system

Abstract

We provide the system. [Solution] A user terminal for uploading advertising content, A preprocessing means for converting uploaded advertisement content into text data, A generative artificial intelligence for reviewing text data from a legal compliance perspective, A means of notifying the user of the review results, A system that includes this.

Description

Technical Field

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

Background Art

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

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] The work of manually checking whether the current advertisement content conflicts with the Scenery Act or the Pharmaceutical Machinery Act requires time and effort, and since subjectivity is likely to enter the review conducted by humans, it is difficult to conduct a consistent review based on certain criteria. Furthermore, since the advertisement content exists in various formats such as images and HTML, a method for efficiently processing these and accurately confirming compliance with laws and regulations is required.

Means for Solving the Problems

[0005] The present invention solves the above problem by providing a system that includes a user terminal for uploading advertising content, a preprocessing means for converting the uploaded advertising content into text data, a generative artificial intelligence for reviewing the text data from a legal compliance perspective, and a means for notifying the user of the review results. The preprocessing means has a function to convert the advertising content into text data using optical character recognition (OCR) technology when the advertising content is an image file, and a function to extract the text portion and convert it into text data when it is in HTML format. The generative artificial intelligence reviews the extracted text data in accordance with the Act against Unjustifiable Premiums and Misleading Representations and the Pharmaceuticals and Medical Devices Act, and evaluates the risk of violating the law. The review results are also notified to the user by email or as an in-system message. The user terminal also has a function to perform user authentication at the same time as uploading the advertising content.

[0006] A "user terminal" is an electronic device used by a user to input and upload advertising content.

[0007] "Preprocessing means" refers to a function that determines the format of uploaded advertisement content and converts it into text data as necessary.

[0008] "Generative artificial intelligence" is an artificial intelligence technology that uses given text data to examine whether its content violates any laws or regulations.

[0009] "Means of notifying users of the review results" refers to communication methods for conveying the review results obtained by generative artificial intelligence to users, and includes methods such as email and in-system messages.

[0010] Optical Character Recognition (OCR) technology is a technology used to convert text within an image into digital text.

[0011] "HTML format" is a markup language format widely used to describe web pages, defining the structure of content that includes text and various media.

[0012] The "Act on Unfair Promotion of Premiums and Representations" is an abbreviation for the Premiums and Representations Act, a Japanese law aimed at preventing competition through unfair customer inducement.

[0013] The "Pharmaceuticals and Medical Devices Act" is an abbreviation for the Act on Securing Quality, Efficacy and Safety of Pharmaceuticals, Medical Devices, etc., and is a law that regulates the handling of pharmaceuticals and medical devices in Japan.

[0014] "Legal compliance" refers to a state in which the content of an advertisement does not violate any applicable laws or regulations.

[0015] User authentication is a procedure to verify that a user accessing a system is a legitimate user. [Brief explanation of the drawing]

[0016] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] Shows an emotion map where multiple emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Embodiment 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Embodiment 2 when combined with an emotion engine. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when combined with an emotion engine.

Mode for Carrying Out the Invention

[0017] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0018] First, the terms used in the following description will be explained.

[0019] In the following embodiments, a numbered processor (hereinafter simply referred to as "processor") may be a single arithmetic unit, or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit, or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0020] In the following embodiments, a numbered RAM (Random Access Memory) is a memory where information is temporarily stored and is used as a work memory by the processor.

[0021] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0022] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0023] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0024] [First Embodiment]

[0025] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0026] As shown in Figure 1, the 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.

[0027] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0028] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0029] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and 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.

[0030] 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 perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0031] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0032] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] As shown in Figure 2, in the data processing device 12, a specific processing 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" related to the technology of this 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 according to the specific processing program 56 executed on the RAM 30.

[0034] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0035] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0036] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0037] This invention is a system for reviewing whether advertising content violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act. Users upload advertising content, a server converts that content into text data, reviews it using generative artificial intelligence, and notifies the user of the results.

[0038] Processing at the user terminal

[0039] When a user logs in, they enter their user ID and password to access the system. Upon successful login, the user gains access to an interface for uploading advertising content. Once the user uploads the content, the file is sent to the server in the specified format.

[0040] Server-side preprocessing

[0041] The server receives files uploaded by users and saves them to the appropriate directory. Next, it determines the format of the uploaded file and converts it to text data if necessary. For example, it uses optical character recognition (OCR) technology to extract text from image files and extracts the text portion from HTML files.

[0042] Specific examples based on the server's processing targets:

[0043] For image files:

[0044] User-uploaded image files may contain screenshots of advertisements. The server analyzes these images using OCR technology and extracts text information from within them.

[0045] For HTML files:

[0046] The HTML file uploaded by the user contains an advertising landing page. The server extracts the text portion from the HTML file and analyzes it.

[0047] Review process using generative artificial intelligence

[0048] The converted text data is sent to a generative artificial intelligence system. The generative AI analyzes the text data and evaluates whether it violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act. The generative AI returns its evaluation results to the server. These results include details on which laws the advertisement content may violate, or whether it is otherwise problematic.

[0049] Examples of the review process:

[0050] If there is a conflict:

[0051] When a generative artificial intelligence detects a statement like "You will lose weight just by taking this supplement," it determines that the statement may violate the Pharmaceutical Affairs Law and returns a detailed assessment of the risk of such a violation.

[0052] Notification means

[0053] The server receives the review results from the generative artificial intelligence and notifies the user. Notification is sent via email or in-system message. Users can check whether their uploaded advertisements violate the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceutical Affairs Act, and identify any problematic aspects.

[0054] Examples of notifications:

[0055] The user receives an email stating that the review results indicate that the phrase "You can lose weight just by drinking this" may violate the Pharmaceutical Affairs Law. Based on this information, the user can modify the advertisement content.

[0056] In this way, the system of the present invention makes it possible to efficiently review advertising content for legal compliance and notify users of the results in a user-friendly manner.

[0057] The following describes the processing flow.

[0058] Step 1:

[0059] The user logs into the system. The user enters their user ID and password, and login authentication is performed. If authentication is successful, the user can access the interface for uploading ad content.

[0060] Step 2:

[0061] The user uploads the ad content. They select a text file, image file, or HTML file as the ad content and send it to the system. The file is sent to the server and saved in the specified directory.

[0062] Step 3:

[0063] The server receives the uploaded file and determines its file format. Here, it determines whether it is a text file, an image file, or an HTML file.

[0064] Step 4:

[0065] The server performs preprocessing according to the file format. If the advertisement content is a text file, it is used and saved as text data. If the advertisement content is an image file, optical character recognition (OCR) technology is used to extract the text within the image. If the advertisement content is an HTML file, the text portion is extracted from the HTML.

[0066] Step 5:

[0067] The server sends the pre-processed text data to a generative artificial intelligence (AI). The AI ​​analyzes the text data and reviews its content in accordance with the Act against Unjustifiable Premiums and Misleading Representations and the Pharmaceuticals and Medical Devices Act. Specifically, it evaluates whether it contains any illegal or misleading expressions.

[0068] Step 6:

[0069] The generative artificial intelligence returns the review results to the server. The review results include which parts may violate the law, along with the reasons why.

[0070] Step 7:

[0071] The server notifies the user of the review results. The review results are sent to the email address registered by the user or displayed as a system message. This allows the user to check whether the advertisement content poses a risk of violating the law.

[0072] Step 8:

[0073] The user receives the review results and modifies the ad content as needed. If modifications are required, the user can re-upload the ad content and repeat the same process until the issue is resolved.

[0074] (Example 1)

[0075] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0076] One challenge is the time-consuming process of reviewing advertisements to ensure they comply with laws such as the Premiums and Representations Act and the Pharmaceuticals and Medical Devices Act. Manual review is time-consuming and labor-intensive, and increases the risk of legal violations. Furthermore, because technical knowledge is required, there is a need for a system that anyone can easily use.

[0077] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0078] In this invention, the server includes terminal means for uploading advertising content, preprocessing means for converting the uploaded advertising content into text data, generative artificial intelligence means for reviewing the text data from a legal compliance perspective, and means for notifying the user of the review results. This makes it possible to automate the legal compliance review of advertising content and to notify the user of the review results quickly and accurately.

[0079] "Terminal means" refers to the device that a user uses to upload advertising content.

[0080] "Preprocessing means" refers to a device or software that performs processing to convert uploaded advertisement content into text data.

[0081] A "generative artificial intelligence method" is an artificial intelligence model that analyzes text data and performs reviews from a legal compliance perspective.

[0082] "Means for notifying users of the review results" refers to a method or system for informing users of the review results performed by a generative artificial intelligence.

[0083] "Optical character recognition technology" is a technology for extracting text information from image-based advertisements.

[0084] "Web page format" refers to advertising content expressed in formats such as HTML.

[0085] This invention is a system for reviewing whether advertising content violates the Premiums and Representations Act or the Pharmaceuticals and Medical Devices Act. The system works by having a user upload advertising content, a server converting that content into text data, a review being performed using generative artificial intelligence, and then notifying the user of the results.

[0086] Users access the system using their user terminal and log in by entering their user ID and password. After logging in, users select a file containing the advertisement content and upload it through the interface. This uploaded file is sent to the server in the specified format.

[0087] The server receives files sent by users and saves them to the appropriate directory. The server determines the format of the received files and converts them to text data as needed. For example, in the case of image files, optical character recognition (OCR) technology is used to extract the text information. One possible technique for this is using the Tesseract OCR engine. In the case of HTML files, libraries such as BeautifulSoup or html.parser are used to extract the text portion.

[0088] The converted text data is sent to a generative artificial intelligence (AI) system. This AI model is pre-trained to detect risks of violations of laws such as the Premiums and Representations Act and the Pharmaceuticals and Medical Devices Act. The AI ​​analyzes and evaluates the text data and returns the evaluation results to the server. These results include details on which laws the advertisement content may violate, or whether it is otherwise problematic.

[0089] For example, if the text data contains a phrase like "You will lose weight just by taking this supplement," the generative artificial intelligence will determine that this phrase may violate the Pharmaceutical Affairs Law and return that as the evaluation result.

[0090] The server notifies the user of the review results received from the generative artificial intelligence. Notification is sent via email or through an in-system message. Based on these results, the user can check whether the advertisement content complies with the law, identify any problems, and make corrections as needed.

[0091] Examples of prompt statements include the following:

[0092] "Please check if the copy in this supplement advertisement violates the Pharmaceutical Affairs Law. The copy says, 'You will lose weight just by taking this supplement.'"

[0093] The above describes a specific embodiment for carrying out this invention. This makes it possible to automate the legal compliance review of advertising content and notify users quickly and accurately.

[0094] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0095] Step 1:

[0096] The user accesses the system from their terminal and logs in by entering their user ID and password. Upon successful login, an interface for uploading advertising content is displayed. The user selects the file containing the advertising content and presses the upload button. This action sends the advertising content file from the user's terminal to the server. The input consists of the user ID, password, and the advertising content file, while the output is the advertising content file sent to the server.

[0097] Step 2:

[0098] The server receives files sent from the user's terminal and saves them to the appropriate directory. It identifies the format of the received file (image file, HTML file, etc.) and converts it to text data as needed. This process yields text data for further analysis. Specifically, it checks the file extension and MIME type. The input is the received file, and the output is text data.

[0099] Step 3:

[0100] For image files, the server extracts text information using Optical Character Recognition (OCR) technology. For example, the Tesseract OCR engine is used. For HTML files, libraries such as BeautifulSoup or html.parser are used to extract the text portion. The input is file data corresponding to the identified file format, and the output is the extracted text data. Specifically, for image files, the OCR engine is called to analyze the characters, and for HTML files, a text extraction library is used to parse the text portion.

[0101] Step 4:

[0102] The server sends the extracted text data to a generative artificial intelligence system. This generative AI is pre-trained to detect risks of violations related to the Premiums and Representations Act and the Pharmaceuticals and Medical Devices Act. The input is the extracted text data, and the output is the text data sent to the generative AI system.

[0103] Step 5:

[0104] The generative artificial intelligence system analyzes transmitted text data and evaluates whether it contains any expressions that violate laws and regulations. The evaluation results include detailed information on which specific parts may violate which laws, or whether there are any problems. The input is text data transmitted from the server, and the output is the evaluation result. For example, if the expression "You will lose weight just by taking this supplement" is detected, the evaluation result will be generated indicating that this expression may violate the Pharmaceuticals and Medical Devices Act.

[0105] Step 6:

[0106] The server notifies the user of the evaluation results received from the generative artificial intelligence system. Notification methods include generating and sending emails, or using in-system messages. Based on these evaluation results, the user can check whether the advertisement content complies with the law and identify any problematic areas. The input is the evaluation results from the generative artificial intelligence system, and the output is the notification to the user. Specifically, the server either sends an email based on the evaluation results or provides information to the user through the system's notification function.

[0107] The above outlines the processing steps of this system's program, providing a detailed explanation of the data processing and calculations performed at each step and the resulting outcomes.

[0108] (Application Example 1)

[0109] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0110] The goal is to reduce the risk of advertising content violating the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act, while also providing advertising creators and marketing personnel with a quick and efficient method for reviewing compliance with laws and regulations. Furthermore, there is a need to obtain the results of advertising content reviews in real time, and it is necessary to solve the problems of time and cost that conventional methods entail.

[0111] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0112] In this invention, the server includes a user terminal for uploading advertising content, preprocessing means for converting the uploaded advertising content into text data, a generative artificial intelligence for reviewing the text data from a legal compliance standpoint, means for notifying the user of the review results, and an application installed on a smartphone for taking photos of or uploading advertising content and providing review results in real time. This makes it possible to quickly review whether advertising content violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act and provide results in real time.

[0113] "Advertising content" refers to text, images, or other information created by users and offered to the public for marketing or promotional purposes.

[0114] A "user device" refers to an internet-connected device used to upload advertising content, and includes smartphones, tablets, and personal computers.

[0115] "Preprocessing means" refers to a function that performs the necessary processing to convert uploaded advertisement content into text data.

[0116] Optical Character Recognition (OCR) technology is a technology that analyzes text within an image and converts it into digital text.

[0117] "Generative artificial intelligence" refers to machine learning models and algorithms used to automatically review advertising content for compliance with various laws and regulations.

[0118] "Means of notifying users of the review results" refers to communication methods such as email, in-system messages, and push notifications used to notify users of the review results.

[0119] An "application installed on a smartphone" is software that runs on a smartphone and enables the shooting or uploading of advertising content and real-time review.

[0120] "Providing results in real time" means processing the review results immediately so that users can check the results without waiting.

[0121] This invention is a system for reviewing whether advertising content violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act. The system includes the following components:

[0122] 1. User terminal:

[0123] This provides an interface for users to log in and upload advertising content. Users can submit text, images, or HTML files of their advertisements to the server through this interface. For example, a user could take a picture of their advertisement content using their smartphone and upload that image.

[0124] 2. Pre-treatment means:

[0125] The server receives files uploaded by users and converts them to the appropriate format. This process uses optical character recognition (OCR) technology and HTML text extraction technology. For OCR technology, the "pytesseract" library is used to extract text from image files. For HTML files, an HTML parsing library such as "BeautifulSoup" is used to extract the text portion.

[0126] 3. Generative Artificial Intelligence:

[0127] The converted text data is sent to a generative artificial intelligence (AI) system. The AI ​​system uses the "transformers" library, and specifically the "BERT-base-japanese" model, to analyze the text data from a legal compliance perspective. The analysis results evaluate whether the advertising content may violate the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act.

[0128] 4. Means of notifying the results of the review:

[0129] The server notifies the user of the review results obtained from the generative artificial intelligence. Notification is sent via in-system messages or email. The user can review the review results and modify the advertisement content as needed.

[0130] This system allows users to quickly and efficiently review the legal compliance of advertising content. The program's processing is explained below in natural language.

[0131] When the server receives an advertisement image uploaded by a user, it first uses optical character recognition (OCR) technology to extract the text within the image. Next, it uses a generative artificial intelligence model (e.g., BERT-base-japanese) to analyze whether the extracted text violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act. The analysis results are then notified to the user as a review result.

[0132] As a concrete example, suppose a user takes a picture of an advertisement with their smartphone and uploads it. This image is sent to a server, and the text is extracted using OCR technology. A generative artificial intelligence model analyzes this text, and if it determines that the expression "You will lose weight just by taking this supplement" violates the Pharmaceuticals and Medical Devices Act, a notification is sent to the user stating that "This expression may violate the Pharmaceuticals and Medical Devices Act."

[0133] The following is an example of a prompt message:

[0134] "I took an image: 'advertisement_image.jpg'. Please extract and review the ad text: 'You will lose weight just by taking this supplement.' Please output the results and notify me."

[0135] In this way, the system of the present invention makes it possible to efficiently review the legal compliance of advertising content and notify users of the results in a user-friendly manner.

[0136] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0137] Step 1:

[0138] Users take pictures of advertisements with their smartphones and upload those images to a server through the application.

[0139] Input: Advertisement image file

[0140] Output: Advertisement image file sent to the server

[0141] Step 2:

[0142] The server receives the uploaded ad image files and saves them to the appropriate directory. Then, optical character recognition (OCR) technology is used to extract text from the images.

[0143] Input: Advertisement image file

[0144] Data processing: Text extraction using Optical Character Recognition (OCR) technology

[0145] Output: Text data

[0146] Step 3:

[0147] The server sends the extracted text data to a generative artificial intelligence model for analysis and review from a legal compliance perspective. The generative AI model used is "BERT-base-japanese" to determine whether the advertising content violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act.

[0148] Input: Text data

[0149] Data processing: Text analysis using generative artificial intelligence models

[0150] Output: Review result data

[0151] Step 4:

[0152] The server notifies the user of the evaluation results based on the evaluation result data obtained from the generative artificial intelligence model. Notification is sent via in-system messages or email.

[0153] Input: Review result data

[0154] Data processing: Formatting of review results and notification settings

[0155] Output: Notification to the user

[0156] Step 5:

[0157] Users can check the review results received via in-system message or email. If necessary, they can revise the ad content and request a review again.

[0158] Input: Notification of review results

[0159] Output: Confirmation of review results and possibility of modifying ad content as the next action.

[0160] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0161] This invention combines a system for reviewing whether advertising content violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act with an emotion engine that recognizes user emotions, thereby achieving a more user-friendly review and notification process. Specifically, the system comprises a user terminal, pre-processing means, generative artificial intelligence, means for notifying the user of the review results, and an emotion engine.

[0162] Processing at the user terminal

[0163] Users log in to the system and upload ad content. A user ID and password are required for login, and upon successful authentication, users can access the interface for uploading ad content. As users upload ad content, the system uses the camera and microphone on the user's device to analyze the user's facial expressions and voice in real time, and an emotion engine recognizes the user's emotions.

[0164] Server-side preprocessing

[0165] The server receives files uploaded by users and saves them to the appropriate directory. Next, it determines the format of the uploaded file and converts it to text data if necessary. For image files, optical character recognition (OCR) technology is used, and for HTML files, the text portion is extracted.

[0166] Specific examples based on the server's processing targets:

[0167] For image files:

[0168] If the user has uploaded a screenshot of an advertisement, the server will analyze the image using OCR technology and extract the text information within the image.

[0169] For HTML files:

[0170] If the user has uploaded an ad landing page, the server will extract the text portion from the HTML file and analyze it.

[0171] Review process using generative artificial intelligence

[0172] The text data obtained during preprocessing is then sent to a generative artificial intelligence (AI). The AI ​​analyzes the text data and conducts a review based on the Act against Unjustifiable Premiums and Misleading Representations and the Pharmaceuticals and Medical Devices Act. As a result, it performs a risk assessment for violations of the law and returns the review results to the server. Specifically, it evaluates whether the text contains illegal or misleading expressions.

[0173] Examples of the review process:

[0174] If there is a conflict:

[0175] If the generative artificial intelligence detects a statement such as "You will lose weight just by taking this supplement," it will determine that the statement may violate the Pharmaceutical Affairs Law and return a specific risk assessment result to the server.

[0176] Notification methods and emotion engines

[0177] The server receives the evaluation results from the generative artificial intelligence and, considering the user's emotional data recognized by the emotion engine, notifies the user of the evaluation results. Notification is sent via email or in-system message. For example, if the user is experiencing stress, the notification can be made using gentler language.

[0178] Examples of notifications:

[0179] If the user is feeling stressed:

[0180] If the emotion engine determines from the user's facial expressions and voice that they are experiencing stress, the server will notify the user via email or in-system message, using milder language such as "Caution advised." The user can then modify the ad content based on this information.

[0181] Suggestions for improving advertising content

[0182] The emotion engine can also generate suggestions for improving ad content based on user emotion data. For example, if a user is feeling down, the engine might suggest "changing the ad content to a more positive tone."

[0183] In this way, the system of the present invention makes it possible to efficiently perform legal compliance reviews of advertising content, provide notifications and improvement suggestions that take into account user sentiment data, and deliver results in a user-friendly manner.

[0184] The following describes the processing flow.

[0185] Step 1:

[0186] The user logs into the system. The user enters their user ID and password, and login authentication is performed. If authentication is successful, the user can access the interface for uploading ad content.

[0187] Step 2:

[0188] The user uploads the ad content. This ad content can be a text file, image file, or HTML file sent to the system. The file is sent to the server and saved in the specified directory.

[0189] Step 3:

[0190] The device's camera and microphone are used to analyze the user's facial expressions and voice in real time as they upload ad content. An emotion engine receives this data and recognizes the user's emotions.

[0191] Step 4:

[0192] The server receives the uploaded file and determines its file format. Here, it determines whether it is a text file, an image file, or an HTML file.

[0193] Step 5:

[0194] The server performs preprocessing according to the file format. If the advertisement content is a text file, it is saved as text data as is. If the advertisement content is an image file, it is converted to text data using optical character recognition (OCR) technology. If the advertisement content is an HTML file, the text portion is extracted from the HTML.

[0195] Step 6:

[0196] The server sends the pre-processed text data to the generative artificial intelligence. The generative AI analyzes the text data and reviews its content in accordance with the Act against Unjustifiable Premiums and Misleading Representations and the Pharmaceuticals and Medical Devices Act.

[0197] Step 7:

[0198] The generative artificial intelligence returns the review results to the server. The review results include which parts may violate the law, along with the reasons why.

[0199] Step 8:

[0200] The server notifies the user of the review results. The notification method and wording are adjusted based on the user's emotional data recognized by the emotion engine. Notification methods include email or in-system messages.

[0201] Step 9:

[0202] Users who receive the review results can check the ad content and make corrections as needed. By using the improvement suggestions provided by the emotion engine as a reference, users can re-upload the ad content and repeat the same process to create a legally compliant ad.

[0203] In this way, the system of the present invention makes it possible to efficiently perform legal compliance reviews of advertising content, provide notifications and improvement suggestions that take into account user sentiment data, and deliver results in a user-friendly manner.

[0204] (Example 2)

[0205] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0206] Traditional advertising content review systems check for legal compliance, but rarely provide notifications or feedback that take user emotions into consideration. As a result, users may experience stress or resentment, leading to a decline in the user experience.

[0207] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0208] In this invention, the server includes a user terminal for uploading advertising content, preprocessing means for converting the uploaded advertising content into text data, a generative artificial intelligence for reviewing the text data from a legal compliance perspective, means for notifying the user of the review results, an emotion engine for recognizing the user's emotions, and means for adjusting the notification content based on the user's emotional state. This makes it possible to provide notifications and feedback that take into account the user's emotional state while simultaneously reviewing for legal compliance.

[0209] A "user terminal" is a device that a user uses to upload advertising content.

[0210] "Preprocessing means" refers to a function that performs the necessary processing to convert uploaded advertisement content into text data.

[0211] "Generative artificial intelligence" refers to an artificial intelligence system that analyzes text data to review advertising content from a legal compliance perspective.

[0212] "Notification method" refers to a means of communication used to inform users of the review results, and includes email and in-system messages.

[0213] The "emotion engine" is a function that identifies the user's emotions and uses that data to inform the system's operations.

[0214] "Means for adjusting notification content" refers to a function that processes data to appropriately change the content and wording of notifications based on the user's emotional state.

[0215] This invention is a user-friendly system for reviewing whether advertising content violates relevant laws and regulations. Specifically, it is configured as a system comprising a user terminal, preprocessing means, generative artificial intelligence, means for notifying the user of the review results, and an emotion engine.

[0216] Use on user terminals

[0217] Users first log in to the system on their own devices. Logging in requires a user ID and password. Upon successful authentication, users gain access to an interface for uploading ad content. As users upload ad content, the system uses the device's camera and microphone to analyze facial expressions and voice, and an emotion engine recognizes the user's emotions in real time.

[0218] Server-side processing and preprocessing methods

[0219] The server receives the advertising content files uploaded from the user's terminal and saves them to the appropriate directory. Next, it determines the file format and converts it to text data as needed. In the case of image files, optical character recognition (OCR) technology is used to extract the text data. For example, a screenshot of an advertisement uploaded by a user is analyzed using OCR to obtain the text information within the image. In the case of HTML files, the server extracts the text portion from the HTML code.

[0220] Review by generative artificial intelligence

[0221] The text data obtained by the preprocessing means is sent to the generative artificial intelligence on the server side. The generative artificial intelligence analyzes the received text data and reviews the advertisement content based on the Act against Unjustifiable Premiums and Misleading Representations and the Pharmaceuticals and Medical Devices Act. If there is a risk of violating specific laws, it returns the evaluation result to the server. As a specific example of the review, if the expression "You will lose weight just by taking this supplement" is detected, it is determined that this may violate the Pharmaceuticals and Medical Devices Act and sends a specific risk assessment to the server.

[0222] Examples of prompts generated using AI

[0223] "Please review the following text data to determine if it violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceutical and Medical Devices Act. Please also provide specific points for users to modify the advertisement content."

[0224] Emotion engine and notifications

[0225] Upon receiving the review results, the server adjusts the notification content based on the user's emotional data, which is recognized in real time by the emotion engine. For example, if it determines that the user is experiencing stress, the review results will be communicated to the user via email or in-system messages, using milder language such as "Caution is advised."

[0226] Suggestions for improving advertising content

[0227] Furthermore, the emotion engine can generate suggestions for improving ad content when a user is feeling down, for example. Specifically, it might suggest "changing the ad content to be more positive." This suggestion is communicated to the user via a notification system from the server, and the user can modify the ad content based on it.

[0228] As described above, the system of this invention provides a more user-friendly review system by incorporating notifications and improvement suggestions that take user sentiment data into the legal compliance review of advertising content.

[0229] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0230] System processing steps

[0231] Step 1:

[0232] The user logs into the system.

[0233] Input: User ID and password

[0234] Process: The user enters their ID and password and clicks the login button. The terminal sends the authentication information to the server. The server authenticates the user based on the authentication information.

[0235] Output: Authentication success or failure

[0236] Step 2:

[0237] Users upload ad content.

[0238] Input: Authentication successful, ad file

[0239] Process: Upon successful login, the device displays an interface for uploading ad content. The user selects a file and clicks the upload button. The ad file is then sent from the device to the server.

[0240] Output: Uploaded ad file

[0241] Step 3:

[0242] The emotion engine recognizes the user's emotions.

[0243] Input: User's facial expression data, voice data

[0244] Processing: The system collects the user's facial expressions and voice in real time through the user's device's camera and microphone. The emotion engine analyzes this data to determine the user's emotional state.

[0245] Output: User emotion data (e.g., stress, depression, etc.)

[0246] Step 4:

[0247] The server receives the uploaded files.

[0248] Input: Uploaded ad file

[0249] Processing: The server receives the advertising file sent from the user's terminal and saves it to the appropriate directory.

[0250] Output: Path to the saved ad file

[0251] Step 5:

[0252] Identify the file format and perform the necessary conversions.

[0253] Input: Path to the saved ad file

[0254] Processing: The server determines the file format (image, HTML, etc.) and performs appropriate preprocessing. For image files, it uses OCR technology to extract text data. For HTML files, it extracts the text portion. For example, for image files, it obtains text information through OCR processing. For HTML files, it performs tag analysis and extracts text.

[0255] Output: Text data

[0256] Step 6:

[0257] The server sends the preprocessing results to the generative artificial intelligence.

[0258] Input: Text data

[0259] Processing: The server sends the text data obtained in the preprocessing stage to the generative artificial intelligence and requests review based on the prompt message.

[0260] Example prompt: "Please review the following text data to determine if it violates any laws or regulations. Please also provide specific points for the user to modify the advertisement content."

[0261] Output: Review Results

[0262] Step 7:

[0263] Generative artificial intelligence will perform the review.

[0264] Input: Text data, prompt text

[0265] Processing: A generative artificial intelligence analyzes the received text data and checks for parts that may violate the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act. The review results are returned to the server.

[0266] Output: Review results (including specific points of criticism)

[0267] Step 8:

[0268] The server receives the review results.

[0269] Input: Review Result

[0270] Processing: The server receives the review results sent from the generative artificial intelligence. The review results include specific points of concern and risks of legal violations.

[0271] Output: Saving of review results

[0272] Step 9:

[0273] The notification content is adjusted based on information from the emotion engine.

[0274] Input: Review results, user sentiment data

[0275] Processing: The server takes into account the user's emotional state, which the emotion engine has previously recognized, and adjusts the content and tone of the review results. For example, if the user is feeling stressed, it will use calmer language.

[0276] Output: Adjusted notification content

[0277] Step 10:

[0278] The server notifies the user of the review results.

[0279] Input: Adjusted notification content

[0280] Processing: The server notifies the user of the adjusted review results. Notification methods include email and in-system messages.

[0281] Output: Notification to the user

[0282] Step 11:

[0283] The emotion engine generates improvement suggestions for the advertisement content

[0284] Input: User's emotion data, review results

[0285] Process: Based on the user's emotional state, the emotion engine generates improvement suggestions for the advertisement content. For example, when the user is feeling down, it suggests "changing the advertisement content to a more positive expression".

[0286] Output: Improvement suggestions

[0287] Step 12:

[0288] Notify the user of the improvement suggestions

[0289] Input: Improvement suggestions

[0290] Process: The server summarizes the suggestions of the emotion engine and notifies the user. The user can refer to these suggestions to modify the advertisement content

[0291] Output: Notification to the user, improvement of the advertisement content

[0292] (Application Example 2)

[0293] Next, Application Example 2 will be described. In the following description, the data processing device 12 is referred to as the "server", and the smart device 14 is referred to as the "terminal".

[0294] [[ID=4३]] In recent years, compliance with advertising laws has become increasingly important. However, the review process for users to create advertisements that do not violate the laws is very complex and can be stressful for users. Also, there is an issue of insufficient feedback for improving the advertisement content in addition to simply complying with the laws. Furthermore, there is a need for a method that takes into account the user's emotions and provides a more user-friendly response in this process

[0295] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0296] In this invention, the server includes means for notifying the user of the review results using an emotion engine that recognizes the user's emotions, means for generating suggestions for improving the advertisement content, and preprocessing means. This enables efficient review of legal compliance of advertisements and the generation of improvement suggestions, as well as notifications that take the user's emotions into consideration.

[0297] A "user terminal" refers to a device operated by a user to input or upload advertising content, and generally refers to a smartphone, tablet, or computer.

[0298] "Preprocessing means" refers to a system or software technology for converting uploaded data into an appropriate format for analyzing advertising content, and includes optical character recognition (OCR) and HTML text extraction.

[0299] "Generative artificial intelligence" refers to artificial intelligence technology that analyzes advertising content and conducts reviews in accordance with laws and regulations such as the Premiums and Representations Act and the Pharmaceuticals and Medical Devices Act.

[0300] An "emotion engine" refers to a system or software technology that recognizes and analyzes a user's emotional state in real time from their facial expressions and voice.

[0301] "Notification means" refers to a system or software technology used to communicate review results or improvement suggestions to users, and includes email and in-system messages.

[0302] "Means for generating improvement suggestions" refers to a system or software technology that automatically analyzes areas for improvement in advertising content and provides suggestions to users.

[0303] Optical Character Recognition (OCR) is a technology that extracts text information contained within an image file and converts it into text data.

[0304] "HTML text extraction" refers to the technology of extracting text data from HTML-formatted files and converting it into an analyzable format.

[0305] Advertising review system

[0306] In addition to the legal compliance review of advertising content, the system of the present invention provides notifications and improvement proposals considering the emotions of users. The specific configuration and processing will be described below.

[0307] User terminal

[0308] It provides an interface for users to upload advertising content. Specifically, a smartphone, tablet, or computer is used. When a user logs in and uploads advertising content, the camera and microphone of the terminal are used to analyze the facial expressions and voices of the user in real time, and the emotion engine recognizes the emotions.

[0309] Server

[0310] The server receives the advertising content uploaded by the user and performs preprocessing. If the advertisement is an image file, it is converted into text data by optical character recognition (OCR) technology. If the advertisement is in HTML format, the text part is extracted and converted into text data. The text data obtained by this preprocessing is sent to the generative artificial intelligence.

[0311] Generative artificial intelligence

[0312] Based on the text data sent by the server, the generative artificial intelligence conducts a review from the perspective of legal compliance. It analyzes whether the advertising content contains expressions that violate the law or cause misunderstandings based on the Premium Display Law and the Pharmaceutical Affairs Law, and conducts a risk assessment. The result is returned to the server.

[0313] Emotion engine and notification means

[0314] The review results received from the generative artificial intelligence, along with the user's emotional data recognized by the emotion engine, are notified to the user via a notification system. If the user is experiencing stress, the review results are conveyed in milder language. Furthermore, if improvements are needed, suggestions for improving the ad content are also generated.

[0315] Hardware and software to use

[0316] Hardware: Cameras and microphones on smartphones, tablets, and computers.

[0317] software:

[0318] cv2 (OpenCV): Camera data capture

[0319] transformers (Hugging Face): Emotion recognition model

[0320] requests: HTTP requests for uploads and notifications

[0321] email_sender (custom module): Email notification

[0322] Example of a prompt

[0323] Examples of prompt statements are as follows:

[0324] Ad text: "You will lose weight just by taking this supplement."

[0325] Detected risk: Potential violation of the Pharmaceuticals and Medical Devices Act

[0326] Suggested revision: "This supplement is expected to be more effective when used in combination with exercise."

[0327] Based on this prompt, generative artificial intelligence can perform appropriate reviews and make improvement suggestions. This allows advertisers to create better advertisements while complying with the law.

[0328] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0329] Step 1:

[0330] Users upload ad content.

[0331] Users enter ad content using their devices and log in to the system. A user ID and password are required for login. Upon successful authentication, users gain access to an interface for uploading ad content.

[0332] Input: User ID, password, advertisement content (image or HTML file)

[0333] Output: Authentication success notification, ad content data

[0334] Step 2:

[0335] The user's device captures emotional data.

[0336] When a user uploads ad content, the device's camera and microphone are activated to analyze the user's facial expressions and voice in real time.

[0337] Input: User video, audio

[0338] Output: Sentiment data

[0339] Step 3:

[0340] Emotional data is analyzed by an emotion engine.

[0341] Real-time facial and voice data is sent to the emotion engine, which analyzes the user's emotional state.

[0342] Input: User video, audio

[0343] Output: Analyzed emotional data (e.g., stress, relaxation, etc.)

[0344] Step 4:

[0345] The advertisement content is uploaded to the server.

[0346] The device sends the advertisement content file to the server. The server saves this advertisement content to the appropriate directory.

[0347] Input: Ad content data

[0348] Output: File saved to server, file path

[0349] Step 5:

[0350] The server preprocesses the ad content.

[0351] The system identifies the format of uploaded files and converts them to text data as needed. For image files, it uses OCR technology; for HTML files, it extracts the text portion.

[0352] Input: Ad content file, file format

[0353] Output: Text data

[0354] Step 6:

[0355] Generative artificial intelligence will review the content of the advertisements.

[0356] The pre-processed text data is sent to a generative artificial intelligence system for review from a legal compliance perspective. Based on the Premiums and Representations Act and the Pharmaceuticals and Medical Devices Act, illegal or misleading expressions are detected, and a risk assessment is performed.

[0357] Input: Text data

[0358] Output: Review results, risk assessment

[0359] Step 7:

[0360] Based on the review results and sentiment data, the notification system informs the user of the outcome.

[0361] The server receives the evaluation results from the generative AI and emotional data from the emotion engine, and notifies the user in appropriate language. If the user is experiencing stress, the results are conveyed in calmer language.

[0362] Input: Review results, sentiment data

[0363] Output: Notification of review results (email or in-system message)

[0364] Step 8:

[0365] The server generates suggestions for improving the ad content.

[0366] Based on data from an emotion engine and generative artificial intelligence, it smoothly suggests improvements to ad content. For example, if a user is feeling down, it suggests revising the ad to use more positive language.

[0367] Input: Review results, sentiment data

[0368] Output: Improvement suggestions

[0369] The specific actions the system takes at each step were explained in detail. This ensures a clear understanding of the role of each processing step and the resulting flow of operations.

[0370] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0371] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0372] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0373] [Second Embodiment]

[0374] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0375] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0376] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0377] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0378] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0379] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0380] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0381] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0382] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0383] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0384] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0385] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. 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".

[0386] This invention is a system for reviewing whether advertising content violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act. Users upload advertising content, a server converts that content into text data, reviews it using generative artificial intelligence, and notifies the user of the results.

[0387] Processing at the user terminal

[0388] When a user logs in, they enter their user ID and password to access the system. Upon successful login, the user gains access to an interface for uploading advertising content. Once the user uploads the content, the file is sent to the server in the specified format.

[0389] Server-side preprocessing

[0390] The server receives files uploaded by users and saves them to the appropriate directory. Next, it determines the format of the uploaded file and converts it to text data if necessary. For example, it uses optical character recognition (OCR) technology to extract text from image files and extracts the text portion from HTML files.

[0391] Specific examples based on the server's processing targets:

[0392] For image files:

[0393] User-uploaded image files may contain screenshots of advertisements. The server analyzes these images using OCR technology and extracts text information from within them.

[0394] For HTML files:

[0395] The HTML file uploaded by the user contains an advertising landing page. The server extracts the text portion from the HTML file and analyzes it.

[0396] Review process using generative artificial intelligence

[0397] The converted text data is sent to a generative artificial intelligence system. The generative AI analyzes the text data and evaluates whether it violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act. The generative AI returns its evaluation results to the server. These results include details on which laws the advertisement content may violate, or whether it is otherwise problematic.

[0398] Examples of the review process:

[0399] If there is a conflict:

[0400] When a generative artificial intelligence detects a statement like "You will lose weight just by taking this supplement," it determines that the statement may violate the Pharmaceutical Affairs Law and returns a detailed assessment of the risk of such a violation.

[0401] Notification means

[0402] The server receives the review results from the generative artificial intelligence and notifies the user. Notification is sent via email or in-system message. Users can check whether their uploaded advertisements violate the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceutical Affairs Act, and identify any problematic aspects.

[0403] Examples of notifications:

[0404] The user receives an email stating that the review results indicate that the phrase "You can lose weight just by drinking this" may violate the Pharmaceutical Affairs Law. Based on this information, the user can modify the advertisement content.

[0405] In this way, the system of the present invention makes it possible to efficiently review advertising content for legal compliance and notify users of the results in a user-friendly manner.

[0406] The following describes the processing flow.

[0407] Step 1:

[0408] The user logs into the system. The user enters their user ID and password, and login authentication is performed. If authentication is successful, the user can access the interface for uploading ad content.

[0409] Step 2:

[0410] The user uploads the ad content. They select a text file, image file, or HTML file as the ad content and send it to the system. The file is sent to the server and saved in the specified directory.

[0411] Step 3:

[0412] The server receives the uploaded file and determines its file format. Here, it determines whether it is a text file, an image file, or an HTML file.

[0413] Step 4:

[0414] The server performs preprocessing according to the file format. If the advertisement content is a text file, it is used and saved as text data. If the advertisement content is an image file, optical character recognition (OCR) technology is used to extract the text within the image. If the advertisement content is an HTML file, the text portion is extracted from the HTML.

[0415] Step 5:

[0416] The server sends the pre-processed text data to a generative artificial intelligence (AI). The AI ​​analyzes the text data and reviews its content in accordance with the Act against Unjustifiable Premiums and Misleading Representations and the Pharmaceuticals and Medical Devices Act. Specifically, it evaluates whether it contains any illegal or misleading expressions.

[0417] Step 6:

[0418] The generative artificial intelligence returns the review results to the server. The review results include which parts may violate the law, along with the reasons why.

[0419] Step 7:

[0420] The server notifies the user of the review results. The review results are sent to the email address registered by the user or displayed as a system message. This allows the user to check whether the advertisement content poses a risk of violating the law.

[0421] Step 8:

[0422] The user receives the review results and modifies the ad content as needed. If modifications are required, the user can re-upload the ad content and repeat the same process until the issue is resolved.

[0423] (Example 1)

[0424] Next, we will describe Example 1. 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."

[0425] One challenge is the time-consuming process of reviewing advertisements to ensure they comply with laws such as the Premiums and Representations Act and the Pharmaceuticals and Medical Devices Act. Manual review is time-consuming and labor-intensive, and increases the risk of legal violations. Furthermore, because technical knowledge is required, there is a need for a system that anyone can easily use.

[0426] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0427] In this invention, the server includes terminal means for uploading advertising content, preprocessing means for converting the uploaded advertising content into text data, generative artificial intelligence means for reviewing the text data from a legal compliance perspective, and means for notifying the user of the review results. This makes it possible to automate the legal compliance review of advertising content and to notify the user of the review results quickly and accurately.

[0428] "Terminal means" refers to the device that a user uses to upload advertising content.

[0429] "Preprocessing means" refers to a device or software that performs processing to convert uploaded advertisement content into text data.

[0430] A "generative artificial intelligence method" is an artificial intelligence model that analyzes text data and performs reviews from a legal compliance perspective.

[0431] "Means for notifying users of the review results" refers to a method or system for informing users of the review results performed by a generative artificial intelligence.

[0432] "Optical character recognition technology" is a technology for extracting text information from image-based advertisements.

[0433] "Web page format" refers to advertising content expressed in formats such as HTML.

[0434] This invention is a system for reviewing whether advertising content violates the Premiums and Representations Act or the Pharmaceuticals and Medical Devices Act. The system works by having a user upload advertising content, a server converting that content into text data, a review being performed using generative artificial intelligence, and then notifying the user of the results.

[0435] Users access the system using their user terminal and log in by entering their user ID and password. After logging in, users select a file containing the advertisement content and upload it through the interface. This uploaded file is sent to the server in the specified format.

[0436] The server receives files sent by users and saves them to the appropriate directory. The server determines the format of the received files and converts them to text data as needed. For example, in the case of image files, optical character recognition (OCR) technology is used to extract the text information. One possible technique for this is using the Tesseract OCR engine. In the case of HTML files, libraries such as BeautifulSoup or html.parser are used to extract the text portion.

[0437] The converted text data is sent to a generative artificial intelligence (AI) system. This AI model is pre-trained to detect risks of violations of laws such as the Premiums and Representations Act and the Pharmaceuticals and Medical Devices Act. The AI ​​analyzes and evaluates the text data and returns the evaluation results to the server. These results include details on which laws the advertisement content may violate, or whether it is otherwise problematic.

[0438] For example, if the text data contains a phrase like "You will lose weight just by taking this supplement," the generative artificial intelligence will determine that this phrase may violate the Pharmaceutical Affairs Law and return that as the evaluation result.

[0439] The server notifies the user of the review results received from the generative artificial intelligence. Notification is sent via email or through an in-system message. Based on these results, the user can check whether the advertisement content complies with the law, identify any problems, and make corrections as needed.

[0440] Examples of prompt statements include the following:

[0441] "Please check if the copy in this supplement advertisement violates the Pharmaceutical Affairs Law. The copy says, 'You will lose weight just by taking this supplement.'"

[0442] The above describes a specific embodiment for carrying out this invention. This makes it possible to automate the legal compliance review of advertising content and notify users quickly and accurately.

[0443] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0444] Step 1:

[0445] The user accesses the system from their terminal and logs in by entering their user ID and password. Upon successful login, an interface for uploading advertising content is displayed. The user selects the file containing the advertising content and presses the upload button. This action sends the advertising content file from the user's terminal to the server. The input consists of the user ID, password, and the advertising content file, while the output is the advertising content file sent to the server.

[0446] Step 2:

[0447] The server receives files sent from the user's terminal and saves them to the appropriate directory. It identifies the format of the received file (image file, HTML file, etc.) and converts it to text data as needed. This process yields text data for further analysis. Specifically, it checks the file extension and MIME type. The input is the received file, and the output is text data.

[0448] Step 3:

[0449] For image files, the server extracts text information using Optical Character Recognition (OCR) technology. For example, the Tesseract OCR engine is used. For HTML files, libraries such as BeautifulSoup or html.parser are used to extract the text portion. The input is file data corresponding to the identified file format, and the output is the extracted text data. Specifically, for image files, the OCR engine is called to analyze the characters, and for HTML files, a text extraction library is used to parse the text portion.

[0450] Step 4:

[0451] The server sends the extracted text data to a generative artificial intelligence system. This generative AI is pre-trained to detect risks of violations related to the Premiums and Representations Act and the Pharmaceuticals and Medical Devices Act. The input is the extracted text data, and the output is the text data sent to the generative AI system.

[0452] Step 5:

[0453] The generative artificial intelligence system analyzes transmitted text data and evaluates whether it contains any expressions that violate laws and regulations. The evaluation results include detailed information on which specific parts may violate which laws, or whether there are any problems. The input is text data transmitted from the server, and the output is the evaluation result. For example, if the expression "You will lose weight just by taking this supplement" is detected, the evaluation result will be generated indicating that this expression may violate the Pharmaceuticals and Medical Devices Act.

[0454] Step 6:

[0455] The server notifies the user of the evaluation results received from the generative artificial intelligence system. Notification methods include generating and sending emails, or using in-system messages. Based on these evaluation results, the user can check whether the advertisement content complies with the law and identify any problematic areas. The input is the evaluation results from the generative artificial intelligence system, and the output is the notification to the user. Specifically, the server either sends an email based on the evaluation results or provides information to the user through the system's notification function.

[0456] The above outlines the processing steps of this system's program, providing a detailed explanation of the data processing and calculations performed at each step and the resulting outcomes.

[0457] (Application Example 1)

[0458] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0459] The goal is to reduce the risk of advertising content violating the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act, while also providing advertising creators and marketing personnel with a quick and efficient method for reviewing compliance with laws and regulations. Furthermore, there is a need to obtain the results of advertising content reviews in real time, and it is necessary to solve the problems of time and cost that conventional methods entail.

[0460] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0461] In this invention, the server includes a user terminal for uploading advertising content, preprocessing means for converting the uploaded advertising content into text data, a generative artificial intelligence for reviewing the text data from a legal compliance standpoint, means for notifying the user of the review results, and an application installed on a smartphone for taking photos of or uploading advertising content and providing review results in real time. This makes it possible to quickly review whether advertising content violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act and provide results in real time.

[0462] "Advertising content" refers to text, images, or other information created by users and offered to the public for marketing or promotional purposes.

[0463] A "user device" refers to an internet-connected device used to upload advertising content, and includes smartphones, tablets, and personal computers.

[0464] "Preprocessing means" refers to a function that performs the necessary processing to convert uploaded advertisement content into text data.

[0465] Optical Character Recognition (OCR) technology is a technology that analyzes text within an image and converts it into digital text.

[0466] "Generative artificial intelligence" refers to machine learning models and algorithms used to automatically review advertising content for compliance with various laws and regulations.

[0467] "Means of notifying users of the review results" refers to communication methods such as email, in-system messages, and push notifications used to notify users of the review results.

[0468] An "application installed on a smartphone" is software that runs on a smartphone and enables the shooting or uploading of advertising content and real-time review.

[0469] "Providing results in real time" means processing the review results immediately so that users can check the results without waiting.

[0470] This invention is a system for reviewing whether advertising content violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act. The system includes the following components:

[0471] 1. User terminal:

[0472] This provides an interface for users to log in and upload advertising content. Users can submit text, images, or HTML files of their advertisements to the server through this interface. For example, a user could take a picture of their advertisement content using their smartphone and upload that image.

[0473] 2. Pre-treatment means:

[0474] The server receives files uploaded by users and converts them to the appropriate format. This process uses optical character recognition (OCR) technology and HTML text extraction technology. For OCR technology, the "pytesseract" library is used to extract text from image files. For HTML files, an HTML parsing library such as "BeautifulSoup" is used to extract the text portion.

[0475] 3. Generative Artificial Intelligence:

[0476] The converted text data is sent to a generative artificial intelligence (AI) system. The AI ​​system uses the "transformers" library, and specifically the "BERT-base-japanese" model, to analyze the text data from a legal compliance perspective. The analysis results evaluate whether the advertising content may violate the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act.

[0477] 4. Means of notifying the results of the review:

[0478] The server notifies the user of the review results obtained from the generative artificial intelligence. Notification is sent via in-system messages or email. The user can review the review results and modify the advertisement content as needed.

[0479] This system allows users to quickly and efficiently review the legal compliance of advertising content. The program's processing is explained below in natural language.

[0480] When the server receives an advertisement image uploaded by a user, it first uses optical character recognition (OCR) technology to extract the text within the image. Next, it uses a generative artificial intelligence model (e.g., BERT-base-japanese) to analyze whether the extracted text violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act. The analysis results are then notified to the user as a review result.

[0481] As a concrete example, suppose a user takes a picture of an advertisement with their smartphone and uploads it. This image is sent to a server, and the text is extracted using OCR technology. A generative artificial intelligence model analyzes this text, and if it determines that the expression "You will lose weight just by taking this supplement" violates the Pharmaceuticals and Medical Devices Act, a notification is sent to the user stating that "This expression may violate the Pharmaceuticals and Medical Devices Act."

[0482] The following is an example of a prompt message:

[0483] "I took an image: 'advertisement_image.jpg'. Please extract and review the ad text: 'You will lose weight just by taking this supplement.' Please output the results and notify me."

[0484] In this way, the system of the present invention makes it possible to efficiently review the legal compliance of advertising content and notify users of the results in a user-friendly manner.

[0485] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0486] Step 1:

[0487] Users take pictures of advertisements with their smartphones and upload those images to a server through the application.

[0488] Input: Advertisement image file

[0489] Output: Advertisement image file sent to the server

[0490] Step 2:

[0491] The server receives the uploaded ad image files and saves them to the appropriate directory. Then, optical character recognition (OCR) technology is used to extract text from the images.

[0492] Input: Advertisement image file

[0493] Data processing: Text extraction using Optical Character Recognition (OCR) technology

[0494] Output: Text data

[0495] Step 3:

[0496] The server sends the extracted text data to a generative artificial intelligence model for analysis and review from a legal compliance perspective. The generative AI model used is "BERT-base-japanese" to determine whether the advertising content violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act.

[0497] Input: Text data

[0498] Data processing: Text analysis using generative artificial intelligence models

[0499] Output: Review result data

[0500] Step 4:

[0501] The server notifies the user of the evaluation results based on the evaluation result data obtained from the generative artificial intelligence model. Notification is sent via in-system messages or email.

[0502] Input: Review result data

[0503] Data processing: Formatting of review results and notification settings

[0504] Output: Notification to the user

[0505] Step 5:

[0506] Users can check the review results received via in-system message or email. If necessary, they can revise the ad content and request a review again.

[0507] Input: Notification of review results

[0508] Output: Confirmation of review results and possibility of modifying ad content as the next action.

[0509] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0510] This invention combines a system for reviewing whether advertising content violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act with an emotion engine that recognizes user emotions, thereby achieving a more user-friendly review and notification process. Specifically, the system comprises a user terminal, pre-processing means, generative artificial intelligence, means for notifying the user of the review results, and an emotion engine.

[0511] Processing at the user terminal

[0512] Users log in to the system and upload ad content. A user ID and password are required for login, and upon successful authentication, users can access the interface for uploading ad content. As users upload ad content, the system uses the camera and microphone on the user's device to analyze the user's facial expressions and voice in real time, and an emotion engine recognizes the user's emotions.

[0513] Server-side preprocessing

[0514] The server receives files uploaded by users and saves them to the appropriate directory. Next, it determines the format of the uploaded file and converts it to text data if necessary. For image files, optical character recognition (OCR) technology is used, and for HTML files, the text portion is extracted.

[0515] Specific examples based on the server's processing targets:

[0516] For image files:

[0517] If the user has uploaded a screenshot of an advertisement, the server will analyze the image using OCR technology and extract the text information within the image.

[0518] For HTML files:

[0519] If the user has uploaded an ad landing page, the server will extract the text portion from the HTML file and analyze it.

[0520] Review process using generative artificial intelligence

[0521] The text data obtained during preprocessing is then sent to a generative artificial intelligence (AI). The AI ​​analyzes the text data and conducts a review based on the Act against Unjustifiable Premiums and Misleading Representations and the Pharmaceuticals and Medical Devices Act. As a result, it performs a risk assessment for violations of the law and returns the review results to the server. Specifically, it evaluates whether the text contains illegal or misleading expressions.

[0522] Examples of the review process:

[0523] If there is a conflict:

[0524] If the generative artificial intelligence detects a statement such as "You will lose weight just by taking this supplement," it will determine that the statement may violate the Pharmaceutical Affairs Law and return a specific risk assessment result to the server.

[0525] Notification methods and emotion engines

[0526] The server receives the evaluation results from the generative artificial intelligence and, considering the user's emotional data recognized by the emotion engine, notifies the user of the evaluation results. Notification is sent via email or in-system message. For example, if the user is experiencing stress, the notification can be made using gentler language.

[0527] Examples of notifications:

[0528] If the user is feeling stressed:

[0529] If the emotion engine determines from the user's facial expressions and voice that they are experiencing stress, the server will notify the user via email or in-system message, using milder language such as "Caution advised." The user can then modify the ad content based on this information.

[0530] Suggestions for improving advertising content

[0531] The emotion engine can also generate suggestions for improving ad content based on user emotion data. For example, if a user is feeling down, the engine might suggest "changing the ad content to a more positive tone."

[0532] In this way, the system of the present invention makes it possible to efficiently perform legal compliance reviews of advertising content, provide notifications and improvement suggestions that take into account user sentiment data, and deliver results in a user-friendly manner.

[0533] The following describes the processing flow.

[0534] Step 1:

[0535] The user logs into the system. The user enters their user ID and password, and login authentication is performed. If authentication is successful, the user can access the interface for uploading ad content.

[0536] Step 2:

[0537] The user uploads the ad content. This ad content can be a text file, image file, or HTML file sent to the system. The file is sent to the server and saved in the specified directory.

[0538] Step 3:

[0539] The device's camera and microphone are used to analyze the user's facial expressions and voice in real time as they upload ad content. An emotion engine receives this data and recognizes the user's emotions.

[0540] Step 4:

[0541] The server receives the uploaded file and determines its file format. Here, it determines whether it is a text file, an image file, or an HTML file.

[0542] Step 5:

[0543] The server performs preprocessing according to the file format. If the advertisement content is a text file, it is saved as text data as is. If the advertisement content is an image file, it is converted to text data using optical character recognition (OCR) technology. If the advertisement content is an HTML file, the text portion is extracted from the HTML.

[0544] Step 6:

[0545] The server sends the pre-processed text data to the generative artificial intelligence. The generative AI analyzes the text data and reviews its content in accordance with the Act against Unjustifiable Premiums and Misleading Representations and the Pharmaceuticals and Medical Devices Act.

[0546] Step 7:

[0547] The generative artificial intelligence returns the review results to the server. The review results include which parts may violate the law, along with the reasons why.

[0548] Step 8:

[0549] The server notifies the user of the review results. The notification method and wording are adjusted based on the user's emotional data recognized by the emotion engine. Notification methods include email or in-system messages.

[0550] Step 9:

[0551] Users who receive the review results can check the ad content and make corrections as needed. By using the improvement suggestions provided by the emotion engine as a reference, users can re-upload the ad content and repeat the same process to create a legally compliant ad.

[0552] In this way, the system of the present invention makes it possible to efficiently perform legal compliance reviews of advertising content, provide notifications and improvement suggestions that take into account user sentiment data, and deliver results in a user-friendly manner.

[0553] (Example 2)

[0554] Next, we will describe Example 2. 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".

[0555] Traditional advertising content review systems check for legal compliance, but rarely provide notifications or feedback that take user emotions into consideration. As a result, users may experience stress or resentment, leading to a decline in the user experience.

[0556] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0557] In this invention, the server includes a user terminal for uploading advertising content, preprocessing means for converting the uploaded advertising content into text data, a generative artificial intelligence for reviewing the text data from a legal compliance perspective, means for notifying the user of the review results, an emotion engine for recognizing the user's emotions, and means for adjusting the notification content based on the user's emotional state. This makes it possible to provide notifications and feedback that take into account the user's emotional state while simultaneously reviewing for legal compliance.

[0558] A "user terminal" is a device that a user uses to upload advertising content.

[0559] "Preprocessing means" refers to a function that performs the necessary processing to convert uploaded advertisement content into text data.

[0560] "Generative artificial intelligence" refers to an artificial intelligence system that analyzes text data to review advertising content from a legal compliance perspective.

[0561] "Notification method" refers to a means of communication used to inform users of the review results, and includes email and in-system messages.

[0562] The "emotion engine" is a function that identifies the user's emotions and uses that data to inform the system's operations.

[0563] "Means for adjusting notification content" refers to a function that processes data to appropriately change the content and wording of notifications based on the user's emotional state.

[0564] This invention is a user-friendly system for reviewing whether advertising content violates relevant laws and regulations. Specifically, it is configured as a system comprising a user terminal, preprocessing means, generative artificial intelligence, means for notifying the user of the review results, and an emotion engine.

[0565] Use on user terminals

[0566] Users first log in to the system on their own devices. Logging in requires a user ID and password. Upon successful authentication, users gain access to an interface for uploading ad content. As users upload ad content, the system uses the device's camera and microphone to analyze facial expressions and voice, and an emotion engine recognizes the user's emotions in real time.

[0567] Server-side processing and preprocessing methods

[0568] The server receives the advertising content files uploaded from the user's terminal and saves them to the appropriate directory. Next, it determines the file format and converts it to text data as needed. In the case of image files, optical character recognition (OCR) technology is used to extract the text data. For example, a screenshot of an advertisement uploaded by a user is analyzed using OCR to obtain the text information within the image. In the case of HTML files, the server extracts the text portion from the HTML code.

[0569] Review by generative artificial intelligence

[0570] The text data obtained by the preprocessing means is sent to the generative artificial intelligence on the server side. The generative artificial intelligence analyzes the received text data and reviews the advertisement content based on the Act against Unjustifiable Premiums and Misleading Representations and the Pharmaceuticals and Medical Devices Act. If there is a risk of violating specific laws, it returns the evaluation result to the server. As a specific example of the review, if the expression "You will lose weight just by taking this supplement" is detected, it is determined that this may violate the Pharmaceuticals and Medical Devices Act and sends a specific risk assessment to the server.

[0571] Examples of prompts generated using AI

[0572] "Please review the following text data to determine if it violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceutical and Medical Devices Act. Please also provide specific points for users to modify the advertisement content."

[0573] Emotion engine and notifications

[0574] Upon receiving the review results, the server adjusts the notification content based on the user's emotional data, which is recognized in real time by the emotion engine. For example, if it determines that the user is experiencing stress, the review results will be communicated to the user via email or in-system messages, using milder language such as "Caution is advised."

[0575] Suggestions for improving advertising content

[0576] Furthermore, the emotion engine can also generate suggestions for improving ad content when a user is feeling down. Specifically, it might suggest "changing the ad content to be more positive." This suggestion is communicated to the user via a notification system from the server, and the user can modify the ad content based on it.

[0577] As described above, the system of this invention provides a more user-friendly review system by incorporating notifications and improvement suggestions that take user sentiment data into the legal compliance review of advertising content.

[0578] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0579] System processing steps

[0580] Step 1:

[0581] The user logs into the system.

[0582] Input: User ID and password

[0583] Process: The user enters their ID and password and clicks the login button. The terminal sends the authentication information to the server. The server authenticates the user based on the authentication information.

[0584] Output: Authentication success or failure

[0585] Step 2:

[0586] Users upload ad content.

[0587] Input: Authentication successful, ad file

[0588] Process: Upon successful login, the device displays an interface for uploading ad content. The user selects a file and clicks the upload button. The ad file is then sent from the device to the server.

[0589] Output: Uploaded ad file

[0590] Step 3:

[0591] The emotion engine recognizes the user's emotions.

[0592] Input: User's facial expression data, voice data

[0593] Processing: The system collects the user's facial expressions and voice in real time through the user's device's camera and microphone. The emotion engine analyzes this data to determine the user's emotional state.

[0594] Output: User emotion data (e.g., stress, depression, etc.)

[0595] Step 4:

[0596] The server receives the uploaded files.

[0597] Input: Uploaded ad file

[0598] Processing: The server receives the advertising file sent from the user's terminal and saves it to the appropriate directory.

[0599] Output: Path to the saved ad file

[0600] Step 5:

[0601] Identify the file format and perform the necessary conversions.

[0602] Input: Path to the saved ad file

[0603] Processing: The server determines the file format (image, HTML, etc.) and performs appropriate preprocessing. For image files, it uses OCR technology to extract text data. For HTML files, it extracts the text portion. For example, for image files, it obtains text information through OCR processing. For HTML files, it performs tag analysis and extracts text.

[0604] Output: Text data

[0605] Step 6:

[0606] The server sends the preprocessing results to the generative artificial intelligence.

[0607] Input: Text data

[0608] Processing: The server sends the text data obtained in the preprocessing stage to the generative artificial intelligence and requests review based on the prompt message.

[0609] Example prompt: "Please review the following text data to determine if it violates any laws or regulations. Please also provide specific points for the user to modify the advertisement content."

[0610] Output: Review Results

[0611] Step 7:

[0612] Generative artificial intelligence will perform the review.

[0613] Input: Text data, prompt text

[0614] Processing: A generative artificial intelligence analyzes the received text data and checks for parts that may violate the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act. The review results are returned to the server.

[0615] Output: Review results (including specific points of criticism)

[0616] Step 8:

[0617] The server receives the review results.

[0618] Input: Review Result

[0619] Processing: The server receives the review results sent from the generative artificial intelligence. The review results include specific points of concern and risks of legal violations.

[0620] Output: Saving of review results

[0621] Step 9:

[0622] The notification content is adjusted based on information from the emotion engine.

[0623] Input: Review results, user sentiment data

[0624] Processing: The server takes into account the user's emotional state, which the emotion engine has previously recognized, and adjusts the content and tone of the review results. For example, if the user is feeling stressed, it will use calmer language.

[0625] Output: Adjusted notification content

[0626] Step 10:

[0627] The server notifies the user of the review results.

[0628] Input: Adjusted notification content

[0629] Processing: The server notifies the user of the adjusted review results. Notification methods include email and in-system messages.

[0630] Output: Notification to the user

[0631] Step 11:

[0632] The emotion engine generates suggestions for improving ad content.

[0633] Input: User sentiment data, review results

[0634] Processing: The emotion engine generates suggestions for improving the ad content based on the user's emotional state. For example, if the user is feeling down, it might suggest "changing the ad content to a more positive tone."

[0635] Output: Improvement suggestions

[0636] Step 12:

[0637] Notify users of improvement suggestions.

[0638] Input: Improvement suggestion

[0639] Processing: The server compiles suggestions from the sentiment engine and notifies the user. The user can then use these suggestions to modify the ad content.

[0640] Output: User notifications, improved ad content

[0641] (Application Example 2)

[0642] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0643] In recent years, compliance with advertising regulations has become increasingly important. However, the review process for creating advertisements that do not violate the law is extremely complex and can be stressful for users. Furthermore, a lack of feedback for improving ad content, beyond simply complying with the law, is a challenge. In addition, there is a need for a more user-friendly approach that takes user emotions into consideration during this process.

[0644] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0645] In this invention, the server includes means for notifying the user of the review results using an emotion engine that recognizes the user's emotions, means for generating suggestions for improving the advertisement content, and preprocessing means. This enables efficient review of legal compliance of advertisements and the generation of improvement suggestions, as well as notifications that take the user's emotions into consideration.

[0646] A "user terminal" refers to a device operated by a user to input or upload advertising content, and generally refers to a smartphone, tablet, or computer.

[0647] "Preprocessing means" refers to a system or software technology for converting uploaded data into an appropriate format for analyzing advertising content, and includes optical character recognition (OCR) and HTML text extraction.

[0648] "Generative artificial intelligence" refers to artificial intelligence technology that analyzes advertising content and conducts reviews in accordance with laws and regulations such as the Premiums and Representations Act and the Pharmaceuticals and Medical Devices Act.

[0649] An "emotion engine" refers to a system or software technology that recognizes and analyzes a user's emotional state in real time from their facial expressions and voice.

[0650] "Notification means" refers to a system or software technology used to communicate review results or improvement suggestions to users, and includes email and in-system messages.

[0651] "Means for generating improvement suggestions" refers to a system or software technology that automatically analyzes areas for improvement in advertising content and provides suggestions to users.

[0652] Optical Character Recognition (OCR) is a technology that extracts text information contained within an image file and converts it into text data.

[0653] "HTML text extraction" is a technology that extracts text data from HTML files and converts it into a parseable format.

[0654] Ad review system

[0655] The system of this invention not only reviews advertising content for legal compliance but also provides notifications and improvement suggestions that take into account user emotions. Its specific configuration and processing are described below.

[0656] User terminal

[0657] It provides an interface for users to upload advertising content. Specifically, a smartphone, tablet, or computer is used. When a user logs in and uploads advertising content, the device's camera and microphone are used to analyze the user's facial expressions and voice in real time, and an emotion engine recognizes emotions.

[0658] server

[0659] The server receives the ad content uploaded by the user and performs preprocessing. If the ad is an image file, it is converted into text data using optical character recognition (OCR) technology. If the ad is in HTML format, the text portion is extracted and converted into text data. The text data obtained through this preprocessing is then sent to a generative artificial intelligence system.

[0660] Generative artificial intelligence

[0661] Based on the text data transmitted by the server, a generative artificial intelligence performs a review from a legal compliance perspective. It analyzes whether the advertisement content contains illegal or misleading expressions based on the Premiums and Representations Act and the Pharmaceuticals and Medical Devices Act, and performs a risk assessment. The results are then returned to the server.

[0662] Emotion engine and notification methods

[0663] The review results received from the generative artificial intelligence, along with the user's emotional data recognized by the emotion engine, are notified to the user via a notification system. If the user is experiencing stress, the review results are conveyed in milder language. Furthermore, if improvements are needed, suggestions for improving the ad content are also generated.

[0664] Hardware and software to use

[0665] Hardware: Cameras and microphones on smartphones, tablets, and computers.

[0666] software:

[0667] cv2 (OpenCV): Camera data capture

[0668] transformers (Hugging Face): Emotion recognition model

[0669] requests: HTTP requests for uploads and notifications

[0670] email_sender (custom module): Email notification

[0671] Example of a prompt

[0672] Examples of prompt statements are as follows:

[0673] Ad text: "You will lose weight just by taking this supplement."

[0674] Detected risk: Potential violation of the Pharmaceuticals and Medical Devices Act

[0675] Suggested revision: "This supplement is expected to be more effective when used in combination with exercise."

[0676] Based on this prompt, generative artificial intelligence can perform appropriate reviews and make improvement suggestions. This allows advertisers to create better advertisements while complying with the law.

[0677] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0678] Step 1:

[0679] Users upload ad content.

[0680] Users enter ad content using their devices and log in to the system. A user ID and password are required for login. Upon successful authentication, users gain access to an interface for uploading ad content.

[0681] Input: User ID, password, advertisement content (image or HTML file)

[0682] Output: Authentication success notification, ad content data

[0683] Step 2:

[0684] The user's device captures emotional data.

[0685] When a user uploads ad content, the device's camera and microphone are activated to analyze the user's facial expressions and voice in real time.

[0686] Input: User video, audio

[0687] Output: Sentiment data

[0688] Step 3:

[0689] Emotional data is analyzed by an emotion engine.

[0690] Real-time facial and voice data is sent to the emotion engine, which analyzes the user's emotional state.

[0691] Input: User video, audio

[0692] Output: Analyzed emotional data (e.g., stress, relaxation, etc.)

[0693] Step 4:

[0694] The advertisement content is uploaded to the server.

[0695] The device sends the advertisement content file to the server. The server saves this advertisement content to the appropriate directory.

[0696] Input: Ad content data

[0697] Output: File saved to server, file path

[0698] Step 5:

[0699] The server preprocesses the ad content.

[0700] The system identifies the format of uploaded files and converts them to text data as needed. For image files, it uses OCR technology; for HTML files, it extracts the text portion.

[0701] Input: Ad content file, file format

[0702] Output: Text data

[0703] Step 6:

[0704] Generative artificial intelligence will review the content of the advertisements.

[0705] The pre-processed text data is sent to a generative artificial intelligence system for review from a legal compliance perspective. Based on the Premiums and Representations Act and the Pharmaceuticals and Medical Devices Act, illegal or misleading expressions are detected, and a risk assessment is performed.

[0706] Input: Text data

[0707] Output: Review results, risk assessment

[0708] Step 7:

[0709] Based on the review results and sentiment data, the notification system informs the user of the outcome.

[0710] The server receives the evaluation results from the generative AI and emotional data from the emotion engine, and notifies the user in appropriate language. If the user is experiencing stress, the results are conveyed in calmer language.

[0711] Input: Review results, sentiment data

[0712] Output: Notification of review results (email or in-system message)

[0713] Step 8:

[0714] The server generates suggestions for improving the ad content.

[0715] Based on data from an emotion engine and generative artificial intelligence, it smoothly suggests improvements to ad content. For example, if a user is feeling down, it suggests revising the ad to use more positive language.

[0716] Input: Review results, sentiment data

[0717] Output: Improvement suggestions

[0718] The specific actions the system takes at each step were explained in detail. This ensures a clear understanding of the role of each processing step and the resulting flow of operations.

[0719] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0720] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0721] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0722] [Third Embodiment]

[0723] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0724] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0725] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0726] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0727] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0728] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0729] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0730] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0731] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0732] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0733] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0734] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0735] This invention is a system for reviewing whether advertising content violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act. Users upload advertising content, a server converts that content into text data, reviews it using generative artificial intelligence, and notifies the user of the results.

[0736] Processing at the user terminal

[0737] When a user logs in, they enter their user ID and password to access the system. Upon successful login, the user gains access to an interface for uploading advertising content. Once the user uploads the content, the file is sent to the server in the specified format.

[0738] Server-side preprocessing

[0739] The server receives files uploaded by users and saves them to the appropriate directory. Next, it determines the format of the uploaded file and converts it to text data if necessary. For example, it uses optical character recognition (OCR) technology to extract text from image files and extracts the text portion from HTML files.

[0740] Specific examples based on the server's processing targets:

[0741] For image files:

[0742] User-uploaded image files may contain screenshots of advertisements. The server analyzes these images using OCR technology and extracts text information from within them.

[0743] For HTML files:

[0744] The HTML file uploaded by the user contains an advertising landing page. The server extracts the text portion from the HTML file and analyzes it.

[0745] Review process using generative artificial intelligence

[0746] The converted text data is sent to a generative artificial intelligence system. The generative AI analyzes the text data and evaluates whether it violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act. The generative AI returns its evaluation results to the server. These results include details on which laws the advertisement content may violate, or whether it is otherwise problematic.

[0747] Examples of the review process:

[0748] If there is a conflict:

[0749] When a generative artificial intelligence detects a statement like "You will lose weight just by taking this supplement," it determines that the statement may violate the Pharmaceutical Affairs Law and returns a detailed assessment of the risk of such a violation.

[0750] Notification means

[0751] The server receives the review results from the generative artificial intelligence and notifies the user. Notification is sent via email or in-system message. Users can check whether their uploaded advertisements violate the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceutical Affairs Act, and identify any problematic aspects.

[0752] Examples of notifications:

[0753] The user receives an email stating that the review results indicate that the phrase "You can lose weight just by drinking this" may violate the Pharmaceutical Affairs Law. Based on this information, the user can modify the advertisement content.

[0754] In this way, the system of the present invention makes it possible to efficiently review advertising content for legal compliance and notify users of the results in a user-friendly manner.

[0755] The following describes the processing flow.

[0756] Step 1:

[0757] The user logs into the system. The user enters their user ID and password, and login authentication is performed. If authentication is successful, the user can access the interface for uploading ad content.

[0758] Step 2:

[0759] The user uploads the ad content. They select a text file, image file, or HTML file as the ad content and send it to the system. The file is sent to the server and saved in the specified directory.

[0760] Step 3:

[0761] The server receives the uploaded file and determines its file format. Here, it determines whether it is a text file, an image file, or an HTML file.

[0762] Step 4:

[0763] The server performs preprocessing according to the file format. If the advertisement content is a text file, it is used and saved as text data. If the advertisement content is an image file, optical character recognition (OCR) technology is used to extract the text within the image. If the advertisement content is an HTML file, the text portion is extracted from the HTML.

[0764] Step 5:

[0765] The server sends the pre-processed text data to a generative artificial intelligence (AI). The AI ​​analyzes the text data and reviews its content in accordance with the Act against Unjustifiable Premiums and Misleading Representations and the Pharmaceuticals and Medical Devices Act. Specifically, it evaluates whether it contains any illegal or misleading expressions.

[0766] Step 6:

[0767] The generative artificial intelligence returns the review results to the server. The review results include which parts may violate the law, along with the reasons why.

[0768] Step 7:

[0769] The server notifies the user of the review results. The review results are sent to the email address registered by the user or displayed as a system message. This allows the user to check whether the advertisement content poses a risk of violating the law.

[0770] Step 8:

[0771] The user receives the review results and modifies the ad content as needed. If modifications are required, the user can re-upload the ad content and repeat the same process until the issue is resolved.

[0772] (Example 1)

[0773] Next, we will describe Example 1. 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."

[0774] One challenge is the time-consuming process of reviewing advertisements to ensure they comply with laws such as the Premiums and Representations Act and the Pharmaceuticals and Medical Devices Act. Manual review is time-consuming and labor-intensive, and increases the risk of legal violations. Furthermore, because technical knowledge is required, there is a need for a system that anyone can easily use.

[0775] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0776] In this invention, the server includes terminal means for uploading advertising content, preprocessing means for converting the uploaded advertising content into text data, generative artificial intelligence means for reviewing the text data from a legal compliance perspective, and means for notifying the user of the review results. This makes it possible to automate the legal compliance review of advertising content and to notify the user of the review results quickly and accurately.

[0777] "Terminal means" refers to the device that a user uses to upload advertising content.

[0778] "Preprocessing means" refers to a device or software that performs processing to convert uploaded advertisement content into text data.

[0779] A "generative artificial intelligence method" is an artificial intelligence model that analyzes text data and performs reviews from a legal compliance perspective.

[0780] "Means for notifying users of the review results" refers to a method or system for informing users of the review results performed by a generative artificial intelligence.

[0781] "Optical character recognition technology" is a technology for extracting text information from image-based advertisements.

[0782] "Web page format" refers to advertising content expressed in formats such as HTML.

[0783] This invention is a system for reviewing whether advertising content violates the Premiums and Representations Act or the Pharmaceuticals and Medical Devices Act. The system works by having a user upload advertising content, a server converting that content into text data, a review being performed using generative artificial intelligence, and then notifying the user of the results.

[0784] Users access the system using their user terminal and log in by entering their user ID and password. After logging in, users select a file containing the advertisement content and upload it through the interface. This uploaded file is sent to the server in the specified format.

[0785] The server receives files sent by users and saves them to the appropriate directory. The server determines the format of the received files and converts them to text data as needed. For example, in the case of image files, optical character recognition (OCR) technology is used to extract the text information. One possible technique for this is using the Tesseract OCR engine. In the case of HTML files, libraries such as BeautifulSoup or html.parser are used to extract the text portion.

[0786] The converted text data is sent to a generative artificial intelligence (AI) system. This AI model is pre-trained to detect risks of violations of laws such as the Premiums and Representations Act and the Pharmaceuticals and Medical Devices Act. The AI ​​analyzes and evaluates the text data and returns the evaluation results to the server. These results include details on which laws the advertisement content may violate, or whether it is otherwise problematic.

[0787] For example, if the text data contains a phrase like "You will lose weight just by taking this supplement," the generative artificial intelligence will determine that this phrase may violate the Pharmaceutical Affairs Law and return that as the evaluation result.

[0788] The server notifies the user of the review results received from the generative artificial intelligence. Notification is sent via email or through an in-system message. Based on these results, the user can check whether the advertisement content complies with the law, identify any problems, and make corrections as needed.

[0789] Examples of prompt statements include the following:

[0790] "Please check if the copy in this supplement advertisement violates the Pharmaceutical Affairs Law. The copy says, 'You will lose weight just by taking this supplement.'"

[0791] The above describes a specific embodiment for carrying out this invention. This makes it possible to automate the legal compliance review of advertising content and notify users quickly and accurately.

[0792] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0793] Step 1:

[0794] The user accesses the system from their terminal and logs in by entering their user ID and password. Upon successful login, an interface for uploading advertising content is displayed. The user selects the file containing the advertising content and presses the upload button. This action sends the advertising content file from the user's terminal to the server. The input consists of the user ID, password, and the advertising content file, while the output is the advertising content file sent to the server.

[0795] Step 2:

[0796] The server receives files sent from the user's terminal and saves them to the appropriate directory. It identifies the format of the received file (image file, HTML file, etc.) and converts it to text data as needed. This process yields text data for further analysis. Specifically, it checks the file extension and MIME type. The input is the received file, and the output is text data.

[0797] Step 3:

[0798] For image files, the server extracts text information using Optical Character Recognition (OCR) technology. For example, the Tesseract OCR engine is used. For HTML files, libraries such as BeautifulSoup or html.parser are used to extract the text portion. The input is file data corresponding to the identified file format, and the output is the extracted text data. Specifically, for image files, the OCR engine is called to analyze the characters, and for HTML files, a text extraction library is used to parse the text portion.

[0799] Step 4:

[0800] The server sends the extracted text data to a generative artificial intelligence system. This generative AI is pre-trained to detect risks of violations related to the Premiums and Representations Act and the Pharmaceuticals and Medical Devices Act. The input is the extracted text data, and the output is the text data sent to the generative AI system.

[0801] Step 5:

[0802] The generative artificial intelligence system analyzes transmitted text data and evaluates whether it contains any expressions that violate laws and regulations. The evaluation results include detailed information on which specific parts may violate which laws, or whether there are any problems. The input is text data transmitted from the server, and the output is the evaluation result. For example, if the expression "You will lose weight just by taking this supplement" is detected, the evaluation result will be generated indicating that this expression may violate the Pharmaceuticals and Medical Devices Act.

[0803] Step 6:

[0804] The server notifies the user of the evaluation results received from the generative artificial intelligence system. Notification methods include generating and sending emails, or using in-system messages. Based on these evaluation results, the user can check whether the advertisement content complies with the law and identify any problematic areas. The input is the evaluation results from the generative artificial intelligence system, and the output is the notification to the user. Specifically, the server either sends an email based on the evaluation results or provides information to the user through the system's notification function.

[0805] The above outlines the processing steps of this system's program, providing a detailed explanation of the data processing and calculations performed at each step and the resulting outcomes.

[0806] (Application Example 1)

[0807] Next, we will explain Application Example 1. In the following explanation, 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."

[0808] The goal is to reduce the risk of advertising content violating the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act, while also providing advertising creators and marketing personnel with a quick and efficient method for reviewing compliance with laws and regulations. Furthermore, there is a need to obtain the results of advertising content reviews in real time, and it is necessary to solve the problems of time and cost that conventional methods entail.

[0809] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0810] In this invention, the server includes a user terminal for uploading advertising content, preprocessing means for converting the uploaded advertising content into text data, a generative artificial intelligence for reviewing the text data from a legal compliance standpoint, means for notifying the user of the review results, and an application installed on a smartphone for taking photos of or uploading advertising content and providing review results in real time. This makes it possible to quickly review whether advertising content violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act and provide results in real time.

[0811] "Advertising content" refers to text, images, or other information created by users and offered to the public for marketing or promotional purposes.

[0812] A "user device" refers to an internet-connected device used to upload advertising content, and includes smartphones, tablets, and personal computers.

[0813] "Preprocessing means" refers to a function that performs the necessary processing to convert uploaded advertisement content into text data.

[0814] Optical Character Recognition (OCR) technology is a technology that analyzes text within an image and converts it into digital text.

[0815] "Generative artificial intelligence" refers to machine learning models and algorithms used to automatically review advertising content for compliance with various laws and regulations.

[0816] "Means of notifying users of the review results" refers to communication methods such as email, in-system messages, and push notifications used to notify users of the review results.

[0817] An "application installed on a smartphone" is software that runs on a smartphone and enables the shooting or uploading of advertising content and real-time review.

[0818] "Providing results in real time" means processing the review results immediately so that users can check the results without waiting.

[0819] This invention is a system for reviewing whether advertising content violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act. The system includes the following components:

[0820] 1. User terminal:

[0821] This provides an interface for users to log in and upload advertising content. Users can submit text, images, or HTML files of their advertisements to the server through this interface. For example, a user could take a picture of their advertisement content using their smartphone and upload that image.

[0822] 2. Pre-treatment means:

[0823] The server receives files uploaded by users and converts them to the appropriate format. This process uses optical character recognition (OCR) technology and HTML text extraction technology. For OCR technology, the "pytesseract" library is used to extract text from image files. For HTML files, an HTML parsing library such as "BeautifulSoup" is used to extract the text portion.

[0824] 3. Generative Artificial Intelligence:

[0825] The converted text data is sent to a generative artificial intelligence (AI) system. The AI ​​system uses the "transformers" library, and specifically the "BERT-base-japanese" model, to analyze the text data from a legal compliance perspective. The analysis results evaluate whether the advertising content may violate the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act.

[0826] 4. Means of notifying the results of the review:

[0827] The server notifies the user of the review results obtained from the generative artificial intelligence. Notification is sent via in-system messages or email. The user can review the review results and modify the advertisement content as needed.

[0828] This system allows users to quickly and efficiently review the legal compliance of advertising content. The program's processing is explained below in natural language.

[0829] When the server receives an advertisement image uploaded by a user, it first uses optical character recognition (OCR) technology to extract the text within the image. Next, it uses a generative artificial intelligence model (e.g., BERT-base-japanese) to analyze whether the extracted text violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act. The analysis results are then notified to the user as a review result.

[0830] As a concrete example, suppose a user takes a picture of an advertisement with their smartphone and uploads it. This image is sent to a server, and the text is extracted using OCR technology. A generative artificial intelligence model analyzes this text, and if it determines that the expression "You will lose weight just by taking this supplement" violates the Pharmaceuticals and Medical Devices Act, a notification is sent to the user stating that "This expression may violate the Pharmaceuticals and Medical Devices Act."

[0831] The following is an example of a prompt message:

[0832] "I took an image: 'advertisement_image.jpg'. Please extract and review the ad text: 'You will lose weight just by taking this supplement.' Please output the results and notify me."

[0833] In this way, the system of the present invention makes it possible to efficiently review the legal compliance of advertising content and notify users of the results in a user-friendly manner.

[0834] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0835] Step 1:

[0836] Users take pictures of advertisements with their smartphones and upload those images to a server through the application.

[0837] Input: Advertisement image file

[0838] Output: Advertisement image file sent to the server

[0839] Step 2:

[0840] The server receives the uploaded ad image files and saves them to the appropriate directory. Then, optical character recognition (OCR) technology is used to extract text from the images.

[0841] Input: Advertisement image file

[0842] Data processing: Text extraction using Optical Character Recognition (OCR) technology

[0843] Output: Text data

[0844] Step 3:

[0845] The server sends the extracted text data to a generative artificial intelligence model for analysis and review from a legal compliance perspective. The generative AI model used is "BERT-base-japanese" to determine whether the advertising content violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act.

[0846] Input: Text data

[0847] Data processing: Text analysis using generative artificial intelligence models

[0848] Output: Review result data

[0849] Step 4:

[0850] The server notifies the user of the evaluation results based on the evaluation result data obtained from the generative artificial intelligence model. Notification is sent via in-system messages or email.

[0851] Input: Review result data

[0852] Data processing: Formatting of review results and notification settings

[0853] Output: Notification to the user

[0854] Step 5:

[0855] Users can check the review results received via in-system message or email. If necessary, they can revise the ad content and request a review again.

[0856] Input: Notification of review results

[0857] Output: Confirmation of review results and possibility of modifying ad content as the next action.

[0858] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0859] This invention combines a system for reviewing whether advertising content violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act with an emotion engine that recognizes user emotions, thereby achieving a more user-friendly review and notification process. Specifically, the system comprises a user terminal, pre-processing means, generative artificial intelligence, means for notifying the user of the review results, and an emotion engine.

[0860] Processing at the user terminal

[0861] Users log in to the system and upload ad content. A user ID and password are required for login, and upon successful authentication, users can access the interface for uploading ad content. As users upload ad content, the system uses the camera and microphone on the user's device to analyze the user's facial expressions and voice in real time, and an emotion engine recognizes the user's emotions.

[0862] Server-side preprocessing

[0863] The server receives files uploaded by users and saves them to the appropriate directory. Next, it determines the format of the uploaded file and converts it to text data if necessary. For image files, optical character recognition (OCR) technology is used, and for HTML files, the text portion is extracted.

[0864] Specific examples based on the server's processing targets:

[0865] For image files:

[0866] If the user has uploaded a screenshot of an advertisement, the server will analyze the image using OCR technology and extract the text information within the image.

[0867] For HTML files:

[0868] If the user has uploaded an ad landing page, the server will extract the text portion from the HTML file and analyze it.

[0869] Review process using generative artificial intelligence

[0870] The text data obtained during preprocessing is then sent to a generative artificial intelligence (AI). The AI ​​analyzes the text data and conducts a review based on the Act against Unjustifiable Premiums and Misleading Representations and the Pharmaceuticals and Medical Devices Act. As a result, it performs a risk assessment for violations of the law and returns the review results to the server. Specifically, it evaluates whether the text contains illegal or misleading expressions.

[0871] Examples of the review process:

[0872] If there is a conflict:

[0873] If the generative artificial intelligence detects a statement such as "You will lose weight just by taking this supplement," it will determine that the statement may violate the Pharmaceutical Affairs Law and return a specific risk assessment result to the server.

[0874] Notification methods and emotion engines

[0875] The server receives the evaluation results from the generative artificial intelligence and, considering the user's emotional data recognized by the emotion engine, notifies the user of the evaluation results. Notification is sent via email or in-system message. For example, if the user is experiencing stress, the notification can be made using gentler language.

[0876] Examples of notifications:

[0877] If the user is feeling stressed:

[0878] If the emotion engine determines from the user's facial expressions and voice that they are experiencing stress, the server will notify the user via email or in-system message, using milder language such as "Caution advised." The user can then modify the ad content based on this information.

[0879] Suggestions for improving advertising content

[0880] The emotion engine can also generate suggestions for improving ad content based on user emotion data. For example, if a user is feeling down, the engine might suggest "changing the ad content to a more positive tone."

[0881] In this way, the system of the present invention makes it possible to efficiently perform legal compliance reviews of advertising content, provide notifications and improvement suggestions that take into account user sentiment data, and deliver results in a user-friendly manner.

[0882] The following describes the processing flow.

[0883] Step 1:

[0884] The user logs into the system. The user enters their user ID and password, and login authentication is performed. If authentication is successful, the user can access the interface for uploading ad content.

[0885] Step 2:

[0886] The user uploads the ad content. This ad content can be a text file, image file, or HTML file sent to the system. The file is sent to the server and saved in the specified directory.

[0887] Step 3:

[0888] The device's camera and microphone are used to analyze the user's facial expressions and voice in real time as they upload ad content. An emotion engine receives this data and recognizes the user's emotions.

[0889] Step 4:

[0890] The server receives the uploaded file and determines its file format. Here, it determines whether it is a text file, an image file, or an HTML file.

[0891] Step 5:

[0892] The server performs preprocessing according to the file format. If the advertisement content is a text file, it is saved as text data as is. If the advertisement content is an image file, it is converted to text data using optical character recognition (OCR) technology. If the advertisement content is an HTML file, the text portion is extracted from the HTML.

[0893] Step 6:

[0894] The server sends the pre-processed text data to the generative artificial intelligence. The generative AI analyzes the text data and reviews its content in accordance with the Act against Unjustifiable Premiums and Misleading Representations and the Pharmaceuticals and Medical Devices Act.

[0895] Step 7:

[0896] The generative artificial intelligence returns the review results to the server. The review results include which parts may violate the law, along with the reasons why.

[0897] Step 8:

[0898] The server notifies the user of the review results. The notification method and wording are adjusted based on the user's emotional data recognized by the emotion engine. Notification methods include email or in-system messages.

[0899] Step 9:

[0900] Users who receive the review results can check the ad content and make corrections as needed. By using the improvement suggestions provided by the emotion engine as a reference, users can re-upload the ad content and repeat the same process to create a legally compliant ad.

[0901] In this way, the system of the present invention makes it possible to efficiently perform legal compliance reviews of advertising content, provide notifications and improvement suggestions that take into account user sentiment data, and deliver results in a user-friendly manner.

[0902] (Example 2)

[0903] Next, we will describe Example 2. 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."

[0904] Traditional advertising content review systems check for legal compliance, but rarely provide notifications or feedback that take user emotions into consideration. As a result, users may experience stress or resentment, leading to a decline in the user experience.

[0905] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0906] In this invention, the server includes a user terminal for uploading advertising content, preprocessing means for converting the uploaded advertising content into text data, a generative artificial intelligence for reviewing the text data from a legal compliance perspective, means for notifying the user of the review results, an emotion engine for recognizing the user's emotions, and means for adjusting the notification content based on the user's emotional state. This makes it possible to provide notifications and feedback that take into account the user's emotional state while simultaneously reviewing for legal compliance.

[0907] A "user terminal" is a device that a user uses to upload advertising content.

[0908] "Preprocessing means" refers to a function that performs the necessary processing to convert uploaded advertisement content into text data.

[0909] "Generative artificial intelligence" refers to an artificial intelligence system that analyzes text data to review advertising content from a legal compliance perspective.

[0910] "Notification method" refers to a means of communication used to inform users of the review results, and includes email and in-system messages.

[0911] The "emotion engine" is a function that identifies the user's emotions and uses that data to inform the system's operations.

[0912] "Means for adjusting notification content" refers to a function that processes data to appropriately change the content and wording of notifications based on the user's emotional state.

[0913] This invention is a user-friendly system for reviewing whether advertising content violates relevant laws and regulations. Specifically, it is configured as a system comprising a user terminal, preprocessing means, generative artificial intelligence, means for notifying the user of the review results, and an emotion engine.

[0914] Use on user terminals

[0915] Users first log in to the system on their own devices. Logging in requires a user ID and password. Upon successful authentication, users gain access to an interface for uploading ad content. As users upload ad content, the system uses the device's camera and microphone to analyze facial expressions and voice, and an emotion engine recognizes the user's emotions in real time.

[0916] Server-side processing and preprocessing methods

[0917] The server receives the advertising content files uploaded from the user's terminal and saves them to the appropriate directory. Next, it determines the file format and converts it to text data as needed. In the case of image files, optical character recognition (OCR) technology is used to extract the text data. For example, a screenshot of an advertisement uploaded by a user is analyzed using OCR to obtain the text information within the image. In the case of HTML files, the server extracts the text portion from the HTML code.

[0918] Review by generative artificial intelligence

[0919] The text data obtained by the preprocessing means is sent to the generative artificial intelligence on the server side. The generative artificial intelligence analyzes the received text data and reviews the advertisement content based on the Act against Unjustifiable Premiums and Misleading Representations and the Pharmaceuticals and Medical Devices Act. If there is a risk of violating specific laws, it returns the evaluation result to the server. As a specific example of the review, if the expression "You will lose weight just by taking this supplement" is detected, it is determined that this may violate the Pharmaceuticals and Medical Devices Act and sends a specific risk assessment to the server.

[0920] Examples of prompts generated using AI

[0921] "Please review the following text data to determine if it violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceutical and Medical Devices Act. Please also provide specific points for users to modify the advertisement content."

[0922] Emotion engine and notifications

[0923] Upon receiving the review results, the server adjusts the notification content based on the user's emotional data, which is recognized in real time by the emotion engine. For example, if it determines that the user is experiencing stress, the review results will be communicated to the user via email or in-system messages, using milder language such as "Caution is advised."

[0924] Suggestions for improving advertising content

[0925] Furthermore, the emotion engine can also generate suggestions for improving ad content when a user is feeling down. Specifically, it might suggest "changing the ad content to be more positive." This suggestion is communicated to the user via a notification system from the server, and the user can modify the ad content based on it.

[0926] As described above, the system of this invention provides a more user-friendly review system by incorporating notifications and improvement suggestions that take user sentiment data into the legal compliance review of advertising content.

[0927] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0928] System processing steps

[0929] Step 1:

[0930] The user logs into the system.

[0931] Input: User ID and password

[0932] Process: The user enters their ID and password and clicks the login button. The terminal sends the authentication information to the server. The server authenticates the user based on the authentication information.

[0933] Output: Authentication success or failure

[0934] Step 2:

[0935] Users upload ad content.

[0936] Input: Authentication successful, ad file

[0937] Process: Upon successful login, the device displays an interface for uploading ad content. The user selects a file and clicks the upload button. The ad file is then sent from the device to the server.

[0938] Output: Uploaded ad file

[0939] Step 3:

[0940] The emotion engine recognizes the user's emotions.

[0941] Input: User's facial expression data, voice data

[0942] Processing: The system collects the user's facial expressions and voice in real time through the user's device's camera and microphone. The emotion engine analyzes this data to determine the user's emotional state.

[0943] Output: User emotion data (e.g., stress, depression, etc.)

[0944] Step 4:

[0945] The server receives the uploaded files.

[0946] Input: Uploaded ad file

[0947] Processing: The server receives the advertising file sent from the user's terminal and saves it to the appropriate directory.

[0948] Output: Path to the saved ad file

[0949] Step 5:

[0950] Identify the file format and perform the necessary conversions.

[0951] Input: Path to the saved ad file

[0952] Processing: The server determines the file format (image, HTML, etc.) and performs appropriate preprocessing. For image files, it uses OCR technology to extract text data. For HTML files, it extracts the text portion. For example, for image files, it obtains text information through OCR processing. For HTML files, it performs tag analysis and extracts text.

[0953] Output: Text data

[0954] Step 6:

[0955] The server sends the preprocessing results to the generative artificial intelligence.

[0956] Input: Text data

[0957] Processing: The server sends the text data obtained in the preprocessing stage to the generative artificial intelligence and requests review based on the prompt message.

[0958] Example prompt: "Please review the following text data to determine if it violates any laws or regulations. Please also provide specific points for the user to modify the advertisement content."

[0959] Output: Review Results

[0960] Step 7:

[0961] Generative artificial intelligence will perform the review.

[0962] Input: Text data, prompt text

[0963] Processing: A generative artificial intelligence analyzes the received text data and checks for parts that may violate the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act. The review results are returned to the server.

[0964] Output: Review results (including specific points of criticism)

[0965] Step 8:

[0966] The server receives the review results.

[0967] Input: Review Result

[0968] Processing: The server receives the review results sent from the generative artificial intelligence. The review results include specific points of concern and risks of legal violations.

[0969] Output: Saving of review results

[0970] Step 9:

[0971] The notification content is adjusted based on information from the emotion engine.

[0972] Input: Review results, user sentiment data

[0973] Processing: The server takes into account the user's emotional state, which the emotion engine has previously recognized, and adjusts the content and tone of the review results. For example, if the user is feeling stressed, it will use calmer language.

[0974] Output: Adjusted notification content

[0975] Step 10:

[0976] The server notifies the user of the review results.

[0977] Input: Adjusted notification content

[0978] Processing: The server notifies the user of the adjusted review results. Notification methods include email and in-system messages.

[0979] Output: Notification to the user

[0980] Step 11:

[0981] The emotion engine generates suggestions for improving ad content.

[0982] Input: User sentiment data, review results

[0983] Processing: The emotion engine generates suggestions for improving the ad content based on the user's emotional state. For example, if the user is feeling down, it might suggest "changing the ad content to a more positive tone."

[0984] Output: Improvement suggestions

[0985] Step 12:

[0986] Notify users of improvement suggestions.

[0987] Input: Improvement suggestion

[0988] Processing: The server compiles suggestions from the sentiment engine and notifies the user. The user can then use these suggestions to modify the ad content.

[0989] Output: User notifications, improved ad content

[0990] (Application Example 2)

[0991] Next, we will explain application example 2. In the following explanation, 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."

[0992] In recent years, compliance with advertising regulations has become increasingly important. However, the review process for creating advertisements that do not violate the law is extremely complex and can be stressful for users. Furthermore, a lack of feedback for improving ad content, beyond simply complying with the law, is a challenge. In addition, there is a need for a more user-friendly approach that takes user emotions into consideration during this process.

[0993] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0994] In this invention, the server includes means for notifying the user of the review results using an emotion engine that recognizes the user's emotions, means for generating suggestions for improving the advertisement content, and preprocessing means. This enables efficient review of legal compliance of advertisements and the generation of improvement suggestions, as well as notifications that take the user's emotions into consideration.

[0995] A "user terminal" refers to a device operated by a user to input or upload advertising content, and generally refers to a smartphone, tablet, or computer.

[0996] "Preprocessing means" refers to a system or software technology for converting uploaded data into an appropriate format for analyzing advertising content, and includes optical character recognition (OCR) and HTML text extraction.

[0997] "Generative artificial intelligence" refers to artificial intelligence technology that analyzes advertising content and conducts reviews in accordance with laws and regulations such as the Premiums and Representations Act and the Pharmaceuticals and Medical Devices Act.

[0998] An "emotion engine" refers to a system or software technology that recognizes and analyzes a user's emotional state in real time from their facial expressions and voice.

[0999] "Notification means" refers to a system or software technology used to communicate review results or improvement suggestions to users, and includes email and in-system messages.

[1000] "Means for generating improvement suggestions" refers to a system or software technology that automatically analyzes areas for improvement in advertising content and provides suggestions to users.

[1001] Optical Character Recognition (OCR) is a technology that extracts text information contained within an image file and converts it into text data.

[1002] "HTML text extraction" is a technology that extracts text data from HTML files and converts it into a parseable format.

[1003] Ad review system

[1004] The system of this invention not only reviews advertising content for legal compliance but also provides notifications and improvement suggestions that take into account user emotions. Its specific configuration and processing are described below.

[1005] User terminal

[1006] It provides an interface for users to upload advertising content. Specifically, a smartphone, tablet, or computer is used. When a user logs in and uploads advertising content, the device's camera and microphone are used to analyze the user's facial expressions and voice in real time, and an emotion engine recognizes emotions.

[1007] server

[1008] The server receives the ad content uploaded by the user and performs preprocessing. If the ad is an image file, it is converted into text data using optical character recognition (OCR) technology. If the ad is in HTML format, the text portion is extracted and converted into text data. The text data obtained through this preprocessing is then sent to a generative artificial intelligence system.

[1009] Generative artificial intelligence

[1010] Based on the text data transmitted by the server, a generative artificial intelligence performs a review from a legal compliance perspective. It analyzes whether the advertisement content contains illegal or misleading expressions based on the Premiums and Representations Act and the Pharmaceuticals and Medical Devices Act, and performs a risk assessment. The results are then returned to the server.

[1011] Emotion engine and notification methods

[1012] The review results received from the generative artificial intelligence, along with the user's emotional data recognized by the emotion engine, are notified to the user via a notification system. If the user is experiencing stress, the review results are conveyed in milder language. Furthermore, if improvements are needed, suggestions for improving the ad content are also generated.

[1013] Hardware and software to use

[1014] Hardware: Cameras and microphones on smartphones, tablets, and computers.

[1015] software:

[1016] cv2 (OpenCV): Camera data capture

[1017] transformers (Hugging Face): Emotion recognition model

[1018] requests: HTTP requests for uploads and notifications

[1019] email_sender (custom module): Email notification

[1020] Example of a prompt

[1021] Examples of prompt statements are as follows:

[1022] Ad text: "You will lose weight just by taking this supplement."

[1023] Detected risk: Potential violation of the Pharmaceuticals and Medical Devices Act

[1024] Suggested revision: "This supplement is expected to be more effective when used in combination with exercise."

[1025] Based on this prompt, generative artificial intelligence can perform appropriate reviews and make improvement suggestions. This allows advertisers to create better advertisements while complying with the law.

[1026] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1027] Step 1:

[1028] Users upload ad content.

[1029] Users enter ad content using their devices and log in to the system. A user ID and password are required for login. Upon successful authentication, users gain access to an interface for uploading ad content.

[1030] Input: User ID, password, advertisement content (image or HTML file)

[1031] Output: Authentication success notification, ad content data

[1032] Step 2:

[1033] The user's device captures emotional data.

[1034] When a user uploads ad content, the device's camera and microphone are activated to analyze the user's facial expressions and voice in real time.

[1035] Input: User video, audio

[1036] Output: Sentiment data

[1037] Step 3:

[1038] Emotional data is analyzed by an emotion engine.

[1039] Real-time facial and voice data is sent to the emotion engine, which analyzes the user's emotional state.

[1040] Input: User video, audio

[1041] Output: Analyzed emotional data (e.g., stress, relaxation, etc.)

[1042] Step 4:

[1043] The advertisement content is uploaded to the server.

[1044] The device sends the advertisement content file to the server. The server saves this advertisement content to the appropriate directory.

[1045] Input: Ad content data

[1046] Output: File saved to server, file path

[1047] Step 5:

[1048] The server preprocesses the ad content.

[1049] The system identifies the format of uploaded files and converts them to text data as needed. For image files, it uses OCR technology; for HTML files, it extracts the text portion.

[1050] Input: Ad content file, file format

[1051] Output: Text data

[1052] Step 6:

[1053] Generative artificial intelligence will review the content of the advertisements.

[1054] The pre-processed text data is sent to a generative artificial intelligence system for review from a legal compliance perspective. Based on the Premiums and Representations Act and the Pharmaceuticals and Medical Devices Act, illegal or misleading expressions are detected, and a risk assessment is performed.

[1055] Input: Text data

[1056] Output: Review results, risk assessment

[1057] Step 7:

[1058] Based on the review results and sentiment data, the notification system informs the user of the outcome.

[1059] The server receives the evaluation results from the generative AI and emotional data from the emotion engine, and notifies the user in appropriate language. If the user is experiencing stress, the results are conveyed in calmer language.

[1060] Input: Review results, sentiment data

[1061] Output: Notification of review results (email or in-system message)

[1062] Step 8:

[1063] The server generates suggestions for improving the ad content.

[1064] Based on data from an emotion engine and generative artificial intelligence, it smoothly suggests improvements to ad content. For example, if a user is feeling down, it suggests revising the ad to use more positive language.

[1065] Input: Review results, sentiment data

[1066] Output: Improvement suggestions

[1067] The specific actions the system takes at each step were explained in detail. This ensures a clear understanding of the role of each processing step and the resulting flow of operations.

[1068] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1069] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1070] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[1071] [Fourth Embodiment]

[1072] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1073] As shown in Figure 7, the 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.

[1074] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1075] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[1076] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1077] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1078] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1079] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive 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 robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[1080] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1081] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[1082] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1083] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1084] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1085] This invention is a system for reviewing whether advertising content violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act. Users upload advertising content, a server converts that content into text data, reviews it using generative artificial intelligence, and notifies the user of the results.

[1086] Processing at the user terminal

[1087] When a user logs in, they enter their user ID and password to access the system. Upon successful login, the user gains access to an interface for uploading advertising content. Once the user uploads the content, the file is sent to the server in the specified format.

[1088] Server-side preprocessing

[1089] The server receives files uploaded by users and saves them to the appropriate directory. Next, it determines the format of the uploaded file and converts it to text data if necessary. For example, it uses optical character recognition (OCR) technology to extract text from image files and extracts the text portion from HTML files.

[1090] Specific examples based on the server's processing targets:

[1091] For image files:

[1092] User-uploaded image files may contain screenshots of advertisements. The server analyzes these images using OCR technology and extracts text information from within them.

[1093] For HTML files:

[1094] The HTML file uploaded by the user contains an advertising landing page. The server extracts the text portion from the HTML file and analyzes it.

[1095] Review process using generative artificial intelligence

[1096] The converted text data is sent to a generative artificial intelligence system. The generative AI analyzes the text data and evaluates whether it violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act. The generative AI returns its evaluation results to the server. These results include details on which laws the advertisement content may violate, or whether it is otherwise problematic.

[1097] Examples of the review process:

[1098] If there is a conflict:

[1099] When a generative artificial intelligence detects a statement like "You will lose weight just by taking this supplement," it determines that the statement may violate the Pharmaceutical Affairs Law and returns a detailed assessment of the risk of such a violation.

[1100] Notification means

[1101] The server receives the review results from the generative artificial intelligence and notifies the user. Notification is sent via email or in-system message. Users can check whether their uploaded advertisements violate the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceutical Affairs Act, and identify any problematic aspects.

[1102] Examples of notifications:

[1103] The user receives an email stating that the review results indicate that the phrase "You can lose weight just by drinking this" may violate the Pharmaceutical Affairs Law. Based on this information, the user can modify the advertisement content.

[1104] In this way, the system of the present invention makes it possible to efficiently review advertising content for legal compliance and notify users of the results in a user-friendly manner.

[1105] The following describes the processing flow.

[1106] Step 1:

[1107] The user logs into the system. The user enters their user ID and password, and login authentication is performed. If authentication is successful, the user can access the interface for uploading ad content.

[1108] Step 2:

[1109] The user uploads the ad content. They select a text file, image file, or HTML file as the ad content and send it to the system. The file is sent to the server and saved in the specified directory.

[1110] Step 3:

[1111] The server receives the uploaded file and determines its file format. Here, it determines whether it is a text file, an image file, or an HTML file.

[1112] Step 4:

[1113] The server performs preprocessing according to the file format. If the advertisement content is a text file, it is used and saved as text data. If the advertisement content is an image file, optical character recognition (OCR) technology is used to extract the text within the image. If the advertisement content is an HTML file, the text portion is extracted from the HTML.

[1114] Step 5:

[1115] The server sends the pre-processed text data to a generative artificial intelligence (AI). The AI ​​analyzes the text data and reviews its content in accordance with the Act against Unjustifiable Premiums and Misleading Representations and the Pharmaceuticals and Medical Devices Act. Specifically, it evaluates whether it contains any illegal or misleading expressions.

[1116] Step 6:

[1117] The generative artificial intelligence returns the review results to the server. The review results include which parts may violate the law, along with the reasons why.

[1118] Step 7:

[1119] The server notifies the user of the review results. The review results are sent to the email address registered by the user or displayed as a system message. This allows the user to check whether the advertisement content poses a risk of violating the law.

[1120] Step 8:

[1121] The user receives the review results and modifies the ad content as needed. If modifications are required, the user can re-upload the ad content and repeat the same process until the issue is resolved.

[1122] (Example 1)

[1123] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1124] One challenge is the time-consuming process of reviewing advertisements to ensure they comply with laws such as the Premiums and Representations Act and the Pharmaceuticals and Medical Devices Act. Manual review is time-consuming and labor-intensive, and increases the risk of legal violations. Furthermore, because technical knowledge is required, there is a need for a system that anyone can easily use.

[1125] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1126] In this invention, the server includes terminal means for uploading advertising content, preprocessing means for converting the uploaded advertising content into text data, generative artificial intelligence means for reviewing the text data from a legal compliance perspective, and means for notifying the user of the review results. This makes it possible to automate the legal compliance review of advertising content and to notify the user of the review results quickly and accurately.

[1127] "Terminal means" refers to the device that a user uses to upload advertising content.

[1128] "Preprocessing means" refers to a device or software that performs processing to convert uploaded advertisement content into text data.

[1129] A "generative artificial intelligence method" is an artificial intelligence model that analyzes text data and performs reviews from a legal compliance perspective.

[1130] "Means for notifying users of the review results" refers to a method or system for informing users of the review results performed by a generative artificial intelligence.

[1131] "Optical character recognition technology" is a technology for extracting text information from image-based advertisements.

[1132] "Web page format" refers to advertising content expressed in formats such as HTML.

[1133] This invention is a system for reviewing whether advertising content violates the Premiums and Representations Act or the Pharmaceuticals and Medical Devices Act. The system works by having a user upload advertising content, a server converting that content into text data, a review being performed using generative artificial intelligence, and then notifying the user of the results.

[1134] Users access the system using their user terminal and log in by entering their user ID and password. After logging in, users select a file containing the advertisement content and upload it through the interface. This uploaded file is sent to the server in the specified format.

[1135] The server receives files sent by users and saves them to the appropriate directory. The server determines the format of the received files and converts them to text data as needed. For example, in the case of image files, optical character recognition (OCR) technology is used to extract the text information. One possible technique for this is using the Tesseract OCR engine. In the case of HTML files, libraries such as BeautifulSoup or html.parser are used to extract the text portion.

[1136] The converted text data is sent to a generative artificial intelligence (AI) system. This AI model is pre-trained to detect risks of violations of laws such as the Premiums and Representations Act and the Pharmaceuticals and Medical Devices Act. The AI ​​analyzes and evaluates the text data and returns the evaluation results to the server. These results include details on which laws the advertisement content may violate, or whether it is otherwise problematic.

[1137] For example, if the text data contains a phrase like "You will lose weight just by taking this supplement," the generative artificial intelligence will determine that this phrase may violate the Pharmaceutical Affairs Law and return that as the evaluation result.

[1138] The server notifies the user of the review results received from the generative artificial intelligence. Notification is sent via email or through an in-system message. Based on these results, the user can check whether the advertisement content complies with the law, identify any problems, and make corrections as needed.

[1139] Examples of prompt statements include the following:

[1140] "Please check if the copy in this supplement advertisement violates the Pharmaceutical Affairs Law. The copy says, 'You will lose weight just by taking this supplement.'"

[1141] The above describes a specific embodiment for carrying out this invention. This makes it possible to automate the legal compliance review of advertising content and notify users quickly and accurately.

[1142] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1143] Step 1:

[1144] The user accesses the system from their terminal and logs in by entering their user ID and password. Upon successful login, an interface for uploading advertising content is displayed. The user selects the file containing the advertising content and presses the upload button. This action sends the advertising content file from the user's terminal to the server. The input consists of the user ID, password, and the advertising content file, while the output is the advertising content file sent to the server.

[1145] Step 2:

[1146] The server receives files sent from the user's terminal and saves them to the appropriate directory. It identifies the format of the received file (image file, HTML file, etc.) and converts it to text data as needed. This process yields text data for further analysis. Specifically, it checks the file extension and MIME type. The input is the received file, and the output is text data.

[1147] Step 3:

[1148] For image files, the server extracts text information using Optical Character Recognition (OCR) technology. For example, the Tesseract OCR engine is used. For HTML files, libraries such as BeautifulSoup or html.parser are used to extract the text portion. The input is file data corresponding to the identified file format, and the output is the extracted text data. Specifically, for image files, the OCR engine is called to analyze the characters, and for HTML files, a text extraction library is used to parse the text portion.

[1149] Step 4:

[1150] The server sends the extracted text data to a generative artificial intelligence system. This generative AI is pre-trained to detect risks of violations related to the Premiums and Representations Act and the Pharmaceuticals and Medical Devices Act. The input is the extracted text data, and the output is the text data sent to the generative AI system.

[1151] Step 5:

[1152] The generative artificial intelligence system analyzes transmitted text data and evaluates whether it contains any expressions that violate laws and regulations. The evaluation results include detailed information on which specific parts may violate which laws, or whether there are any problems. The input is text data transmitted from the server, and the output is the evaluation result. For example, if the expression "You will lose weight just by taking this supplement" is detected, the evaluation result will be generated indicating that this expression may violate the Pharmaceuticals and Medical Devices Act.

[1153] Step 6:

[1154] The server notifies the user of the evaluation results received from the generative artificial intelligence system. Notification methods include generating and sending emails, or using in-system messages. Based on these evaluation results, the user can check whether the advertisement content complies with the law and identify any problematic areas. The input is the evaluation results from the generative artificial intelligence system, and the output is the notification to the user. Specifically, the server either sends an email based on the evaluation results or provides information to the user through the system's notification function.

[1155] The above outlines the processing steps of this system's program, providing a detailed explanation of the data processing and calculations performed at each step and the resulting outcomes.

[1156] (Application Example 1)

[1157] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1158] The goal is to reduce the risk of advertising content violating the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act, while also providing advertising creators and marketing personnel with a quick and efficient method for reviewing compliance with laws and regulations. Furthermore, there is a need to obtain the results of advertising content reviews in real time, and it is necessary to solve the problems of time and cost that conventional methods entail.

[1159] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1160] In this invention, the server includes a user terminal for uploading advertising content, preprocessing means for converting the uploaded advertising content into text data, a generative artificial intelligence for reviewing the text data from a legal compliance standpoint, means for notifying the user of the review results, and an application installed on a smartphone for taking photos of or uploading advertising content and providing review results in real time. This makes it possible to quickly review whether advertising content violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act and provide results in real time.

[1161] "Advertising content" refers to text, images, or other information created by users and offered to the public for marketing or promotional purposes.

[1162] A "user device" refers to an internet-connected device used to upload advertising content, and includes smartphones, tablets, and personal computers.

[1163] "Preprocessing means" refers to a function that performs the necessary processing to convert uploaded advertisement content into text data.

[1164] Optical Character Recognition (OCR) technology is a technology that analyzes text within an image and converts it into digital text.

[1165] "Generative artificial intelligence" refers to machine learning models and algorithms used to automatically review advertising content for compliance with various laws and regulations.

[1166] "Means of notifying users of the review results" refers to communication methods such as email, in-system messages, and push notifications used to notify users of the review results.

[1167] An "application installed on a smartphone" is software that runs on a smartphone and enables the shooting or uploading of advertising content and real-time review.

[1168] "Providing results in real time" means processing the review results immediately so that users can check the results without waiting.

[1169] This invention is a system for reviewing whether advertising content violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act. The system includes the following components:

[1170] 1. User terminal:

[1171] This provides an interface for users to log in and upload advertising content. Users can submit text, images, or HTML files of their advertisements to the server through this interface. For example, a user could take a picture of their advertisement content using their smartphone and upload that image.

[1172] 2. Pre-treatment means:

[1173] The server receives files uploaded by users and converts them to the appropriate format. This process uses optical character recognition (OCR) technology and HTML text extraction technology. For OCR technology, the "pytesseract" library is used to extract text from image files. For HTML files, an HTML parsing library such as "BeautifulSoup" is used to extract the text portion.

[1174] 3. Generative Artificial Intelligence:

[1175] The converted text data is sent to a generative artificial intelligence (AI) system. The AI ​​system uses the "transformers" library, and specifically the "BERT-base-japanese" model, to analyze the text data from a legal compliance perspective. The analysis results evaluate whether the advertising content may violate the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act.

[1176] 4. Means of notifying the results of the review:

[1177] The server notifies the user of the review results obtained from the generative artificial intelligence. Notification is sent via in-system messages or email. The user can review the review results and modify the advertisement content as needed.

[1178] This system allows users to quickly and efficiently review the legal compliance of advertising content. The program's processing is explained below in natural language.

[1179] When the server receives an advertisement image uploaded by a user, it first uses optical character recognition (OCR) technology to extract the text within the image. Next, it uses a generative artificial intelligence model (e.g., BERT-base-japanese) to analyze whether the extracted text violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act. The analysis results are then notified to the user as a review result.

[1180] As a concrete example, suppose a user takes a picture of an advertisement with their smartphone and uploads it. This image is sent to a server, and the text is extracted using OCR technology. A generative artificial intelligence model analyzes this text, and if it determines that the expression "You will lose weight just by taking this supplement" violates the Pharmaceuticals and Medical Devices Act, a notification is sent to the user stating that "This expression may violate the Pharmaceuticals and Medical Devices Act."

[1181] The following is an example of a prompt message:

[1182] "I took an image: 'advertisement_image.jpg'. Please extract and review the ad text: 'You will lose weight just by taking this supplement.' Please output the results and notify me."

[1183] In this way, the system of the present invention makes it possible to efficiently review the legal compliance of advertising content and notify users of the results in a user-friendly manner.

[1184] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1185] Step 1:

[1186] Users take pictures of advertisements with their smartphones and upload those images to a server through the application.

[1187] Input: Advertisement image file

[1188] Output: Advertisement image file sent to the server

[1189] Step 2:

[1190] The server receives the uploaded ad image files and saves them to the appropriate directory. Then, optical character recognition (OCR) technology is used to extract text from the images.

[1191] Input: Advertisement image file

[1192] Data processing: Text extraction using Optical Character Recognition (OCR) technology

[1193] Output: Text data

[1194] Step 3:

[1195] The server sends the extracted text data to a generative artificial intelligence model for analysis and review from a legal compliance perspective. The generative AI model used is "BERT-base-japanese" to determine whether the advertising content violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act.

[1196] Input: Text data

[1197] Data processing: Text analysis using generative artificial intelligence models

[1198] Output: Review result data

[1199] Step 4:

[1200] The server notifies the user of the evaluation results based on the evaluation result data obtained from the generative artificial intelligence model. Notification is sent via in-system messages or email.

[1201] Input: Review result data

[1202] Data processing: Formatting of review results and notification settings

[1203] Output: Notification to the user

[1204] Step 5:

[1205] Users can check the review results received via in-system message or email. If necessary, they can revise the ad content and request a review again.

[1206] Input: Notification of review results

[1207] Output: Confirmation of review results and possibility of modifying ad content as the next action.

[1208] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1209] This invention combines a system for reviewing whether advertising content violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act with an emotion engine that recognizes user emotions, thereby achieving a more user-friendly review and notification process. Specifically, the system comprises a user terminal, pre-processing means, generative artificial intelligence, means for notifying the user of the review results, and an emotion engine.

[1210] Processing at the user terminal

[1211] Users log in to the system and upload ad content. A user ID and password are required for login, and upon successful authentication, users can access the interface for uploading ad content. As users upload ad content, the system uses the camera and microphone on the user's device to analyze the user's facial expressions and voice in real time, and an emotion engine recognizes the user's emotions.

[1212] Server-side preprocessing

[1213] The server receives files uploaded by users and saves them to the appropriate directory. Next, it determines the format of the uploaded file and converts it to text data if necessary. For image files, optical character recognition (OCR) technology is used, and for HTML files, the text portion is extracted.

[1214] Specific examples based on the server's processing targets:

[1215] For image files:

[1216] If the user has uploaded a screenshot of an advertisement, the server will analyze the image using OCR technology and extract the text information within the image.

[1217] For HTML files:

[1218] If the user has uploaded an ad landing page, the server will extract the text portion from the HTML file and analyze it.

[1219] Review process using generative artificial intelligence

[1220] The text data obtained during preprocessing is then sent to a generative artificial intelligence (AI). The AI ​​analyzes the text data and conducts a review based on the Act against Unjustifiable Premiums and Misleading Representations and the Pharmaceuticals and Medical Devices Act. As a result, it performs a risk assessment for violations of the law and returns the review results to the server. Specifically, it evaluates whether the text contains illegal or misleading expressions.

[1221] Examples of the review process:

[1222] If there is a conflict:

[1223] If the generative artificial intelligence detects a statement such as "You will lose weight just by taking this supplement," it will determine that the statement may violate the Pharmaceutical Affairs Law and return a specific risk assessment result to the server.

[1224] Notification methods and emotion engines

[1225] The server receives the evaluation results from the generative artificial intelligence and, considering the user's emotional data recognized by the emotion engine, notifies the user of the evaluation results. Notification is sent via email or in-system message. For example, if the user is experiencing stress, the notification can be made using gentler language.

[1226] Examples of notifications:

[1227] If the user is feeling stressed:

[1228] If the emotion engine determines from the user's facial expressions and voice that they are experiencing stress, the server will notify the user via email or in-system message, using milder language such as "Caution advised." The user can then modify the ad content based on this information.

[1229] Suggestions for improving advertising content

[1230] The emotion engine can also generate suggestions for improving ad content based on user emotion data. For example, if a user is feeling down, the engine might suggest "changing the ad content to a more positive tone."

[1231] In this way, the system of the present invention makes it possible to efficiently perform legal compliance reviews of advertising content, provide notifications and improvement suggestions that take into account user sentiment data, and deliver results in a user-friendly manner.

[1232] The following describes the processing flow.

[1233] Step 1:

[1234] The user logs into the system. The user enters their user ID and password, and login authentication is performed. If authentication is successful, the user can access the interface for uploading ad content.

[1235] Step 2:

[1236] The user uploads the ad content. This ad content can be a text file, image file, or HTML file sent to the system. The file is sent to the server and saved in the specified directory.

[1237] Step 3:

[1238] The device's camera and microphone are used to analyze the user's facial expressions and voice in real time as they upload ad content. An emotion engine receives this data and recognizes the user's emotions.

[1239] Step 4:

[1240] The server receives the uploaded file and determines its file format. Here, it determines whether it is a text file, an image file, or an HTML file.

[1241] Step 5:

[1242] The server performs preprocessing according to the file format. If the advertisement content is a text file, it is saved as text data as is. If the advertisement content is an image file, it is converted to text data using optical character recognition (OCR) technology. If the advertisement content is an HTML file, the text portion is extracted from the HTML.

[1243] Step 6:

[1244] The server sends the pre-processed text data to the generative artificial intelligence. The generative AI analyzes the text data and reviews its content in accordance with the Act against Unjustifiable Premiums and Misleading Representations and the Pharmaceuticals and Medical Devices Act.

[1245] Step 7:

[1246] The generative artificial intelligence returns the review results to the server. The review results include which parts may violate the law, along with the reasons why.

[1247] Step 8:

[1248] The server notifies the user of the review results. The notification method and wording are adjusted based on the user's emotional data recognized by the emotion engine. Notification methods include email or in-system messages.

[1249] Step 9:

[1250] Users who receive the review results can check the ad content and make corrections as needed. By using the improvement suggestions provided by the emotion engine as a reference, users can re-upload the ad content and repeat the same process to create a legally compliant ad.

[1251] In this way, the system of the present invention makes it possible to efficiently perform legal compliance reviews of advertising content, provide notifications and improvement suggestions that take into account user sentiment data, and deliver results in a user-friendly manner.

[1252] (Example 2)

[1253] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1254] Traditional advertising content review systems check for legal compliance, but rarely provide notifications or feedback that take user emotions into consideration. As a result, users may experience stress or resentment, leading to a decline in the user experience.

[1255] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1256] In this invention, the server includes a user terminal for uploading advertising content, preprocessing means for converting the uploaded advertising content into text data, a generative artificial intelligence for reviewing the text data from a legal compliance perspective, means for notifying the user of the review results, an emotion engine for recognizing the user's emotions, and means for adjusting the notification content based on the user's emotional state. This makes it possible to provide notifications and feedback that take into account the user's emotional state while simultaneously reviewing for legal compliance.

[1257] A "user terminal" is a device that a user uses to upload advertising content.

[1258] "Preprocessing means" refers to a function that performs the necessary processing to convert uploaded advertisement content into text data.

[1259] "Generative artificial intelligence" refers to an artificial intelligence system that analyzes text data to review advertising content from a legal compliance perspective.

[1260] "Notification method" refers to a means of communication used to inform users of the review results, and includes email and in-system messages.

[1261] The "emotion engine" is a function that identifies the user's emotions and uses that data to inform the system's operations.

[1262] "Means for adjusting notification content" refers to a function that processes data to appropriately change the content and wording of notifications based on the user's emotional state.

[1263] This invention is a user-friendly system for reviewing whether advertising content violates relevant laws and regulations. Specifically, it is configured as a system comprising a user terminal, preprocessing means, generative artificial intelligence, means for notifying the user of the review results, and an emotion engine.

[1264] Use on user terminals

[1265] Users first log in to the system on their own devices. Logging in requires a user ID and password. Upon successful authentication, users gain access to an interface for uploading ad content. As users upload ad content, the system uses the device's camera and microphone to analyze facial expressions and voice, and an emotion engine recognizes the user's emotions in real time.

[1266] Server-side processing and preprocessing methods

[1267] The server receives the advertising content files uploaded from the user's terminal and saves them to the appropriate directory. Next, it determines the file format and converts it to text data as needed. In the case of image files, optical character recognition (OCR) technology is used to extract the text data. For example, a screenshot of an advertisement uploaded by a user is analyzed using OCR to obtain the text information within the image. In the case of HTML files, the server extracts the text portion from the HTML code.

[1268] Review by generative artificial intelligence

[1269] The text data obtained by the preprocessing means is sent to the generative artificial intelligence on the server side. The generative artificial intelligence analyzes the received text data and reviews the advertisement content based on the Act against Unjustifiable Premiums and Misleading Representations and the Pharmaceuticals and Medical Devices Act. If there is a risk of violating specific laws, it returns the evaluation result to the server. As a specific example of the review, if the expression "You will lose weight just by taking this supplement" is detected, it is determined that this may violate the Pharmaceuticals and Medical Devices Act and sends a specific risk assessment to the server.

[1270] Examples of prompts generated using AI

[1271] "Please review the following text data to determine if it violates the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceutical and Medical Devices Act. Please also provide specific points for users to modify the advertisement content."

[1272] Emotion engine and notifications

[1273] Upon receiving the review results, the server adjusts the notification content based on the user's emotional data, which is recognized in real time by the emotion engine. For example, if it determines that the user is experiencing stress, the review results will be communicated to the user via email or in-system messages, using milder language such as "Caution is advised."

[1274] Suggestions for improving advertising content

[1275] Furthermore, the emotion engine can also generate suggestions for improving ad content when a user is feeling down. Specifically, it might suggest "changing the ad content to be more positive." This suggestion is communicated to the user via a notification system from the server, and the user can modify the ad content based on it.

[1276] As described above, the system of this invention provides a more user-friendly review system by incorporating notifications and improvement suggestions that take user sentiment data into the legal compliance review of advertising content.

[1277] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1278] System processing steps

[1279] Step 1:

[1280] The user logs into the system.

[1281] Input: User ID and password

[1282] Process: The user enters their ID and password and clicks the login button. The terminal sends the authentication information to the server. The server authenticates the user based on the authentication information.

[1283] Output: Authentication success or failure

[1284] Step 2:

[1285] Users upload ad content.

[1286] Input: Authentication successful, ad file

[1287] Process: Upon successful login, the device displays an interface for uploading ad content. The user selects a file and clicks the upload button. The ad file is then sent from the device to the server.

[1288] Output: Uploaded ad file

[1289] Step 3:

[1290] The emotion engine recognizes the user's emotions.

[1291] Input: User's facial expression data, voice data

[1292] Processing: The system collects the user's facial expressions and voice in real time through the user's device's camera and microphone. The emotion engine analyzes this data to determine the user's emotional state.

[1293] Output: User emotion data (e.g., stress, depression, etc.)

[1294] Step 4:

[1295] The server receives the uploaded files.

[1296] Input: Uploaded ad file

[1297] Processing: The server receives the advertising file sent from the user's terminal and saves it to the appropriate directory.

[1298] Output: Path to the saved ad file

[1299] Step 5:

[1300] Identify the file format and perform the necessary conversions.

[1301] Input: Path to the saved ad file

[1302] Processing: The server determines the file format (image, HTML, etc.) and performs appropriate preprocessing. For image files, it uses OCR technology to extract text data. For HTML files, it extracts the text portion. For example, for image files, it obtains text information through OCR processing. For HTML files, it performs tag analysis and extracts text.

[1303] Output: Text data

[1304] Step 6:

[1305] The server sends the preprocessing results to the generative artificial intelligence.

[1306] Input: Text data

[1307] Processing: The server sends the text data obtained in the preprocessing stage to the generative artificial intelligence and requests review based on the prompt message.

[1308] Example prompt: "Please review the following text data to determine if it violates any laws or regulations. Please also provide specific points for the user to modify the advertisement content."

[1309] Output: Review Results

[1310] Step 7:

[1311] Generative artificial intelligence will perform the review.

[1312] Input: Text data, prompt text

[1313] Processing: A generative artificial intelligence analyzes the received text data and checks for parts that may violate the Act against Unjustifiable Premiums and Misleading Representations or the Pharmaceuticals and Medical Devices Act. The review results are returned to the server.

[1314] Output: Review results (including specific points of criticism)

[1315] Step 8:

[1316] The server receives the review results.

[1317] Input: Review Result

[1318] Processing: The server receives the review results sent from the generative artificial intelligence. The review results include specific points of concern and risks of legal violations.

[1319] Output: Saving of review results

[1320] Step 9:

[1321] The notification content is adjusted based on information from the emotion engine.

[1322] Input: Review results, user sentiment data

[1323] Processing: The server takes into account the user's emotional state, which the emotion engine has previously recognized, and adjusts the content and tone of the review results. For example, if the user is feeling stressed, it will use calmer language.

[1324] Output: Adjusted notification content

[1325] Step 10:

[1326] The server notifies the user of the review results.

[1327] Input: Adjusted notification content

[1328] Processing: The server notifies the user of the adjusted review results. Notification methods include email and in-system messages.

[1329] Output: Notification to the user

[1330] Step 11:

[1331] The emotion engine generates suggestions for improving ad content.

[1332] Input: User sentiment data, review results

[1333] Processing: The emotion engine generates suggestions for improving the ad content based on the user's emotional state. For example, if the user is feeling down, it might suggest "changing the ad content to a more positive tone."

[1334] Output: Improvement suggestions

[1335] Step 12:

[1336] Notify users of improvement suggestions.

[1337] Input: Improvement suggestion

[1338] Processing: The server compiles suggestions from the sentiment engine and notifies the user. The user can then use these suggestions to modify the ad content.

[1339] Output: User notifications, improved ad content

[1340] (Application Example 2)

[1341] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1342] In recent years, compliance with advertising regulations has become increasingly important. However, the review process for creating advertisements that do not violate the law is extremely complex and can be stressful for users. Furthermore, a lack of feedback for improving ad content, beyond simply complying with the law, is a challenge. In addition, there is a need for a more user-friendly approach that takes user emotions into consideration during this process.

[1343] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1344] In this invention, the server includes means for notifying the user of the review results using an emotion engine that recognizes the user's emotions, means for generating suggestions for improving the advertisement content, and preprocessing means. This enables efficient review of legal compliance of advertisements and the generation of improvement suggestions, as well as notifications that take the user's emotions into consideration.

[1345] A "user terminal" refers to a device operated by a user to input or upload advertising content, and generally refers to a smartphone, tablet, or computer.

[1346] "Preprocessing means" refers to a system or software technology for converting uploaded data into an appropriate format for analyzing advertising content, and includes optical character recognition (OCR) and HTML text extraction.

[1347] "Generative artificial intelligence" refers to artificial intelligence technology that analyzes advertising content and conducts reviews in accordance with laws and regulations such as the Premiums and Representations Act and the Pharmaceuticals and Medical Devices Act.

[1348] An "emotion engine" refers to a system or software technology that recognizes and analyzes a user's emotional state in real time from their facial expressions and voice.

[1349] "Notification means" refers to a system or software technology used to communicate review results or improvement suggestions to users, and includes email and in-system messages.

[1350] "Means for generating improvement suggestions" refers to a system or software technology that automatically analyzes areas for improvement in advertising content and provides suggestions to users.

[1351] Optical Character Recognition (OCR) is a technology that extracts text information contained within an image file and converts it into text data.

[1352] "HTML text extraction" is a technology that extracts text data from HTML files and converts it into a parseable format.

[1353] Ad review system

[1354] The system of this invention not only reviews advertising content for legal compliance but also provides notifications and improvement suggestions that take into account user emotions. Its specific configuration and processing are described below.

[1355] User terminal

[1356] It provides an interface for users to upload advertising content. Specifically, a smartphone, tablet, or computer is used. When a user logs in and uploads advertising content, the device's camera and microphone are used to analyze the user's facial expressions and voice in real time, and an emotion engine recognizes emotions.

[1357] server

[1358] The server receives the ad content uploaded by the user and performs preprocessing. If the ad is an image file, it is converted into text data using optical character recognition (OCR) technology. If the ad is in HTML format, the text portion is extracted and converted into text data. The text data obtained through this preprocessing is then sent to a generative artificial intelligence system.

[1359] Generative artificial intelligence

[1360] Based on the text data transmitted by the server, a generative artificial intelligence performs a review from a legal compliance perspective. It analyzes whether the advertisement content contains illegal or misleading expressions based on the Premiums and Representations Act and the Pharmaceuticals and Medical Devices Act, and performs a risk assessment. The results are then returned to the server.

[1361] Emotion engine and notification methods

[1362] The review results received from the generative artificial intelligence, along with the user's emotional data recognized by the emotion engine, are notified to the user via a notification system. If the user is experiencing stress, the review results are conveyed in milder language. Furthermore, if improvements are needed, suggestions for improving the ad content are also generated.

[1363] Hardware and software to use

[1364] Hardware: Cameras and microphones on smartphones, tablets, and computers.

[1365] software:

[1366] cv2 (OpenCV): Camera data capture

[1367] transformers (Hugging Face): Emotion recognition model

[1368] requests: HTTP requests for uploads and notifications

[1369] email_sender (custom module): Email notification

[1370] Example of a prompt

[1371] Examples of prompt statements are as follows:

[1372] Ad text: "You will lose weight just by taking this supplement."

[1373] Detected risk: Potential violation of the Pharmaceuticals and Medical Devices Act

[1374] Suggested revision: "This supplement is expected to be more effective when used in combination with exercise."

[1375] Based on this prompt, generative artificial intelligence can perform appropriate reviews and make improvement suggestions. This allows advertisers to create better advertisements while complying with the law.

[1376] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1377] Step 1:

[1378] Users upload ad content.

[1379] Users enter ad content using their devices and log in to the system. A user ID and password are required for login. Upon successful authentication, users gain access to an interface for uploading ad content.

[1380] Input: User ID, password, advertisement content (image or HTML file)

[1381] Output: Authentication success notification, ad content data

[1382] Step 2:

[1383] The user's device captures emotional data.

[1384] When a user uploads ad content, the device's camera and microphone are activated to analyze the user's facial expressions and voice in real time.

[1385] Input: User video, audio

[1386] Output: Sentiment data

[1387] Step 3:

[1388] Emotional data is analyzed by an emotion engine.

[1389] Real-time facial and voice data is sent to the emotion engine, which analyzes the user's emotional state.

[1390] Input: User video, audio

[1391] Output: Analyzed emotional data (e.g., stress, relaxation, etc.)

[1392] Step 4:

[1393] The advertisement content is uploaded to the server.

[1394] The device sends the advertisement content file to the server. The server saves this advertisement content to the appropriate directory.

[1395] Input: Ad content data

[1396] Output: File saved to server, file path

[1397] Step 5:

[1398] The server preprocesses the ad content.

[1399] The system identifies the format of uploaded files and converts them to text data as needed. For image files, it uses OCR technology; for HTML files, it extracts the text portion.

[1400] Input: Ad content file, file format

[1401] Output: Text data

[1402] Step 6:

[1403] Generative artificial intelligence will review the content of the advertisements.

[1404] The pre-processed text data is sent to a generative artificial intelligence system for review from a legal compliance perspective. Based on the Premiums and Representations Act and the Pharmaceuticals and Medical Devices Act, illegal or misleading expressions are detected, and a risk assessment is performed.

[1405] Input: Text data

[1406] Output: Review results, risk assessment

[1407] Step 7:

[1408] Based on the review results and sentiment data, the notification system informs the user of the outcome.

[1409] The server receives the evaluation results from the generative AI and emotional data from the emotion engine, and notifies the user in appropriate language. If the user is experiencing stress, the results are conveyed in calmer language.

[1410] Input: Review results, sentiment data

[1411] Output: Notification of review results (email or in-system message)

[1412] Step 8:

[1413] The server generates suggestions for improving the ad content.

[1414] Based on data from an emotion engine and generative artificial intelligence, it smoothly suggests improvements to ad content. For example, if a user is feeling down, it suggests revising the ad to use more positive language.

[1415] Input: Review results, sentiment data

[1416] Output: Improvement suggestions

[1417] The specific actions the system takes at each step were explained in detail. This ensures a clear understanding of the role of each processing step and the resulting flow of operations.

[1418] 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 controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1419] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1420] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[1421] Furthermore, the emotion identification model 59, acting 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 a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1422] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[1423] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[1424] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[1425] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[1426] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is 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 the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[1427] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[1428] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[1429] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[1430] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[1431] 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.

[1432] Furthermore, it is not necessary to store the entirety 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 the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[1433] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[1434] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of 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). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[1435] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[1436] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[1437] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[1438] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted as being incorporated by reference.

[1439] The following is further disclosed regarding the embodiments described above.

[1440] (Claim 1)

[1441] A user terminal for uploading advertising content,

[1442] A preprocessing means for converting uploaded advertisement content into text data,

[1443] A generative artificial intelligence for reviewing text data from a legal compliance perspective,

[1444] A means of notifying the user of the review results,

[1445] A system that includes this.

[1446] (Claim 2)

[1447] The system according to claim 1, wherein the preprocessing means has a function to convert the advertisement content into text data using optical character recognition (OCR) technology when the advertisement content is an image file.

[1448] (Claim 3)

[1449] The system according to claim 1, wherein the preprocessing means has a function to extract the text portion and convert it into text data when the advertisement content is in HTML format.

[1450] (Claim 4)

[1451] The system according to claim 1, comprising a generative artificial intelligence that examines extracted text data in accordance with the Act against Unjustifiable Premiums and Misleading Representations and the Pharmaceuticals and Medical Devices Act, and has a function to assess the risk of violating the law.

[1452] (Claim 5)

[1453] The system according to claim 1, wherein the means for notifying the user of the review results includes a function to send the review results as email or an in-system message.

[1454] (Claim 6)

[1455] The system according to claim 1, wherein the user terminal has a function to perform user authentication simultaneously with uploading advertising content.

[1456] "Example 1"

[1457] (Claim 1)

[1458] A terminal device for uploading advertising content,

[1459] A preprocessing means for converting uploaded advertisement content into text data,

[1460] A generative artificial intelligence system for reviewing text data from a legal compliance perspective,

[1461] A means of notifying the user of the review results,

[1462] A system that includes this.

[1463] (Claim 2)

[1464] The system according to claim 1, wherein the preprocessing means has a function to convert the advertisement content into text data using optical character recognition technology when the advertisement content is in image format.

[1465] (Claim 3)

[1466] The system according to claim 1, wherein the preprocessing means has a function to extract the text portion and convert it into text data when the advertisement content is in web page format.

[1467] "Application Example 1"

[1468] (Claim 1)

[1469] A user terminal for uploading advertising content,

[1470] A preprocessing means for converting uploaded advertisement content into text data,

[1471] A generative artificial intelligence for reviewing text data from a legal compliance perspective,

[1472] A means of notifying the user of the review results,

[1473] An application installed on a smartphone for taking or uploading advertising content and providing real-time review results,

[1474] A system that includes this.

[1475] (Claim 2)

[1476] The system according to claim 1, wherein the preprocessing means has a function to convert the advertisement content into text data using optical character recognition (OCR) technology when the advertisement content is an image file.

[1477] (Claim 3)

[1478] The system according to claim 1, wherein the preprocessing means has a function to extract the text portion and convert it into text data when the advertisement content is in HTML format.

[1479] "Example 2 of combining an emotion engine"

[1480] (Claim 1)

[1481] A user terminal for uploading advertising content,

[1482] A preprocessing means for converting uploaded advertisement content into text data,

[1483] A generative artificial intelligence for reviewing text data from a legal compliance perspective,

[1484] A means of notifying the user of the review results,

[1485] An emotion engine that recognizes the user's emotions,

[1486] A means of adjusting notification content based on the user's emotional state,

[1487] A system that includes this.

[1488] (Claim 2)

[1489] The system according to claim 1, wherein the preprocessing means has a function to convert the advertisement content into text data using optical character recognition (OCR) technology when the advertisement content is an image file.

[1490] (Claim 3)

[1491] The system according to claim 1, wherein the preprocessing means has a function to extract the text portion and convert it into text data when the advertisement content is in HTML format.

[1492] "Application example 2 when combining with an emotional engine"

[1493] (Claim 1)

[1494] A user terminal for uploading advertising content,

[1495] A preprocessing means for converting uploaded advertisement content into text data,

[1496] A generative artificial intelligence for reviewing text data from a legal compliance perspective,

[1497] An emotion engine that recognizes the user's emotions,

[1498] A means of notifying users of the review results based on their emotions,

[1499] A means of generating suggestions for improving advertising content,

[1500] A system that includes this.

[1501] (Claim 2)

[1502] The system according to claim 1, wherein the preprocessing means has a function to convert the advertisement content into text data using optical character recognition (OCR) technology when the advertisement content is an image file.

[1503] (Claim 3)

[1504] The system according to claim 1, wherein the preprocessing means has a function to extract the text portion and convert it into text data when the advertisement content is in HTML format. [Explanation of Symbols]

[1505] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A user terminal for uploading advertising content, A preprocessing means for converting uploaded advertisement content into text data, A generative artificial intelligence for reviewing text data from a legal compliance perspective, A means of notifying the user of the review results, A system that includes this.

2. The system according to claim 1, wherein the preprocessing means has a function to convert the advertisement content into text data using optical character recognition (OCR) technology when the advertisement content is an image file.

3. The system according to claim 1, wherein the preprocessing means has a function to extract the text portion and convert it into text data when the advertisement content is in HTML format.

4. The system according to claim 1, comprising a generative artificial intelligence that examines extracted text data in accordance with the Act against Unjustifiable Premiums and Misleading Representations and the Pharmaceuticals and Medical Devices Act, and has a function to assess the risk of violating the law.

5. The system according to claim 1, wherein the means for notifying the user of the review results includes a function to send the review results as email or an in-system message.

6. The system according to claim 1, wherein the user terminal has a function to perform user authentication simultaneously with uploading advertising content.

Images