their advertising operations, and plan their next advertising strategy based on the report contents.

[1584] The above are the specific processing steps for implementing the present invention in combination with an emotion engine. This system allows users to efficiently create and manage high-quality, effective video advertisements, and by utilizing the emotion engine, highly personalized advertisements can be achieved.

[1585] Example 2

[1586] Next, a description will be given of 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."

[1587] Conventional advertising video production and operation systems lacked personalization, making it difficult to optimize by incorporating each user's emotions and real-time feedback. Furthermore, the process of measuring effectiveness and analyzing feedback was complicated, making it difficult to respond quickly to improve advertising effectiveness. This resulted in problems such as a lack of improvement in user experience and reduced advertising effectiveness.

[1588] 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.

[1589] In this invention, the server includes means for generating videos for advertisements using a generative model, means for analyzing big data to generate advertising scenarios, means for operating the generated videos and measuring their effectiveness, means for recognizing emotions from a user's facial expressions and voice and dynamically adjusting advertising content based on the emotion data, means for analyzing the content of feedback and regenerating videos based on the analysis results, and means for aggregating the results of measuring the effectiveness of advertising operation, generating detailed reports, and providing them to users. This makes it possible to highly personalize advertisements and optimize them in real time.

[1590] A "generative model" is a machine learning model that uses algorithms to automatically generate content.

[1591] "Big data" is a collection of large amounts of diverse data, the analysis of which makes it possible to discover new insights and patterns.

[1592] An "advertising scenario" is a detailed design document for planning the composition and development of an advertising video.

[1593] "Emotion recognition" is a technology that analyzes non-verbal information such as a user's facial expressions and voice to recognize their emotional state.

[1594] "Dynamic adjustment" is the process of instantly changing advertising content based on real-time data.

[1595] "Feedback" refers to the evaluations and opinions provided by users, and the system is improved and adjusted based on them.

[1596] "Effectiveness measurement" is the process of collecting and analyzing data to verify whether advertising is having the intended effect.

[1597] A "report" is a document that summarizes the results of effectiveness measurement and analysis, and is used as a guideline for the next strategy planning and improvement.

[1598] "Advertising Requirements" are detailed instructions that set out the conditions and goals necessary to run an advertising campaign.

[1599] "Operational optimization" refers to adjustments and improvement activities carried out to maximize advertising operational performance.

[1600] "Performance Data" means data used to evaluate the results of advertising operations, such as the number of times an advertisement is displayed, click-through rate, and number of conversions.

[1601] The present invention provides a video advertising production and management system that combines an emotion engine that recognizes user emotions to achieve advanced personalization. The following describes an embodiment of the system.

[1602] System configuration

[1603] This system consists of a server, a terminal, and a user. The specific hardware and software used are as follows:

[1604] 1. Server:

[1605] Generative model: Uses OpenAI's GPT model.

[1606] Big Data Analytics Unit: Uses AWS SageMaker and Google Cloud BigQuery.

[1607] Performance Measurement Unit: A Python script for collecting and analyzing performance data.

[1608] Optimization unit: A script that adjusts ad operations.

[1609] Emotion Engine: Emotion recognition software for recognizing the user's facial expressions and voice.

[1610] 2. Terminal:

[1611] User interface: Built with HTML / CSS and JavaScript, accepts input from the user.

[1612] Emotion recognition interface: Recognizes the user's emotions using a camera and microphone.

[1613] Dashboard: An interface that displays operational data and effectiveness measurement results.

[1614] 3. User:

[1615] Provide your advertising requirements: Enter your advertising campaign details.

[1616] Providing feedback: Provide feedback such as your ratings and opinions on the provided advertising video.

[1617] System operation explanation

[1618] Video ad requirements definition

[1619] User: Logs in to the portal site and enters details such as the target demographic, product information, and advertising objectives of the advertising campaign. For example, the user might enter, "The target demographic is young people in their 20s, and the purpose is to raise awareness of a new smartphone product."

[1620] Scenario generation and video generation

[1621] Server: Analyzes the requirements data sent by the user and passes it to the generative model. A Python script is used for this preprocessing. The generative model (e.g., GPT-3) generates effective advertising scenarios.

[1622] Server: Based on the generated scenario, images, audio, and text materials are collected and a video ad is automatically generated using software such as Adobe Premiere Pro or Final Cut Pro. The generated video is then encoded and a preview link is generated.

[1623] User: Check the generated video via the preview link and provide feedback if necessary. For example, "I'd like the user's voice to be more prominent in the final scene."

[1624] Improving video quality and preparing for operation

[1625] Server: Receives user feedback, analyzes the data using the big data analysis unit, refines the scenario based on the analysis results, and regenerates the video ad using the generative model.

[1626] Server: Encodes the regenerated video and notifies the user that it is ready for production.

[1627] Terminal: The device will notify the user that it is ready for operation, and the user can click the "Start operation" button to start advertising operations.

[1628] Effect monitoring, optimization, and emotion recognition

[1629] Server: Once the ad is launched, real-time performance data such as impressions, click-through rates, and conversions are collected.

[1630] Server: The emotion engine recognizes emotions from facial expressions and voice when users view advertisements, and dynamically adjusts the content of advertisements based on that data.

[1631] Server: Uses generative models to regenerate and optimize ads when performance is poor, and also dynamically adjusts ad timing and targeting.

[1632] Measurement and reporting

[1633] Server: After the advertising period ends, all data is aggregated and effectiveness is measured. A detailed report is generated based on the analysis results and provided to the user.

[1634] Users: They can review the provided reports and plan their next advertising strategy. For example, they can decide to target women in their early 20s in their next campaign.

[1635] Example prompts to input to the generative AI model

[1636] "The target demographic is young people in their 20s, and the goal is to raise awareness of a new smartphone product. Please generate an effective advertising scenario for this target demographic."

[1637] This system allows users to efficiently create and operate high-quality, effective video ads. In particular, by utilizing the emotion engine, it is possible to provide the optimal advertising experience for each user.

[1638] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1639] Step 1:

[1640] User: Enter your advertising requirements

[1641] Users log in to the portal site and enter the target demographic, product information, and purpose of the ad campaign. For example, they enter specific information such as "The target demographic is young people in their 20s, and the purpose is to raise awareness of a new smartphone product," and click the submit button. The entered data is sent to the server.

[1642] Input: Advertising requirement data

[1643] Output: Ad requirements data sent to the server

[1644] Step 2:

[1645] Server: Preprocessing requirements data

[1646] The server parses the ad requirement data received from users and converts it into a format suitable for the generative model. This preprocessing is done using Python scripts, such as cleaning and formatting the text data.

[1647] Input: Ad requirements data sent by the user

[1648] Output: Data converted into a format suitable for feeding into a generative model

[1649] Step 3:

[1650] Server: Automatic scenario generation

[1651] The server passes the preprocessed data to a generative model, which automatically generates advertising scenarios. Here, a generative AI model (e.g., GPT-3) is used. The generated scenarios are stored on the server.

[1652] Input: Data converted to a format suitable for feeding into a generative model

[1653] Output: Generated advertising scenario

[1654] Step 4:

[1655] Server: Collects advertising materials and generates videos

[1656] The server collects the necessary image, audio, and text materials based on the generated scenario. Using the collected materials, the video ad is automatically generated using software such as Adobe Premiere Pro or Final Cut Pro. The generated video is then encoded and a preview link is generated.

[1657] Input: Generated advertising scenario

[1658] Output: Generated video ad with preview link

[1659] Step 5:

[1660] Users: Video reviews and feedback

[1661] Users can watch the generated video via the provided preview link, review the video, and provide feedback on corrections or additional requirements as needed. For example, they could provide feedback such as "I would like the product name to be more prominent in the final scene."

[1662] Input: Preview link and user feedback

[1663] Output: Feedback data

[1664] Step 6:

[1665] Server: Feedback analysis and scenario refinement

[1666] The server analyzes the feedback received from the user and uses a big data analysis unit to analyze the data. Based on the analysis results, the advertising scenario is improved, for example, by modifying certain parts of the scenario based on the feedback.

[1667] Input: Feedback data

[1668] Output: Improved advertising scenario

[1669] Step 7:

[1670] Server: Regenerate videos and prepare for operation

[1671] The server regenerates the video ad using the generative model based on the improved scenario, re-encodes the video, and notifies the server that it is ready for production.

[1672] Input: Improved advertising scenario

[1673] Output: Regenerated video ads and notifications

[1674] Step 8:

[1675] Terminal: Notification of start of operation

[1676] The device will notify the user that it is ready for operation. The user will receive the notification, check the contents, and click the "Start operation" button to start advertising operations.

[1677] Input: Notification of operational readiness

[1678] Output: Ads go live

[1679] Step 9:

[1680] Server: Performance data collection and analysis

[1681] As soon as the ad starts running, the server collects performance data in real time, such as the number of impressions, click-through rates, and conversions, which are then analyzed by the effectiveness measurement unit.

[1682] Input: Ad performance data

[1683] Output: Parsed performance report

[1684] Step 10:

[1685] Server: Emotion data analysis and dynamic ad adjustment

[1686] The server's emotion engine recognizes emotions in real time from facial expressions and voice as the user views the advertisement. The content of the advertisement is dynamically adjusted based on this emotional data. For example, if the user shows interest, the content will be changed to emphasize that element.

[1687] Input: Emotion data

[1688] Output: Dynamically adjusted ad content

[1689] Step 11:

[1690] Server: Ad regeneration and optimization

[1691] The server uses the generative model to regenerate and optimize ads when performance is poor, and dynamically adjusts ad timing and targeting settings.

[1692] Input: Performance and emotion data

[1693] Output: Regenerated optimized video ad

[1694] Step 12:

[1695] Server: Performance measurement and report generation

[1696] Once the advertising period ends, the server aggregates all data and measures the effectiveness. Based on the results of the analysis, a detailed report is generated and provided to the user.

[1697] Input: All performance data

[1698] Output: Performance measurement results and reports

[1699] Step 13:

[1700] User: Check the report and consider the next strategy

[1701] Users can review the provided reports and evaluate the results of their advertising operations. Based on this evaluation, they can consider their next advertising strategy and come up with further improvements.

[1702] Input: Performance measurement report

[1703] Output: Next advertising strategy plan

[1704] (Application example 2)

[1705] Next, a description will be given of Application 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."

[1706] In the production and provision of modern advertising videos and entertainment content, it is difficult to consider the different emotions and interests of each user, and generic advertising may not be effective for all users. Furthermore, in order to maximize passenger satisfaction, entertainment systems inside autonomous vehicles are required to provide content tailored to passenger emotions in real time. However, current technology does not provide a method to achieve this.

[1707] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for generating videos for advertising using a generative model, means for generating advertising scenarios by analyzing big data, means for operating the generated videos and measuring their effectiveness, means for analyzing user emotions using an emotion recognition engine and dynamically optimizing the video content, and means for providing advertising and entertainment content based on emotion recognition. This makes it possible to provide highly personalized advertising and entertainment experiences according to the emotional state of each passenger.

[1708] A "generative model" is an algorithm that uses artificial intelligence technology to automatically create advertising videos.

[1709] "Big data" refers to a large amount of data collected from a wide variety of sources.

[1710] An "advertising scenario" is a plan or script that determines the content and structure of an advertising video.

[1711] "Operation" refers to the process of actually distributing the generated advertising video and observing its effectiveness.

[1712] "Effectiveness measurement" is an analytical method for evaluating the performance of advertising videos.

[1713] "Optimization" is the process of improving advertising videos and their delivery methods to maximize the desired effect.

[1714] An "emotion recognition engine" is a technology for analyzing emotions from a user's facial expressions and voice.

[1715] "Dynamic optimization" refers to changing the content of the advertising video in real time according to the user's state.

[1716] "Entertainment content" refers to various media materials that provide entertainment and interest to users.

[1717] "User emotional feedback" refers to the emotional response of users when they watch advertising videos or entertainment content.

[1718] The present invention provides a system for providing highly personalized advertising and entertainment experiences within an autonomous vehicle that are tailored to the emotional state of each passenger. Specific embodiments for implementing the present invention will be described below.

[1719] System configuration

[1720] This system consists of three main components: a server, a terminal, and a user.

[1721] 1. Server

[1722] Generative model: An artificial intelligence model for automatically generating advertising videos and entertainment content.

[1723] Big Data Analysis Unit: A unit that analyzes the performance of advertising scenarios and content.

[1724] Effectiveness measurement unit: A unit for measuring the effectiveness of ads and content in real time.

[1725] Optimization unit: A unit for optimizing the delivery method and content of ads and content.

[1726] Emotion engine: An engine for recognizing emotions from the user's facial expressions and voice.

[1727] 2. Terminal

[1728] In-car camera: A camera for capturing facial images of passengers.

[1729] User Interface: A display for displaying advertisements and content.

[1730] Emotion recognition interface: An interface for sending facial images captured by the in-car camera to a server.

[1731] Dashboard: A tool for monitoring the delivery and effectiveness of ads and content.

[1732] 3. Users

[1733] Passengers: Users who watch advertising and entertainment content in the vehicle.

[1734] Processing flow explanation

[1735] 1. Passenger Emotion Recognition

[1736] The device's onboard camera captures the passenger's face and transmits the image in real time to an emotion-recognition interface.

[1737] The emotion engine on the server analyzes the received images and predicts the passenger's emotional state using image processing libraries such as OpenCV and dlib, as well as deep learning models using Keras.

[1738] 2. Dynamic generation and optimization of advertising and entertainment content

[1739] The server's generative model generates advertising videos and entertainment content based on the emotion data obtained from the emotion engine.

[1740] The big data analysis unit creates content scenarios, and the effectiveness measurement unit monitors performance in real time.

[1741] The optimization unit adjusts content to maximize its effectiveness based on emotional state and collected performance data.

[1742] 3. Delivery of optimal content

[1743] The server sends the generated and optimized advertising and entertainment content to the terminal's user interface for display to the passenger.

[1744] Specific examples

[1745] For example, if a passenger makes a "disinterested" facial expression while in the car, the server's emotion engine will recognize that emotion and use a generative model to regenerate advertising content to attract more interest. An optimization unit then monitors and measures the effectiveness of the content used and makes further adjustments if necessary.

[1746] Prompt Sentence Examples

[1747] Design a system that recognizes emotions in real time from facial expressions that indicate the user's emotions when viewing an advertisement, and retrieves and displays the most appropriate video advertisement from the server based on that user's emotional state.

[1748] This enables the system to provide optimal advertising and entertainment experiences for each passenger.

[1749] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1750] Step 1:

[1751] The device's onboard camera captures the passenger's face and transmits the image in real time to an emotion-recognition interface.

[1752] Input: Passenger face image captured by in-car camera

[1753] Processing: Send the facial image to the emotion recognition interface

[1754] Output: Facial image sent to emotion recognition interface

[1755] Step 2:

[1756] The server's emotion engine analyzes the received images and predicts the passenger's emotional state.

[1757] Input: Facial image sent from emotion recognition interface

[1758] Processing: Detect faces using image processing libraries (OpenCV, dlib) and predict emotions using deep learning models (Keras)

[1759] Output: Predicted emotional state (e.g., uninterested, amused)

[1760] Step 3:

[1761] The server's generative model generates advertising videos and entertainment content based on the emotion data obtained from the emotion engine.

[1762] Input: Emotional state data from the emotion engine

[1763] Processing: Generative models use emotional state data to generate appropriate advertising and content scenarios, and collect and edit materials.

[1764] Output: Generated advertising video or entertainment content

[1765] Step 4:

[1766] The server's big data analysis unit creates content scenarios, and the effectiveness measurement unit monitors performance in real time.

[1767] Input: Generated ad video or content, various performance data (number of impressions, click-through rate, emotional feedback)

[1768] Processing: Analyzing performance data, building scenarios and making adjustments

[1769] Output: Improved scenarios and performance measurement data

[1770] Step 5:

[1771] The server's optimization unit adjusts content to maximize its effectiveness based on emotional state and collected performance data.

[1772] Input: Improved scenario, various performance data

[1773] Processing: Adjusting how and what content is delivered based on the optimization algorithms used by the optimization unit

[1774] Output: Optimized advertising videos or entertainment content

[1775] Step 6:

[1776] The server sends the generated and optimized advertising and entertainment content to the terminal's user interface for display to the passenger.

[1777] Input: Optimized ad video or entertainment content

[1778] Processing: Encode the content and send it to the device's user interface

[1779] Output: Advertising videos or entertainment content displayed on the device's display

[1780] Step 7:

[1781] The server's effectiveness measurement unit monitors the delivery status and effectiveness of advertisements and content.

[1782] Input: Various performance data (number of impressions, click-through rate, number of conversions, emotional feedback)

[1783] Processing: Analyzes the collected performance data and generates reports

[1784] Output: Reports provided to users, insights for future advertising operations

[1785] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 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.

[1786] 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> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) 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.

[1787] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1788] [Fourth embodiment]

[1789] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1790] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1791] 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).

[1792] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. 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 control target 443 are also connected to the bus 52.

[1793] 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.

[1794] 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).

[1795] 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.

[1796] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1797] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, 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.

[1798] 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.

[1799] 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.

[1800] In the robot 414, the processor 46 performs the reception output process. 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.

[1801] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1802] The present invention provides a system for efficiently implementing the entire process from video advertising production to operation, and specific embodiments thereof are described below. This system integrates generative models, big data analysis, feedback processing, and optimization functions to effectively produce and operate video advertising.

[1803] System configuration

[1804] The system consists of the following main components:

[1805] 1. Server

[1806] Generative Model

[1807] Big Data Analysis Unit

[1808] Effectiveness Measurement Unit

[1809] Optimization Unit

[1810] 2. Terminal

[1811] User Interface

[1812] Dashboard

[1813] 3. Users

[1814] Advertisement requirements provided

[1815] Providing Feedback

[1816] Program processing flow

[1817] Video ad requirements definition

[1818] User:

[1819] Users first log in to the portal site and enter the requirements for a new advertising campaign, including target demographics, product information, advertising objectives, etc. After entering the requirements, users can provide detailed information by clicking the "Submit" button to send them to the server.

[1820] server:

[1821] The server analyzes the requirements data received from the user and performs preprocessing before passing it to the Generative Model. This preprocessing converts the requirements data into an appropriate format so that the Generative Model can operate efficiently.

[1822] Scenario generation and video generation

[1823] server:

[1824] The server passes the preprocessed data to a generative model to generate an advertising scenario. Based on the generated scenario, the server collects additional materials such as images, audio, and text, and uses the generative model to automatically generate a video advertisement. The generated video is then encoded for initial review, and a preview link is generated.

[1825] User:

[1826] The user can review the generated video by checking the provided preview link. If feedback is required, the user can enter comments or corrections on the platform and send them back to the server.

[1827] Improving video quality and preparing for operation

[1828] server:

[1829] The server receives feedback from users and analyzes the big data to improve the advertising scenario. It uses patterns obtained from the big data to reflect improvements to the scenario. Based on the improved scenario, it encodes the regenerated video and prepares it for deployment. Once the deployment is ready, the server sends a notification to the user.

[1830] Device:

[1831] The terminal notifies the user that the operation preparation is complete. The user confirms the contents and clicks the "Start operation" button to start the advertisement operation.

[1832] Efficacy monitoring and optimization

[1833] server:

[1834] As soon as the server starts running the ad, it collects performance data such as the number of impressions, click-through rates, and number of conversions in real time. The collected data is analyzed by an effectiveness measurement unit, and if performance is poor, the ad is regenerated and optimized using a generative model. The server also adjusts the timing and targeting of ad displays as appropriate.

[1835] Examples:

[1836] Suppose a company creates an advertising video for a new smartphone product and starts running it. If the click-through rate on the first day is low, the server uses a generative model to improve the video and generates a new scenario based on big data analysis. Then, it changes the timing of the ad display to the optimal time and runs it again.

[1837] Measurement and reporting

[1838] server:

[1839] Once the ad run period is over, the server aggregates all data and measures its effectiveness. A detailed report summarizing the analysis results is generated and provided to the user. The report includes detailed results for each performance indicator and insights for future ad runs.

[1840] User:

[1841] Users can receive the provided report, check the results of their advertising operations, and plan their next advertising strategy based on the report contents.

[1842] The above is a specific description of the embodiment of the present invention. The present system enables users to efficiently create and operate high-quality, effective video advertisements.

[1843] The processing flow will be explained below.

[1844] Step 1:

[1845] User:

[1846] Users log in to the portal site and enter the requirements for a new advertising campaign, providing detailed information such as target demographic, product information, and advertising objectives (e.g., brand awareness, increased conversions). They then confirm the information they entered and click the "Submit" button.

[1847] Step 2:

[1848] server:

[1849] The server analyzes the requirements data received from the user and performs preprocessing before passing it to the generative model. Preprocessing converts the requirements data into an appropriate format so that the generative model can operate efficiently.

[1850] Step 3:

[1851] server:

[1852] The server passes the preprocessed data to a generative model to generate advertising scenarios. The generative model automatically generates effective advertising scenarios based on the input data.

[1853] Step 4:

[1854] server:

[1855] Based on the generated scenario, the server collects materials such as images, audio, and text. Using these materials, the generative model automatically creates a video ad. The generated video is encoded for early review and a preview link is generated.

[1856] Step 5:

[1857] User:

[1858] The user can review the generated video content by checking the provided preview link, and provide feedback and input corrections or additional requirements as needed.

[1859] Step 6:

[1860] server:

[1861] The server receives feedback from users and analyzes the big data to improve the advertising scenario. It utilizes patterns obtained from the big data to improve the scenario.

[1862] Step 7:

[1863] server:

[1864] Based on the improved scenario, the generative model is used to regenerate the video ad. The regenerated video is then encoded and made ready for deployment. Once ready, the server sends a notification to the user.

[1865] Step 8:

[1866] Device:

[1867] The device notifies the user that it is ready for operation. The user confirms the information and clicks the "Start operation" button to start advertising operations.

[1868] Step 9:

[1869] server:

[1870] As soon as the ad starts running, the server collects performance data in real time, such as the number of impressions, click-through rates, and conversions, which are then analyzed by the effectiveness measurement unit.

[1871] Step 10:

[1872] server:

[1873] If performance is poor, the server uses the generative model to regenerate and optimize the ads, adjusting the timing and targeting of the ads accordingly.

[1874] Step 11:

[1875] server:

[1876] At the end of the advertising period, the server aggregates all data and measures the effectiveness. Based on the analysis results, a detailed report is generated and provided to the user.

[1877] Step 12:

[1878] User:

[1879] Users can receive the provided report, check the results of their advertising operations, and plan their next advertising strategy based on the report contents.

[1880] The above are the specific processing steps for carrying out the present invention. This system allows users to efficiently create and operate high-quality, effective video advertisements.

[1881] Example 1

[1882] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1883] Traditional advertising production and management processes are time-consuming, resource-intensive, and inefficient. In particular, they lack flexibility and make it difficult to measure and optimize advertising effectiveness in real time. Another issue is the difficulty of quickly incorporating user feedback and quickly improving advertising content.

[1884] 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.

[1885] In this invention, the server includes means for generating videos for advertisements using a generative AI model, means for analyzing large-scale data to generate advertising scenarios, means for operating the generated videos and measuring their effectiveness, means for optimizing the operation of advertisements based on the measurement results, means for receiving and preprocessing advertising requirements from users, means for receiving user feedback on the generated videos, and means for improving and regenerating the videos based on the feedback. This streamlines the process of advertisement production and operation, enables real-time measurement and optimization of effectiveness, and enables rapid improvement that reflects user feedback.

[1886] A "generative AI model" is an artificial intelligence model that automatically generates advertising videos and scenarios based on specified prompt text.

[1887] "Large-scale data" refers to large amounts of data, such as advertising operations and user feedback, and is a collection of information obtained by analyzing this data.

[1888] "Scenario generation" is the process of designing the structure, content, and storyboard of an advertising video.

[1889] "Effectiveness measurement" is the process of collecting data such as the number of impressions, click rates, and number of conversions obtained during advertising operations and evaluating the performance of the advertisements.

[1890] "Optimization" refers to the process of adjusting advertising operations based on the results of effectiveness measurement to improve performance.

[1891] "User feedback" refers to comments and correction instructions given by the user to the generated advertising video.

[1892] "Regeneration" is the process of recreating advertising videos and scenarios based on feedback and effectiveness measurement results.

[1893] "Preprocessing" is the initial process of analyzing the advertising requirement data received from the user and converting it into a format that is easy for the generative AI model to process.

[1894] The present invention is a system for efficiently implementing the entire process from video advertising production to operation, and specific embodiments thereof are described below. This system integrates generative AI models, large-scale data analysis, feedback processing, and optimization functions to effectively produce and operate video advertising.

[1895] System configuration

[1896] The system consists of the following main components:

[1897] 1. Server

[1898] Generative AI Models

[1899] Large-Scale Data Analysis Unit

[1900] Effectiveness Measurement Unit

[1901] Optimization Unit

[1902] 2. Terminal

[1903] User Interface

[1904] Dashboard

[1905] 3. Users

[1906] Advertisement requirements provided

[1907] Providing Feedback

[1908] Video ad requirements definition

[1909] User:

[1910] Users log in to the portal site and enter the requirements for a new advertising campaign. These requirements include target demographics, product information, advertising objectives, etc., and users can provide detailed information. For example, they can enter a requirement such as "Raising awareness of a new headphone for music lovers aged 25-35." After entering the information, users click the "Submit" button to send it to the server.

[1911] server:

[1912] The server analyzes the requirements data received from the user and performs preprocessing before passing it to the generative AI model. This preprocessing converts the requirements data into an appropriate format, allowing the generative AI model to operate efficiently.

[1913] Scenario generation and video generation

[1914] server:

[1915] The server then passes the preprocessed data to a generative AI model to generate an advertising scenario. For example, it uses a prompt such as, "Generate a scenario that will resonate with music lovers aged 25-35." Based on the generated scenario, the server collects additional materials such as images, audio, and text, and uses the generative model to automatically generate a video ad. The generated video is then encoded for initial review, and a preview link is generated.

[1916] User:

[1917] The user opens the provided preview link to check the content of the generated video, and if necessary, enters comments or corrections on the platform and sends it back to the server.

[1918] Improving video quality and preparing for operation

[1919] server:

[1920] The server receives user feedback, analyzes large-scale data, and refines the advertising scenario. It uses patterns extracted from the large-scale data to improve the scenario. Based on the refined scenario, the video is regenerated and ready for deployment.

[1921] Device:

[1922] The device will notify the user that it is ready for operation. The user can confirm the details and click the "Start operation" button to start advertising operations.

[1923] Efficacy monitoring and optimization

[1924] server:

[1925] As soon as the server starts running the ad, it collects performance data such as the number of impressions, click-through rates, and number of conversions in real time. The collected data is analyzed by an effectiveness measurement unit, and if performance is poor, the ad is regenerated and optimized using a generative model. The timing and targeting of ad displays are also adjusted as appropriate.

[1926] Examples:

[1927] For example, if a company is running an ad for a new smartphone product and the click-through rate on the first day is low, the server will use the generative model to generate new scenarios, improve the video, optimize the timing of the landing page, and adjust the display time of the ad.

[1928] Measurement and reporting

[1929] server:

[1930] Once the ad run period is over, the server aggregates all data and measures its effectiveness. A detailed report summarizing the analysis results is generated and provided to the user. The report includes detailed results for each performance indicator and insights for future ad runs.

[1931] User:

[1932] Users can receive the provided report, check the results of their advertising operations, and plan their next advertising strategy based on the report contents.

[1933] This system streamlines the video ad production and operation process, enabling real-time measurement and optimization of effectiveness. It also allows for quick reflection of user feedback, making it possible to constantly improve ad content to keep it up-to-date and optimal.

[1934] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1935] Step 1: Enter user requirements

[1936] User:

[1937] Input: Users log in to the portal site and enter the requirements for a new advertising campaign, providing detailed information in text format, such as the target demographic (e.g., "music lovers aged 25-35"), product information, and advertising objectives (e.g., "increase awareness of a new product").

[1938] Output: User-entered requirements data.

[1939] Action: The user clicks the "Submit" button to send the entered requirements data to the server.

[1940] Step 2: Preprocessing requirements data

[1941] server:

[1942] Input: Requirement data submitted by the user.

[1943] Output: Preprocessed requirements data (structured data).

[1944] How it works: The server parses the received requirements data and preprocesses it to be passed to the generative AI model. Specifically, it structures the text data and splits it into necessary fields.

[1945] Step 3: Scenario generation

[1946] server:

[1947] Input: Preprocessed requirements data.

[1948] Output: The generated advertising scenario.

[1949] How it works: The server inputs the preprocessed data into a generative AI model and generates advertising scenarios using prompts, such as "Generate a scenario that will resonate with music lovers aged 25-35."

[1950] Step 4: Gathering materials and creating a video

[1951] server:

[1952] Input: The generated ad scenario.

[1953] Output: The generated video ad file.

[1954] How it works: Based on the generated scenario, the server collects the necessary images, audio, and text materials, and automatically generates a video ad using a generative AI model. The generated video file is encoded and a preview link is generated.

[1955] Step 5: Video preview and feedback

[1956] User:

[1957] Input: The video generated using the preview link.

[1958] Output: User feedback (comments and suggestions for corrections).

[1959] How it works: The user opens the provided preview link, checks the generated video, and optionally enters comments and corrections on the platform, which then sends the video back to the server.

[1960] Step 6: Receive and analyze feedback

[1961] server:

[1962] Input: User feedback.

[1963] Output: The scenario with the improvements.

[1964] How it works: The server receives feedback from users and analyzes it using a large-scale data analysis unit. Based on the analysis results, it extracts areas for improvement in the scenario and generates a new scenario using a generative AI model.

[1965] Step 7: Scenario refinement and video regeneration

[1966] server:

[1967] Input: The scenario with the improvements.

[1968] Output: Regenerated video ad files.

[1969] Operation: The server gathers additional material and regenerates the video based on the improved scenario. The regenerated video file is then re-encoded and ready for production.

[1970] Step 8: Notification of operational readiness

[1971] server:

[1972] Input: The regenerated video ad file.

[1973] Output: Operational readiness notification.

[1974] Action: The server notifies the user that it is ready for operation.

[1975] Step 9: Go live

[1976] Device:

[1977] Input: Operational readiness notification.

[1978] Output: Advertising videos are launched.

[1979] Operation: The device displays a notification to the user that preparation for operation is complete. The user clicks the "Start operation" button to start ad operation.

[1980] Step 10: Collecting performance measurement data

[1981] server:

[1982] Input: Ad video performance data (number of impressions, click-through rate, number of conversions, etc.).

[1983] Output: Performance measurement data.

[1984] How it works: The server starts running ads and collects real-time performance data such as impressions, click-through rates, and conversions.

[1985] Step 11: Effectiveness analysis and optimization

[1986] server:

[1987] Input: Performance measurement data.

[1988] Output: Improved and optimized ad video.

[1989] How it works: The server analyzes the collected data using an effectiveness measurement unit, and if performance is poor, it uses a generative model to regenerate ads, adjusting the timing and targeting of ads accordingly.

[1990] Step 12: Measurement and reporting

[1991] server:

[1992] Input: Aggregated performance measurement data.

[1993] Output: Detailed performance measurement report.

[1994] How it works: After the ad campaign period ends, the server aggregates all data, measures the effectiveness, and generates a detailed report summarizing the analysis results, which is then provided to the user.

[1995] Step 13: Review reports and develop strategies

[1996] User:

[1997] Input: Detailed performance measurement report.

[1998] Output: Next advertising strategy.

[1999] How it works: Users can receive the provided report, check the results of their advertising campaigns, and plan their next advertising strategy based on the report.

[2000] (Application example 1)

[2001] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[2002] Traditional video ad production systems have the problem that the processes of ad creation, operation, effectiveness measurement, and optimization are fragmented, making efficient operation difficult. Furthermore, there is a lack of a system that can optimize ads in real time and immediately reflect user feedback. As a result, advertising effectiveness is reduced and user satisfaction is low.

[2003] 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.

[2004] In this invention, the server includes a means for generating videos for advertisements using a generative model, a means for analyzing big data to generate advertising scenarios, a means for operating the generated videos and measuring their effectiveness, a means for optimizing the operation of advertisements based on the measurement results, a means for displaying the optimized video advertisements in real time using smart glasses, and a means for regenerating and optimizing advertisements based on collected data. This enables the entire process from the production of video advertisements to their operation and feedback to be integrated and operated efficiently and effectively.

[2005] A "generative model" is an algorithm that automatically generates scenarios and videos for advertising.

[2006] "Big data" refers to large amounts of data, a group of data that is used to generate and optimize advertising scenarios.

[2007] An "advertising scenario" is a plan that defines the content and progression of a video advertisement.

[2008] "Operation" refers to the process of actually delivering the generated video advertisements and displaying them to users.

[2009] "Effectiveness measurement" is the process of evaluating the results of advertising and collecting and analyzing performance data.

[2010] "Optimization" refers to making adjustments based on collected data to improve the performance of your ads.

[2011] "Smart glasses" are a type of wearable device that visually displays information to the user while they are wearing them.

[2012] "Real-time" means that processing and reaction are immediate and based on the current time zone.

[2013] "Feedback" means reviews and correction instructions provided by users.

[2014] An "interface" is a point of contact or means by which a user interacts with a system.

[2015] "Data" refers to a set of information or numbers that a system processes.

[2016] "Regeneration" is the process of creating a new version of an ad based on previously generated content.

[2017] The present invention provides a system for efficiently implementing the entire process from video advertising production to operation, and is characterized by its ability to display optimized video advertising in real time using smart glasses. Specific embodiments are described below. The system integrates generative models, big data analysis, feedback processing, and optimization functions to effectively produce and operate video advertising.

[2018] System configuration

[2019] The system consists of the following main components:

[2020] 1. Server:

[2021] Generative model: An algorithm that automatically generates scenarios and videos for advertising.

[2022] Big Data Analysis Unit: A data set used to analyze large amounts of data and measure and optimize advertising effectiveness.

[2023] Effectiveness Measurement Unit: Measures the effectiveness of advertising and analyzes the collected data.

[2024] Optimization unit: Optimizes your ads based on the data collected to improve their performance.

[2025] 2. Terminal:

[2026] Smart glasses: Glasses that visually display information while worn by the user (e.g., Google Glass, Vuzix Blade).

[2027] User Interface: An interface for entering ad requirements, previewing, and providing feedback.

[2028] Dashboard: A management screen for displaying advertising results and effectiveness measurement results.

[2029] 3. User:

[2030] Provide Ad Requirements: Enter your requirements for a new ad campaign.

[2031] Providing feedback: Review the content of the generated video and provide feedback.

[2032] System Operation

[2033] 1. Video ad requirements definition:

[2034] Users input their advertising requirements through the smart glasses interface, including target demographic, product information, and advertising objectives, and then the requirements are sent to the server.

[2035] 2. Scenario and video generation:

[2036] The server analyzes the received requirements data and passes it to the generative model. The generative model generates a scenario, collects the necessary materials (images, audio, text), and automatically generates a video advertisement. The generated video is encoded for initial review and streamed to the smart glasses.

[2037] 3. Viewing videos and providing feedback:

[2038] Users can view the generated video through smart glasses and provide feedback using voice commands or touch gestures, which is then sent to the server, which regenerates and optimizes the video as needed.

[2039] 4. Effect monitoring and optimization:

[2040] The server monitors the effectiveness of the ad (number of views, viewing time, user response) in real time. If the effectiveness is low, it will be regenerated and optimized to improve the ad performance.

[2041] 5. Report Generation:

[2042] The server aggregates all data, measures effectiveness, and generates detailed reports to provide users with insights to help them plan their next advertising strategy.

[2043] Specific examples

[2044] For example, consider a company that is creating and running an advertising video for a new smartphone product. The target demographic is technology-conscious people in their 20s and 30s, and the goal of the ad is to stimulate purchasing desire and increase brand awareness. Based on this requirement, the following prompt sentences are input into the generative AI model:

[2045] Prompt statement:

[2046] The target demographic for a new smartphone is tech-savvy people in their 20s and 30s. The advertising objective is to stimulate purchase intent and increase brand awareness. Based on this requirement, generate the following scenario and creative. The ad should include high-performance camera features, a long battery life, and a seamless user experience.

[2047] As described above, this system allows users to view high-quality, effective video advertisements in real time using smart glasses, and allows for immediate feedback.

[2048] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[2049] Step 1: User Input

[2050] Through the smart glasses' user interface, users input requirements for a new advertising campaign, including target demographic, product information, and advertising objectives. The input is made on the interface and transmitted to the server. The input data also specifies specific elements and conditions that should be included in the advertisement.

[2051] Input: target demographic, product information, advertising objectives

[2052] Output: Requirements data sent to the server

[2053] Step 2: Requirement Data Analysis

[2054] The server analyzes the requirements data received from the user, and the analysis results are converted into a format that can be passed to the generative AI model. This preprocessing allows the generative model to operate efficiently.

[2055] Input: Requirement data

[2056] Output: Data converted into a format to be passed to the generative model

[2057] Step 3: Scenario generation

[2058] The server passes the preprocessed data to the generative model to generate an advertising scenario. The generative model constructs a scenario based on the prompt sentence and determines the story of the advertising video.

[2059] Input: Preprocessed data, generative AI model

[2060] Output: Generated scenario

[2061] Step 4: Gathering materials and creating a video

[2062] The server collects materials such as images, audio, and text based on the generated scenario, then uses the generative model to automatically generate advertising videos, which are then encoded and prepared for streaming to the smart glasses.

[2063] Input: Generated scenario, material data

[2064] Output: Generated video ad

[2065] Step 5: Advertising and Feedback

[2066] The user watches the generated video advertisement through the smart glasses. If feedback is required, the user can provide it using voice commands or touch gestures. The feedback is then sent to the server.

[2067] Input: Generated video ad

[2068] Output: User feedback

[2069] Step 6: Feedback analysis and video regeneration

[2070] The server analyzes the feedback received from the user, refines the content of the video advertisement based on the feedback, and regenerates the video, which is then re-encoded and streamed to the smart glasses.

[2071] Input: User feedback

[2072] Output: Regenerated video ad

[2073] Step 7: Measure your results

[2074] The server starts running the ad and simultaneously collects viewing data (number of views, viewing time, user response) in real time. The collected data is analyzed by the effectiveness measurement unit and the ad is optimized as necessary.

[2075] Input: Viewing data

[2076] Output: Effectiveness measurement report, optimization instructions

[2077] Step 8: Generate reports

[2078] After the ad run period ends, the server aggregates all data and generates a detailed performance measurement report, which is provided to the user and contains insights for planning their next advertising strategy.

[2079] Input: Aggregated operational data

[2080] Output: Performance measurement report

[2081] Through the above process, the present invention makes it possible to efficiently and effectively produce, operate, collect feedback, and optimize advertising videos.

[2082] 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.

[2083] The present invention is a system for efficiently and effectively implementing the entire process from the production to the operation of video advertisements, and in particular, by combining an emotion engine that recognizes user emotions, it achieves a high level of personalization. Below, we will explain the form for implementing the present invention and the specific processing content thereof.

[2084] System configuration

[2085] The system consists of the following main components:

[2086] 1. Server

[2087] Generative Model

[2088] Big Data Analysis Unit

[2089] Effectiveness Measurement Unit

[2090] Optimization Unit

[2091] Emotion Engine

[2092] 2. Terminal

[2093] User Interface

[2094] Emotion Recognition Interface

[2095] Dashboard

[2096] 3. Users

[2097] Advertisement requirements provided

[2098] Providing Feedback

[2099] Program processing flow

[2100] Video ad requirements definition

[2101] User:

[2102] First, users log in to the portal site and enter the requirements for a new advertising campaign, providing detailed information such as target demographic, product information, and advertising objectives (e.g., brand awareness, increased conversions). They then confirm the information they entered and click the "Submit" button.

[2103] server:

[2104] The server analyzes the requirements data received from the user and performs preprocessing before passing it to the Generative Model. This preprocessing converts the requirements data into an appropriate format so that the Generative Model can operate efficiently.

[2105] Scenario generation and video generation

[2106] server:

[2107] The server passes the preprocessed data to a generative model to generate advertising scenarios. The generative model automatically generates effective advertising scenarios based on the input data.

[2108] server:

[2109] Based on the generated scenario, the server collects materials such as images, audio, and text. Using these materials, the generative model automatically creates a video ad. The generated video is encoded for early review and a preview link is generated.

[2110] User:

[2111] The user can review the generated video content by checking the provided preview link, and provide feedback and input corrections or additional requirements as needed.

[2112] Improving video quality and preparing for operation

[2113] server:

[2114] The server receives feedback from users and analyzes the big data to improve the advertising scenario. It utilizes patterns obtained from the big data to improve the scenario.

[2115] server:

[2116] Based on the improved scenario, the generative model is used to regenerate the video ad. The regenerated video is then encoded and made ready for deployment. Once ready, the server sends a notification to the user.

[2117] Device:

[2118] The device notifies the user that it is ready for operation. The user confirms the information and clicks the "Start operation" button to start advertising operations.

[2119] Effect monitoring, optimization, and emotion recognition

[2120] server:

[2121] As soon as the ad starts running, the server collects performance data in real time, such as the number of impressions, click-through rates, and conversions, which are then analyzed by the effectiveness measurement unit.

[2122] server:

[2123] The emotion engine recognizes emotions in real time from facial expressions and voice when a user views an ad, and dynamically adjusts the content of the ad based on that data. For example, if a user has a happy expression while viewing an ad, the engine can add more engaging elements to match that positive emotion.

[2124] server:

[2125] If performance is poor, the server uses the generative model to regenerate and optimize the ads, adjusting the timing and targeting of the ads accordingly.

[2126] Examples:

[2127] Suppose a company creates and starts using an advertising video for a new smartphone product. If many users who watch the ad feel uninterested, the server uses an emotion engine to analyze the emotion data, improves the advertising scenario, and regenerates it to make it more interesting.

[2128] Measurement and reporting

[2129] server:

[2130] Once the ad campaign period ends, the server aggregates all data and measures its effectiveness. Based on the analysis results, a detailed report is generated and provided to the user. The report includes detailed results for each performance indicator and insights for future ad campaigns.

[2131] User:

[2132] Users can receive the provided report, check the results of their advertising operations, and plan their next advertising strategy based on the report contents.

[2133] The above is a specific description of an embodiment of the present invention. This system allows users to efficiently create and manage high-quality, effective video advertisements. Furthermore, by utilizing the emotion engine, it is possible to achieve a high level of personalization of advertisements and provide each user with an optimal advertising experience.

[2134] The processing flow will be explained below.

[2135] Step 1:

[2136] User:

[2137] Users log in to the portal site and enter the requirements for a new advertising campaign, including detailed information such as target demographic, product information, and advertising objectives (e.g., brand awareness, increased conversions). They confirm the information they entered and click the "Submit" button.

[2138] Step 2:

[2139] server:

[2140] The server analyzes the requirements data received from the user and performs preprocessing before passing it to the Generative Model. This preprocessing converts the provided data into an appropriate format so that the Generative Model can operate efficiently.

[2141] Step 3:

[2142] server:

[2143] The server passes the preprocessed data to a generative model to generate advertising scenarios. The generative model automatically generates effective advertising scenarios based on the input data.

[2144] Step 4:

[2145] server:

[2146] Based on the generated scenario, the server collects materials such as images, audio, and text. Using these materials, the generative model automatically creates a video ad. The generated video is encoded for early review and a preview link is generated.

[2147] Step 5:

[2148] User:

[2149] The user can review the generated video content by checking the provided preview link, and provide feedback and input corrections or additional requirements as needed.

[2150] Step 6:

[2151] server:

[2152] The server receives feedback from users and analyzes the big data to improve the advertising scenario. It utilizes patterns obtained from the big data to improve the scenario.

[2153] Step 7:

[2154] server:

[2155] Based on the improved scenario, the generative model is used to regenerate the video ad. The regenerated video is then encoded and made ready for deployment. Once ready, the server sends a notification to the user.

[2156] Step 8:

[2157] Device:

[2158] The device notifies the user that it is ready for operation. The user confirms the information and clicks the "Start operation" button to start advertising operations.

[2159] Step 9:

[2160] server:

[2161] As soon as the ad starts running, performance data such as the number of impressions, click-through rate, and number of conversions is collected in real time, and the collected data is analyzed by the effectiveness measurement unit.

[2162] Step 10:

[2163] server:

[2164] The emotion engine recognizes emotions in real time from facial expressions and voice when a user views an advertisement, and dynamically adjusts the content of the advertisement based on that data. For example, if a user is not interested in an advertisement, the scenario will be improved by reflecting that emotional data.

[2165] Step 11:

[2166] server:

[2167] If performance is poor, the server uses the generative model to regenerate and optimize the ads, adjusting the timing and targeting of the ads accordingly.

[2168] Step 12:

[2169] server:

[2170] Once the advertising period ends, the server aggregates all data and measures the effectiveness. Based on the analysis results, a detailed report is generated and provided to the user.

[2171] Step 13:

[2172] User:

[2173] Users can receive the provided report, check the results of their advertising operations, and plan their next advertising strategy based on the report contents.

[2174] The above are the specific processing steps for implementing the present invention in combination with an emotion engine. This system allows users to efficiently create and manage high-quality, effective video advertisements, and by utilizing the emotion engine, highly personalized advertisements can be achieved.

[2175] Example 2

[2176] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[2177] Conventional advertising video production and operation systems lacked personalization, making it difficult to optimize by incorporating each user's emotions and real-time feedback. Furthermore, the process of measuring effectiveness and analyzing feedback was complicated, making it difficult to respond quickly to improve advertising effectiveness. This resulted in problems such as a lack of improvement in user experience and reduced advertising effectiveness.

[2178] 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.

[2179] In this invention, the server includes means for generating videos for advertisements using a generative model, means for analyzing big data to generate advertising scenarios, means for operating the generated videos and measuring their effectiveness, means for recognizing emotions from a user's facial expressions and voice and dynamically adjusting advertising content based on the emotion data, means for analyzing the content of feedback and regenerating videos based on the analysis results, and means for aggregating the results of measuring the effectiveness of advertising operation, generating detailed reports, and providing them to users. This makes it possible to highly personalize advertisements and optimize them in real time.

[2180] A "generative model" is a machine learning model that uses algorithms to automatically generate content.

[2181] "Big data" is a collection of large amounts of diverse data, the analysis of which makes it possible to discover new insights and patterns.

[2182] An "advertising scenario" is a detailed design document for planning the composition and development of an advertising video.

[2183] "Emotion recognition" is a technology that analyzes non-verbal information such as a user's facial expressions and voice to recognize their emotional state.

[2184] "Dynamic adjustment" is the process of instantly changing advertising content based on real-time data.

[2185] "Feedback" refers to the evaluations and opinions provided by users, and the system is improved and adjusted based on them.

[2186] "Effectiveness measurement" is the process of collecting and analyzing data to verify whether advertising is having the intended effect.

[2187] A "report" is a document that summarizes the results of effectiveness measurement and analysis, and is used as a guideline for the next strategy planning and improvement.

[2188] "Advertising Requirements" are detailed instructions that set out the conditions and goals necessary to run an advertising campaign.

[2189] "Operational optimization" refers to adjustments and improvement activities carried out to maximize advertising operational performance.

[2190] "Performance Data" means data used to evaluate the results of advertising operations, such as the number of times an advertisement is displayed, click-through rate, and number of conversions.

[2191] The present invention provides a video advertising production and management system that combines an emotion engine that recognizes user emotions to achieve advanced personalization. The following describes an embodiment of the system.

[2192] System configuration

[2193] This system consists of a server, a terminal, and a user. The specific hardware and software used are as follows:

[2194] 1. Server:

[2195] Generative model: Uses OpenAI's GPT model.

[2196] Big Data Analytics Unit: Uses AWS SageMaker and Google Cloud BigQuery.

[2197] Performance Measurement Unit: A Python script for collecting and analyzing performance data.

[2198] Optimization unit: A script that adjusts ad operations.

[2199] Emotion Engine: Emotion recognition software for recognizing the user's facial expressions and voice.

[2200] 2. Terminal:

[2201] User interface: Built with HTML / CSS and JavaScript, accepts input from the user.

[2202] Emotion recognition interface: Recognizes the user's emotions using a camera and microphone.

[2203] Dashboard: An interface that displays operational data and effectiveness measurement results.

[2204] 3. User:

[2205] Provide your advertising requirements: Enter your advertising campaign details.

[2206] Providing feedback: Provide feedback such as your ratings and opinions on the provided advertising video.

[2207] System operation explanation

[2208] Video ad requirements definition

[2209] User: Logs in to the portal site and enters details such as the target demographic, product information, and advertising objectives of the advertising campaign. For example, the user might enter, "The target demographic is young people in their 20s, and the purpose is to raise awareness of a new smartphone product."

[2210] Scenario generation and video generation

[2211] Server: Analyzes the requirements data sent by the user and passes it to the generative model. A Python script is used for this preprocessing. The generative model (e.g., GPT-3) generates effective advertising scenarios.

[2212] Server: Based on the generated scenario, images, audio, and text materials are collected and a video ad is automatically generated using software such as Adobe Premiere Pro or Final Cut Pro. The generated video is then encoded and a preview link is generated.

[2213] User: Check the generated video via the preview link and provide feedback if necessary. For example, "I'd like the user's voice to be more prominent in the final scene."

[2214] Improving video quality and preparing for operation

[2215] Server: Receives user feedback, analyzes the data using the big data analysis unit, refines the scenario based on the analysis results, and regenerates the video ad using the generative model.

[2216] Server: Encodes the regenerated video and notifies the user that it is ready for production.

[2217] Terminal: The device will notify the user that it is ready for operation, and the user can click the "Start operation" button to start advertising operations.

[2218] Effect monitoring, optimization, and emotion recognition

[2219] Server: Once the ad is launched, real-time performance data such as impressions, click-through rates, and conversions are collected.

[2220] Server: The emotion engine recognizes emotions from facial expressions and voice when users view advertisements, and dynamically adjusts the content of advertisements based on that data.

[2221] Server: Uses generative models to regenerate and optimize ads when performance is poor, and also dynamically adjusts ad timing and targeting.

[2222] Measurement and reporting

[2223] Server: After the advertising period ends, all data is aggregated and effectiveness is measured. A detailed report is generated based on the analysis results and provided to the user.

[2224] Users: They can review the provided reports and plan their next advertising strategy. For example, they can decide to target women in their early 20s in their next campaign.

[2225] Example prompts to input to the generative AI model

[2226] "The target demographic is young people in their 20s, and the goal is to raise awareness of a new smartphone product. Please generate an effective advertising scenario for this target demographic."

[2227] This system allows users to efficiently create and operate high-quality, effective video ads. In particular, by utilizing the emotion engine, it is possible to provide the optimal advertising experience for each user.

[2228] The flow of the identification process in the second embodiment will be described with reference to FIG.

[2229] Step 1:

[2230] User: Enter your advertising requirements

[2231] Users log in to the portal site and enter the target demographic, product information, and purpose of the ad campaign. For example, they enter specific information such as "The target demographic is young people in their 20s, and the purpose is to raise awareness of a new smartphone product," and click the submit button. The entered data is sent to the server.

[2232] Input: Advertising requirement data

[2233] Output: Ad requirements data sent to the server

[2234] Step 2:

[2235] Server: Preprocessing requirements data

[2236] The server parses the ad requirement data received from users and converts it into a format suitable for the generative model. This preprocessing is done using Python scripts, such as cleaning and formatting the text data.

[2237] Input: Ad requirements data sent by the user

[2238] Output: Data converted into a format suitable for feeding into a generative model

[2239] Step 3:

[2240] Server: Automatic scenario generation

[2241] The server passes the preprocessed data to a generative model, which automatically generates advertising scenarios. Here, a generative AI model (e.g., GPT-3) is used. The generated scenarios are stored on the server.

[2242] Input: Data converted to a format suitable for feeding into a generative model

[2243] Output: Generated advertising scenario

[2244] Step 4:

[2245] Server: Collects advertising materials and generates videos

[2246] The server collects the necessary image, audio, and text materials based on the generated scenario. Using the collected materials, the video ad is automatically generated using software such as Adobe Premiere Pro or Final Cut Pro. The generated video is then encoded and a preview link is generated.

[2247] Input: Generated advertising scenario

[2248] Output: Generated video ad with preview link

[2249] Step 5:

[2250] Users: Video reviews and feedback

[2251] Users can watch the generated video via the provided preview link, review the video, and provide feedback on corrections or additional requirements as needed. For example, they could provide feedback such as "I would like the product name to be more prominent in the final scene."

[2252] Input: Preview link and user feedback

[2253] Output: Feedback data

[2254] Step 6:

[2255] Server: Feedback analysis and scenario refinement

[2256] The server analyzes the feedback received from the user and uses a big data analysis unit to analyze the data. Based on the analysis results, the advertising scenario is improved, for example, by modifying certain parts of the scenario based on the feedback.

[2257] Input: Feedback data

[2258] Output: Improved advertising scenario

[2259] Step 7:

[2260] Server: Regenerate videos and prepare for operation

[2261] The server regenerates the video ad using the generative model based on the improved scenario, re-encodes the video, and notifies the server that it is ready for production.

[2262] Input: Improved advertising scenario

[2263] Output: Regenerated video ads and notifications

[2264] Step 8:

[2265] Terminal: Notification of start of operation

[2266] The device will notify the user that it is ready for operation. The user will receive the notification, check the contents, and click the "Start operation" button to start advertising operations.

[2267] Input: Notification of operational readiness

[2268] Output: Ads go live

[2269] Step 9:

[2270] Server: Performance data collection and analysis

[2271] As soon as the ad starts running, the server collects performance data in real time, such as the number of impressions, click-through rates, and conversions, which are then analyzed by the effectiveness measurement unit.

[2272] Input: Ad performance data

[2273] Output: Parsed performance report

[2274] Step 10:

[2275] Server: Emotion data analysis and dynamic ad adjustment

[2276] The server's emotion engine recognizes emotions in real time from facial expressions and voice as the user views the advertisement. The content of the advertisement is dynamically adjusted based on this emotional data. For example, if the user shows interest, the content will be changed to emphasize that element.

[2277] Input: Emotion data

[2278] Output: Dynamically adjusted ad content

[2279] Step 11:

[2280] Server: Ad regeneration and optimization

[2281] The server uses the generative model to regenerate and optimize ads when performance is poor, and dynamically adjusts ad timing and targeting settings.

[2282] Input: Performance and emotion data

[2283] Output: Regenerated optimized video ad

[2284] Step 12:

[2285] Server: Performance measurement and report generation

[2286] Once the advertising period ends, the server aggregates all data and measures the effectiveness. Based on the results of the analysis, a detailed report is generated and provided to the user.

[2287] Input: All performance data

[2288] Output: Performance measurement results and reports

[2289] Step 13:

[2290] User: Check the report and consider the next strategy

[2291] Users can review the provided reports and evaluate the results of their advertising operations. Based on this evaluation, they can consider their next advertising strategy and come up with further improvements.

[2292] Input: Performance measurement report

[2293] Output: Next advertising strategy plan

[2294] (Application example 2)

[2295] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[2296] In the production and provision of modern advertising videos and entertainment content, it is difficult to consider the different emotions and interests of each user, and generic advertising may not be effective for all users. Furthermore, in order to maximize passenger satisfaction, entertainment systems inside autonomous vehicles are required to provide content tailored to passenger emotions in real time. However, current technology does not provide a method to achieve this.

[2297] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for generating videos for advertising using a generative model, means for generating advertising scenarios by analyzing big data, means for operating the generated videos and measuring their effectiveness, means for analyzing user emotions using an emotion recognition engine and dynamically optimizing the video content, and means for providing advertising and entertainment content based on emotion recognition. This makes it possible to provide highly personalized advertising and entertainment experiences according to the emotional state of each passenger.

[2298] A "generative model" is an algorithm that uses artificial intelligence technology to automatically create advertising videos.

[2299] "Big data" refers to a large amount of data collected from a wide variety of sources.

[2300] An "advertising scenario" is a plan or script that determines the content and structure of an advertising video.

[2301] "Operation" refers to the process of actually distributing the generated advertising video and observing its effectiveness.

[2302] "Effectiveness measurement" is an analytical method for evaluating the performance of advertising videos.

[2303] "Optimization" is the process of improving advertising videos and their delivery methods to maximize the desired effect.

[2304] An "emotion recognition engine" is a technology for analyzing emotions from a user's facial expressions and voice.

[2305] "Dynamic optimization" refers to changing the content of the advertising video in real time according to the user's state.

[2306] "Entertainment content" refers to various media materials that provide entertainment and interest to users.

[2307] "User emotional feedback" refers to the emotional response of users when they watch advertising videos or entertainment content.

[2308] The present invention provides a system for providing highly personalized advertising and entertainment experiences within an autonomous vehicle that are tailored to the emotional state of each passenger. Specific embodiments for implementing the present invention will be described below.

[2309] System configuration

[2310] This system consists of three main components: a server, a terminal, and a user.

[2311] 1. Server

[2312] Generative model: An artificial intelligence model for automatically generating advertising videos and entertainment content.

[2313] Big Data Analysis Unit: A unit that analyzes the performance of advertising scenarios and content.

[2314] Effectiveness measurement unit: A unit for measuring the effectiveness of ads and content in real time.

[2315] Optimization unit: A unit for optimizing the delivery method and content of ads and content.

[2316] Emotion engine: An engine for recognizing emotions from the user's facial expressions and voice.

[2317] 2. Terminal

[2318] In-car camera: A camera for capturing facial images of passengers.

[2319] User Interface: A display for displaying advertisements and content.

[2320] Emotion recognition interface: An interface for sending facial images captured by the in-car camera to a server.

[2321] Dashboard: A tool for monitoring the delivery and effectiveness of ads and content.

[2322] 3. Users

[2323] Passengers: Users who watch advertising and entertainment content in the vehicle.

[2324] Processing flow explanation

[2325] 1. Passenger Emotion Recognition

[2326] The device's onboard camera captures the passenger's face and transmits the image in real time to an emotion-recognition interface.

[2327] The emotion engine on the server analyzes the received images and predicts the passenger's emotional state using image processing libraries such as OpenCV and dlib, as well as deep learning models using Keras.

[2328] 2. Dynamic generation and optimization of advertising and entertainment content

[2329] The server's generative model generates advertising videos and entertainment content based on the emotion data obtained from the emotion engine.

[2330] The big data analysis unit creates content scenarios, and the effectiveness measurement unit monitors performance in real time.

[2331] The optimization unit adjusts content to maximize its effectiveness based on emotional state and collected performance data.

[2332] 3. Delivery of optimal content

[2333] The server sends the generated and optimized advertising and entertainment content to the terminal's user interface for display to the passenger.

[2334] Specific examples

[2335] For example, if a passenger makes a "disinterested" facial expression while in the car, the server's emotion engine will recognize that emotion and use a generative model to regenerate advertising content to attract more interest. An optimization unit then monitors and measures the effectiveness of the content used and makes further adjustments if necessary.

[2336] Prompt Sentence Examples

[2337] Design a system that recognizes emotions in real time from facial expressions that indicate the user's emotions when viewing an advertisement, and retrieves and displays the most appropriate video advertisement from the server based on that user's emotional state.

[2338] This enables the system to provide optimal advertising and entertainment experiences for each passenger.

[2339] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[2340] Step 1:

[2341] The device's onboard camera captures the passenger's face and transmits the image in real time to an emotion-recognition interface.

[2342] Input: Passenger face image captured by in-car camera

[2343] Processing: Send the facial image to the emotion recognition interface

[2344] Output: Facial image sent to emotion recognition interface

[2345] Step 2:

[2346] The server's emotion engine analyzes the received images and predicts the passenger's emotional state.

[2347] Input: Facial image sent from emotion recognition interface

[2348] Processing: Detect faces using image processing libraries (OpenCV, dlib) and predict emotions using deep learning models (Keras)

[2349] Output: Predicted emotional state (e.g., uninterested, amused)

[2350] Step 3:

[2351] The server's generative model generates advertising videos and entertainment content based on the emotion data obtained from the emotion engine.

[2352] Input: Emotional state data from the emotion engine

[2353] Processing: Generative models use emotional state data to generate appropriate advertising and content scenarios, and collect and edit materials.

[2354] Output: Generated advertising video or entertainment content

[2355] Step 4:

[2356] The server's big data analysis unit creates content scenarios, and the effectiveness measurement unit monitors performance in real time.

[2357] Input: Generated ad video or content, various performance data (number of impressions, click-through rate, emotional feedback)

[2358] Processing: Analyzing performance data, building scenarios and making adjustments

[2359] Output: Improved scenarios and performance measurement data

[2360] Step 5:

[2361] The server's optimization unit adjusts content to maximize its effectiveness based on emotional state and collected performance data.

[2362] Input: Improved scenario, various performance data

[2363] Processing: Adjusting how and what content is delivered based on the optimization algorithms used by the optimization unit

[2364] Output: Optimized advertising videos or entertainment content

[2365] Step 6:

[2366] The server sends the generated and optimized advertising and entertainment content to the terminal's user interface for display to the passenger.

[2367] Input: Optimized ad video or entertainment content

[2368] Processing: Encode the content and send it to the device's user interface

[2369] Output: Advertising videos or entertainment content displayed on the device's display

[2370] Step 7:

[2371] The server's effectiveness measurement unit monitors the delivery status and effectiveness of advertisements and content.

[2372] Input: Various performance data (number of impressions, click-through rate, number of conversions, emotional feedback)

[2373] Processing: Analyzes the collected performance data and generates reports

[2374] Output: Reports provided to users, insights for future advertising operations

[2375] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice 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 voice data.

[2376] 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> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) 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.

[2377] 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 robot 414.

[2378] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[2379] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[2380] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[2381] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[2382] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[2383] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[2384] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[2385] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[2386] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[2387] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[2388] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[2389] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[2390] The hardware resource for executing a specific process can be any of the following pr...

Claims

1. A means for generating a video for an advertisement using the generative model; A means for analyzing big data and generating advertising scenarios; A means to operate the generated videos and measure their effectiveness, A means for optimizing advertising operations based on the measurement results; A system including:

2. a means for modifying a video for advertising using the generative model; A means for improving advertising scenarios based on big data analysis; The system of claim 1 further comprising:

3. means for receiving advertising requirements from a user; means for providing a video generated based on the received requirements to a user and receiving feedback; means for improving and regenerating the video based on said feedback; The system of claim 1 further comprising:

Images