System

Abstract

A system is provided.SOLUTION: A system comprising: means for receiving detailed information of a commodity from a user; means for inputting the received detailed information to a generated AI model and automatically generating an advertisement image; means for providing the generated advertisement image to the user; and means for feeding back advertisement effect information collected after advertisement placement to the generated AI model and causing the generated machine learning model to learn the advertisement effect information.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] In the advertising market, technical expertise, high costs, and hassle are barriers to entry. This often discourages companies and individuals from placing ads. This limits the expansion of the advertising market and prevents the potential effectiveness of advertising from being realized. This invention aims to lower the barriers to entry that would expand the advertising market by reducing the effort and cost of creating ads and making it easier to create and place ads. [Means for solving the problem]

[0005] To solve this problem, the present invention provides the following means. First, a means is provided for accepting detailed product information from a user, and these details are input into a generative AI model to automatically generate an advertising image. Next, the generated advertising image is provided to the user, who uses it to post an advertisement. Furthermore, a means is provided for feeding back advertising effectiveness data collected after the advertisement is posted to the generative AI model, allowing it to continuously learn. Specifically, the generative AI model generates advertising images using a neural network, and an API is provided for automatically collecting advertising effectiveness data, thereby improving the accuracy of advertisement creation. This results in a system that streamlines the process of creating and posting advertisements and provides high convenience to users.

[0006] "Product Details" refers to information about a particular product, such as its name, price, description, and associated images.

[0007] A "generative AI model" refers to an artificial intelligence model that automatically generates advertising images from given data.

[0008] "Advertising Image" refers to a visual representation used to promote a product or service.

[0009] "User" refers to an individual or company that uses this system to generate advertising images and place advertisements.

[0010] "Advertising effectiveness data" refers to data used to evaluate the performance of an advertisement, and specifically refers to information such as the number of clicks, number of impressions, and conversion rate.

[0011] A "neural network" is a type of algorithm used in artificial intelligence, and refers to a technology that performs pattern recognition and learning by imitating the behavior of nerve cells in the human brain.

[0012] "API" refers to a technology that provides an interface for different software systems to communicate with each other. [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] A specific embodiment for carrying out the present invention will be described below. A user, a terminal, and a server work together based on the following steps.

[0035] User Interface (UI) Design

[0036] User: First, the user accesses the system on a terminal. The UI on the screen displays fields for entering the product name, price, description, and image. The user enters the product details in these fields.

[0037] Sending product information

[0038] Terminal: Collects product information entered by the user and sends it to the server in JSON format, combining the entered product name, price, description, and uploaded image into a single data object.

[0039] Generating advertising images

[0040] Server: Once product information is received, the server passes it to the generative AI model, which uses a neural network to generate advertising images based on the input information. This generation process is automated, and the generated advertising images are stored on the server.

[0041] Sending and displaying advertising images

[0042] Server: Obtain the storage path or URL of the generated ad image and return it to the user's device.

[0043] Terminal: The URL of the ad image received from the server is acquired and displayed on the user's screen. The user can then check the generated ad image and use it to place an ad.

[0044] Advertising and data collection

[0045] User: Uses the generated ad image to post an ad on an advertising platform such as Yahoo! or LINE. At this time, the user collects performance data (number of clicks, number of impressions, conversion rate, etc.) related to the posted ad.

[0046] Data feedback and AI model retraining

[0047] Server: After the ad is published, the server receives the ad effectiveness data provided by the user. The server feeds this data back into the generative AI model and retrains the model. This feedback loop improves the accuracy of the AI ​​model's ad image generation.

[0048] As a concrete example, consider a user creating an advertisement for a new smartphone. First, the user enters the product name "Latest Model Smartphone," the price "50,000 yen," the product description "High-capacity battery, fast charging compatible," and a product image into the system. The device sends this information to the server, and the server-side generative AI model automatically generates the advertisement image. The generated advertisement image is sent back to the user's device, and the user uses it in their own advertising campaign. At a later date, the user sends the advertisement's performance data to the server, and the AI ​​model retrains to further improve accuracy. This series of processes allows users to easily generate and publish high-quality advertisements, even without specialized knowledge.

[0049] The processing flow will be explained below.

[0050] Step 1: The user accesses the system on a terminal and enters product details such as product name, price, description, and image.

[0051] Step 2: The user's device compiles the entered product information into a JavaScript object, converts it to JSON format, and sends an HTTP POST request to the server.

[0052] Step 3: The server receives the HTTP request, parses the product information from the request body, converts the product information into an appropriate format, and passes it to the generative AI model.

[0053] Step 4: The server's AI model generates advertising images based on the product information. The generated advertising images are stored on the server.

[0054] Step 5: The server obtains the storage path or URL of the generated ad image and returns it to the user's device in JSON format.

[0055] Step 6: The user's device retrieves the URL of the ad image received from the server, sets it as the src attribute of the image element, and displays it on the user's screen. The user then checks the generated ad image.

[0056] Step 7: The user uploads the confirmed ad image to an advertising platform such as Yahoo or LINE, sets up an advertising campaign, and publishes it.

[0057] Step 8: During the advertising campaign, the user collects advertising effectiveness data such as clicks, impressions, and conversion rates from the advertising platform.

[0058] Step 9: The user provides the collected advertising effectiveness data to the server using the API provided by the server.

[0059] Step 10: The server stores the received advertising effectiveness data in a database and feeds it back to the generative AI model.

[0060] Step 11: The server's generated AI model is retrained based on the feedback data to improve the accuracy of advertising image generation.

[0061] These steps realize a system in which users, terminals, and servers work together to automatically generate advertising images and utilize advertising effectiveness data to continuously improve accuracy.

[0062] Example 1

[0063] 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."

[0064] Conventional advertising image generation systems require specialized knowledge and lack a means to automate the creation of advertising images and their effectiveness analysis. Furthermore, they lack a feedback mechanism to improve the accuracy of advertising image generation. This makes it difficult for average users to create high-quality advertising images and effectively display advertisements.

[0065] 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.

[0066] In this invention, the server includes: means for a user to input detailed product information using an information processing terminal; means for transmitting the input detailed product information to the server in data format; means for the server to receive the detailed product information and input the information into a generative AI model to automatically generate an advertising image; means for the server to return the generated advertising image to the user's information processing terminal; means for the information processing terminal to display the generated advertising image to the user; means for the user to post an advertisement and collect advertising effectiveness data; and means for the server to receive the collected advertising effectiveness data and feed it back to the generative AI model for learning. This enables users without specialized knowledge to easily generate high-quality advertising images, post advertisements effectively, and evolve the generative AI model based on the effectiveness data.

[0067] "Information processing terminal" refers to a device used by a user to input detailed product information, and specifically includes a personal computer or smartphone.

[0068] "Product details information" refers to information that a user inputs to generate an advertising image, such as the product name, price, description, and image.

[0069] "Data format" refers to the standardized format, such as JSON or XML, used when sending product details to the server.

[0070] "Server" refers to the computer system that receives product details, generates advertising images using a generative AI model, and returns the results to the user.

[0071] A "generative AI model" refers to an algorithm that automatically generates advertising images from input product details using machine learning techniques such as neural networks.

[0072] "Advertising images" refers to visual content that is automatically generated by a generative AI model and is intended for use by users as advertising.

[0073] "Advertising effectiveness data" refers to data such as the number of clicks, impressions, and conversion rates obtained as a result of users placing advertisements.

[0074] "Feedback" refers to the process of re-inputting collected advertising effectiveness data into the generative AI model to improve the model's performance.

[0075] "Auto-generation" refers to the process by which a generative AI model generates advertising images without human intervention.

[0076] A specific embodiment for carrying out the present invention will be described. In this system, users, terminals, and servers work in cooperation with each other. Each element of the system and its role will be specifically described below.

[0077] First, the user accesses the system's web application from an information processing terminal. Typical information processing terminals are personal computers or smartphones. The web application is composed of HTML, CSS, and JavaScript, and displays a form for entering product details. This form includes fields for product name, price, description, and image upload.

[0078] Next, the user enters the product name, price, description, and image into the input fields and clicks the submit button. The terminal converts the entered product details into JSON format and sends it to the server as an HTTP POST request. The technologies used include JavaScript, JSON, and the HTTP protocol.

[0079] The server runs on Node.js and processes the received product details in JSON format. Python is used on the server to pass the information to a generative AI model made up of a neural network. The generative AI model generates advertising images based on the product details and saves the image data in storage on the server. The technologies used include Node.js, Python, and generative AI models (e.g., GAN).

[0080] The generated ad image is sent back from the server to the user's device. Specifically, the URL or path of the generated ad image is sent as an HTTP response. The device receives this and displays the ad image using JavaScript. The user can then view the generated ad image in their browser.

[0081] Users use the generated ad images to place ads on advertising platforms (typically search engines or social media advertising platforms). They then collect advertising effectiveness data, such as the number of clicks, impressions, and conversion rates. At this time, users can upload this advertising effectiveness data to a server. The collected effectiveness data is used as feedback to retrain the generative AI model.

[0082] As a concrete example, consider a scenario in which a user creates an advertisement for a new smartphone. The user inputs the product name "Latest Model Smartphone," the price "50,000 yen," the product description "Large Capacity Battery, Fast Charging Supported," and a product image into the system. The device sends this information to the server, and the server's generative AI model automatically generates the advertisement image. The generated advertisement image is sent back to the user's device, and the user uses it in their own advertising campaign. At a later date, the user sends the advertisement's performance data to the server, and the AI ​​model is retrained to further improve its accuracy.

[0083] Examples of prompts include:

[0084] "Generate an ad for a new smartphone. Product name: Latest model smartphone, Price: 50,000 yen, Description: Large capacity battery, Fast charging supported, Product image: [Image URL]."

[0085] This process allows users to generate high-quality advertising images without specialized knowledge, place ads effectively, and use the results data to improve the generative AI model.

[0086] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0087] Step 1:

[0088] User Interface Display

[0089] User: Accesses the system's web application from an information processing terminal (personal computer or smartphone).

[0090] Input: A URL request made by a web browser.

[0091] Output: An HTML form is displayed to enter product details.

[0092] What happens: The browser makes a server request, the server responds with HTML, CSS, and JavaScript, and the screen displays the product name, price, description, and a field for uploading an image.

[0093] Step 2:

[0094] Enter and submit product details

[0095] User: Enters product name, price, description, and image and clicks submit.

[0096] Input: Product details (product name, price, description, image).

[0097] Output: The data is converted to JSON format and sent to the server.

[0098] What it does: JavaScript takes the input, converts it to JSON format, and sends that data to the server as an HTTP POST request.

[0099] Step 3:

[0100] Receiving and analyzing product details

[0101] Server: The server receives the HTTP request and parses the JSON formatted data.

[0102] Input: Product details in JSON format sent as an HTTP POST request.

[0103] Output: Structured data with extracted product name, price, description, and image path.

[0104] What it does: A Node.js backend receives requests and parses the data, which is then passed to a Python generative AI model.

[0105] Step 4:

[0106] Generating advertising images

[0107] Server: The generative AI model on the server automatically generates advertising images based on product information.

[0108] Input: Structured product details (product name, price, description, image path).

[0109] Output: The generated ad image.

[0110] How it works: A Python script calls a generative AI model (e.g., GAN) and generates advertising images using product information as input. The generated images are then stored in the server's storage.

[0111] Step 5:

[0112] Return and display of generated advertising images

[0113] Server: Obtains the URL or path of the generated ad image and returns it to the user's device as an HTTP response.

[0114] Input: The path to the generated ad image.

[0115] Output: HTTP response containing the image URL or path.

[0116] Specific operation: Node.js obtains the storage path of the generated ad image and returns it to the user's device in JSON format.

[0117] Terminal: Receives the response and displays the ad image.

[0118] Input: The URL of the ad image returned by the server.

[0119] Output: The ad image that is displayed in the user's browser.

[0120] What it does: JavaScript parses the response, gets the image URL and inserts it into the HTML.

[0121] Step 6:

[0122] Advertisement placement and collection of effectiveness data

[0123] User: Posts the generated ad image on an advertising platform and collects performance data.

[0124] Input: Generated ad image and ad campaign configuration information.

[0125] Output: Ad effectiveness data (clicks, impressions, conversion rate, etc.).

[0126] Specific operations: Users log in to the advertising platform and set up campaigns using ad images. After publishing, performance data is collected.

[0127] Step 7:

[0128] Re-learning based on effectiveness data

[0129] User: Uploads advertising effectiveness data to the server.

[0130] Input: Advertising effectiveness data (e.g., CSV format).

[0131] Output: Data uploaded to the server.

[0132] Specific action: The user enters effect data into a web form and presses the submit button.

[0133] Server: Analyzes the received effect data and retrains the generative AI model.

[0134] Input: Uploaded ad effectiveness data.

[0135] Output: A generative AI model with improved accuracy.

[0136] What it does: A Python script uses the effects data to retrain the generative AI model.

[0137] (Application example 1)

[0138] 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."

[0139] Currently, companies and individuals running online stores and e-commerce sites need advanced design skills and expertise to generate and utilize effective advertising images. Without these skills, it is difficult to create high-quality advertising images. Furthermore, there is a lack of mechanisms for effectively utilizing advertising performance data to continuously improve ad generation AI models. For these reasons, there is a need for a system that can automatically and efficiently generate ads and optimize their effectiveness.

[0140] 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.

[0141] In this invention, the server includes means for receiving detailed product information from a user, means for inputting the received detailed information into a generation AI model to automatically generate advertising images, means for providing the generated advertising images to the user, means for feeding back advertising effectiveness data collected after the advertisement is posted to the generation AI model for learning, means for using a database for saving the advertising effectiveness data, means for re-learning the AI ​​model, and means for using a generation engine for improving the quality of the advertising images based on the collected advertising effectiveness data. This makes it possible to generate high-quality advertising images and optimize advertising effectiveness without any special expertise.

[0142] The "means for accepting detailed product information from the user" is a function for providing an interface that allows the user to input information about the product, such as the product name, price, description, and image, and inputting that information into the system.

[0143] "Means for automatically generating advertising images by inputting received detailed information into a generative AI model" refers to a function that sends product information entered by the user to a server, and the generative AI model automatically generates advertising images based on that information.

[0144] "Means for providing generated advertising images to users" refers to a function for providing advertising images created by a generative AI model to users' devices so that users can view and use them.

[0145] "Means of feeding back advertising effectiveness data collected after an advertisement is posted to the AI ​​generation model to allow it to learn" refers to a function that feeds back advertising effectiveness data collected after a user posts an advertisement, such as the number of clicks, number of impressions, and conversion rate, to the AI ​​generation model and uses it as learning data for the model.

[0146] The "means for using a database for storing advertising effectiveness data" is a function for storing collected advertising effectiveness data in a database within the system so that it can be used for later analysis and model re-learning.

[0147] "Means for retraining the AI ​​model" refers to a function that retrains the generation AI model based on advertising effectiveness data to improve the accuracy of advertising image generation.

[0148] "Means for using a generation engine to improve the quality of advertising images based on collected advertising effectiveness data" refers to a function for generating higher quality advertising images by analyzing collected advertising effectiveness data and updating and improving the generation AI model based on the results.

[0149] A specific embodiment for carrying out the present invention will be described below. A user, a terminal, and a server work together based on the following steps.

[0150] User Interface (UI) Design

[0151] The server provides a user interface for users to enter product information. The UI displays fields for entering the product name, price, description, and image. The user enters product details in these fields. For example, the user enters the product name as "Latest Model Smartphone," the price as "50,000 yen," the product description as "Large capacity battery, fast charging compatible," and uploads a product image.

[0152] Sending product information

[0153] The terminal collects the product information entered by the user and sends it to the server in JSON format, where the entered product name, price, description, and uploaded image are compiled into a single data object.

[0154] Generating advertising images

[0155] The server passes the received product information to a generative AI model, which uses a neural network to generate advertising images based on the input information. This generation process is automatic, and the generated advertising images are stored on the server.

[0156] Sending and displaying advertising images

[0157] The server obtains the storage path or URL of the generated ad image and returns it to the user's device. The device then obtains the URL of the ad image received from the server and displays it on the user's screen. The user can then check the generated ad image and use it to place an ad on the advertising platform.

[0158] Advertising and data collection

[0159] The user uses the generated advertising image to post an advertisement on the advertising platform, and in this case, the user collects performance data (number of clicks, number of impressions, conversion rate, etc.) related to the posted advertisement.

[0160] Data feedback and AI model retraining

[0161] The server receives advertising effectiveness data provided by users after the advertisement is posted. The server feeds this data back into the generative AI model and retrains the AI ​​model. This feedback loop improves the accuracy of the generative AI model in generating advertising images.

[0162] Hardware and Software

[0163] To realize this system, the following hardware and software are used:

[0164] Server: Receives product information, generates advertising images using a generative AI model, and sends them back to the user's device. The generative AI model is installed on the server, and the advertising images are generated using that model.

[0165] Terminal: The user enters product information and sends it to the server. The server returns an advertisement image and displays it.

[0166] Generative AI model: A neural network model that creates advertising images based on product information. It is expected to use OpenAI APIs, etc.

[0167] Database: Stores advertising effectiveness data and uses it to retrain AI models.

[0168] Examples of specific examples and prompts

[0169] For example, if a user were to create an ad for a new smartphone, they would enter the following information:

[0170] Product name: Latest model smartphone

[0171] Price: \50,000

[0172] Description: Large capacity battery, fast charging

[0173] Image: base64 encoded image data

[0174] Example prompt for a generative AI model:

[0175] Please generate an ad image based on the following product information:

[0176] Product name: Latest model smartphone

[0177] Price: \50,000

[0178] Description: Large capacity battery, fast charging

[0179] Image: base64 encoded image data

[0180] In this way, a system can be constructed that can easily generate high-quality advertising images without requiring specialized knowledge and can optimize the effectiveness of advertising.

[0181] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0182] Step 1:

[0183] The user enters product information using a device. The device's user interface displays fields for entering the product name, price, description, and image. The user enters the required information in these fields and uploads an image. The entered information is temporarily stored on the device.

[0184] Input: User input of product name, price, description, and image

[0185] Output: Product information saved on the device

[0186] Step 2:

[0187] The device collects product information in JSON format and sends it to the server. The product information is formatted as a single data object. The server receives this and saves it as product information.

[0188] Input: Product information saved on the device

[0189] Output: JSON formatted data object, product information sent to the server

[0190] Step 3:

[0191] The server passes the received product information to the generative AI model, which generates advertising images based on input information such as product name, price, description, and image. The generative AI model uses a neural network to automatically create advertising images, using the input information as prompts during this process.

[0192] Input: A JSON formatted data object

[0193] Output: Advertising image generated by the generative AI model

[0194] Step 4:

[0195] The server obtains the storage path or URL of the generated advertisement image and returns it to the terminal. The terminal obtains the URL of the received advertisement image and displays it on the user's screen.

[0196] Input: Generated ad image

[0197] Output: Ad image storage path or URL, ad image displayed on the user's screen

[0198] Step 5:

[0199] Users use the generated ad images to place ads on the advertising platform, and performance data on the placed ads, such as the number of clicks, impressions, and conversion rates, is collected and later sent to the server.

[0200] Input: Ad image, Ad placement on advertising platform

[0201] Output: Collected advertising effectiveness data

[0202] Step 6:

[0203] The server receives advertising effectiveness data provided by users and feeds it back to the generative AI model. This data is stored in a database and used to retrain the AI ​​model, allowing the generative AI model to improve the accuracy of generating advertising images.

[0204] Input: Advertising effectiveness data

[0205] Output: Advertising effectiveness data stored in a database, and a generative AI model that receives feedback.

[0206] Step 7:

[0207] The server periodically retrains the AI ​​model based on collected advertising effectiveness data, improving the quality of the generated advertising images and helping with future ad generation.

[0208] Input: Advertising effectiveness data stored in the database

[0209] Output: The generative AI model updated through retraining

[0210] These steps realize a system that allows users to easily generate high-quality advertising images without specialized knowledge and optimize the effectiveness of advertising.

[0211] 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.

[0212] A specific embodiment for carrying out the present invention will be described below. The user, terminal, server, and emotion engine work together based on the following steps.

[0213] User Interface (UI) Design

[0214] User: First, the user accesses the system on their device. The UI on the screen displays fields for entering the product name, price, description, and image. The user enters product details in these fields. At this time, an emotion engine that recognizes the user's emotional state in real time is activated, and the user's emotional data is collected.

[0215] Sending product information

[0216] Terminal: Collects product information entered by the user and sends it to the server in JSON format. The entered product name, price, description, and uploaded image are compiled into a single data object. In addition, the user's emotion data collected by the emotion engine is also sent.

[0217] Generating advertising images

[0218] Server: After receiving the product information and emotion data, the server passes it to the generative AI model. The generative AI model uses a neural network to generate advertising images based on the input information. This generation process is automated, and the generated advertising images are stored on the server.

[0219] Sending and displaying advertising images

[0220] Server: Obtain the storage path or URL of the generated ad image and return it to the user's device.

[0221] Device: The URL of the ad image received from the server is obtained, set as the src attribute of the image element, and displayed on the user's screen. The user checks the generated ad image. At this time, the emotion engine recognizes the user's reaction in real time and collects emotional feedback data.

[0222] Advertising and data collection

[0223] User: Using the generated ad image, the user sets up an ad campaign and places ads on advertising platforms such as Yahoo! and LINE. At this time, the user collects performance data (number of clicks, number of impressions, conversion rate, etc.) related to the ads they place.

[0224] Data feedback and AI model retraining

[0225] Server: After the ad is displayed, the server receives the ad effectiveness data provided by the user and the emotion data recognized by the emotion engine. The server feeds this data back into the generative AI model and retrains the model. This feedback loop improves the accuracy of the AI ​​model's ad image generation.

[0226] As a concrete example, consider a user creating an advertisement for a new smartphone. When the user enters the product name "Latest Model Smartphone," the price "50,000 yen," the product description "High-capacity battery, fast charging compatible," and a product image into the system, the emotion engine monitors the user's emotional state in real time and collects data. The device sends this information to the server, and the server-side generative AI model automatically generates an advertisement image. The generated advertisement image is sent back to the user's device, where the user reviews it, and the emotion engine again analyzes the user's reaction. The user runs an advertising campaign using the generated advertisement image, and then provides the collected advertising effectiveness data and emotion data to the server. The generative AI model re-trains using the information obtained in this way, improving the accuracy of future advertisement image generation. This system enables more efficient and accurate advertisement creation, providing greater convenience to users.

[0227] The processing flow will be explained below.

[0228] Step 1: The user accesses the system on their device and inputs product details such as product name, price, description, and image. During input, the emotion engine recognizes the user's emotional state in real time and collects emotion data.

[0229] Step 2: The user's device compiles the entered product information and the emotion data collected by the emotion engine into a JavaScript object, converts it into JSON format, and sends an HTTP POST request to the server.

[0230] Step 3: The server receives the HTTP request, parses the product information and sentiment data from the request body, converts the product information into an appropriate format, and passes it to the generative AI model.

[0231] Step 4: The generative AI model generates advertising images based on product information and emotion data. This generation process is carried out using a neural network. The generated advertising images are stored on the server.

[0232] Step 5: The server obtains the storage path or URL of the generated ad image and returns it to the user's device in JSON format.

[0233] Step 6: The user's device retrieves the URL of the ad image received from the server, sets it as the src attribute of the image element, and displays it on the user's screen. The user checks the generated ad image. At this time, the emotion engine recognizes the user's reaction in real time and collects emotional feedback data.

[0234] Step 7: The user uploads the confirmed ad image to an advertising platform such as Yahoo or LINE, sets up an advertising campaign, and publishes it.

[0235] Step 8: During the advertising campaign, the user collects advertising effectiveness data such as clicks, impressions, and conversion rates from the advertising platform.

[0236] Step 9: The user provides the collected advertising effectiveness data to the server using the API provided by the server.

[0237] Step 10: The server stores the received advertising effectiveness data and the emotion data recognized by the emotion engine in a database and feeds it back to the generative AI model.

[0238] Step 11: The generative AI model is retrained based on the feedback data to improve the accuracy of advertising image generation.

[0239] These steps realize a system in which the user, terminal, server, and emotion engine work together to automatically generate advertising images and continuously improve accuracy by utilizing advertising effectiveness data and emotion data.

[0240] Example 2

[0241] 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."

[0242] In today's online advertising market, the generation and optimization of effective ad images is crucial, but the process of users manually creating and optimizing ad images is time-consuming and labor-intensive. Furthermore, while generating ad images that take users' emotional states into account would improve the user experience, existing technologies have not automated this process. Furthermore, there are insufficient mechanisms for effectively collecting post-campaign effectiveness data and incorporating it into re-learning. To address these challenges, it is necessary to automate ad image generation, collect real-time user emotional data, generate ad images that reflect that data, and establish a feedback loop for ad effectiveness data.

[0243] 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.

[0244] In this invention, the server includes a means for providing a user interface for users to input product names, prices, descriptions, and images; a means for collecting the input product information and real-time emotion data; a means for converting the collected data into JSON format and transmitting it to the server; a means for the server to input the received data into a generative AI model and automatically generate advertising images; a means for providing the generated advertising images to users; a means for collecting advertising effectiveness data after the advertising campaign is implemented; a means for feeding back the collected advertising effectiveness data and emotion data to the generative AI model for learning; and a means for the terminal to recognize the user's emotional state in real time and collect emotion data. This enables the automation and optimization of advertising image generation, enabling advertising image generation that takes user emotion data into account. Furthermore, the feedback loop of advertising effectiveness data improves the accuracy of the generative AI model, enabling more effective advertising campaign design.

[0245] A "user interface" is a system component that includes an input screen for a user to input product names, prices, descriptions, and images.

[0246] "Emotion engine" refers to a piece of software or hardware that recognizes a user's emotional state in real time and collects that data.

[0247] "JSON format" is an abbreviation for JavaScript Object Notation, and is a format for structuring and expressing data in text format.

[0248] A "generative AI model" is a model with automatic generation capabilities using artificial intelligence, which generates advertising images using neural networks.

[0249] "Advertising images" are visual advertising materials generated by a generative AI model based on product information and emotional data.

[0250] "Advertising effectiveness data" means data collected after an advertising campaign is conducted, including performance indicators such as clicks, impressions, and conversion rates.

[0251] "Feedback" is the process of providing advertising effectiveness data and emotional data to the generative AI model in order to retrain the model and improve its accuracy.

[0252] "API" stands for Application Programming Interface, an interface that enables data exchange between different software systems.

[0253] The present invention relates to a system for automatically generating and optimizing advertising images, in which a user interface, an emotion engine, a server, and a generative AI model work in cooperation with each other. Specific embodiments for carrying out the present invention will be described below.

[0254] User Interface Design

[0255] Terminal: First, the user accesses the system through a terminal. The user interface displays fields for entering the product name, price, description, and image. These fields are designed to be intuitive, allowing the user to easily enter product details.

[0256] Entering product information and collecting sentiment data

[0257] User: When a user enters product information, the emotion engine works in the background to recognize the user's emotional state in real time through the user's camera. For example, if the user is smiling, the system will recognize the emotion as "positive" and collect this data.

[0258] Product information and emotion data sent to server

[0259] Terminal: After the user finishes entering information, the terminal converts the product information and emotion data into JSON format and sends it to the server. The transmitted data includes the product name, price, description, image, and emotion data.

[0260] Generating advertising images

[0261] Server: The server analyzes the received data and passes it to a generative AI model. This generative AI model uses a neural network to automatically generate advertising images based on product information and emotional data. The generated advertising images are stored on the server.

[0262] Providing advertising images

[0263] Server: Obtains the storage path or URL of the generated ad image and returns it to the user's device. The user can view the ad image on their screen.

[0264] Running an advertising campaign

[0265] User: Using the generated ad image, the user sets up an advertising campaign. The ad is then placed on advertising platforms such as Yahoo! and LINE, and information such as target demographic, period, and budget is set.

[0266] Collection and feedback of advertising effectiveness data

[0267] User: After an advertising campaign is run, the user collects advertising effectiveness data such as clicks, impressions, and conversion rates.

[0268] Terminal: Collected advertising effectiveness data and emotion data are compiled and sent back to the server.

[0269] Server: The server feeds this data back into the generative AI model and retrains it, improving the accuracy of subsequent ad image generation.

[0270] Specific examples

[0271] For example, if a user is creating an ad for a new smartphone, they might input the following prompt into the generative AI model:

[0272] Product name: Latest model smartphone

[0273] Price: \50,000

[0274] Product description: Large capacity battery, fast charging compatible

[0275] Product Image: Image URL (e.g. www.example.com / image.jpg)

[0276] In this way, users can quickly and efficiently generate highly effective advertising images. This system is designed to significantly reduce the effort required for creating advertisements and maximize their effectiveness.

[0277] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0278] Step 1:

[0279] User Interface Display

[0280] Device: An application or web browser is launched that displays a screen that provides fields for entering product name, price, description, and image. The screen is designed to make it easy for the user to enter information visually. Once the input fields are visible, this constitutes input.

[0281] Step 2:

[0282] Entering product information and collecting sentiment data

[0283] User: Enters product name, price, description, and image into input fields on the device. This is the input data. Meanwhile, the emotion engine runs in the background, recognizing the user's emotional state (e.g., positive, negative) in real time through the user's camera and collecting emotion data. The emotion engine captures frames from the camera and uses a facial expression analysis algorithm to categorize the user's emotional state.

[0284] Step 3:

[0285] Product information and emotion data sent to server

[0286] Terminal: Once the user has completed the input, the terminal will combine the product name, price, description, uploaded image, and emotion data into a single data object. This is then serialized into JSON format and sent to the server via a secure protocol (e.g., HTTPS). Once the data has been sent to the server, the server receives it.

[0287] Step 4:

[0288] Generating advertising images

[0289] Server: Analyzes the received JSON data and passes the data object to the generative AI model. The generative AI model uses deep learning technology to generate advertising images using product information and emotional data as input. Specifically, a neural network processes the input data and creates advertising images containing appropriate visual elements and text designs. The generated advertising images are stored on the server. The output is the generated advertising image data.

[0290] Step 5:

[0291] Providing advertising images

[0292] Server: The output is to obtain the storage path or URL of the generated ad image and return it to the user's device in JSON format.

[0293] On the device: The received ad image URL is parsed and set as the src attribute of an HTML image element, which displays the image on the screen. This allows the user to check the generated ad image. The emotion engine then runs again, analyzing the user's reactions in real time and collecting new emotion data.

[0294] Step 6:

[0295] Running an advertising campaign

[0296] User: Uses the generated ad image to set up an ad campaign. The user starts the ad campaign by entering information about the target demographic, period, budget, and ad platform (e.g., Yahoo, LINE). This setting information is the input data. The ad campaign is then carried out, and the actual ads are displayed by the ad platform.

[0297] Step 7:

[0298] Collection and transmission of advertising effectiveness data

[0299] User: After running an advertising campaign, you collect advertising effectiveness data such as clicks, impressions, and conversion rates from the advertising platform. This is input data.

[0300] Terminal: The output is to compile the collected advertising effectiveness data and the user feedback data recorded by the emotion engine, serialize it in JSON format, and send it back to the server.

[0301] Step 8:

[0302] Data feedback and AI model retraining

[0303] Server: Analyzes the received advertising effectiveness data and emotion data and feeds it back to the generative AI model. This allows the generative AI model to be retrained using this data. In particular, the generative AI model uses deep learning algorithms such as backpropagation to adjust model parameters and improve the accuracy of subsequent ad image generation. This is the final output.

[0304] (Application example 2)

[0305] 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."

[0306] Conventional ad generation systems generate ad images without considering user emotions, resulting in poor ad effectiveness. Furthermore, the process of feeding back ad effectiveness data and retraining the AI ​​model is cumbersome, making it difficult to improve the accuracy of generating effective ads. This results in problems such as the inability to create effective ads that users desire, and reduced efficiency of advertising campaigns.

[0307] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0308] In this invention, the server includes means for receiving detailed product information from a user, means for inputting the received detailed information and emotion data into a generative AI model to automatically generate advertising images, means for providing the generated advertising images to the user, an emotion engine for collecting user emotional responses in real time, and means for feeding back advertising effectiveness data collected after the advertisement is posted to the generative AI model for learning. This enables automatic generation of advertising images that take user emotions into consideration and re-training of the AI ​​model based on the advertising effectiveness data.

[0309] "Product details" are basic attributes of a product provided by the user, such as its name, price, description, and image.

[0310] "Emotional data" refers to data that measures and collects a user's emotional state in real time.

[0311] A "generative AI model" is an artificial intelligence model that automatically generates advertising images based on input information.

[0312] "Advertising images" are visual promotional materials generated based on detailed product information and emotional data provided by users.

[0313] The "emotion engine" is a system for collecting and analyzing users' emotional responses in real time.

[0314] "Advertising effectiveness data" refers to data on performance indicators such as the number of clicks, impressions, and conversion rates in advertising campaigns.

[0315] "Feedback" is the process of retraining the generative AI model based on collected data to improve the accuracy of ad generation from the next time onwards.

[0316] An "application program interface" is a standardized means for exchanging data between different software systems.

[0317] This invention is a system that automatically generates advertising images based on input of product information, and improves the accuracy of advertisement generation by grasping user emotions in real time and obtaining feedback. This system is implemented on both the front end and the server side.

[0318] front end

[0319] On the front end, users access the interface using their smartphones. Specifically, they input product names, prices, descriptions, and image URLs through an application built with React Native. An emotion engine also detects the user's facial expressions in real time and collects emotional data. This collected data is sent to the server in JSON format.

[0320] Specific examples

[0321] When a user creates an ad for a new smartwatch, they enter the following information:

[0322] Product Name: Latest Smartwatch

[0323] Price: \30,000

[0324] Product description: Heart rate sensor, high-precision GPS

[0325] Product image URL: https: / / example.com / images / smartwatch.jpg

[0326] Server Side

[0327] On the server side, the received data is processed using Node.js and Express. Specifically, the received JSON data is input into a generative AI model to generate ad images. This generative AI model uses TensorFlow.js, which generates ad images using a neural network. The generated ad images are stored on the server, and their URLs are returned to the user's device. In addition, user emotional responses and ad effectiveness data (number of clicks, number of impressions, conversion rate, etc.) are collected in real time and fed back to the server. This data is passed to the generative AI model and used for re-training.

[0328] Specific examples

[0329] The user checks the ad image and runs an advertising campaign. For example, the generated ad image is used to post on an advertising platform, and the AI ​​model is retrained using the advertising effectiveness data obtained later. An example of the prompt sentence in this case is as follows:

[0330] Prompt Sentence Examples

[0331] Based on the information received from the user, the generative AI model generates advertising images as follows:

[0332] Prompt text: Product name "Latest smartwatch", price "30,000 yen", product description "Heart rate sensor, high-precision GPS", image URL "https: / / example.com / images / smartwatch.jpg", emotion data "Happiness level: 80%, Neutrality level: 20%".

[0333] In this way, it is possible to improve the accuracy of advertising images generated based on user emotions and advertising effectiveness data, and to realize efficient advertising campaigns.

[0334] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0335] Step 1:

[0336] A user uses a smartphone to access an application built with React Native.

[0337] Users enter the product name, price, description, and image URL in the corresponding input fields, and the emotion engine collects emotional data from the user's facial expressions, such as happiness, surprise, anger, etc. These details and emotional data are then compiled and converted into JSON format.

[0338] input:

[0339] Product name, price, description, image URL

[0340] Emotional data (happiness, surprise, anger, etc.)

[0341] output:

[0342] Detailed information and emotion data in JSON format

[0343] Step 2:

[0344] The device sends the collected details and emotion data in JSON format to the server.

[0345] This data is sent as an HTTP POST request.

[0346] input:

[0347] Detailed information and emotion data in JSON format

[0348] output:

[0349] HTTP POST request to the server

[0350] Step 3:

[0351] The server analyzes the received JSON-formatted details and emotion data and prepares them for input into the generative AI model.

[0352] Specifically, it formats each input item (product name, price, description, image URL, emotion data).

[0353] input:

[0354] Detailed information and emotion data in JSON format

[0355] output:

[0356] Formatted details and sentiment data

[0357] Step 4:

[0358] The server inputs the formatted details and emotional data into a generative AI model to generate advertising images.

[0359] The generative AI model uses TensorFlow.js and employs a neural network to generate advertising images from input data.

[0360] input:

[0361] Formatted details and sentiment data

[0362] output:

[0363] Generated advertising image

[0364] Step 5:

[0365] The server stores the generated advertisement image therein and returns the URL to the terminal.

[0366] The URL of the returned ad image is displayed in the client-side interface for the user to view.

[0367] input:

[0368] Generated advertising image

[0369] output:

[0370] Ad image URL

[0371] Step 6:

[0372] The user can review the generated advertising images and use them for their advertising campaigns.

[0373] In addition, the emotion engine records the user's emotional reactions in real time and sends them to the server.

[0374] input:

[0375] Advertising images

[0376] output:

[0377] User emotional response data

[0378] Step 7:

[0379] The server feeds back the emotional response data collected from users and advertising effectiveness data (number of clicks, number of impressions, conversion rate, etc.) to the generative AI model.

[0380] The generated AI model is retrained based on this data.

[0381] input:

[0382] Emotional response data

[0383] Advertising effectiveness data

[0384] output:

[0385] Generative AI model with improved accuracy through retraining

[0386] This makes it possible to generate advertising images based on user emotions and advertising effectiveness data, enabling efficient advertising campaign management.

[0387] 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.

[0388] 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.

[0389] 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.

[0390] [Second embodiment]

[0391] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0392] 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.

[0393] 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).

[0394] 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.

[0395] 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.

[0396] 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).

[0397] 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. 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.

[0398] 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.

[0399] 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.

[0400] 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.

[0401] In the smart glasses 214, 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.

[0402] 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."

[0403] A specific embodiment for carrying out the present invention will be described below. A user, a terminal, and a server work together based on the following steps.

[0404] User Interface (UI) Design

[0405] User: First, the user accesses the system on a terminal. The UI on the screen displays fields for entering the product name, price, description, and image. The user enters the product details in these fields.

[0406] Sending product information

[0407] Terminal: Collects product information entered by the user and sends it to the server in JSON format, combining the entered product name, price, description, and uploaded image into a single data object.

[0408] Generating advertising images

[0409] Server: Once product information is received, the server passes it to the generative AI model, which uses a neural network to generate advertising images based on the input information. This generation process is automated, and the generated advertising images are stored on the server.

[0410] Sending and displaying advertising images

[0411] Server: Obtain the storage path or URL of the generated ad image and return it to the user's device.

[0412] Terminal: The URL of the ad image received from the server is acquired and displayed on the user's screen. The user can then check the generated ad image and use it to place an ad.

[0413] Advertising and data collection

[0414] User: Uses the generated ad image to post an ad on an advertising platform such as Yahoo! or LINE. At this time, the user collects performance data (number of clicks, number of impressions, conversion rate, etc.) related to the posted ad.

[0415] Data feedback and AI model retraining

[0416] Server: After the ad is published, the server receives the ad effectiveness data provided by the user. The server feeds this data back into the generative AI model and retrains the model. This feedback loop improves the accuracy of the AI ​​model's ad image generation.

[0417] As a concrete example, consider a user creating an advertisement for a new smartphone. First, the user enters the product name "Latest Model Smartphone," the price "50,000 yen," the product description "High-capacity battery, fast charging compatible," and a product image into the system. The device sends this information to the server, and the server-side generative AI model automatically generates the advertisement image. The generated advertisement image is sent back to the user's device, and the user uses it in their own advertising campaign. At a later date, the user sends the advertisement's performance data to the server, and the AI ​​model retrains to further improve accuracy. This series of processes allows users to easily generate and publish high-quality advertisements, even without specialized knowledge.

[0418] The processing flow will be explained below.

[0419] Step 1: The user accesses the system on a terminal and enters product details such as product name, price, description, and image.

[0420] Step 2: The user's device compiles the entered product information into a JavaScript object, converts it to JSON format, and sends an HTTP POST request to the server.

[0421] Step 3: The server receives the HTTP request, parses the product information from the request body, converts the product information into an appropriate format, and passes it to the generative AI model.

[0422] Step 4: The server's AI model generates advertising images based on the product information. The generated advertising images are stored on the server.

[0423] Step 5: The server obtains the storage path or URL of the generated ad image and returns it to the user's device in JSON format.

[0424] Step 6: The user's device retrieves the URL of the ad image received from the server, sets it as the src attribute of the image element, and displays it on the user's screen. The user then checks the generated ad image.

[0425] Step 7: The user uploads the confirmed ad image to an advertising platform such as Yahoo or LINE, sets up an advertising campaign, and publishes it.

[0426] Step 8: During the advertising campaign, the user collects advertising effectiveness data such as clicks, impressions, and conversion rates from the advertising platform.

[0427] Step 9: The user provides the collected advertising effectiveness data to the server using the API provided by the server.

[0428] Step 10: The server stores the received advertising effectiveness data in a database and feeds it back to the generative AI model.

[0429] Step 11: The server's generated AI model is retrained based on the feedback data to improve the accuracy of advertising image generation.

[0430] These steps realize a system in which users, terminals, and servers work together to automatically generate advertising images and utilize advertising effectiveness data to continuously improve accuracy.

[0431] Example 1

[0432] 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."

[0433] Conventional advertising image generation systems require specialized knowledge and lack a means to automate the creation of advertising images and their effectiveness analysis. Furthermore, they lack a feedback mechanism to improve the accuracy of advertising image generation. This makes it difficult for average users to create high-quality advertising images and effectively display advertisements.

[0434] 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.

[0435] In this invention, the server includes: means for a user to input detailed product information using an information processing terminal; means for transmitting the input detailed product information to the server in data format; means for the server to receive the detailed product information and input the information into a generative AI model to automatically generate an advertising image; means for the server to return the generated advertising image to the user's information processing terminal; means for the information processing terminal to display the generated advertising image to the user; means for the user to post an advertisement and collect advertising effectiveness data; and means for the server to receive the collected advertising effectiveness data and feed it back to the generative AI model for learning. This enables users without specialized knowledge to easily generate high-quality advertising images, post advertisements effectively, and evolve the generative AI model based on the effectiveness data.

[0436] "Information processing terminal" refers to a device used by a user to input detailed product information, and specifically includes a personal computer or smartphone.

[0437] "Product details information" refers to information that a user inputs to generate an advertising image, such as the product name, price, description, and image.

[0438] "Data format" refers to the standardized format, such as JSON or XML, used when sending product details to the server.

[0439] "Server" refers to the computer system that receives product details, generates advertising images using a generative AI model, and returns the results to the user.

[0440] A "generative AI model" refers to an algorithm that automatically generates advertising images from input product details using machine learning techniques such as neural networks.

[0441] "Advertising images" refers to visual content that is automatically generated by a generative AI model and is intended for use by users as advertising.

[0442] "Advertising effectiveness data" refers to data such as the number of clicks, impressions, and conversion rates obtained as a result of users placing advertisements.

[0443] "Feedback" refers to the process of re-inputting collected advertising effectiveness data into the generative AI model to improve the model's performance.

[0444] "Auto-generation" refers to the process by which a generative AI model generates advertising images without human intervention.

[0445] A specific embodiment for carrying out the present invention will be described. In this system, users, terminals, and servers work in cooperation with each other. Each element of the system and its role will be specifically described below.

[0446] First, the user accesses the system's web application from an information processing terminal. Typical information processing terminals are personal computers or smartphones. The web application is composed of HTML, CSS, and JavaScript, and displays a form for entering product details. This form includes fields for product name, price, description, and image upload.

[0447] Next, the user enters the product name, price, description, and image into the input fields and clicks the submit button. The terminal converts the entered product details into JSON format and sends it to the server as an HTTP POST request. The technologies used include JavaScript, JSON, and the HTTP protocol.

[0448] The server runs on Node.js and processes the received product details in JSON format. Python is used on the server to pass the information to a generative AI model made up of a neural network. The generative AI model generates advertising images based on the product details and saves the image data in storage on the server. The technologies used include Node.js, Python, and generative AI models (e.g., GAN).

[0449] The generated ad image is sent back from the server to the user's device. Specifically, the URL or path of the generated ad image is sent as an HTTP response. The device receives this and displays the ad image using JavaScript. The user can then view the generated ad image in their browser.

[0450] Users use the generated ad images to place ads on advertising platforms (typically search engines or social media advertising platforms). They then collect advertising effectiveness data, such as the number of clicks, impressions, and conversion rates. At this time, users can upload this advertising effectiveness data to a server. The collected effectiveness data is used as feedback to retrain the generative AI model.

[0451] As a concrete example, consider a scenario in which a user creates an advertisement for a new smartphone. The user inputs the product name "Latest Model Smartphone," the price "50,000 yen," the product description "Large Capacity Battery, Fast Charging Supported," and a product image into the system. The device sends this information to the server, and the server's generative AI model automatically generates the advertisement image. The generated advertisement image is sent back to the user's device, and the user uses it in their own advertising campaign. At a later date, the user sends the advertisement's performance data to the server, and the AI ​​model is retrained to further improve its accuracy.

[0452] Examples of prompts include:

[0453] "Generate an ad for a new smartphone. Product name: Latest model smartphone, Price: 50,000 yen, Description: Large capacity battery, Fast charging supported, Product image: [Image URL]."

[0454] This process allows users to generate high-quality advertising images without specialized knowledge, place ads effectively, and use the results data to improve the generative AI model.

[0455] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0456] Step 1:

[0457] User Interface Display

[0458] User: Accesses the system's web application from an information processing terminal (personal computer or smartphone).

[0459] Input: A URL request made by a web browser.

[0460] Output: An HTML form is displayed to enter product details.

[0461] What happens: The browser makes a server request, the server responds with HTML, CSS, and JavaScript, and the screen displays the product name, price, description, and a field for uploading an image.

[0462] Step 2:

[0463] Enter and submit product details

[0464] User: Enters product name, price, description, and image and clicks submit.

[0465] Input: Product details (product name, price, description, image).

[0466] Output: The data is converted to JSON format and sent to the server.

[0467] What it does: JavaScript takes the input, converts it to JSON format, and sends that data to the server as an HTTP POST request.

[0468] Step 3:

[0469] Receiving and analyzing product details

[0470] Server: The server receives the HTTP request and parses the JSON formatted data.

[0471] Input: Product details in JSON format sent as an HTTP POST request.

[0472] Output: Structured data with extracted product name, price, description, and image path.

[0473] What it does: A Node.js backend receives requests and parses the data, which is then passed to a Python generative AI model.

[0474] Step 4:

[0475] Generating advertising images

[0476] Server: The generative AI model on the server automatically generates advertising images based on product information.

[0477] Input: Structured product details (product name, price, description, image path).

[0478] Output: The generated ad image.

[0479] How it works: A Python script calls a generative AI model (e.g., GAN) and generates advertising images using product information as input. The generated images are then stored in the server's storage.

[0480] Step 5:

[0481] Return and display of generated advertising images

[0482] Server: Obtains the URL or path of the generated ad image and returns it to the user's device as an HTTP response.

[0483] Input: The path to the generated ad image.

[0484] Output: HTTP response containing the image URL or path.

[0485] Specific operation: Node.js obtains the storage path of the generated ad image and returns it to the user's device in JSON format.

[0486] Terminal: Receives the response and displays the ad image.

[0487] Input: The URL of the ad image returned by the server.

[0488] Output: The ad image that is displayed in the user's browser.

[0489] What it does: JavaScript parses the response, gets the image URL and inserts it into the HTML.

[0490] Step 6:

[0491] Advertisement placement and collection of effectiveness data

[0492] User: Posts the generated ad image on an advertising platform and collects performance data.

[0493] Input: Generated ad image and ad campaign configuration information.

[0494] Output: Ad effectiveness data (clicks, impressions, conversion rate, etc.).

[0495] Specific operations: Users log in to the advertising platform and set up campaigns using ad images. After publishing, performance data is collected.

[0496] Step 7:

[0497] Re-learning based on effectiveness data

[0498] User: Uploads advertising effectiveness data to the server.

[0499] Input: Advertising effectiveness data (e.g., CSV format).

[0500] Output: Data uploaded to the server.

[0501] Specific action: The user enters effect data into a web form and presses the submit button.

[0502] Server: Analyzes the received effect data and retrains the generative AI model.

[0503] Input: Uploaded ad effectiveness data.

[0504] Output: A generative AI model with improved accuracy.

[0505] What it does: A Python script uses the effects data to retrain the generative AI model.

[0506] (Application example 1)

[0507] 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."

[0508] Currently, companies and individuals running online stores and e-commerce sites need advanced design skills and expertise to generate and utilize effective advertising images. Without these skills, it is difficult to create high-quality advertising images. Furthermore, there is a lack of mechanisms for effectively utilizing advertising performance data to continuously improve ad generation AI models. For these reasons, there is a need for a system that can automatically and efficiently generate ads and optimize their effectiveness.

[0509] 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.

[0510] In this invention, the server includes means for receiving detailed product information from a user, means for inputting the received detailed information into a generation AI model to automatically generate advertising images, means for providing the generated advertising images to the user, means for feeding back advertising effectiveness data collected after the advertisement is posted to the generation AI model for learning, means for using a database for saving the advertising effectiveness data, means for re-learning the AI ​​model, and means for using a generation engine for improving the quality of the advertising images based on the collected advertising effectiveness data. This makes it possible to generate high-quality advertising images and optimize advertising effectiveness without any special expertise.

[0511] The "means for accepting detailed product information from the user" is a function for providing an interface that allows the user to input information about the product, such as the product name, price, description, and image, and inputting that information into the system.

[0512] "Means for automatically generating advertising images by inputting received detailed information into a generative AI model" refers to a function that sends product information entered by the user to a server, and the generative AI model automatically generates advertising images based on that information.

[0513] "Means for providing generated advertising images to users" refers to a function for providing advertising images created by a generative AI model to users' devices so that users can view and use them.

[0514] "Means of feeding back advertising effectiveness data collected after an advertisement is posted to the AI ​​generation model to allow it to learn" refers to a function that feeds back advertising effectiveness data collected after a user posts an advertisement, such as the number of clicks, number of impressions, and conversion rate, to the AI ​​generation model and uses it as learning data for the model.

[0515] The "means for using a database for storing advertising effectiveness data" is a function for storing collected advertising effectiveness data in a database within the system so that it can be used for later analysis and model re-learning.

[0516] "Means for retraining the AI ​​model" refers to a function that retrains the generation AI model based on advertising effectiveness data to improve the accuracy of advertising image generation.

[0517] "Means for using a generation engine to improve the quality of advertising images based on collected advertising effectiveness data" refers to a function for generating higher quality advertising images by analyzing collected advertising effectiveness data and updating and improving the generation AI model based on the results.

[0518] A specific embodiment for carrying out the present invention will be described below. A user, a terminal, and a server work together based on the following steps.

[0519] User Interface (UI) Design

[0520] The server provides a user interface for users to enter product information. The UI displays fields for entering the product name, price, description, and image. The user enters product details in these fields. For example, the user enters the product name as "Latest Model Smartphone," the price as "50,000 yen," the product description as "Large capacity battery, fast charging compatible," and uploads a product image.

[0521] Sending product information

[0522] The terminal collects the product information entered by the user and sends it to the server in JSON format, where the entered product name, price, description, and uploaded image are compiled into a single data object.

[0523] Generating advertising images

[0524] The server passes the received product information to a generative AI model, which uses a neural network to generate advertising images based on the input information. This generation process is automatic, and the generated advertising images are stored on the server.

[0525] Sending and displaying advertising images

[0526] The server obtains the storage path or URL of the generated ad image and returns it to the user's device. The device then obtains the URL of the ad image received from the server and displays it on the user's screen. The user can then check the generated ad image and use it to place an ad on the advertising platform.

[0527] Advertising and data collection

[0528] The user uses the generated advertising image to post an advertisement on the advertising platform, and in this case, the user collects performance data (number of clicks, number of impressions, conversion rate, etc.) related to the posted advertisement.

[0529] Data feedback and AI model retraining

[0530] The server receives advertising effectiveness data provided by users after the advertisement is posted. The server feeds this data back into the generative AI model and retrains the AI ​​model. This feedback loop improves the accuracy of the generative AI model in generating advertising images.

[0531] Hardware and Software

[0532] To realize this system, the following hardware and software are used:

[0533] Server: Receives product information, generates advertising images using a generative AI model, and sends them back to the user's device. The generative AI model is installed on the server, and the advertising images are generated using that model.

[0534] Terminal: The user enters product information and sends it to the server. The server returns an advertisement image and displays it.

[0535] Generative AI model: A neural network model that creates advertising images based on product information. It is expected to use OpenAI APIs, etc.

[0536] Database: Stores advertising effectiveness data and uses it to retrain AI models.

[0537] Examples of specific examples and prompts

[0538] For example, if a user were to create an ad for a new smartphone, they would enter the following information:

[0539] Product name: Latest model smartphone

[0540] Price: \50,000

[0541] Description: Large capacity battery, fast charging

[0542] Image: base64 encoded image data

[0543] Example prompt for a generative AI model:

[0544] Please generate an ad image based on the following product information:

[0545] Product name: Latest model smartphone

[0546] Price: \50,000

[0547] Description: Large capacity battery, fast charging

[0548] Image: base64 encoded image data

[0549] In this way, a system can be constructed that can easily generate high-quality advertising images without requiring specialized knowledge and can optimize the effectiveness of advertising.

[0550] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0551] Step 1:

[0552] The user enters product information using a device. The device's user interface displays fields for entering the product name, price, description, and image. The user enters the required information in these fields and uploads an image. The entered information is temporarily stored on the device.

[0553] Input: User input of product name, price, description, and image

[0554] Output: Product information saved on the device

[0555] Step 2:

[0556] The device collects product information in JSON format and sends it to the server. The product information is formatted as a single data object. The server receives this and saves it as product information.

[0557] Input: Product information saved on the device

[0558] Output: JSON formatted data object, product information sent to the server

[0559] Step 3:

[0560] The server passes the received product information to the generative AI model, which generates advertising images based on input information such as product name, price, description, and image. The generative AI model uses a neural network to automatically create advertising images, using the input information as prompts during this process.

[0561] Input: A JSON formatted data object

[0562] Output: Advertising image generated by the generative AI model

[0563] Step 4:

[0564] The server obtains the storage path or URL of the generated advertisement image and returns it to the terminal. The terminal obtains the URL of the received advertisement image and displays it on the user's screen.

[0565] Input: Generated ad image

[0566] Output: Ad image storage path or URL, ad image displayed on the user's screen

[0567] Step 5:

[0568] Users use the generated ad images to place ads on the advertising platform, and performance data on the posted ads, such as the number of clicks, impressions, and conversion rates, is collected and later sent to the server.

[0569] Input: Ad image, Ad placement on advertising platform

[0570] Output: Collected advertising effectiveness data

[0571] Step 6:

[0572] The server receives advertising effectiveness data provided by users and feeds it back to the generative AI model. This data is stored in a database and used to retrain the AI ​​model, allowing the generative AI model to improve the accuracy of generating advertising images.

[0573] Input: Advertising effectiveness data

[0574] Output: Advertising effectiveness data stored in a database, and a generative AI model that receives feedback.

[0575] Step 7:

[0576] The server periodically retrains the AI ​​model based on collected advertising effectiveness data, improving the quality of the generated advertising images and helping with future ad generation.

[0577] Input: Advertising effectiveness data stored in the database

[0578] Output: The generative AI model updated through retraining

[0579] These steps realize a system that allows users to easily generate high-quality advertising images without specialized knowledge and optimize the effectiveness of advertising.

[0580] 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.

[0581] A specific embodiment for carrying out the present invention will be described below. The user, terminal, server, and emotion engine work together based on the following steps.

[0582] User Interface (UI) Design

[0583] User: First, the user accesses the system on their device. The UI on the screen displays fields for entering the product name, price, description, and image. The user enters product details in these fields. At this time, an emotion engine that recognizes the user's emotional state in real time is activated, and the user's emotional data is collected.

[0584] Sending product information

[0585] Terminal: Collects product information entered by the user and sends it to the server in JSON format. The entered product name, price, description, and uploaded image are compiled into a single data object. In addition, the user's emotion data collected by the emotion engine is also sent.

[0586] Generating advertising images

[0587] Server: After receiving the product information and emotion data, the server passes it to the generative AI model. The generative AI model uses a neural network to generate advertising images based on the input information. This generation process is automated, and the generated advertising images are stored on the server.

[0588] Sending and displaying advertising images

[0589] Server: Obtain the storage path or URL of the generated ad image and return it to the user's device.

[0590] Device: The URL of the ad image received from the server is obtained, set as the src attribute of the image element, and displayed on the user's screen. The user checks the generated ad image. At this time, the emotion engine recognizes the user's reaction in real time and collects emotional feedback data.

[0591] Advertising and data collection

[0592] User: Using the generated ad image, the user sets up an ad campaign and places ads on advertising platforms such as Yahoo! and LINE. At this time, the user collects performance data (number of clicks, number of impressions, conversion rate, etc.) related to the ads they place.

[0593] Data feedback and AI model retraining

[0594] Server: After the ad is displayed, the server receives the ad effectiveness data provided by the user and the emotion data recognized by the emotion engine. The server feeds this data back into the generative AI model and retrains the model. This feedback loop improves the accuracy of the AI ​​model's ad image generation.

[0595] As a concrete example, consider a user creating an advertisement for a new smartphone. When the user enters the product name "Latest Model Smartphone," the price "50,000 yen," the product description "High-capacity battery, fast charging compatible," and a product image into the system, the emotion engine monitors the user's emotional state in real time and collects data. The device sends this information to the server, and the server-side generative AI model automatically generates an advertisement image. The generated advertisement image is sent back to the user's device, where the user reviews it, and the emotion engine again analyzes the user's reaction. The user runs an advertising campaign using the generated advertisement image, and then provides the collected advertising effectiveness data and emotion data to the server. The generative AI model re-trains using the information obtained in this way, improving the accuracy of future advertisement image generation. This system enables more efficient and accurate advertisement creation, providing greater convenience to users.

[0596] The processing flow will be explained below.

[0597] Step 1: The user accesses the system on their device and inputs product details such as product name, price, description, and image. During input, the emotion engine recognizes the user's emotional state in real time and collects emotion data.

[0598] Step 2: The user's device compiles the entered product information and the emotion data collected by the emotion engine into a JavaScript object, converts it into JSON format, and sends an HTTP POST request to the server.

[0599] Step 3: The server receives the HTTP request, parses the product information and sentiment data from the request body, converts the product information into an appropriate format, and passes it to the generative AI model.

[0600] Step 4: The generative AI model generates advertising images based on product information and emotion data. This generation process is carried out using a neural network. The generated advertising images are stored on the server.

[0601] Step 5: The server obtains the storage path or URL of the generated ad image and returns it to the user's device in JSON format.

[0602] Step 6: The user's device retrieves the URL of the ad image received from the server, sets it as the src attribute of the image element, and displays it on the user's screen. The user checks the generated ad image. At this time, the emotion engine recognizes the user's reaction in real time and collects emotional feedback data.

[0603] Step 7: The user uploads the confirmed ad image to an advertising platform such as Yahoo or LINE, sets up an advertising campaign, and publishes it.

[0604] Step 8: During the advertising campaign, the user collects advertising effectiveness data such as clicks, impressions, and conversion rates from the advertising platform.

[0605] Step 9: The user provides the collected advertising effectiveness data to the server using the API provided by the server.

[0606] Step 10: The server stores the received advertising effectiveness data and the emotion data recognized by the emotion engine in a database and feeds it back to the generative AI model.

[0607] Step 11: The generative AI model is retrained based on the feedback data to improve the accuracy of advertising image generation.

[0608] These steps realize a system in which the user, terminal, server, and emotion engine work together to automatically generate advertising images and continuously improve accuracy by utilizing advertising effectiveness data and emotion data.

[0609] Example 2

[0610] 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."

[0611] In today's online advertising market, the generation and optimization of effective ad images is crucial, but the process of users manually creating and optimizing ad images is time-consuming and labor-intensive. Furthermore, while generating ad images that take users' emotional states into account would improve the user experience, existing technologies have not automated this process. Furthermore, there are insufficient mechanisms for effectively collecting post-campaign effectiveness data and incorporating it into re-learning. To address these challenges, it is necessary to automate ad image generation, collect real-time user emotional data, generate ad images that reflect that data, and establish a feedback loop for ad effectiveness data.

[0612] 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.

[0613] In this invention, the server includes a means for providing a user interface for users to input product names, prices, descriptions, and images; a means for collecting the input product information and real-time emotion data; a means for converting the collected data into JSON format and transmitting it to the server; a means for the server to input the received data into a generative AI model and automatically generate advertising images; a means for providing the generated advertising images to users; a means for collecting advertising effectiveness data after the advertising campaign is implemented; a means for feeding back the collected advertising effectiveness data and emotion data to the generative AI model for learning; and a means for the terminal to recognize the user's emotional state in real time and collect emotion data. This enables the automation and optimization of advertising image generation, enabling advertising image generation that takes user emotion data into account. Furthermore, the feedback loop of advertising effectiveness data improves the accuracy of the generative AI model, enabling more effective advertising campaign design.

[0614] A "user interface" is a system component that includes an input screen for a user to input product names, prices, descriptions, and images.

[0615] "Emotion engine" refers to a piece of software or hardware that recognizes a user's emotional state in real time and collects that data.

[0616] "JSON format" is an abbreviation for JavaScript Object Notation, and is a format for structuring and expressing data in text format.

[0617] A "generative AI model" is a model with automatic generation capabilities using artificial intelligence, which generates advertising images using neural networks.

[0618] "Advertising images" are visual advertising materials generated by a generative AI model based on product information and emotional data.

[0619] "Advertising effectiveness data" means data collected after an advertising campaign is conducted, including performance indicators such as clicks, impressions, and conversion rates.

[0620] "Feedback" is the process of providing advertising effectiveness data and emotional data to the generative AI model in order to retrain the model and improve its accuracy.

[0621] "API" stands for Application Programming Interface, an interface that enables data exchange between different software systems.

[0622] The present invention relates to a system for automatically generating and optimizing advertising images, in which a user interface, an emotion engine, a server, and a generative AI model work in cooperation with each other. Specific embodiments for carrying out the present invention will be described below.

[0623] User Interface Design

[0624] Terminal: First, the user accesses the system through a terminal. The user interface displays fields for entering the product name, price, description, and image. These fields are designed to be intuitive, allowing the user to easily enter product details.

[0625] Entering product information and collecting sentiment data

[0626] User: When a user enters product information, the emotion engine works in the background to recognize the user's emotional state in real time through the user's camera. For example, if the user is smiling, the system will recognize the emotion as "positive" and collect this data.

[0627] Product information and emotion data sent to server

[0628] Terminal: After the user finishes entering information, the terminal converts the product information and emotion data into JSON format and sends it to the server. The transmitted data includes the product name, price, description, image, and emotion data.

[0629] Generating advertising images

[0630] Server: The server analyzes the received data and passes it to a generative AI model. This generative AI model uses a neural network to automatically generate advertising images based on product information and emotional data. The generated advertising images are stored on the server.

[0631] Providing advertising images

[0632] Server: Obtains the storage path or URL of the generated ad image and returns it to the user's device. The user can view the ad image on their screen.

[0633] Running an advertising campaign

[0634] User: Using the generated ad image, the user sets up an advertising campaign. The ad is then placed on advertising platforms such as Yahoo! and LINE, and information such as target demographic, period, and budget is set.

[0635] Collection and feedback of advertising effectiveness data

[0636] User: After an advertising campaign is run, the user collects advertising effectiveness data such as clicks, impressions, and conversion rates.

[0637] Terminal: Collected advertising effectiveness data and emotion data are compiled and sent back to the server.

[0638] Server: The server feeds this data back into the generative AI model and retrains it, improving the accuracy of subsequent ad image generation.

[0639] Specific examples

[0640] For example, if a user is creating an ad for a new smartphone, they might input the following prompt into the generative AI model:

[0641] Product name: Latest model smartphone

[0642] Price: \50,000

[0643] Product description: Large capacity battery, fast charging compatible

[0644] Product Image: Image URL (e.g. www.example.com / image.jpg)

[0645] In this way, users can quickly and efficiently generate highly effective advertising images. This system is designed to significantly reduce the effort required for creating advertisements and maximize their effectiveness.

[0646] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0647] Step 1:

[0648] User Interface Display

[0649] Device: An application or web browser is launched that displays a screen that provides fields for entering product name, price, description, and image. The screen is designed to make it easy for the user to enter information visually. Once the input fields are visible, this constitutes input.

[0650] Step 2:

[0651] Entering product information and collecting sentiment data

[0652] User: Enters product name, price, description, and image into input fields on the device. This is the input data. Meanwhile, the emotion engine runs in the background, recognizing the user's emotional state (e.g., positive, negative) in real time through the user's camera and collecting emotion data. The emotion engine captures frames from the camera and uses a facial expression analysis algorithm to categorize the user's emotional state.

[0653] Step 3:

[0654] Product information and emotion data sent to server

[0655] Terminal: Once the user has completed the input, the terminal will combine the product name, price, description, uploaded image, and emotion data into a single data object. This is then serialized into JSON format and sent to the server via a secure protocol (e.g., HTTPS). Once the data has been sent to the server, the server receives it.

[0656] Step 4:

[0657] Generating advertising images

[0658] Server: Analyzes the received JSON data and passes the data object to the generative AI model. The generative AI model uses deep learning technology to generate advertising images using product information and emotional data as input. Specifically, a neural network processes the input data and creates advertising images containing appropriate visual elements and text designs. The generated advertising images are stored on the server. The output is the generated advertising image data.

[0659] Step 5:

[0660] Providing advertising images

[0661] Server: The output is to obtain the storage path or URL of the generated ad image and return it to the user's device in JSON format.

[0662] On the device: The received ad image URL is parsed and set as the src attribute of an HTML image element, which displays the image on the screen. This allows the user to check the generated ad image. The emotion engine then runs again, analyzing the user's reactions in real time and collecting new emotion data.

[0663] Step 6:

[0664] Running an advertising campaign

[0665] User: Uses the generated ad image to set up an ad campaign. The user starts the ad campaign by entering information about the target demographic, period, budget, and ad platform (e.g., Yahoo, LINE). This setting information is the input data. The ad campaign is then carried out, and the actual ads are displayed by the ad platform.

[0666] Step 7:

[0667] Collection and transmission of advertising effectiveness data

[0668] User: After running an advertising campaign, you collect advertising effectiveness data such as clicks, impressions, and conversion rates from the advertising platform. This is input data.

[0669] Terminal: The output is to compile the collected advertising effectiveness data and the user feedback data recorded by the emotion engine, serialize it in JSON format, and send it back to the server.

[0670] Step 8:

[0671] Data feedback and AI model retraining

[0672] Server: Analyzes the received advertising effectiveness data and emotion data and feeds it back to the generative AI model. This allows the generative AI model to be retrained using this data. In particular, the generative AI model uses deep learning algorithms such as backpropagation to adjust model parameters and improve the accuracy of subsequent ad image generation. This is the final output.

[0673] (Application example 2)

[0674] 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."

[0675] Conventional ad generation systems generate ad images without considering user emotions, resulting in poor ad effectiveness. Furthermore, the process of feeding back ad effectiveness data and retraining the AI ​​model is cumbersome, making it difficult to improve the accuracy of generating effective ads. This results in problems such as the inability to create effective ads that users desire, and reduced efficiency of advertising campaigns.

[0676] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0677] In this invention, the server includes means for receiving detailed product information from a user, means for inputting the received detailed information and emotion data into a generative AI model to automatically generate advertising images, means for providing the generated advertising images to the user, an emotion engine for collecting user emotional responses in real time, and means for feeding back advertising effectiveness data collected after the advertisement is posted to the generative AI model for learning. This enables automatic generation of advertising images that take user emotions into consideration and re-training of the AI ​​model based on the advertising effectiveness data.

[0678] "Product details" are basic attributes of a product provided by the user, such as its name, price, description, and image.

[0679] "Emotional data" refers to data that measures and collects a user's emotional state in real time.

[0680] A "generative AI model" is an artificial intelligence model that automatically generates advertising images based on input information.

[0681] "Advertising images" are visual promotional materials generated based on detailed product information and emotional data provided by users.

[0682] The "emotion engine" is a system for collecting and analyzing users' emotional responses in real time.

[0683] "Advertising effectiveness data" refers to data on performance indicators such as the number of clicks, impressions, and conversion rates in advertising campaigns.

[0684] "Feedback" is the process of retraining the generative AI model based on collected data to improve the accuracy of ad generation from the next time onwards.

[0685] An "application program interface" is a standardized means for exchanging data between different software systems.

[0686] This invention is a system that automatically generates advertising images based on input of product information, and improves the accuracy of advertisement generation by grasping user emotions in real time and obtaining feedback. This system is implemented on both the front end and the server side.

[0687] front end

[0688] On the front end, users access the interface using their smartphones. Specifically, they input product names, prices, descriptions, and image URLs through an application built with React Native. An emotion engine also detects the user's facial expressions in real time and collects emotional data. This collected data is sent to the server in JSON format.

[0689] Specific examples

[0690] When a user creates an ad for a new smartwatch, they enter the following information:

[0691] Product Name: Latest Smartwatch

[0692] Price: \30,000

[0693] Product description: Heart rate sensor, high-precision GPS

[0694] Product image URL: https: / / example.com / images / smartwatch.jpg

[0695] Server Side

[0696] On the server side, the received data is processed using Node.js and Express. Specifically, the received JSON data is input into a generative AI model to generate ad images. This generative AI model uses TensorFlow.js, which generates ad images using a neural network. The generated ad images are stored on the server, and their URLs are returned to the user's device. In addition, user emotional responses and ad effectiveness data (number of clicks, number of impressions, conversion rate, etc.) are collected in real time and fed back to the server. This data is passed to the generative AI model and used for re-training.

[0697] Specific examples

[0698] The user checks the ad image and runs an advertising campaign. For example, the generated ad image is used to post on an advertising platform, and the AI ​​model is retrained using the advertising effectiveness data obtained later. An example of the prompt sentence in this case is as follows:

[0699] Prompt Sentence Examples

[0700] Based on the information received from the user, the generative AI model generates advertising images as follows:

[0701] Prompt text: Product name "Latest smartwatch", price "30,000 yen", product description "Heart rate sensor, high-precision GPS", image URL "https: / / example.com / images / smartwatch.jpg", emotion data "Happiness level: 80%, Neutrality level: 20%".

[0702] In this way, it is possible to improve the accuracy of advertising images generated based on user emotions and advertising effectiveness data, and to realize efficient advertising campaigns.

[0703] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0704] Step 1:

[0705] A user uses a smartphone to access an application built with React Native.

[0706] Users enter the product name, price, description, and image URL in the corresponding input fields, and the emotion engine collects emotional data from the user's facial expressions, such as happiness, surprise, anger, etc. These details and emotional data are then compiled and converted into JSON format.

[0707] input:

[0708] Product name, price, description, image URL

[0709] Emotional data (happiness, surprise, anger, etc.)

[0710] output:

[0711] Detailed information and emotion data in JSON format

[0712] Step 2:

[0713] The device sends the collected details and emotion data in JSON format to the server.

[0714] This data is sent as an HTTP POST request.

[0715] input:

[0716] Detailed information and emotion data in JSON format

[0717] output:

[0718] HTTP POST request to the server

[0719] Step 3:

[0720] The server analyzes the received JSON-formatted details and emotion data and prepares them for input into the generative AI model.

[0721] Specifically, it formats each input item (product name, price, description, image URL, emotion data).

[0722] input:

[0723] Detailed information and emotion data in JSON format

[0724] output:

[0725] Formatted details and sentiment data

[0726] Step 4:

[0727] The server inputs the formatted details and emotional data into a generative AI model to generate advertising images.

[0728] The generative AI model uses TensorFlow.js and employs a neural network to generate advertising images from input data.

[0729] input:

[0730] Formatted details and sentiment data

[0731] output:

[0732] Generated advertising image

[0733] Step 5:

[0734] The server stores the generated advertisement image therein and returns the URL to the terminal.

[0735] The URL of the returned ad image is displayed in the client-side interface for the user to view.

[0736] input:

[0737] Generated advertising image

[0738] output:

[0739] Ad image URL

[0740] Step 6:

[0741] The user can review the generated advertising images and use them for their advertising campaigns.

[0742] In addition, the emotion engine records the user's emotional reactions in real time and sends them to the server.

[0743] input:

[0744] Advertising images

[0745] output:

[0746] User emotional response data

[0747] Step 7:

[0748] The server feeds back the emotional response data collected from users and advertising effectiveness data (number of clicks, number of impressions, conversion rate, etc.) to the generative AI model.

[0749] The generated AI model is retrained based on this data.

[0750] input:

[0751] Emotional response data

[0752] Advertising effectiveness data

[0753] output:

[0754] Generative AI model with improved accuracy through retraining

[0755] This makes it possible to generate advertising images based on user emotions and advertising effectiveness data, enabling efficient advertising campaign management.

[0756] 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.

[0757] 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.

[0758] 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.

[0759] [Third embodiment]

[0760] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0761] 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.

[0762] 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).

[0763] 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.

[0764] 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.

[0765] 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).

[0766] 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. 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.

[0767] 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.

[0768] 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.

[0769] 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.

[0770] 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.

[0771] 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."

[0772] A specific embodiment for carrying out the present invention will be described below. A user, a terminal, and a server work together based on the following steps.

[0773] User Interface (UI) Design

[0774] User: First, the user accesses the system on a terminal. The UI on the screen displays fields for entering the product name, price, description, and image. The user enters the product details in these fields.

[0775] Sending product information

[0776] Terminal: Collects product information entered by the user and sends it to the server in JSON format, combining the entered product name, price, description, and uploaded image into a single data object.

[0777] Generating advertising images

[0778] Server: Once product information is received, the server passes it to the generative AI model, which uses a neural network to generate advertising images based on the input information. This generation process is automated, and the generated advertising images are stored on the server.

[0779] Sending and displaying advertising images

[0780] Server: Obtain the storage path or URL of the generated ad image and return it to the user's device.

[0781] Terminal: The URL of the ad image received from the server is acquired and displayed on the user's screen. The user can then check the generated ad image and use it to place an ad.

[0782] Advertising and data collection

[0783] User: Uses the generated ad image to post an ad on an advertising platform such as Yahoo! or LINE. At this time, the user collects performance data (number of clicks, number of impressions, conversion rate, etc.) related to the posted ad.

[0784] Data feedback and AI model retraining

[0785] Server: After the ad is published, the server receives the ad effectiveness data provided by the user. The server feeds this data back into the generative AI model and retrains the model. This feedback loop improves the accuracy of the AI ​​model's ad image generation.

[0786] As a concrete example, consider a user creating an advertisement for a new smartphone. First, the user enters the product name "Latest Model Smartphone," the price "50,000 yen," the product description "High-capacity battery, fast charging compatible," and a product image into the system. The device sends this information to the server, and the server-side generative AI model automatically generates the advertisement image. The generated advertisement image is sent back to the user's device, and the user uses it in their own advertising campaign. At a later date, the user sends the advertisement's performance data to the server, and the AI ​​model retrains to further improve accuracy. This series of processes allows users to easily generate and publish high-quality advertisements, even without specialized knowledge.

[0787] The processing flow will be explained below.

[0788] Step 1: The user accesses the system on a terminal and enters product details such as product name, price, description, and image.

[0789] Step 2: The user's device compiles the entered product information into a JavaScript object, converts it to JSON format, and sends an HTTP POST request to the server.

[0790] Step 3: The server receives the HTTP request, parses the product information from the request body, converts the product information into an appropriate format, and passes it to the generative AI model.

[0791] Step 4: The server's AI model generates advertising images based on the product information. The generated advertising images are stored on the server.

[0792] Step 5: The server obtains the storage path or URL of the generated ad image and returns it to the user's device in JSON format.

[0793] Step 6: The user's device retrieves the URL of the ad image received from the server, sets it as the src attribute of the image element, and displays it on the user's screen. The user then checks the generated ad image.

[0794] Step 7: The user uploads the confirmed ad image to an advertising platform such as Yahoo or LINE, sets up an advertising campaign, and publishes it.

[0795] Step 8: During the advertising campaign, the user collects advertising effectiveness data such as clicks, impressions, and conversion rates from the advertising platform.

[0796] Step 9: The user provides the collected advertising effectiveness data to the server using the API provided by the server.

[0797] Step 10: The server stores the received advertising effectiveness data in a database and feeds it back to the generative AI model.

[0798] Step 11: The server's generated AI model is retrained based on the feedback data to improve the accuracy of advertising image generation.

[0799] These steps realize a system in which users, terminals, and servers work together to automatically generate advertising images and utilize advertising effectiveness data to continuously improve accuracy.

[0800] Example 1

[0801] 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."

[0802] Conventional advertising image generation systems require specialized knowledge and lack a means to automate the creation of advertising images and their effectiveness analysis. Furthermore, they lack a feedback mechanism to improve the accuracy of advertising image generation. This makes it difficult for average users to create high-quality advertising images and effectively display advertisements.

[0803] 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.

[0804] In this invention, the server includes: means for a user to input detailed product information using an information processing terminal; means for transmitting the input detailed product information to the server in data format; means for the server to receive the detailed product information and input the information into a generative AI model to automatically generate an advertising image; means for the server to return the generated advertising image to the user's information processing terminal; means for the information processing terminal to display the generated advertising image to the user; means for the user to post an advertisement and collect advertising effectiveness data; and means for the server to receive the collected advertising effectiveness data and feed it back to the generative AI model for learning. This enables users without specialized knowledge to easily generate high-quality advertising images, post advertisements effectively, and evolve the generative AI model based on the effectiveness data.

[0805] "Information processing terminal" refers to a device used by a user to input detailed product information, and specifically includes a personal computer or smartphone.

[0806] "Product details information" refers to information that a user inputs to generate an advertising image, such as the product name, price, description, and image.

[0807] "Data format" refers to the standardized format, such as JSON or XML, used when sending product details to the server.

[0808] "Server" refers to the computer system that receives product details, generates advertising images using a generative AI model, and returns the results to the user.

[0809] A "generative AI model" refers to an algorithm that automatically generates advertising images from input product details using machine learning techniques such as neural networks.

[0810] "Advertising images" refers to visual content that is automatically generated by a generative AI model and is intended for use by users as advertising.

[0811] "Advertising effectiveness data" refers to data such as the number of clicks, impressions, and conversion rates obtained as a result of users placing advertisements.

[0812] "Feedback" refers to the process of re-inputting collected advertising effectiveness data into the generative AI model to improve the model's performance.

[0813] "Auto-generation" refers to the process by which a generative AI model generates advertising images without human intervention.

[0814] A specific embodiment for carrying out the present invention will be described. In this system, users, terminals, and servers work in cooperation with each other. Each element of the system and its role will be specifically described below.

[0815] First, the user accesses the system's web application from an information processing terminal. Typical information processing terminals are personal computers or smartphones. The web application is composed of HTML, CSS, and JavaScript, and displays a form for entering product details. This form includes fields for product name, price, description, and image upload.

[0816] Next, the user enters the product name, price, description, and image into the input fields and clicks the submit button. The terminal converts the entered product details into JSON format and sends it to the server as an HTTP POST request. The technologies used include JavaScript, JSON, and the HTTP protocol.

[0817] The server runs on Node.js and processes the received product details in JSON format. Python is used on the server to pass the information to a generative AI model made up of a neural network. The generative AI model generates advertising images based on the product details and saves the image data in storage on the server. The technologies used include Node.js, Python, and generative AI models (e.g., GAN).

[0818] The generated ad image is sent back from the server to the user's device. Specifically, the URL or path of the generated ad image is sent as an HTTP response. The device receives this and displays the ad image using JavaScript. The user can then view the generated ad image in their browser.

[0819] Users use the generated ad images to place ads on advertising platforms (typically search engines or social media advertising platforms). They then collect advertising effectiveness data, such as the number of clicks, impressions, and conversion rates. At this time, users can upload this advertising effectiveness data to a server. The collected effectiveness data is used as feedback to retrain the generative AI model.

[0820] As a concrete example, consider a scenario in which a user creates an advertisement for a new smartphone. The user inputs the product name "Latest Model Smartphone," the price "50,000 yen," the product description "Large Capacity Battery, Fast Charging Supported," and a product image into the system. The device sends this information to the server, and the server's generative AI model automatically generates the advertisement image. The generated advertisement image is sent back to the user's device, and the user uses it in their own advertising campaign. At a later date, the user sends the advertisement's performance data to the server, and the AI ​​model is retrained to further improve its accuracy.

[0821] Examples of prompts include:

[0822] "Generate an ad for a new smartphone. Product name: Latest model smartphone, Price: 50,000 yen, Description: Large capacity battery, Fast charging supported, Product image: [Image URL]."

[0823] This process allows users to generate high-quality advertising images without specialized knowledge, place ads effectively, and use the results data to improve the generative AI model.

[0824] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0825] Step 1:

[0826] User Interface Display

[0827] User: Accesses the system's web application from an information processing terminal (personal computer or smartphone).

[0828] Input: A URL request made by a web browser.

[0829] Output: An HTML form is displayed to enter product details.

[0830] What happens: The browser makes a server request, the server responds with HTML, CSS, and JavaScript, and the screen displays the product name, price, description, and a field for uploading an image.

[0831] Step 2:

[0832] Enter and submit product details

[0833] User: Enters product name, price, description, and image and clicks submit.

[0834] Input: Product details (product name, price, description, image).

[0835] Output: The data is converted to JSON format and sent to the server.

[0836] What it does: JavaScript takes the input, converts it to JSON format, and sends that data to the server as an HTTP POST request.

[0837] Step 3:

[0838] Receiving and analyzing product details

[0839] Server: The server receives the HTTP request and parses the JSON formatted data.

[0840] Input: Product details in JSON format sent as an HTTP POST request.

[0841] Output: Structured data with extracted product name, price, description, and image path.

[0842] What it does: A Node.js backend receives requests and parses the data, which is then passed to a Python generative AI model.

[0843] Step 4:

[0844] Generating advertising images

[0845] Server: The generative AI model on the server automatically generates advertising images based on product information.

[0846] Input: Structured product details (product name, price, description, image path).

[0847] Output: The generated ad image.

[0848] How it works: A Python script calls a generative AI model (e.g., GAN) and generates advertising images using product information as input. The generated images are then stored in the server's storage.

[0849] Step 5:

[0850] Return and display of generated advertising images

[0851] Server: Obtains the URL or path of the generated ad image and returns it to the user's device as an HTTP response.

[0852] Input: The path to the generated ad image.

[0853] Output: HTTP response containing the image URL or path.

[0854] Specific operation: Node.js obtains the storage path of the generated ad image and returns it to the user's device in JSON format.

[0855] Terminal: Receives the response and displays the ad image.

[0856] Input: The URL of the ad image returned by the server.

[0857] Output: The ad image that is displayed in the user's browser.

[0858] What it does: JavaScript parses the response, gets the image URL and inserts it into the HTML.

[0859] Step 6:

[0860] Advertisement placement and collection of effectiveness data

[0861] User: Posts the generated ad image on an advertising platform and collects performance data.

[0862] Input: Generated ad image and ad campaign configuration information.

[0863] Output: Ad effectiveness data (clicks, impressions, conversion rate, etc.).

[0864] Specific operations: Users log in to the advertising platform and set up campaigns using ad images. After publishing, performance data is collected.

[0865] Step 7:

[0866] Re-learning based on effectiveness data

[0867] User: Uploads advertising effectiveness data to the server.

[0868] Input: Advertising effectiveness data (e.g., CSV format).

[0869] Output: Data uploaded to the server.

[0870] Specific action: The user enters effect data into a web form and presses the submit button.

[0871] Server: Analyzes the received effect data and retrains the generative AI model.

[0872] Input: Uploaded ad effectiveness data.

[0873] Output: A generative AI model with improved accuracy.

[0874] What it does: A Python script uses the effects data to retrain the generative AI model.

[0875] (Application example 1)

[0876] 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."

[0877] Currently, companies and individuals running online stores and e-commerce sites need advanced design skills and expertise to generate and utilize effective advertising images. Without these skills, it is difficult to create high-quality advertising images. Furthermore, there is a lack of mechanisms for effectively utilizing advertising performance data to continuously improve ad generation AI models. For these reasons, there is a need for a system that can automatically and efficiently generate ads and optimize their effectiveness.

[0878] 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.

[0879] In this invention, the server includes means for receiving detailed product information from a user, means for inputting the received detailed information into a generation AI model to automatically generate advertising images, means for providing the generated advertising images to the user, means for feeding back advertising effectiveness data collected after the advertisement is posted to the generation AI model for learning, means for using a database for saving the advertising effectiveness data, means for re-learning the AI ​​model, and means for using a generation engine for improving the quality of the advertising images based on the collected advertising effectiveness data. This makes it possible to generate high-quality advertising images and optimize advertising effectiveness without any special expertise.

[0880] The "means for accepting detailed product information from the user" is a function for providing an interface that allows the user to input information about the product, such as the product name, price, description, and image, and inputting that information into the system.

[0881] "Means for automatically generating advertising images by inputting received detailed information into a generative AI model" refers to a function that sends product information entered by the user to a server, and the generative AI model automatically generates advertising images based on that information.

[0882] "Means for providing generated advertising images to users" refers to a function for providing advertising images created by a generative AI model to users' devices so that users can view and use them.

[0883] "Means of feeding back advertising effectiveness data collected after an advertisement is posted to the AI ​​generation model to allow it to learn" refers to a function that feeds back advertising effectiveness data collected after a user posts an advertisement, such as the number of clicks, number of impressions, and conversion rate, to the AI ​​generation model and uses it as learning data for the model.

[0884] The "means for using a database for storing advertising effectiveness data" is a function for storing collected advertising effectiveness data in a database within the system so that it can be used for later analysis and model re-learning.

[0885] "Means for retraining the AI ​​model" refers to a function that retrains the generation AI model based on advertising effectiveness data to improve the accuracy of advertising image generation.

[0886] "Means for using a generation engine to improve the quality of advertising images based on collected advertising effectiveness data" refers to a function for generating higher quality advertising images by analyzing collected advertising effectiveness data and updating and improving the generation AI model based on the results.

[0887] A specific embodiment for carrying out the present invention will be described below. A user, a terminal, and a server work together based on the following steps.

[0888] User Interface (UI) Design

[0889] The server provides a user interface for users to enter product information. The UI displays fields for entering the product name, price, description, and image. The user enters product details in these fields. For example, the user enters the product name as "Latest Model Smartphone," the price as "50,000 yen," the product description as "Large capacity battery, fast charging compatible," and uploads a product image.

[0890] Sending product information

[0891] The terminal collects the product information entered by the user and sends it to the server in JSON format, where the entered product name, price, description, and uploaded image are compiled into a single data object.

[0892] Generating advertising images

[0893] The server passes the received product information to a generative AI model, which uses a neural network to generate advertising images based on the input information. This generation process is automatic, and the generated advertising images are stored on the server.

[0894] Sending and displaying advertising images

[0895] The server obtains the storage path or URL of the generated ad image and returns it to the user's device. The device then obtains the URL of the ad image received from the server and displays it on the user's screen. The user can then check the generated ad image and use it to place an ad on the advertising platform.

[0896] Advertising and data collection

[0897] The user uses the generated advertising image to post an advertisement on the advertising platform, and in this case, the user collects performance data (number of clicks, number of impressions, conversion rate, etc.) related to the posted advertisement.

[0898] Data feedback and AI model retraining

[0899] The server receives advertising effectiveness data provided by users after the advertisement is posted. The server feeds this data back into the generative AI model and retrains the AI ​​model. This feedback loop improves the accuracy of the generative AI model in generating advertising images.

[0900] Hardware and Software

[0901] To realize this system, the following hardware and software are used:

[0902] Server: Receives product information, generates advertising images using a generative AI model, and sends them back to the user's device. The generative AI model is installed on the server, and the advertising images are generated using that model.

[0903] Terminal: The user enters product information and sends it to the server. The server returns an advertisement image and displays it.

[0904] Generative AI model: A neural network model that creates advertising images based on product information. It is expected to use OpenAI APIs, etc.

[0905] Database: Stores advertising effectiveness data and uses it to retrain AI models.

[0906] Examples of specific examples and prompts

[0907] For example, if a user were to create an ad for a new smartphone, they would enter the following information:

[0908] Product name: Latest model smartphone

[0909] Price: \50,000

[0910] Description: Large capacity battery, fast charging

[0911] Image: base64 encoded image data

[0912] Example prompt for a generative AI model:

[0913] Please generate an ad image based on the following product information:

[0914] Product name: Latest model smartphone

[0915] Price: \50,000

[0916] Description: Large capacity battery, fast charging

[0917] Image: base64 encoded image data

[0918] In this way, a system can be constructed that can easily generate high-quality advertising images without requiring specialized knowledge and can optimize the effectiveness of advertising.

[0919] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0920] Step 1:

[0921] The user enters product information using a device. The device's user interface displays fields for entering the product name, price, description, and image. The user enters the required information in these fields and uploads an image. The entered information is temporarily stored on the device.

[0922] Input: User input of product name, price, description, and image

[0923] Output: Product information saved on the device

[0924] Step 2:

[0925] The device collects product information in JSON format and sends it to the server. The product information is formatted as a single data object. The server receives this and saves it as product information.

[0926] Input: Product information saved on the device

[0927] Output: JSON formatted data object, product information sent to the server

[0928] Step 3:

[0929] The server passes the received product information to the generative AI model, which generates advertising images based on input information such as product name, price, description, and image. The generative AI model uses a neural network to automatically create advertising images, using the input information as prompts during this process.

[0930] Input: A JSON formatted data object

[0931] Output: Advertising image generated by the generative AI model

[0932] Step 4:

[0933] The server obtains the storage path or URL of the generated advertisement image and returns it to the terminal. The terminal obtains the URL of the received advertisement image and displays it on the user's screen.

[0934] Input: Generated ad image

[0935] Output: Ad image storage path or URL, ad image displayed on the user's screen

[0936] Step 5:

[0937] Users use the generated ad images to place ads on the advertising platform, and performance data on the posted ads, such as the number of clicks, impressions, and conversion rates, is collected and later sent to the server.

[0938] Input: Ad image, Ad placement on advertising platform

[0939] Output: Collected advertising effectiveness data

[0940] Step 6:

[0941] The server receives advertising effectiveness data provided by users and feeds it back to the generative AI model. This data is stored in a database and used to retrain the AI ​​model, allowing the generative AI model to improve the accuracy of generating advertising images.

[0942] Input: Advertising effectiveness data

[0943] Output: Advertising effectiveness data stored in a database, and a generative AI model that receives feedback.

[0944] Step 7:

[0945] The server periodically retrains the AI ​​model based on collected advertising effectiveness data, improving the quality of the generated advertising images and helping with future ad generation.

[0946] Input: Advertising effectiveness data stored in the database

[0947] Output: The generative AI model updated through retraining

[0948] These steps realize a system that allows users to easily generate high-quality advertising images without specialized knowledge and optimize the effectiveness of advertising.

[0949] 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.

[0950] A specific embodiment for carrying out the present invention will be described below. The user, terminal, server, and emotion engine work together based on the following steps.

[0951] User Interface (UI) Design

[0952] User: First, the user accesses the system on their device. The UI on the screen displays fields for entering the product name, price, description, and image. The user enters product details in these fields. At this time, an emotion engine that recognizes the user's emotional state in real time is activated, and the user's emotional data is collected.

[0953] Sending product information

[0954] Terminal: Collects product information entered by the user and sends it to the server in JSON format. The entered product name, price, description, and uploaded image are compiled into a single data object. In addition, the user's emotion data collected by the emotion engine is also sent.

[0955] Generating advertising images

[0956] Server: After receiving the product information and emotion data, the server passes it to the generative AI model. The generative AI model uses a neural network to generate advertising images based on the input information. This generation process is automated, and the generated advertising images are stored on the server.

[0957] Sending and displaying advertising images

[0958] Server: Obtain the storage path or URL of the generated ad image and return it to the user's device.

[0959] Device: The URL of the ad image received from the server is obtained, set as the src attribute of the image element, and displayed on the user's screen. The user checks the generated ad image. At this time, the emotion engine recognizes the user's reaction in real time and collects emotional feedback data.

[0960] Advertising and data collection

[0961] User: Using the generated ad image, the user sets up an ad campaign and places ads on advertising platforms such as Yahoo! and LINE. At this time, the user collects performance data (number of clicks, number of impressions, conversion rate, etc.) related to the ads they place.

[0962] Data feedback and AI model retraining

[0963] Server: After the ad is displayed, the server receives the ad effectiveness data provided by the user and the emotion data recognized by the emotion engine. The server feeds this data back into the generative AI model and retrains the model. This feedback loop improves the accuracy of the AI ​​model's ad image generation.

[0964] As a concrete example, consider a user creating an advertisement for a new smartphone. When the user enters the product name "Latest Model Smartphone," the price "50,000 yen," the product description "High-capacity battery, fast charging compatible," and a product image into the system, the emotion engine monitors the user's emotional state in real time and collects data. The device sends this information to the server, and the server-side generative AI model automatically generates an advertisement image. The generated advertisement image is sent back to the user's device, where the user reviews it, and the emotion engine again analyzes the user's reaction. The user runs an advertising campaign using the generated advertisement image, and then provides the collected advertising effectiveness data and emotion data to the server. The generative AI model re-trains using the information obtained in this way, improving the accuracy of future advertisement image generation. This system enables more efficient and accurate advertisement creation, providing greater convenience to users.

[0965] The processing flow will be explained below.

[0966] Step 1: The user accesses the system on their device and inputs product details such as product name, price, description, and image. During input, the emotion engine recognizes the user's emotional state in real time and collects emotion data.

[0967] Step 2: The user's device compiles the entered product information and the emotion data collected by the emotion engine into a JavaScript object, converts it into JSON format, and sends an HTTP POST request to the server.

[0968] Step 3: The server receives the HTTP request, parses the product information and sentiment data from the request body, converts the product information into an appropriate format, and passes it to the generative AI model.

[0969] Step 4: The generative AI model generates advertising images based on product information and emotion data. This generation process is carried out using a neural network. The generated advertising images are stored on the server.

[0970] Step 5: The server obtains the storage path or URL of the generated ad image and returns it to the user's device in JSON format.

[0971] Step 6: The user's device retrieves the URL of the ad image received from the server, sets it as the src attribute of the image element, and displays it on the user's screen. The user checks the generated ad image. At this time, the emotion engine recognizes the user's reaction in real time and collects emotional feedback data.

[0972] Step 7: The user uploads the confirmed ad image to an advertising platform such as Yahoo or LINE, sets up an advertising campaign, and publishes it.

[0973] Step 8: During the advertising campaign, the user collects advertising effectiveness data such as clicks, impressions, and conversion rates from the advertising platform.

[0974] Step 9: The user provides the collected advertising effectiveness data to the server using the API provided by the server.

[0975] Step 10: The server stores the received advertising effectiveness data and the emotion data recognized by the emotion engine in a database and feeds it back to the generative AI model.

[0976] Step 11: The generative AI model is retrained based on the feedback data to improve the accuracy of advertising image generation.

[0977] These steps realize a system in which the user, terminal, server, and emotion engine work together to automatically generate advertising images and continuously improve accuracy by utilizing advertising effectiveness data and emotion data.

[0978] Example 2

[0979] 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."

[0980] In today's online advertising market, the generation and optimization of effective ad images is crucial, but the process of users manually creating and optimizing ad images is time-consuming and labor-intensive. Furthermore, while generating ad images that take users' emotional states into account would improve the user experience, existing technologies have not automated this process. Furthermore, there are insufficient mechanisms for effectively collecting post-campaign effectiveness data and incorporating it into re-learning. To address these challenges, it is necessary to automate ad image generation, collect real-time user emotional data, generate ad images that reflect that data, and establish a feedback loop for ad effectiveness data.

[0981] 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.

[0982] In this invention, the server includes a means for providing a user interface for users to input product names, prices, descriptions, and images; a means for collecting the input product information and real-time emotion data; a means for converting the collected data into JSON format and transmitting it to the server; a means for the server to input the received data into a generative AI model and automatically generate advertising images; a means for providing the generated advertising images to users; a means for collecting advertising effectiveness data after the advertising campaign is implemented; a means for feeding back the collected advertising effectiveness data and emotion data to the generative AI model for learning; and a means for the terminal to recognize the user's emotional state in real time and collect emotion data. This enables the automation and optimization of advertising image generation, enabling advertising image generation that takes user emotion data into account. Furthermore, the feedback loop of advertising effectiveness data improves the accuracy of the generative AI model, enabling more effective advertising campaign design.

[0983] A "user interface" is a system component that includes an input screen for a user to input product names, prices, descriptions, and images.

[0984] "Emotion engine" refers to a piece of software or hardware that recognizes a user's emotional state in real time and collects that data.

[0985] "JSON format" is an abbreviation for JavaScript Object Notation, and is a format for structuring and expressing data in text format.

[0986] A "generative AI model" is a model with automatic generation capabilities using artificial intelligence, which generates advertising images using neural networks.

[0987] "Advertising images" are visual advertising materials generated by a generative AI model based on product information and emotional data.

[0988] "Advertising effectiveness data" means data collected after an advertising campaign is conducted, including performance indicators such as clicks, impressions, and conversion rates.

[0989] "Feedback" is the process of providing advertising effectiveness data and emotional data to the generative AI model in order to retrain the model and improve its accuracy.

[0990] "API" stands for Application Programming Interface, an interface that enables data exchange between different software systems.

[0991] The present invention relates to a system for automatically generating and optimizing advertising images, in which a user interface, an emotion engine, a server, and a generative AI model work in cooperation with each other. Specific embodiments for carrying out the present invention will be described below.

[0992] User Interface Design

[0993] Terminal: First, the user accesses the system through a terminal. The user interface displays fields for entering the product name, price, description, and image. These fields are designed to be intuitive, allowing the user to easily enter product details.

[0994] Entering product information and collecting sentiment data

[0995] User: When a user enters product information, the emotion engine works in the background to recognize the user's emotional state in real time through the user's camera. For example, if the user is smiling, the system will recognize the emotion as "positive" and collect this data.

[0996] Product information and emotion data sent to server

[0997] Terminal: After the user finishes entering information, the terminal converts the product information and emotion data into JSON format and sends it to the server. The transmitted data includes the product name, price, description, image, and emotion data.

[0998] Generating advertising images

[0999] Server: The server analyzes the received data and passes it to a generative AI model. This generative AI model uses a neural network to automatically generate advertising images based on product information and emotional data. The generated advertising images are stored on the server.

[1000] Providing advertising images

[1001] Server: Obtains the storage path or URL of the generated ad image and returns it to the user's device. The user can view the ad image on their screen.

[1002] Running an advertising campaign

[1003] User: Using the generated ad image, the user sets up an advertising campaign. The ad is then placed on advertising platforms such as Yahoo! and LINE, and information such as target demographic, period, and budget is set.

[1004] Collection and feedback of advertising effectiveness data

[1005] User: After an advertising campaign is run, the user collects advertising effectiveness data such as clicks, impressions, and conversion rates.

[1006] Terminal: Collected advertising effectiveness data and emotion data are compiled and sent back to the server.

[1007] Server: The server feeds this data back into the generative AI model and retrains it, improving the accuracy of subsequent ad image generation.

[1008] Specific examples

[1009] For example, if a user is creating an ad for a new smartphone, they might input the following prompt into the generative AI model:

[1010] Product name: Latest model smartphone

[1011] Price: \50,000

[1012] Product description: Large capacity battery, fast charging compatible

[1013] Product Image: Image URL (e.g. www.example.com / image.jpg)

[1014] In this way, users can quickly and efficiently generate highly effective advertising images. This system is designed to significantly reduce the effort required for creating advertisements and maximize their effectiveness.

[1015] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1016] Step 1:

[1017] User Interface Display

[1018] Device: An application or web browser is launched that displays a screen that provides fields for entering product name, price, description, and image. The screen is designed to make it easy for the user to enter information visually. Once the input fields are visible, this constitutes input.

[1019] Step 2:

[1020] Entering product information and collecting sentiment data

[1021] User: Enters product name, price, description, and image into input fields on the device. This is the input data. Meanwhile, the emotion engine runs in the background, recognizing the user's emotional state (e.g., positive, negative) in real time through the user's camera and collecting emotion data. The emotion engine captures frames from the camera and uses a facial expression analysis algorithm to categorize the user's emotional state.

[1022] Step 3:

[1023] Product information and emotion data sent to server

[1024] Terminal: Once the user has completed the input, the terminal will combine the product name, price, description, uploaded image, and emotion data into a single data object. This is then serialized into JSON format and sent to the server via a secure protocol (e.g., HTTPS). Once the data has been sent to the server, the server receives it.

[1025] Step 4:

[1026] Generating advertising images

[1027] Server: Analyzes the received JSON data and passes the data object to the generative AI model. The generative AI model uses deep learning technology to generate advertising images using product information and emotional data as input. Specifically, a neural network processes the input data and creates advertising images containing appropriate visual elements and text designs. The generated advertising images are stored on the server. The output is the generated advertising image data.

[1028] Step 5:

[1029] Providing advertising images

[1030] Server: The output is to obtain the storage path or URL of the generated ad image and return it to the user's device in JSON format.

[1031] On the device: The received ad image URL is parsed and set as the src attribute of an HTML image element, which displays the image on the screen. This allows the user to check the generated ad image. The emotion engine then runs again, analyzing the user's reactions in real time and collecting new emotion data.

[1032] Step 6:

[1033] Running an advertising campaign

[1034] User: Uses the generated ad image to set up an ad campaign. The user starts the ad campaign by entering information about the target demographic, period, budget, and ad platform (e.g., Yahoo, LINE). This setting information is the input data. The ad campaign is then carried out, and the actual ads are displayed by the ad platform.

[1035] Step 7:

[1036] Collection and transmission of advertising effectiveness data

[1037] User: After running an advertising campaign, you collect advertising effectiveness data such as clicks, impressions, and conversion rates from the advertising platform. This is input data.

[1038] Terminal: The output is to compile the collected advertising effectiveness data and the user feedback data recorded by the emotion engine, serialize it in JSON format, and send it back to the server.

[1039] Step 8:

[1040] Data feedback and AI model retraining

[1041] Server: Analyzes the received advertising effectiveness data and emotion data and feeds it back to the generative AI model. This allows the generative AI model to be retrained using this data. In particular, the generative AI model uses deep learning algorithms such as backpropagation to adjust model parameters and improve the accuracy of subsequent ad image generation. This is the final output.

[1042] (Application example 2)

[1043] 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."

[1044] Conventional ad generation systems generate ad images without considering user emotions, resulting in poor ad effectiveness. Furthermore, the process of feeding back ad effectiveness data and retraining the AI ​​model is cumbersome, making it difficult to improve the accuracy of generating effective ads. This results in problems such as the inability to create effective ads that users desire, and reduced efficiency of advertising campaigns.

[1045] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1046] In this invention, the server includes means for receiving detailed product information from a user, means for inputting the received detailed information and emotion data into a generative AI model to automatically generate advertising images, means for providing the generated advertising images to the user, an emotion engine for collecting user emotional responses in real time, and means for feeding back advertising effectiveness data collected after the advertisement is posted to the generative AI model for learning. This enables automatic generation of advertising images that take user emotions into consideration and re-training of the AI ​​model based on the advertising effectiveness data.

[1047] "Product details" are basic attributes of a product provided by the user, such as its name, price, description, and image.

[1048] "Emotional data" refers to data that measures and collects a user's emotional state in real time.

[1049] A "generative AI model" is an artificial intelligence model that automatically generates advertising images based on input information.

[1050] "Advertising images" are visual promotional materials generated based on detailed product information and emotional data provided by users.

[1051] The "emotion engine" is a system for collecting and analyzing users' emotional responses in real time.

[1052] "Advertising effectiveness data" refers to data on performance indicators such as the number of clicks, impressions, and conversion rates in advertising campaigns.

[1053] "Feedback" is the process of retraining the generative AI model based on collected data to improve the accuracy of ad generation from the next time onwards.

[1054] An "application program interface" is a standardized means for exchanging data between different software systems.

[1055] This invention is a system that automatically generates advertising images based on input of product information, and improves the accuracy of advertisement generation by grasping user emotions in real time and obtaining feedback. This system is implemented on both the front end and the server side.

[1056] front end

[1057] On the front end, users access the interface using their smartphones. Specifically, they input product names, prices, descriptions, and image URLs through an application built with React Native. An emotion engine also detects the user's facial expressions in real time and collects emotional data. This collected data is sent to the server in JSON format.

[1058] Specific examples

[1059] When a user creates an ad for a new smartwatch, they enter the following information:

[1060] Product Name: Latest Smartwatch

[1061] Price: \30,000

[1062] Product description: Heart rate sensor, high-precision GPS

[1063] Product image URL: https: / / example.com / images / smartwatch.jpg

[1064] Server Side

[1065] On the server side, the received data is processed using Node.js and Express. Specifically, the received JSON data is input into a generative AI model to generate ad images. This generative AI model uses TensorFlow.js, which generates ad images using a neural network. The generated ad images are stored on the server, and their URLs are returned to the user's device. In addition, user emotional responses and ad effectiveness data (number of clicks, number of impressions, conversion rate, etc.) are collected in real time and fed back to the server. This data is passed to the generative AI model and used for re-training.

[1066] Specific examples

[1067] The user checks the ad image and runs an advertising campaign. For example, the generated ad image is used to post on an advertising platform, and the AI ​​model is retrained using the advertising effectiveness data obtained later. An example of the prompt sentence in this case is as follows:

[1068] Prompt Sentence Examples

[1069] Based on the information received from the user, the generative AI model generates advertising images as follows:

[1070] Prompt text: Product name "Latest smartwatch", price "30,000 yen", product description "Heart rate sensor, high-precision GPS", image URL "https: / / example.com / images / smartwatch.jpg", emotion data "Happiness level: 80%, Neutrality level: 20%".

[1071] In this way, it is possible to improve the accuracy of advertising images generated based on user emotions and advertising effectiveness data, and to realize efficient advertising campaigns.

[1072] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1073] Step 1:

[1074] A user uses a smartphone to access an application built with React Native.

[1075] Users enter the product name, price, description, and image URL in the corresponding input fields, and the emotion engine collects emotional data from the user's facial expressions, such as happiness, surprise, anger, etc. These details and emotional data are then compiled and converted into JSON format.

[1076] input:

[1077] Product name, price, description, image URL

[1078] Emotional data (happiness, surprise, anger, etc.)

[1079] output:

[1080] Detailed information and emotion data in JSON format

[1081] Step 2:

[1082] The device sends the collected details and emotion data in JSON format to the server.

[1083] This data is sent as an HTTP POST request.

[1084] input:

[1085] Detailed information and emotion data in JSON format

[1086] output:

[1087] HTTP POST request to the server

[1088] Step 3:

[1089] The server analyzes the received JSON-formatted details and emotion data and prepares them for input into the generative AI model.

[1090] Specifically, it formats each input item (product name, price, description, image URL, emotion data).

[1091] input:

[1092] Detailed information and emotion data in JSON format

[1093] output:

[1094] Formatted details and sentiment data

[1095] Step 4:

[1096] The server inputs the formatted details and emotional data into a generative AI model to generate advertising images.

[1097] The generative AI model uses TensorFlow.js and employs a neural network to generate advertising images from input data.

[1098] input:

[1099] Formatted details and sentiment data

[1100] output:

[1101] Generated advertising image

[1102] Step 5:

[1103] The server stores the generated advertisement image therein and returns the URL to the terminal.

[1104] The URL of the returned ad image is displayed in the client-side interface for the user to view.

[1105] input:

[1106] Generated advertising image

[1107] output:

[1108] Ad image URL

[1109] Step 6:

[1110] The user can review the generated advertising images and use them for their advertising campaigns.

[1111] In addition, the emotion engine records the user's emotional reactions in real time and sends them to the server.

[1112] input:

[1113] Advertising images

[1114] output:

[1115] User emotional response data

[1116] Step 7:

[1117] The server feeds back the emotional response data collected from users and advertising effectiveness data (number of clicks, number of impressions, conversion rate, etc.) to the generative AI model.

[1118] The generated AI model is retrained based on this data.

[1119] input:

[1120] Emotional response data

[1121] Advertising effectiveness data

[1122] output:

[1123] Generative AI model with improved accuracy through retraining

[1124] This makes it possible to generate advertising images based on user emotions and advertising effectiveness data, enabling efficient advertising campaign management.

[1125] 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.

[1126] 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.

[1127] 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.

[1128] [Fourth embodiment]

[1129] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1130] 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.

[1131] 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).

[1132] 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.

[1133] 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.

[1134] 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).

[1135] 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. 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.

[1136] 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.

[1137] 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.

[1138] 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.

[1139] 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.

[1140] 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.

[1141] 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."

[1142] A specific embodiment for carrying out the present invention will be described below. A user, a terminal, and a server work together based on the following steps.

[1143] User Interface (UI) Design

[1144] User: First, the user accesses the system on a terminal. The UI on the screen displays fields for entering the product name, price, description, and image. The user enters the product details in these fields.

[1145] Sending product information

[1146] Terminal: Collects product information entered by the user and sends it to the server in JSON format, combining the entered product name, price, description, and uploaded image into a single data object.

[1147] Generating advertising images

[1148] Server: Once product information is received, the server passes it to the generative AI model, which uses a neural network to generate advertising images based on the input information. This generation process is automated, and the generated advertising images are stored on the server.

[1149] Sending and displaying advertising images

[1150] Server: Obtain the storage path or URL of the generated ad image and return it to the user's device.

[1151] Terminal: The URL of the ad image received from the server is acquired and displayed on the user's screen. The user can then check the generated ad image and use it to place an ad.

[1152] Advertising and data collection

[1153] User: Uses the generated ad image to post an ad on an advertising platform such as Yahoo! or LINE. At this time, the user collects performance data (number of clicks, number of impressions, conversion rate, etc.) related to the posted ad.

[1154] Data feedback and AI model retraining

[1155] Server: After the ad is published, the server receives the ad effectiveness data provided by the user. The server feeds this data back into the generative AI model and retrains the model. This feedback loop improves the accuracy of the AI ​​model's ad image generation.

[1156] As a concrete example, consider a user creating an advertisement for a new smartphone. First, the user enters the product name "Latest Model Smartphone," the price "50,000 yen," the product description "High-capacity battery, fast charging compatible," and a product image into the system. The device sends this information to the server, and the server-side generative AI model automatically generates the advertisement image. The generated advertisement image is sent back to the user's device, and the user uses it in their own advertising campaign. At a later date, the user sends the advertisement's performance data to the server, and the AI ​​model retrains to further improve accuracy. This series of processes allows users to easily generate and publish high-quality advertisements, even without specialized knowledge.

[1157] The processing flow will be explained below.

[1158] Step 1: The user accesses the system on a terminal and enters product details such as product name, price, description, and image.

[1159] Step 2: The user's device compiles the entered product information into a JavaScript object, converts it to JSON format, and sends an HTTP POST request to the server.

[1160] Step 3: The server receives the HTTP request, parses the product information from the request body, converts the product information into an appropriate format, and passes it to the generative AI model.

[1161] Step 4: The server's AI model generates advertising images based on the product information. The generated advertising images are stored on the server.

[1162] Step 5: The server obtains the storage path or URL of the generated ad image and returns it to the user's device in JSON format.

[1163] Step 6: The user's device retrieves the URL of the ad image received from the server, sets it as the src attribute of the image element, and displays it on the user's screen. The user then checks the generated ad image.

[1164] Step 7: The user uploads the confirmed ad image to an advertising platform such as Yahoo or LINE, sets up an advertising campaign, and publishes it.

[1165] Step 8: During the advertising campaign, the user collects advertising effectiveness data such as clicks, impressions, and conversion rates from the advertising platform.

[1166] Step 9: The user provides the collected advertising effectiveness data to the server using the API provided by the server.

[1167] Step 10: The server stores the received advertising effectiveness data in a database and feeds it back to the generative AI model.

[1168] Step 11: The server's generated AI model is retrained based on the feedback data to improve the accuracy of advertising image generation.

[1169] These steps realize a system in which users, terminals, and servers work together to automatically generate advertising images and utilize advertising effectiveness data to continuously improve accuracy.

[1170] Example 1

[1171] 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."

[1172] Conventional advertising image generation systems require specialized knowledge and lack a means to automate the creation of advertising images and their effectiveness analysis. Furthermore, they lack a feedback mechanism to improve the accuracy of advertising image generation. This makes it difficult for average users to create high-quality advertising images and effectively display advertisements.

[1173] 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.

[1174] In this invention, the server includes: means for a user to input detailed product information using an information processing terminal; means for transmitting the input detailed product information to the server in data format; means for the server to receive the detailed product information and input the information into a generative AI model to automatically generate an advertising image; means for the server to return the generated advertising image to the user's information processing terminal; means for the information processing terminal to display the generated advertising image to the user; means for the user to post an advertisement and collect advertising effectiveness data; and means for the server to receive the collected advertising effectiveness data and feed it back to the generative AI model for learning. This enables users without specialized knowledge to easily generate high-quality advertising images, post advertisements effectively, and evolve the generative AI model based on the effectiveness data.

[1175] "Information processing terminal" refers to a device used by a user to input detailed product information, and specifically includes a personal computer or smartphone.

[1176] "Product details information" refers to information that a user inputs to generate an advertising image, such as the product name, price, description, and image.

[1177] "Data format" refers to the standardized format, such as JSON or XML, used when sending product details to the server.

[1178] "Server" refers to the computer system that receives product details, generates advertising images using a generative AI model, and returns the results to the user.

[1179] A "generative AI model" refers to an algorithm that automatically generates advertising images from input product details using machine learning techniques such as neural networks.

[1180] "Advertising images" refers to visual content that is automatically generated by a generative AI model and is intended for use by users as advertising.

[1181] "Advertising effectiveness data" refers to data such as the number of clicks, impressions, and conversion rates obtained as a result of users placing advertisements.

[1182] "Feedback" refers to the process of re-inputting collected advertising effectiveness data into the generative AI model to improve the model's performance.

[1183] "Auto-generation" refers to the process by which a generative AI model generates advertising images without human intervention.

[1184] A specific embodiment for carrying out the present invention will be described. In this system, users, terminals, and servers work in cooperation with each other. Each element of the system and its role will be specifically described below.

[1185] First, the user accesses the system's web application from an information processing terminal. Typical information processing terminals are personal computers or smartphones. The web application is composed of HTML, CSS, and JavaScript, and displays a form for entering product details. This form includes fields for product name, price, description, and image upload.

[1186] Next, the user enters the product name, price, description, and image into the input fields and clicks the submit button. The terminal converts the entered product details into JSON format and sends it to the server as an HTTP POST request. The technologies used include JavaScript, JSON, and the HTTP protocol.

[1187] The server runs on Node.js and processes the received product details in JSON format. Python is used on the server to pass the information to a generative AI model made up of a neural network. The generative AI model generates advertising images based on the product details and saves the image data in storage on the server. The technologies used include Node.js, Python, and generative AI models (e.g., GAN).

[1188] The generated ad image is sent back from the server to the user's device. Specifically, the URL or path of the generated ad image is sent as an HTTP response. The device receives this and displays the ad image using JavaScript. The user can then view the generated ad image in their browser.

[1189] Users use the generated ad images to place ads on advertising platforms (typically search engines or social media advertising platforms). They then collect advertising effectiveness data, such as the number of clicks, impressions, and conversion rates. At this time, users can upload this advertising effectiveness data to a server. The collected effectiveness data is used as feedback to retrain the generative AI model.

[1190] As a concrete example, consider a scenario in which a user creates an advertisement for a new smartphone. The user inputs the product name "Latest Model Smartphone," the price "50,000 yen," the product description "Large Capacity Battery, Fast Charging Supported," and a product image into the system. The device sends this information to the server, and the server's generative AI model automatically generates the advertisement image. The generated advertisement image is sent back to the user's device, and the user uses it in their own advertising campaign. At a later date, the user sends the advertisement's performance data to the server, and the AI ​​model is retrained to further improve its accuracy.

[1191] Examples of prompts include:

[1192] "Generate an ad for a new smartphone. Product name: Latest model smartphone, Price: 50,000 yen, Description: Large capacity battery, Fast charging supported, Product image: [Image URL]."

[1193] This process allows users to generate high-quality advertising images without specialized knowledge, place ads effectively, and use the results data to improve the generative AI model.

[1194] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1195] Step 1:

[1196] User Interface Display

[1197] User: Accesses the system's web application from an information processing terminal (personal computer or smartphone).

[1198] Input: A URL request made by a web browser.

[1199] Output: An HTML form is displayed to enter product details.

[1200] What happens: The browser makes a server request, the server responds with HTML, CSS, and JavaScript, and the screen displays the product name, price, description, and a field for uploading an image.

[1201] Step 2:

[1202] Enter and submit product details

[1203] User: Enters product name, price, description, and image and clicks submit.

[1204] Input: Product details (product name, price, description, image).

[1205] Output: The data is converted to JSON format and sent to the server.

[1206] What it does: JavaScript takes the input, converts it to JSON format, and sends that data to the server as an HTTP POST request.

[1207] Step 3:

[1208] Receiving and analyzing product details

[1209] Server: The server receives the HTTP request and parses the JSON formatted data.

[1210] Input: Product details in JSON format sent as an HTTP POST request.

[1211] Output: Structured data with extracted product name, price, description, and image path.

[1212] What it does: A Node.js backend receives requests and parses the data, which is then passed to a Python generative AI model.

[1213] Step 4:

[1214] Generating advertising images

[1215] Server: The generative AI model on the server automatically generates advertising images based on product information.

[1216] Input: Structured product details (product name, price, description, image path).

[1217] Output: The generated ad image.

[1218] How it works: A Python script calls a generative AI model (e.g., GAN) and generates advertising images using product information as input. The generated images are then stored in the server's storage.

[1219] Step 5:

[1220] Return and display of generated advertising images

[1221] Server: Obtains the URL or path of the generated ad image and returns it to the user's device as an HTTP response.

[1222] Input: The path to the generated ad image.

[1223] Output: HTTP response containing the image URL or path.

[1224] Specific operation: Node.js obtains the storage path of the generated ad image and returns it to the user's device in JSON format.

[1225] Terminal: Receives the response and displays the ad image.

[1226] Input: The URL of the ad image returned by the server.

[1227] Output: The ad image that is displayed in the user's browser.

[1228] What it does: JavaScript parses the response, gets the image URL and inserts it into the HTML.

[1229] Step 6:

[1230] Advertisement placement and collection of effectiveness data

[1231] User: Posts the generated ad image on an advertising platform and collects performance data.

[1232] Input: Generated ad image and ad campaign configuration information.

[1233] Output: Ad effectiveness data (clicks, impressions, conversion rate, etc.).

[1234] Specific operations: Users log in to the advertising platform and set up campaigns using ad images. After publishing, performance data is collected.

[1235] Step 7:

[1236] Re-learning based on effectiveness data

[1237] User: Uploads advertising effectiveness data to the server.

[1238] Input: Advertising effectiveness data (e.g., CSV format).

[1239] Output: Data uploaded to the server.

[1240] Specific action: The user enters effect data into a web form and presses the submit button.

[1241] Server: Analyzes the received effect data and retrains the generative AI model.

[1242] Input: Uploaded ad effectiveness data.

[1243] Output: A generative AI model with improved accuracy.

[1244] What it does: A Python script uses the effects data to retrain the generative AI model.

[1245] (Application example 1)

[1246] 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."

[1247] Currently, companies and individuals running online stores and e-commerce sites need advanced design skills and expertise to generate and utilize effective advertising images. Without these skills, it is difficult to create high-quality advertising images. Furthermore, there is a lack of mechanisms for effectively utilizing advertising performance data to continuously improve ad generation AI models. For these reasons, there is a need for a system that can automatically and efficiently generate ads and optimize their effectiveness.

[1248] 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.

[1249] In this invention, the server includes means for receiving detailed product information from a user, means for inputting the received detailed information into a generation AI model to automatically generate advertising images, means for providing the generated advertising images to the user, means for feeding back advertising effectiveness data collected after the advertisement is posted to the generation AI model for learning, means for using a database for saving the advertising effectiveness data, means for re-learning the AI ​​model, and means for using a generation engine for improving the quality of the advertising images based on the collected advertising effectiveness data. This makes it possible to generate high-quality advertising images and optimize advertising effectiveness without any special expertise.

[1250] The "means for accepting detailed product information from the user" is a function for providing an interface that allows the user to input information about the product, such as the product name, price, description, and image, and inputting that information into the system.

[1251] "Means for automatically generating advertising images by inputting received detailed information into a generative AI model" refers to a function that sends product information entered by the user to a server, and the generative AI model automatically generates advertising images based on that information.

[1252] "Means for providing generated advertising images to users" refers to a function for providing advertising images created by a generative AI model to users' devices so that users can view and use them.

[1253] "Means of feeding back advertising effectiveness data collected after an advertisement is posted to the AI ​​generation model to allow it to learn" refers to a function that feeds back advertising effectiveness data collected after a user posts an advertisement, such as the number of clicks, number of impressions, and conversion rate, to the AI ​​generation model and uses it as learning data for the model.

[1254] The "means for using a database for storing advertising effectiveness data" is a function for storing collected advertising effectiveness data in a database within the system so that it can be used for later analysis and model re-learning.

[1255] "Means for retraining the AI ​​model" refers to a function that retrains the generation AI model based on advertising effectiveness data to improve the accuracy of advertising image generation.

[1256] "Means for using a generation engine to improve the quality of advertising images based on collected advertising effectiveness data" refers to a function for generating higher quality advertising images by analyzing collected advertising effectiveness data and updating and improving the generation AI model based on the results.

[1257] A specific embodiment for carrying out the present invention will be described below. A user, a terminal, and a server work together based on the following steps.

[1258] User Interface (UI) Design

[1259] The server provides a user interface for users to enter product information. The UI displays fields for entering the product name, price, description, and image. The user enters product details in these fields. For example, the user enters the product name as "Latest Model Smartphone," the price as "50,000 yen," the product description as "Large capacity battery, fast charging compatible," and uploads a product image.

[1260] Sending product information

[1261] The terminal collects the product information entered by the user and sends it to the server in JSON format, where the entered product name, price, description, and uploaded image are compiled into a single data object.

[1262] Generating advertising images

[1263] The server passes the received product information to a generative AI model, which uses a neural network to generate advertising images based on the input information. This generation process is automatic, and the generated advertising images are stored on the server.

[1264] Sending and displaying advertising images

[1265] The server obtains the storage path or URL of the generated ad image and returns it to the user's device. The device then obtains the URL of the ad image received from the server and displays it on the user's screen. The user can then check the generated ad image and use it to place an ad on the advertising platform.

[1266] Advertising and data collection

[1267] The user uses the generated advertising image to post an advertisement on the advertising platform, and in this case, the user collects performance data (number of clicks, number of impressions, conversion rate, etc.) related to the posted advertisement.

[1268] Data feedback and AI model retraining

[1269] The server receives advertising effectiveness data provided by users after the advertisement is posted. The server feeds this data back into the generative AI model and retrains the AI ​​model. This feedback loop improves the accuracy of the generative AI model in generating advertising images.

[1270] Hardware and Software

[1271] To realize this system, the following hardware and software are used:

[1272] Server: Receives product information, generates advertising images using a generative AI model, and sends them back to the user's device. The generative AI model is installed on the server, and the advertising images are generated using that model.

[1273] Terminal: The user enters product information and sends it to the server. The server returns an advertisement image and displays it.

[1274] Generative AI model: A neural network model that creates advertising images based on product information. It is expected to use OpenAI APIs, etc.

[1275] Database: Stores advertising effectiveness data and uses it to retrain AI models.

[1276] Examples of specific examples and prompts

[1277] For example, if a user were to create an ad for a new smartphone, they would enter the following information:

[1278] Product name: Latest model smartphone

[1279] Price: \50,000

[1280] Description: Large capacity battery, fast charging

[1281] Image: base64 encoded image data

[1282] Example prompt for a generative AI model:

[1283] Please generate an ad image based on the following product information:

[1284] Product name: Latest model smartphone

[1285] Price: \50,000

[1286] Description: Large capacity battery, fast charging

[1287] Image: base64 encoded image data

[1288] In this way, a system can be constructed that can easily generate high-quality advertising images without requiring specialized knowledge and can optimize the effectiveness of advertising.

[1289] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1290] Step 1:

[1291] The user enters product information using a device. The device's user interface displays fields for entering the product name, price, description, and image. The user enters the required information in these fields and uploads an image. The entered information is temporarily stored on the device.

[1292] Input: User input of product name, price, description, and image

[1293] Output: Product information saved on the device

[1294] Step 2:

[1295] The device collects product information in JSON format and sends it to the server. The product information is formatted as a single data object. The server receives this and saves it as product information.

[1296] Input: Product information saved on the device

[1297] Output: JSON formatted data object, product information sent to the server

[1298] Step 3:

[1299] The server passes the received product information to the generative AI model, which generates advertising images based on input information such as product name, price, description, and image. The generative AI model uses a neural network to automatically create advertising images, using the input information as prompts during this process.

[1300] Input: A JSON formatted data object

[1301] Output: Advertising image generated by the generative AI model

[1302] Step 4:

[1303] The server obtains the storage path or URL of the generated advertisement image and returns it to the terminal. The terminal obtains the URL of the received advertisement image and displays it on the user's screen.

[1304] Input: Generated ad image

[1305] Output: Ad image storage path or URL, ad image displayed on the user's screen

[1306] Step 5:

[1307] Users use the generated ad images to place ads on the advertising platform, and performance data on the posted ads, such as the number of clicks, impressions, and conversion rates, is collected and later sent to the server.

[1308] Input: Ad image, Ad placement on advertising platform

[1309] Output: Collected advertising effectiveness data

[1310] Step 6:

[1311] The server receives advertising effectiveness data provided by users and feeds it back to the generative AI model. This data is stored in a database and used to retrain the AI ​​model, allowing the generative AI model to improve the accuracy of generating advertising images.

[1312] Input: Advertising effectiveness data

[1313] Output: Advertising effectiveness data stored in a database, and a generative AI model that receives feedback.

[1314] Step 7:

[1315] The server periodically retrains the AI ​​model based on collected advertising effectiveness data, improving the quality of the generated advertising images and helping with future ad generation.

[1316] Input: Advertising effectiveness data stored in the database

[1317] Output: The generative AI model updated through retraining

[1318] These steps realize a system that allows users to easily generate high-quality advertising images without specialized knowledge and optimize the effectiveness of advertising.

[1319] 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.

[1320] A specific embodiment for carrying out the present invention will be described below. The user, terminal, server, and emotion engine work together based on the following steps.

[1321] User Interface (UI) Design

[1322] User: First, the user accesses the system on their device. The UI on the screen displays fields for entering the product name, price, description, and image. The user enters product details in these fields. At this time, an emotion engine that recognizes the user's emotional state in real time is activated, and the user's emotional data is collected.

[1323] Sending product information

[1324] Terminal: Collects product information entered by the user and sends it to the server in JSON format. The entered product name, price, description, and uploaded image are compiled into a single data object. In addition, the user's emotion data collected by the emotion engine is also sent.

[1325] Generating advertising images

[1326] Server: After receiving the product information and emotion data, the server passes it to the generative AI model. The generative AI model uses a neural network to generate advertising images based on the input information. This generation process is automated, and the generated advertising images are stored on the server.

[1327] Sending and displaying advertising images

[1328] Server: Obtain the storage path or URL of the generated ad image and return it to the user's device.

[1329] Device: The URL of the ad image received from the server is obtained, set as the src attribute of the image element, and displayed on the user's screen. The user checks the generated ad image. At this time, the emotion engine recognizes the user's reaction in real time and collects emotional feedback data.

[1330] Advertising and data collection

[1331] User: Using the generated ad image, the user sets up an ad campaign and places ads on advertising platforms such as Yahoo! and LINE. At this time, the user collects performance data (number of clicks, number of impressions, conversion rate, etc.) related to the ads they place.

[1332] Data feedback and AI model retraining

[1333] Server: After the ad is displayed, the server receives the ad effectiveness data provided by the user and the emotion data recognized by the emotion engine. The server feeds this data back into the generative AI model and retrains the model. This feedback loop improves the accuracy of the AI ​​model's ad image generation.

[1334] As a concrete example, consider a user creating an advertisement for a new smartphone. When the user enters the product name "Latest Model Smartphone," the price "50,000 yen," the product description "High-capacity battery, fast charging compatible," and a product image into the system, the emotion engine monitors the user's emotional state in real time and collects data. The device sends this information to the server, and the server-side generative AI model automatically generates an advertisement image. The generated advertisement image is sent back to the user's device, where the user reviews it, and the emotion engine again analyzes the user's reaction. The user runs an advertising campaign using the generated advertisement image, and then provides the collected advertising effectiveness data and emotion data to the server. The generative AI model re-trains using the information obtained in this way, improving the accuracy of future advertisement image generation. This system enables more efficient and accurate advertisement creation, providing greater convenience to users.

[1335] The processing flow will be explained below.

[1336] Step 1: The user accesses the system on their device and inputs product details such as product name, price, description, and image. During input, the emotion engine recognizes the user's emotional state in real time and collects emotion data.

[1337] Step 2: The user's device compiles the entered product information and the emotion data collected by the emotion engine into a JavaScript object, converts it into JSON format, and sends an HTTP POST request to the server.

[1338] Step 3: The server receives the HTTP request, parses the product information and sentiment data from the request body, converts the product information into an appropriate format, and passes it to the generative AI model.

[1339] Step 4: The generative AI model generates advertising images based on product information and emotion data. This generation process is carried out using a neural network. The generated advertising images are stored on the server.

[1340] Step 5: The server obtains the storage path or URL of the generated ad image and returns it to the user's device in JSON format.

[1341] Step 6: The user's device retrieves the URL of the ad image received from the server, sets it as the src attribute of the image element, and displays it on the user's screen. The user checks the generated ad image. At this time, the emotion engine recognizes the user's reaction in real time and collects emotional feedback data.

[1342] Step 7: The user uploads the confirmed ad image to an advertising platform such as Yahoo or LINE, sets up an advertising campaign, and publishes it.

[1343] Step 8: During the advertising campaign, the user collects advertising effectiveness data such as clicks, impressions, and conversion rates from the advertising platform.

[1344] Step 9: The user provides the collected advertising effectiveness data to the server using the API provided by the server.

[1345] Step 10: The server stores the received advertising effectiveness data and the emotion data recognized by the emotion engine in a database and feeds it back to the generative AI model.

[1346] Step 11: The generative AI model is retrained based on the feedback data to improve the accuracy of advertising image generation.

[1347] These steps realize a system in which the user, terminal, server, and emotion engine work together to automatically generate advertising images and continuously improve accuracy by utilizing advertising effectiveness data and emotion data.

[1348] Example 2

[1349] 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."

[1350] In today's online advertising market, the generation and optimization of effective ad images is crucial, but the process of users manually creating and optimizing ad images is time-consuming and labor-intensive. Furthermore, while generating ad images that take users' emotional states into account would improve the user experience, existing technologies have not automated this process. Furthermore, there are insufficient mechanisms for effectively collecting post-campaign effectiveness data and incorporating it into re-learning. To address these challenges, it is necessary to automate ad image generation, collect real-time user emotional data, generate ad images that reflect that data, and establish a feedback loop for ad effectiveness data.

[1351] 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.

[1352] In this invention, the server includes a means for providing a user interface for users to input product names, prices, descriptions, and images; a means for collecting the input product information and real-time emotion data; a means for converting the collected data into JSON format and transmitting it to the server; a means for the server to input the received data into a generative AI model and automatically generate advertising images; a means for providing the generated advertising images to users; a means for collecting advertising effectiveness data after the advertising campaign is implemented; a means for feeding back the collected advertising effectiveness data and emotion data to the generative AI model for learning; and a means for the terminal to recognize the user's emotional state in real time and collect emotion data. This enables the automation and optimization of advertising image generation, enabling advertising image generation that takes user emotion data into account. Furthermore, the feedback loop of advertising effectiveness data improves the accuracy of the generative AI model, enabling more effective advertising campaign design.

[1353] A "user interface" is a system component that includes an input screen for a user to input product names, prices, descriptions, and images.

[1354] "Emotion engine" refers to a piece of software or hardware that recognizes a user's emotional state in real time and collects that data.

[1355] "JSON format" is an abbreviation for JavaScript Object Notation, and is a format for structuring and expressing data in text format.

[1356] A "generative AI model" is a model with automatic generation capabilities using artificial intelligence, which generates advertising images using neural networks.

[1357] "Advertising images" are visual advertising materials generated by a generative AI model based on product information and emotional data.

[1358] "Advertising effectiveness data" means data collected after an advertising campaign is conducted, including performance indicators such as clicks, impressions, and conversion rates.

[1359] "Feedback" is the process of providing advertising effectiveness data and emotional data to the generative AI model in order to retrain the model and improve its accuracy.

[1360] "API" stands for Application Programming Interface, an interface that enables data exchange between different software systems.

[1361] The present invention relates to a system for automatically generating and optimizing advertising images, in which a user interface, an emotion engine, a server, and a generative AI model work in cooperation with each other. Specific embodiments for carrying out the present invention will be described below.

[1362] User Interface Design

[1363] Terminal: First, the user accesses the system through a terminal. The user interface displays fields for entering the product name, price, description, and image. These fields are designed to be intuitive, allowing the user to easily enter product details.

[1364] Entering product information and collecting sentiment data

[1365] User: When a user enters product information, the emotion engine works in the background to recognize the user's emotional state in real time through the user's camera. For example, if the user is smiling, the system will recognize the emotion as "positive" and collect this data.

[1366] Product information and emotion data sent to server

[1367] Terminal: After the user finishes entering information, the terminal converts the product information and emotion data into JSON format and sends it to the server. The transmitted data includes the product name, price, description, image, and emotion data.

[1368] Generating advertising images

[1369] Server: The server analyzes the received data and passes it to a generative AI model. This generative AI model uses a neural network to automatically generate advertising images based on product information and emotional data. The generated advertising images are stored on the server.

[1370] Providing advertising images

[1371] Server: Obtains the storage path or URL of the generated ad image and returns it to the user's device. The user can view the ad image on their screen.

[1372] Running an advertising campaign

[1373] User: Using the generated ad image, the user sets up an advertising campaign. The ad is then placed on advertising platforms such as Yahoo! and LINE, and information such as target demographic, period, and budget is set.

[1374] Collection and feedback of advertising effectiveness data

[1375] User: After an advertising campaign is run, the user collects advertising effectiveness data such as clicks, impressions, and conversion rates.

[1376] Terminal: Collected advertising effectiveness data and emotion data are compiled and sent back to the server.

[1377] Server: The server feeds this data back into the generative AI model and retrains it, improving the accuracy of subsequent ad image generation.

[1378] Specific examples

[1379] For example, if a user is creating an ad for a new smartphone, they might input the following prompt into the generative AI model:

[1380] Product name: Latest model smartphone

[1381] Price: \50,000

[1382] Product description: Large capacity battery, fast charging compatible

[1383] Product Image: Image URL (e.g. www.example.com / image.jpg)

[1384] In this way, users can quickly and efficiently generate highly effective advertising images. This system is designed to significantly reduce the effort required for creating advertisements and maximize their effectiveness.

[1385] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1386] Step 1:

[1387] User Interface Display

[1388] Device: An application or web browser is launched that displays a screen that provides fields for entering product name, price, description, and image. The screen is designed to make it easy for the user to enter information visually. Once the input fields are visible, this constitutes input.

[1389] Step 2:

[1390] Entering product information and collecting sentiment data

[1391] User: Enters product name, price, description, and image into input fields on the device. This is the input data. Meanwhile, the emotion engine runs in the background, recognizing the user's emotional state (e.g., positive, negative) in real time through the user's camera and collecting emotion data. The emotion engine captures frames from the camera and uses a facial expression analysis algorithm to categorize the user's emotional state.

[1392] Step 3:

[1393] Product information and emotion data sent to server

[1394] Terminal: Once the user has completed the input, the terminal will combine the product name, price, description, uploaded image, and emotion data into a single data object. This is then serialized into JSON format and sent to the server via a secure protocol (e.g., HTTPS). Once the data has been sent to the server, the server receives it.

[1395] Step 4:

[1396] Generating advertising images

[1397] Server: Analyzes the received JSON data and passes the data object to the generative AI model. The generative AI model uses deep learning technology to generate advertising images using product information and emotional data as input. Specifically, a neural network processes the input data and creates advertising images containing appropriate visual elements and text designs. The generated advertising images are stored on the server. The output is the generated advertising image data.

[1398] Step 5:

[1399] Providing advertising images

[1400] Server: The output is to obtain the storage path or URL of the generated ad image and return it to the user's device in JSON format.

[1401] On the device: The received ad image URL is parsed and set as the src attribute of an HTML image element, which displays the image on the screen. This allows the user to check the generated ad image. The emotion engine then runs again, analyzing the user's reactions in real time and collecting new emotion data.

[1402] Step 6:

[1403] Running an advertising campaign

[1404] User: Uses the generated ad image to set up an ad campaign. The user starts the ad campaign by entering information about the target demographic, period, budget, and ad platform (e.g., Yahoo, LINE). This setting information is the input data. The ad campaign is then carried out, and the actual ads are displayed by the ad platform.

[1405] Step 7:

[1406] Collection and transmission of advertising effectiveness data

[1407] User: After running an advertising campaign, you collect advertising effectiveness data such as clicks, impressions, and conversion rates from the advertising platform. This is input data.

[1408] Terminal: The output is to compile the collected advertising effectiveness data and the user feedback data recorded by the emotion engine, serialize it in JSON format, and send it back to the server.

[1409] Step 8:

[1410] Data feedback and AI model retraining

[1411] Server: Analyzes the received advertising effectiveness data and emotion data and feeds it back to the generative AI model. This allows the generative AI model to be retrained using this data. In particular, the generative AI model uses deep learning algorithms such as backpropagation to adjust model parameters and improve the accuracy of subsequent ad image generation. This is the final output.

[1412] (Application example 2)

[1413] 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."

[1414] Conventional ad generation systems generate ad images without considering user emotions, resulting in poor ad effectiveness. Furthermore, the process of feeding back ad effectiveness data and retraining the AI ​​model is cumbersome, making it difficult to improve the accuracy of generating effective ads. This results in problems such as the inability to create effective ads that users desire, and reduced efficiency of advertising campaigns.

[1415] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1416] In this invention, the server includes means for receiving detailed product information from a user, means for inputting the received detailed information and emotion data into a generative AI model to automatically generate advertising images, means for providing the generated advertising images to the user, an emotion engine for collecting user emotional responses in real time, and means for feeding back advertising effectiveness data collected after the advertisement is posted to the generative AI model for learning. This enables automatic generation of advertising images that take user emotions into consideration and re-training of the AI ​​model based on the advertising effectiveness data.

[1417] "Product details" are basic attributes of a product provided by the user, such as its name, price, description, and image.

[1418] "Emotional data" refers to data that measures and collects a user's emotional state in real time.

[1419] A "generative AI model" is an artificial intelligence model that automatically generates advertising images based on input information.

[1420] "Advertising images" are visual promotional materials generated based on detailed product information and emotional data provided by users.

[1421] The "emotion engine" is a system for collecting and analyzing users' emotional responses in real time.

[1422] "Advertising effectiveness data" refers to data on performance indicators such as the number of clicks, impressions, and conversion rates in advertising campaigns.

[1423] "Feedback" is the process of retraining the generative AI model based on collected data to improve the accuracy of ad generation from the next time onwards.

[1424] An "application program interface" is a standardized means for exchanging data between different software systems.

[1425] This invention is a system that automatically generates advertising images based on input of product information, and improves the accuracy of advertisement generation by grasping user emotions in real time and obtaining feedback. This system is implemented on both the front end and the server side.

[1426] front end

[1427] On the front end, users access the interface using their smartphones. Specifically, they input product names, prices, descriptions, and image URLs through an application built with React Native. An emotion engine also detects the user's facial expressions in real time and collects emotional data. This collected data is sent to the server in JSON format.

[1428] Specific examples

[1429] When a user creates an ad for a new smartwatch, they enter the following information:

[1430] Product Name: Latest Smartwatch

[1431] Price: \30,000

[1432] Product description: Heart rate sensor, high-precision GPS

[1433] Product image URL: https: / / example.com / images / smartwatch.jpg

[1434] Server Side

[1435] On the server side, the received data is processed using Node.js and Express. Specifically, the received JSON data is input into a generative AI model to generate ad images. This generative AI model uses TensorFlow.js, which generates ad images using a neural network. The generated ad images are stored on the server, and their URLs are returned to the user's device. In addition, user emotional responses and ad effectiveness data (number of clicks, number of impressions, conversion rate, etc.) are collected in real time and fed back to the server. This data is passed to the generative AI model and used for re-training.

[1436] Specific examples

[1437] The user checks the ad image and runs an advertising campaign. For example, the generated ad image is used to post on an advertising platform, and the AI ​​model is retrained using the advertising effectiveness data obtained later. An example of the prompt sentence in this case is as follows:

[1438] Prompt Sentence Examples

[1439] Based on the information received from the user, the generative AI model generates advertising images as follows:

[1440] Prompt text: Product name "Latest smartwatch", price "30,000 yen", product description "Heart rate sensor, high-precision GPS", image URL "https: / / example.com / images / smartwatch.jpg", emotion data "Happiness level: 80%, Neutrality level: 20%".

[1441] In this way, it is possible to improve the accuracy of advertising images generated based on user emotions and advertising effectiveness data, and to realize efficient advertising campaigns.

[1442] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1443] Step 1:

[1444] A user uses a smartphone to access an application built with React Native.

[1445] Users enter the product name, price, description, and image URL in the corresponding input fields, and the emotion engine collects emotional data from the user's facial expressions, such as happiness, surprise, anger, etc. These details and emotional data are then compiled and converted into JSON format.

[1446] input:

[1447] Product name, price, description, image URL

[1448] Emotional data (happiness, surprise, anger, etc.)

[1449] output:

[1450] Detailed information and emotion data in JSON format

[1451] Step 2:

[1452] The device sends the collected details and emotion data in JSON format to the server.

[1453] This data is sent as an HTTP POST request.

[1454] input:

[1455] Detailed information and emotion data in JSON format

[1456] output:

[1457] HTTP POST request to the server

[1458] Step 3:

[1459] The server analyzes the received JSON-formatted details and emotion data and prepares them for input into the generative AI model.

[1460] Specifically, it formats each input item (product name, price, description, image URL, emotion data).

[1461] input:

[1462] Detailed information and emotion data in JSON format

[1463] output:

[1464] Formatted details and sentiment data

[1465] Step 4:

[1466] The server inputs the formatted details and emotional data into a generative AI model to generate advertising images.

[1467] The generative AI model uses TensorFlow.js and employs a neural network to generate advertising images from input data.

[1468] input:

[1469] Formatted details and sentiment data

[1470] output:

[1471] Generated advertising image

[1472] Step 5:

[1473] The server stores the generated advertisement image therein and returns the URL to the terminal.

[1474] The URL of the returned ad image is displayed in the client-side interface for the user to view.

[1475] input:

[1476] Generated advertising image

[1477] output:

[1478] Ad image URL

[1479] Step 6:

[1480] The user can review the generated advertising images and use them for their advertising campaigns.

[1481] In addition, the emotion engine records the user's emotional reactions in real time and sends them to the server.

[1482] input:

[1483] Advertising images

[1484] output:

[1485] User emotional response data

[1486] Step 7:

[1487] The server feeds back the emotional response data collected from users and advertising effectiveness data (number of clicks, number of impressions, conversion rate, etc.) to the generative AI model.

[1488] The generated AI model is retrained based on this data.

[1489] input:

[1490] Emotional response data

[1491] Advertising effectiveness data

[1492] output:

[1493] Generative AI model with improved accuracy through retraining

[1494] This makes it possible to generate advertising images based on user emotions and advertising effectiveness data, enabling efficient advertising campaign management.

[1495] 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.

[1496] 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.

[1497] 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 robot 414.

[1498] 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.

[1499] 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.

[1500] 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.

[1501] 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).

[1502] 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.

[1503] 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."

[1504] 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.

[1505] 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).

[1506] 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.

[1507] 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.

[1508] 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.

[1509] 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.

[1510] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1511] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1512] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1513] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1514] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, in order to avoid confusion and to facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1515] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1516] The following is further disclosed regarding the above embodiment.

[1517] (Claim 1)

[1518] A means for accepting detailed product information from users;

[1519] A means to input the received detailed information into a generation AI model to automatically generate advertising images;

[1520] A means for providing the generated advertising image to a user;

[1521] A system that includes a means for feeding back advertising effectiveness data collected after advertisements are published into a generative AI model to allow it to learn.

[1522] (Claim 2)

[1523] The system of claim 1, wherein the generative AI model for generating advertising images uses a neural network.

[1524] (Claim 3)

[1525] 10. The system of claim 1, further comprising: means for providing an API for automatically collecting advertising effectiveness data from users.

[1526] "Example 1"

[1527] (Claim 1)

[1528] A means for a user to input detailed product information using an information processing terminal;

[1529] means for transmitting the input product detail information in data format to a server;

[1530] A server receives detailed product information and inputs the information into a generation AI model to automatically generate advertising images;

[1531] A means for the server to return the generated advertising image to the user's information processing terminal;

[1532] a means for displaying the generated advertising image to a user by an information processing terminal;

[1533] A means for users to post advertisements and collect advertising effectiveness data;

[1534] A system that includes a means for a server to receive collected advertising effectiveness data and feed it back to the generative AI model to allow it to learn.

[1535] (Claim 2)

[1536] 2. The system of claim 1, wherein the generative AI model generates advertising images using a neural network.

[1537] (Claim 3)

[1538] 10. The system of claim 1, further comprising: means for providing an API for a user to automatically transmit advertising effectiveness data to the server.

[1539] "Application Example 1"

[1540] (Claim 1)

[1541] A means for accepting detailed product information from users;

[1542] A means to input the received detailed information into a generation AI model to automatically generate advertising images;

[1543] A means for providing the generated advertising image to a user;

[1544] A method for feeding back advertising effectiveness data collected after advertising is published into the AI ​​model to allow it to learn.

[1545] a means for using a database for storing advertising effectiveness data;

[1546] A means to retrain the AI ​​model,

[1547] The system includes means for using a generation engine to improve the quality of advertising images based on collected advertising effectiveness data.

[1548] (Claim 2)

[1549] The system of claim 1, wherein the generative AI model for generating advertising images uses a neural network.

[1550] (Claim 3)

[1551] 10. The system of claim 1, further comprising: means for providing an API for automatically collecting advertising effectiveness data from users.

[1552] "Example 2: Combining Emotion Engines"

[1553] (Claim 1)

[1554] means for providing a user interface for a user to input a product name, price, description, and image;

[1555] a means for collecting input product information and real-time sentiment data;

[1556] A means of converting the collected data into JSON format and sending it to the server;

[1557] A means for inputting the data received by the server into a generation AI model to automatically generate advertising images;

[1558] means for providing the generated advertising image to a user;

[1559] a means for collecting advertising effectiveness data after an advertising campaign is implemented;

[1560] A means for feeding back the collected advertising effect data and emotion data to the generation AI model to make it learn;

[1561] The system includes a means for a device to recognize a user's emotional state in real time and collect emotional data.

[1562] (Claim 2)

[1563] 2. The system of claim 1, wherein the generative AI model generates advertising images using a neural network.

[1564] (Claim 3)

[1565] 10. The system of claim 1, further comprising: means for providing an application programming interface for automatically collecting advertising effectiveness data from users.

[1566] "Application example 2 when combining emotion engines"

[1567] (Claim 1)

[1568] A means for accepting detailed product information from users;

[1569] A means for automatically generating advertising images by inputting the received detailed information and emotional data into a generation AI model;

[1570] A means for providing the generated advertising image to a user;

[1571] An emotion engine that collects users' emotional responses in real time;

[1572] A system that includes a means for feeding back advertising effectiveness data collected after advertisements are published into a generative AI model to allow it to learn.

[1573] (Claim 2)

[1574] The system of claim 1, wherein the generative AI model for generating advertising images uses a neural network.

[1575] (Claim 3)

[1576] 10. The system of claim 1, further comprising: means for providing an application program interface for automatically collecting advertising effectiveness data from users. [Explanation of symbols]

[1577] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for accepting detailed product information from users; A means to input the received detailed information into a generation AI model to automatically generate advertising images; A means for providing the generated advertising image to a user; A system that includes a means for feeding back advertising effectiveness data collected after advertisements are published into a generative AI model to allow it to learn.

2. The system of claim 1, wherein the generative AI model for generating advertising images uses a neural network.

3. The system of claim 1 , further comprising: means for providing an API for automatically collecting advertising effectiveness data from users.

Images