System

Abstract

A system is provided.SOLUTION: A system, comprising: means for collecting information associated with a target user; means for analyzing the collected information to identify demographic characteristics; means for generating content based on the analysis; means for correcting inappropriate expressions in the generated content; and means for outputting the corrected content.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Creating press releases and social media posts requires time and effort, and it is difficult to create content that will resonate with the target audience. Furthermore, if inappropriate language is included, there is a risk of it sparking outrage, which could damage a company's brand image. To solve these issues, there is a need for a method to efficiently and effectively generate content and reduce the risk of outrage. [Means for solving the problem]

[0005] The present invention provides a system that includes a means for collecting information related to target users, a means for automatically analyzing the collected information to identify the characteristics of the user demographic, a means for generating content based on the analysis results, a means for correcting inappropriate expressions in the generated content, and a means for outputting the corrected content. By using natural language processing technology for the analysis and a generative AI model for content generation, the system achieves effective and safe content creation.

[0006] "Target users" refers to a specific group of users who are interested in or related to a particular product or service.

[0007] "Information Collection Method" refers to any technology or method that automatically obtains information from online data sources.

[0008] "Means of analyzing information" refers to techniques or methods for processing collected information and extracting meaningful patterns or features.

[0009] "Means for identifying user demographic characteristics" refers to technology or methods that reveal the interests, concerns, behavioral patterns, etc. of target users based on the analysis results.

[0010] "Means for generating content" refers to technology or methods that automatically generate text or information that effectively reaches target users based on the results of analysis.

[0011] "Profanity Modification Measures" refers to technologies or methods that automatically detect and modify misleading or inflammatory language in Generated Content.

[0012] "Means for outputting content" refers to the technology or method for outputting and displaying modified content in a form that can be viewed by users.

[0013] "Natural language processing technology" refers to artificial intelligence technology that enables computers to understand, interpret, and generate human language.

[0014] "Generative AI model" refers to an artificial intelligence model that is trained to generate new content based on a dataset. [Brief explanation of the drawings]

[0015] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0016] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0017] First, the terms used in the following description will be explained.

[0018] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0019] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0020] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0021] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0022] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0023] [First embodiment]

[0024] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0025] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0026] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0027] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0028] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0029] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0030] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0031] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0032] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0033] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0034] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0035] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0036] This invention is a system that efficiently generates press releases and social media posts that are well-suited to target users. This system achieves effective and safe content creation by using the following methods: information collection, data analysis, content generation, correction of inappropriate expressions, and output.

[0037] 1. System Configuration

[0038] The system mainly consists of the following components:

[0039] 1. Server:

[0040] Data Collection Module

[0041] Data Analysis Module

[0042] Content Generation Module

[0043] Expression Correction Module

[0044] Database

[0045] 2. Terminal:

[0046] User Interface

[0047] Content display and editing function

[0048] 2. Program processing (natural language explanation)

[0049] How information is collected

[0050] The server collects a large amount of information related to the target user from online sources. This process involves gathering data such as news articles, social media posts, and blog posts based on specific keywords (e.g., "environment" or "energy efficiency"). The collected data is then stored in a database.

[0051] A means of analyzing information

[0052] The server's data analysis module analyzes the collected information using natural language processing technology. For example, it extracts topics and keywords that interest the target user demographic and performs sentiment analysis to categorize positive, negative, and neutral expressions. The analysis results are then used in the next content generation module.

[0053] A means of generating content

[0054] The server's content generation module automatically generates drafts of press releases and social media posts based on the analysis results. Specifically, a generative AI model (e.g., GPT-3 (registered trademark)) creates new text from the analyzed data to effectively reach the target user demographic. For example, when creating a press release on the theme of "new product launch," it can include information about environmentally friendly design and energy efficiency.

[0055] How to fix inappropriate language

[0056] The server's expression correction module automatically detects inappropriate language or words that pose a risk of causing controversy in the generated draft and corrects them to appropriate language. For example, it replaces "language attacking competitors" with "neutral language." This process minimizes the inappropriate impact on users and society.

[0057] A means of outputting content

[0058] The device allows users to review and edit the final content. The user interface displays the generated text and includes a function that allows users to make fine adjustments as needed. For example, a marketing person may review a draft of a press release, make final adjustments, and then post it on the company's official page or social media.

[0059] Specific examples

[0060] The server collects data related to "new product launches." Next, the data analysis module analyzes the interests of users and identifies positive keywords related to "environmentally friendly new products." Based on the analysis results, the content generation module generates a draft that reads, "Our new product is designed with environmental considerations in mind and is energy efficient." The expression correction module corrects the excessive expression in this draft, replacing "excellent" with "highly acclaimed." The corrected draft is sent to the user's device, where the user performs a final review and publishes it after approval.

[0061] In this way, the present invention can automatically generate and revise effective, low-risk content, making marketing activities more efficient.

[0062] The processing flow will be explained below.

[0063] Step 1:

[0064] The server collects information related to the target user from online sources, such as news articles, social media posts, past press releases, and relevant blog posts, using web scraping and APIs. The collected data is then stored in a centralized database.

[0065] Step 2:

[0066] The data collected by the server is analyzed by the data analysis module. Specifically, natural language processing technology is used to extract topics and keywords from the collected data and identify the characteristics of the interests and concerns of the target user demographic. Sentiment analysis is also performed to classify expressions in the data as positive, negative, or neutral.

[0067] Step 3:

[0068] Based on the analysis results, the server uses a content generation module to automatically generate drafts of press releases and social media posts. Using a generative AI model (e.g., GPT-3), the server combines the analyzed and identified keywords and concepts into sentences to create content that effectively reaches the target audience.

[0069] Step 4:

[0070] The server then inspects the generated draft with a language correction module, which uses an automatic filtering algorithm to detect inappropriate language or potentially inflammatory words and replace them with appropriate language. For example, it can correct "offensive language" to "neutral language."

[0071] Step 5:

[0072] The server sends the revised draft to the user's device for review and editing. A user interface displays the generated content and allows the user to make final adjustments and fine-tuning.

[0073] Step 6:

[0074] The user performs a final review and approves the revised content. For example, a marketing person reviews the draft, makes any necessary revisions, and then finally approves it for publication.

[0075] Step 7:

[0076] The device will then post the final approved content to the company's official website or social media, making it publicly available and reaching the target audience.

[0077] Example 1

[0078] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0079] Conventional content generation systems are unable to accurately grasp the interests of target users, making it difficult to generate effective content. Furthermore, if the generated content contains inappropriate language, manual correction is required, resulting in inefficiency. Furthermore, there is no environment in place for quickly checking and editing the generated content, which means it takes a long time to reach the final output.

[0080] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0081] In this invention, the server includes means for collecting information related to target users, means for analyzing the collected information to identify user demographic characteristics, means for generating content based on the analysis results, means for correcting inappropriate language in the generated content, and means for outputting the corrected content. This makes it possible to automatically generate effective content tailored to target users and quickly correct inappropriate language. Furthermore, by including a terminal with a user interface for viewing and editing the content, the generated content can be immediately viewed, edited as necessary, and quickly published.

[0082] A "target user" is a group of users who are expected to have an interest in or demand for a particular product or service.

[0083] "Information gathering methods" are the tools and processes used to gather relevant data online based on specific keywords.

[0084] A "database" is a system that stores collected information or data so that it can be accessed and processed at a later time.

[0085] "Means for analyzing information" refers to methods for analyzing collected data using natural language processing technology, etc., to identify the characteristics and interests of user groups.

[0086] "Natural language processing technology" is a technology that enables computers to understand, generate, and manipulate human language.

[0087] "Means of generating content" refers to processes or tools that automatically generate effective text and information for target users based on analysis results.

[0088] A "generative AI model" is an artificial intelligence model that has been pre-trained with large amounts of data and has the ability to generate advanced text based on input.

[0089] A "prompt" is an explanatory or instructional text to be input into a generative AI model, indicating the theme or topic of the content to be generated.

[0090] "Profanity correction methods" are processes or tools that automatically detect profanity or risky language in generated content and correct it to appropriate language.

[0091] "Means for outputting content" refers to the functionality that displays the final revised content and allows users to access, review, and edit it.

[0092] "User interface" refers to the screens and operating means through which a user interacts with the system and checks and edits the generated content.

[0093] "Terminal" means a device that a User uses to view or edit content using the System, including, for example, a PC or tablet.

[0094] This invention is a system that efficiently generates press releases and social media posts that are well-suited to target users. This system achieves effective and safe content creation by using the following methods: information collection, data analysis, content generation, correction of inappropriate expressions, and output.

[0095] System configuration

[0096] The system mainly consists of the following components:

[0097] 1. Server:

[0098] Data Collection Module

[0099] Data Analysis Module

[0100] Content Generation Module

[0101] Expression Correction Module

[0102] Database

[0103] 2. Terminal:

[0104] User Interface

[0105] Content display and editing function

[0106] Methods of collecting information

[0107] A server collects large amounts of information relevant to a target user from online sources. This is done based on specific keywords (e.g., "environment" or "energy efficiency") and includes data such as news articles, social media posts, and blog posts. The collected data is then stored in a database. Technologies used in this process include web scraping tools (e.g., BeautifulSoup) and APIs (e.g., Twitter API).

[0108] Information analysis methods

[0109] The server's data analysis module analyzes the collected information using natural language processing techniques. Specifically, it performs topic modeling (e.g., LDA) to extract topics and keywords that interest the target user demographic. It also uses sentiment analysis tools (e.g., VADER) to classify the sentiment of the collected data into positive, negative, or neutral.

[0110] Means of content generation

[0111] The server's content generation module automatically generates drafts of press releases and social media posts based on the analysis results. This step uses a generative AI model (e.g., GPT-3). For example, the following prompt is used:

[0112] "Write a press release for a new product launch. Your target demographic is environmentally conscious and focused on energy efficiency. Highlight the product's environmentally friendly design and high energy efficiency as features."

[0113] How to correct inappropriate language

[0114] The server's expression correction module automatically detects and corrects inappropriate language and words that pose a risk of causing controversy in the generated draft. For example, it uses a text analysis tool (e.g., TextBlob) to detect and correct "language attacking competitors" and other such words to make them more neutral.

[0115] Content output method

[0116] The device allows users to review and edit the final content. The user interface displays the generated text and includes an editing function for users to make fine adjustments. Specifically, marketers can review the draft press release, make any necessary corrections, and then post it on the company's official website or social media.

[0117] Specific examples

[0118] The server collects data related to "new product launches," and the data analysis module analyzes user demographic interests and identifies positive keywords related to "environmentally friendly new products." Based on the analysis results, the content generation module generates a draft that reads, "Our new product is designed with environmental considerations in mind and is also energy efficient." The expression correction module corrects the excessive expression in this draft, replacing "excellent" with "highly acclaimed." The corrected draft is sent to the user's device, where the user performs a final review and publishes it after approval.

[0119] In this way, the present invention can automatically generate and revise effective, low-risk content, making marketing activities more efficient.

[0120] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0121] Step 1: Gather information

[0122] The server collects information relevant to the target user from the internet. As input, it is given a list of specific keywords (e.g., "environment," "energy efficiency," etc.). Specifically, it uses web scraping tools (e.g., BeautifulSoup) or APIs (e.g., Twitter API) to retrieve data such as news articles, social media posts, and blog posts. The collected data is stored in a database. As output, it obtains a large amount of unanalyzed data stored in the database.

[0123] Step 2: Information analysis

[0124] The server's data analysis module analyzes the information collected in step 1. The input is the unanalyzed data stored in the database. Specifically, natural language processing techniques (e.g., NLTK, Spacy) are used to tokenize the text of the collected data. Next, topic modeling (e.g., LDA) is performed to extract topics and keywords that interest the target user demographic. A sentiment analysis tool (e.g., VADER) is also used to classify the data into positive, negative, and neutral. The output includes the analysis results, which include data on the topics of interest to the target user demographic and sentiment classification.

[0125] Step 3: Content generation

[0126] The server's content generation module automatically generates drafts of press releases and social media posts based on the analysis results obtained in step 2. The input is data on the topics of interest and sentiment classification of the target user demographic. Specifically, the generative AI model (e.g., GPT-3) is given a prompt such as the following: "Please write a press release about the launch of a new product. The target user demographic is concerned with the environment and focuses on energy efficiency. Please emphasize the product's environmentally friendly design and high energy efficiency as features." Based on this prompt, the generative AI model generates text that will effectively reach the target users. The output is a generated draft.

[0127] Step 4: Correct inappropriate language

[0128] The server's expression correction module detects and corrects inappropriate language and words that pose a risk of causing controversy in the draft generated in step 3. The input is the generated draft. Specifically, it uses a text analysis tool (e.g., TextBlob) to detect offensive language and extreme assertions and replaces them with neutral language. For example, it corrects "language attacking competitors" to "neutral language." The output is a draft with inappropriate language corrected.

[0129] Step 5: Content Output

[0130] The user uses the terminal to check the final content corrected in step 4 and edit it as necessary. The input is the corrected draft. Specifically, the generated text is displayed on the user interface, and the user edits the necessary parts with the mouse or keyboard. For example, a marketing person makes a final check of a press release draft, makes any necessary corrections, and then posts it on the company's official website or social media. The output is the final corrected and checked text.

[0131] Through the above steps, the system efficiently generates effective, low-risk content and supports marketing activities.

[0132] (Application example 1)

[0133] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0134] In modern digital marketing, creating content that effectively appeals to target users is extremely important. However, generating effective content efficiently and disseminating information specific to food delivery services while removing inappropriate language is a time-consuming and labor-intensive task. In particular, the data collection and analysis stages from news articles, social media posts, and blog articles require a great deal of effort, which can lead to variations in the quality of the final content.

[0135] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0136] In this invention, the server includes means for automatically collecting news, social media posts, and blog articles related to food delivery, means for performing sentiment analysis on the collected data to identify positive, negative, and neutral expressions, and means for automatically generating drafts of social media posts and press releases related to food delivery services. This makes it possible to efficiently generate content that effectively appeals to target users and disseminate high-quality information by removing inappropriate expressions.

[0137] "Target users" refers to the group of customers who are the primary target of a particular marketing activity, product, or service.

[0138] "Means for collecting information" refers to functions or devices for automatically obtaining information related to the target user from various data sources.

[0139] "Means for analyzing and identifying the characteristics of the user demographic" refers to a function or device for analyzing collected information to understand characteristics such as the interests, behavioral patterns, and emotional tendencies of target users.

[0140] "Means for generating content" refers to a function or device for creating content such as text and images that effectively appeal to target users based on the analysis results.

[0141] "Means for correcting inappropriate language" refers to a function or device that automatically detects misleading or inappropriate language in generated content and changes it to appropriate language.

[0142] "Means for outputting modified content" refers to the function or device that provides the user with the final content with the inappropriate language corrected, and displays it in a format that allows for review and editing.

[0143] "Food delivery" refers to a service that aims to deliver meals ordered by customers via the Internet to a specified location.

[0144] "News, social media posts, and blog posts" refers to types of online public information sources, such as news articles, posts on social networking sites, and content from blogs run by individuals or businesses, that are available to users.

[0145] "Means for sentiment analysis of data" refers to a function or device for analyzing the sentiment of collected text data and identifying its emotional tendencies, such as positive, negative, or neutral.

[0146] "Means for automatically generating drafts of social media posts and press releases" refers to a function or device for automatically creating text for new social media posts or press releases based on the analysis results.

[0147] This invention is a system that efficiently generates content related to effective food delivery services for target users. This system is mainly composed of a server and terminals, and by interoperating with each other's functions, it provides high-quality content to users.

[0148] 1. System Configuration

[0149] The system consists of the following elements:

[0150] server:

[0151] Data Collection Module: Automatically collects online news articles, social media posts, and blog posts related to food delivery. Specifically, this data is obtained using APIs.

[0152] Data Analysis Module: Analyzes collected data using natural language processing techniques to identify sentiment and interests. This process involves using the TextBlob library to perform sentiment analysis, such as positive, negative, or neutral.

[0153] Content generation module: Automatically generates drafts of new social media posts and press releases based on the analysis results. It uses generative AI models (e.g., GPT-3) to create content. For example, it can generate drafts based on a prompt such as, "Please highlight healthy and delicious dishes to introduce the new menu."

[0154] Profanity Correction Module: Automatically corrects profanity and misleading language in the generated drafts by detecting specific keywords and phrases and replacing them with appropriate ones.

[0155] Database: Stores collected data and generated content.

[0156] Device:

[0157] User Interface: Allows users to view, edit, and finalize the generated content. Implemented as a web browser or smartphone application.

[0158] Content display and editing function: Generated content can be displayed and easily edited by the user. After final confirmation, it is possible to post it to social media or the official page.

[0159] 2. Specific Examples

[0160] The system collects data about "new product launches." The data analysis module analyzes the data to identify positive keywords, such as "new healthy menu items," that interest target customers. Based on the analysis, the content generation module generates drafts for social media posts, such as:

[0161] "Our new products are designed to be environmentally friendly and provide healthy and delicious food. Give them a try!"

[0162] The expression correction module inspects this content and corrects it to a more neutral expression, such as changing "Excellent" to "Highly rated." The corrected content is sent to the device, where the user can make a final check and, if necessary, make additional corrections before posting it to social media or the official page.

[0163] The above is a specific embodiment for carrying out the invention, which makes it possible to generate content that appeals to target users effectively and quickly, and to significantly improve the efficiency of marketing activities.

[0164] Hardware and software used

[0165] Data collection: API (e.g. News API)

[0166] Data Analysis: Python's TextBlob Library

[0167] Content generation: OpenAI® GPT models (e.g., GPT-3)

[0168] Profanity fix: Custom Python scripts

[0169] Prompt Sentence Examples

[0170] "Highlight healthy and tasty dishes as an introduction to your new menu."

[0171] "Write a press release that succinctly explains the new features of your food delivery service."

[0172] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0173] Step 1:

[0174] Data collection

[0175] The server automatically collects online news articles, social media posts, and blog posts related to food delivery. The input is specific keywords (e.g., "food delivery," "food delivery," etc.), and the output is a dataset related to these keywords. It uses an API (e.g., News API) to obtain information based on the specific keywords and stores the collected data in a local database.

[0176] Step 2:

[0177] Data analysis

[0178] The data collected by the server is analyzed using natural language processing techniques. In particular, sentiment analysis is performed using Python's TextBlob library. The input is the dataset collected in step 1, and the output is the sentiment classification results (positive, negative, neutral) and keywords of interest. TextBlob is applied to the dataset to extract the sentiment score and keywords for each sentence.

[0179] Step 3:

[0180] Content Generation

[0181] The server automatically generates drafts for social media posts and press releases using a generative AI model (e.g., GPT-3) based on the sentiment analysis results and keywords of interest. The input is the sentiment analysis results and keywords obtained in step 2, as well as a prompt entered by the user (e.g., "Please emphasize healthy and delicious dishes as an introduction to the new menu item."). The output is the generated text. The prompt and analysis results are input into the generative AI model to generate new content.

[0182] Step 4:

[0183] Correction of inappropriate expressions

[0184] The server inspects the generated content and corrects profanity. The input is the text generated in step 3, and the output is the corrected text. It runs a custom script that searches for specific keywords and replaces profanity or offensive phrases with neutral ones.

[0185] Step 5:

[0186] Content Output

[0187] The server sends the final modified content to the terminal, where the user can review and edit the content. The input is the text modified in step 4, and the output is the final content for the user to edit and post. The content is displayed through a user interface, and the user can make fine adjustments as needed.

[0188] Step 6:

[0189] User Verification and Submission

[0190] The user reviews the final content, makes any necessary adjustments, and then posts it to social media or an official page. The input is the final content displayed in step 5, and the output is the published content. When the user presses the confirm button, it is automatically posted to the selected platform.

[0191] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0192] This invention is a system that efficiently generates press releases and social media posts that resonate with target users and adjusts the tone of the content by recognizing user emotions. This system achieves effective and safe content creation by using information collection, data analysis, content generation, correction of inappropriate expressions, output, and an emotion engine.

[0193] 1. System Configuration

[0194] The system mainly consists of the following components:

[0195] 1. Server:

[0196] Data Collection Module

[0197] Data Analysis Module

[0198] Content Generation Module

[0199] Expression Correction Module

[0200] Emotion Engine

[0201] Database

[0202] 2. Terminal:

[0203] User Interface

[0204] Content display and editing function

[0205] Emotion recognition function

[0206] 2. Program processing (natural language explanation)

[0207] How information is collected

[0208] The server collects large amounts of information related to the target user from online sources, such as news articles, social media posts, past press releases, and relevant blog posts, using web scraping and APIs. The collected data is then stored in a centralized database.

[0209] A means of analyzing information

[0210] The server's data analysis module analyzes the collected information using natural language processing technology. For example, it extracts topics and keywords that interest the target user demographic and performs sentiment analysis to categorize positive, negative, and neutral expressions. The analysis results are then used in the next content generation module.

[0211] A means of generating content

[0212] The server's content generation module automatically generates drafts of press releases and social media posts based on the analysis results, using a generative AI model (e.g., GPT-3) to create new text from the analyzed data and create content that effectively reaches the target audience.

[0213] How to fix inappropriate language

[0214] The server's expression correction module automatically detects inappropriate language or words that pose a risk of causing controversy in the generated draft and corrects them to appropriate language. For example, it corrects "offensive language" to "neutral language." This process minimizes the inappropriate impact on users and society.

[0215] Emotion engine that recognizes user emotions

[0216] The server is equipped with an emotion engine that analyzes real-time emotional data from users. For example, it recognizes their emotional state from facial expressions, voice, and text while they are using the system. Based on the emotion recognition results, the content generation module can adjust the tone and expression of the generated text. This enables the creation of more effective content that is tailored to the user's emotions.

[0217] A means of outputting content

[0218] The terminal allows the user to review and edit the revised final content. The user interface displays the generated text and includes functionality that allows the user to make fine adjustments as needed.

[0219] Specific examples

[0220] The server collects data on "new product launches." Next, the data analysis module analyzes user demographic interests and identifies positive keywords related to "environmentally friendly new products." Based on the analysis results, the content generation module generates a draft that reads, "Our new product is designed with environmental considerations in mind and is energy efficient." The expression correction module corrects the overstatement in this draft, replacing "excellent" with "highly acclaimed."

[0221] Furthermore, the emotion engine analyzes the user's real-time emotional data and adjusts the expression to be more positive if the user shows excitement or curiosity. The revised draft is sent to the user's device, where the user performs a final check and is published after approval.

[0222] In this way, the present invention automatically generates and modifies effective, low-risk content, and adjusts the tone to match the user's emotions, thereby streamlining marketing activities.

[0223] The processing flow will be explained below.

[0224] Step 1:

[0225] The server collects information related to the target user from online sources, specifically news articles, social media posts, past press releases, and relevant blog posts, automatically using web scraping and APIs. The collected data is then stored in a centralized database.

[0226] Step 2:

[0227] The data collected by the server is analyzed in the data analysis module. Natural language processing technology is used to extract topics and keywords that interest the target user demographic, and sentiment analysis is performed to classify expressions in the data as positive, negative, or neutral. The analysis results are used in the next content generation module.

[0228] Step 3:

[0229] The server's content generation module automatically generates drafts of press releases and social media posts based on the analysis results. Using a generative AI model (e.g., GPT-3), the analyzed and identified keywords and concepts are compiled into text. At this stage, content is created that effectively reaches the target audience.

[0230] Step 4:

[0231] The server's expression correction module inspects the generated draft. Using an automatic filtering algorithm, it detects inappropriate language or areas at high risk of flaming, and corrects them to appropriate language. For example, by replacing "offensive language" with "neutral language," the risk of misunderstanding or flaming is reduced.

[0232] Step 5:

[0233] The server's emotion engine analyzes the user's real-time emotional data, recognizing their emotional state from facial expressions, voice, and text data, and adjusting the tone of the content as needed. For example, if the user expresses joy or excitement, the tone of the content will be made more positive.

[0234] Step 6:

[0235] The device will then display the corrected and adjusted content to the user, who can then review the generated text through a user interface and make any necessary final adjustments or refinements.

[0236] Step 7:

[0237] The user will then do a final review and approve the revised content. Marketing staff will then review the draft, make any necessary revisions, and approve it for publication on the official website and social media.

[0238] Step 8:

[0239] The device then posts the final approved content to the company's official website or social media, making the generated content public and reaching the target audience.

[0240] Example 2

[0241] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0242] With conventional content generation systems, it was difficult to quickly and accurately generate content that would effectively appeal to target users. Furthermore, the generated content could contain inappropriate language, and the tone of the content was not automatically adjusted to take user emotions into account, making it difficult to maximize marketing effectiveness.

[0243] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0244] In this invention, the server includes means for collecting information related to target users, means for analyzing the collected information to identify the characteristics of the user demographic, means for generating content based on the analysis results, means for correcting inappropriate expressions in the generated content, means for recognizing the emotional state of the user and adjusting the tone and expression of the content, and means for outputting the corrected content. This makes it possible to quickly and efficiently generate content that appeals appropriately to target users and further adjust the tone to match the user's emotions.

[0245] "Target Audience" refers to the specific customer group or individual for whom a particular marketing activity or piece of content is intended.

[0246] "Information collection methods" refers to the technologies and devices used to gather online news articles, social media posts, blog posts, past press releases, etc.

[0247] "Means for analyzing information" refers to technology and software that uses natural language processing technology to analyze collected information and identify the characteristics of user demographics.

[0248] "Means for generating content" refers to software or programs that use generative AI models to create effective press releases, social media posts, and other written content based on the analysis results.

[0249] "Profanity Modification Measures" refers to technology or software that detects and modifies excessive or offensive language in generated content.

[0250] "Means for recognizing a user's emotional state" refers to technologies or devices that analyze a user's emotional data in real time and adjust the tone and expression of content to match that emotion.

[0251] "Means for outputting modified content" refers to the functionality or device that provides the final modified content to the user and allows it to be viewed and edited through a user interface.

[0252] This invention is a system that efficiently generates press releases and social media posts that resonate with target users and adjusts the tone of the content by recognizing user emotions. This system achieves effective and safe content creation by using information collection, data analysis, content generation, correction of inappropriate expressions, output, and an emotion engine.

[0253] System configuration

[0254] The system mainly consists of the following components:

[0255] 1. Server:

[0256] Data Collection Module

[0257] Data Analysis Module

[0258] Content Generation Module

[0259] Expression Correction Module

[0260] Emotion Engine

[0261] Database

[0262] 2. Terminal:

[0263] User Interface

[0264] Content display and editing function

[0265] Emotion recognition function

[0266] Program processing

[0267] How information is collected

[0268] The server collects large amounts of information related to the target user from online sources. Specifically, it uses web scraping and APIs to collect news articles, social media posts, past press releases, relevant blog posts, etc. The collected data is then stored in a centralized database. For example, articles can be extracted from news sites using the Python library "BeautifulSoup."

[0269] A means of analyzing information

[0270] The server's data analysis module analyzes the collected information using natural language processing techniques, specifically using the NLTK library to tokenize the text and extract important topics and keywords, and the Hugging Face sentiment analysis model to classify positive, negative, and neutral expressions.

[0271] A means of generating content

[0272] The server's content generation module automatically generates drafts of press releases and social media posts based on the analysis results. Specifically, it uses a generative AI model (e.g., GPT-3) to create new sentences from the analyzed data. For example, the prompt sentence could be, "Can you give me some ideas for a press release that highlights the features of our new environmentally friendly product?"

[0273] How to fix inappropriate language

[0274] The server's expression correction module automatically detects inappropriate expressions or words that may pose a risk of causing controversy in the generated draft and corrects them to appropriate expressions. For example, it converts "excellent" to "highly acclaimed." This process minimizes the inappropriate impact on users and society.

[0275] Emotion engine that recognizes user emotions

[0276] The server is equipped with an emotion engine that analyzes users' real-time emotional data. Specifically, it uses OpenCV and DeepFace to recognize the user's emotional state from facial expressions, voice, text, and other information while using the system. Based on the emotion recognition results, the content generation module adjusts the tone and expression of the generated text. This enables the creation of more effective content that matches the user's emotions.

[0277] A means of outputting content

[0278] The terminal allows the user to review and edit the final revised content. The user interface displays the generated text and includes a function that allows the user to make fine adjustments as needed. The generated content can be published on social media or as a press release.

[0279] Specific examples

[0280] The server collects data related to "new product launches." Next, the data analysis module analyzes user demographic interests and identifies positive keywords related to "environmentally friendly new products." Based on the analysis results, the content generation module generates a draft that reads, "Our new product is designed with environmental considerations in mind and is energy efficient." The expression correction module corrects the excessive expression in this draft, replacing "excellent" with "highly acclaimed."

[0281] Furthermore, the emotion engine analyzes the user's real-time emotional data and adjusts the expression to be more positive if the user shows excitement or curiosity. The revised draft is sent to the user's device, where the user performs a final check and is published after approval.

[0282] In this way, the present invention can significantly improve the efficiency of marketing activities by automatically generating and modifying effective, low-risk content and adjusting the tone to suit the user's emotions.

[0283] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0284] Step 1: Gather information

[0285] The server collects information related to the target user from online sources. Specifically, it uses Python's BeautifulSoup library to scrape articles from specific news sites and blogs. It also uses the APIs of the target sites to retrieve social media posts and past press releases. This information is then stored in a database.

[0286] Input: URL of a website or API relevant to your target users

[0287] Data processing: web scraping, API calls

[0288] Output: Collected text data

[0289] Step 2: Analyze the information

[0290] The server's data analysis module analyzes the collected information. First, it uses the NLTK library to tokenize the text and extract frequent keywords and important topics. Next, it uses Hugging Face's sentiment analysis model to classify the sentiment of the text as positive, negative, or neutral, which allows it to identify the interests of the target user demographic.

[0291] Input: Collected text data

[0292] Data processing: tokenization, keyword extraction, sentiment analysis

[0293] Output: Analyzed information (keywords, sentiment classification)

[0294] Step 3: Generate content

[0295] The server's content generation module generates prompts based on the analyzed information and uses a generative AI model (e.g., GPT-3) to draft press releases and social media posts. For example, the prompt might be, "Can you give us some ideas for a press release highlighting the features of our new eco-friendly product?"

[0296] Input: Parsed information, prompt

[0297] Data processing: Automatic text generation using AI models

[0298] Output: Generated content draft

[0299] Step 4: Fix inappropriate language

[0300] The server's correction module scans the generated draft for inappropriate or excessive language, for example, converting "excellent" to "highly acclaimed." This process makes the content more neutral and less risky.

[0301] Input: Generated content draft

[0302] Data processing: text analysis, expression correction

[0303] Output: Revised content draft

[0304] Step 5: Recognizing User Emotions

[0305] The server's emotion engine analyzes the user's real-time emotion data, for example, by using OpenCV and DeepFace to capture the user's facial expressions with a camera and recognize their emotional state. Based on this data, the content generation module adjusts the tone of the generated content.

[0306] Input: Real-time user emotion data (facial expressions, voice, text)

[0307] Data processing: facial expression analysis, emotion recognition

[0308] Output: Emotionally adjusted tone-adjusted draft content

[0309] Step 6: Outputting content

[0310] The device allows the user to review and edit the final revised content. A user interface displays the generated content and provides editing functions for the user to make fine adjustments as needed. Once final review is complete, the content is published to social media platforms and websites.

[0311] Input: Revised content draft

[0312] Data processing: Display and editing through the user interface

[0313] Output: Final content approved by the user

[0314] (Application example 2)

[0315] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0316] Conventional content generation systems have limited capabilities for automatically generating effective advertising content for target users, and are unable to adjust the tone of the advertisement based on the user's real-time emotions. As a result, the effectiveness of advertisements is often limited, and real-time adjustments to match user emotions are required.

[0317] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting information, means for analyzing the collected information to identify user demographic characteristics, means for generating content based on the analysis results, means for correcting inappropriate expressions in the generated content, means for outputting the corrected content, and means for acquiring real-time emotional data of the user and adjusting the tone of the content based on the emotional data. This makes it possible to generate advertising content that is effective for target users in real time and adjust the tone to match the user's emotions.

[0318] "Information collection means" refers to the methods and devices used to collect data related to the target user online.

[0319] "Means for analyzing collected information to identify the characteristics of the user demographic" refers to a method or device that analyzes collected data and identifies characteristics such as the interests, concerns, and behavioral patterns of target users.

[0320] "Content generation means" refers to a method or device that automatically creates content such as advertisements, press releases, and social media posts based on the characteristics of the user demographic.

[0321] "Means for correcting inappropriate language in generated content" refers to a method or device that automatically detects and corrects offensive, prejudiced, or inappropriate language in text generated by the system, such as advertisements or articles.

[0322] "Means for outputting modified content" refers to the method or device that ultimately provides users with modified or adjusted content such as advertisements, press releases, and social media posts.

[0323] "Means for acquiring real-time emotional data of a user and adjusting the tone of content based on that emotional data" refers to a method or device that detects a user's emotional state in real time from facial expressions, voice, text, etc., and dynamically adapts the tone and expression of content based on that.

[0324] MODE FOR CARRYING OUT THE INVENTION

[0325] The present invention relates to a system for efficiently and effectively generating advertising content and adjusting the tone based on a user's real-time emotions. The system includes means for information collection, data analysis, content generation, inappropriate expression correction, content output, and emotion recognition, and is capable of effectively and adaptively generating advertising content.

[0326] 1. System Configuration

[0327] server

[0328] Information Collection Module

[0329] The server uses web scraping and APIs to collect information related to the target users, including news articles, social media posts, past press releases, relevant blog posts, etc. This data is then stored in a centralized database.

[0330] Data Analysis Module

[0331] The server analyzes the collected information using natural language processing techniques (e.g., TextBlob) to identify demographic interests and characteristics and categorize positive, negative, and neutral expressions.

[0332] Content Generation Module

[0333] Based on the analysis results, a generative AI model (e.g., OpenAI's GPT-3) is used to automatically generate draft copy for advertising and social media posts. For example, a prompt such as "Generate a social media post about a new product launch with a positive tone" is generated.

[0334] Expression Correction Module

[0335] The server automatically detects inappropriate language or words that pose a risk of causing controversy in the generated draft and edits them to appropriate language, minimizing the impact on users and society.

[0336] Emotion Recognition Module

[0337] The server acquires the user's real-time emotions and adjusts the tone of the generated content accordingly. For example, it analyzes the user's facial expressions, voice, and text data when operating the system, and adjusts the tone according to their emotional state (e.g., excitement, curiosity, confusion, anxiety).

[0338] Terminal

[0339] User Interface

[0340] Users can review and edit the generated content through an on-device user interface, which displays the final content adjusted based on emotion recognition and includes the ability for users to fine-tune it as needed.

[0341] Specific examples

[0342] For example, the server collects data on "new product launches." Then, the data analysis module analyzes the interests of the user demographic and identifies positive keywords related to "new eco-friendly products." Based on the analysis results, the content generation module generates a draft of a social media post like this:

[0343] Generate social media posts about your new product launch. Keep the tone positive.

[0344] The generated content is the following sentence:

[0345] Our new products are designed to be environmentally friendly and are energy efficient! Give them a try!

[0346] The expression correction module corrects the over-expression in this draft, replacing "excellent" with "highly rated." Furthermore, the emotion recognition module analyzes the user's real-time emotional data and adjusts the expression to be more positive if the user shows excitement or curiosity. For example:

[0347] Our new product is designed to be environmentally friendly and has received rave reviews! Give it a try!

[0348] The revised draft is sent to the user's device, where the user reviews it and publishes it after approval. This process allows the invention to effectively reach the target audience and adjust the tone appropriately based on real-time emotions.

[0349] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0350] Program processing flow

[0351] Step 1: Gather information

[0352] The server collects information related to the target users from online sources, such as news articles, blog posts, social media posts, and past press releases, using web scraping or APIs. The input is keywords (e.g., "new product announcement," "environmentally friendly"), and the output is the collected text data.

[0353] Step 2: Data analysis

[0354] The server analyzes the collected information using natural language processing technology. For example, it uses TextBlob to perform sentiment analysis of text and classify it into positive, negative, and neutral expressions. The input is the collected text data, and the output is keywords related to the characteristics of the user demographic.

[0355] Step 3: Content generation

[0356] The server generates drafts of advertising copy and social media posts using a generative AI model (e.g., OpenAI's GPT-3) based on the analysis results. The input is keywords obtained from the analysis results and a prompt (e.g., "Generate a social media post about a new product launch. The tone should be positive."), and the output is the generated content.

[0357] Step 4: Correct inappropriate language

[0358] The server detects inappropriate language in the generated content and corrects it to appropriate language, e.g., changing excessive language to neutral language. The input is the generated draft, and the output is the corrected content.

[0359] Step 5: User Emotion Recognition

[0360] The server acquires the user's real-time emotional data (e.g., facial expressions, voice, text) and analyzes it with an emotion recognition library (e.g., DeepFace). The input is the user's emotional data collected in real time, and the output is the recognized emotional state.

[0361] Step 6: Adjust the tone

[0362] The server adjusts the tone of the generated content based on the recognized emotional state. For example, if the user is excited, it adjusts the tone to a more positive one, and if the user is confused, it adjusts the tone to a more calm, non-sensational one. The input is the modified content and the user's emotional state, and the output is the final adjusted content.

[0363] Step 7: Content Output

[0364] The server sends the final adjusted content to the terminal and displays it on the terminal's user interface. The input is the adjusted content, and the output is the content displayed to the user. The user can then perform further editing or final review as needed.

[0365] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0366] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0367] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0368] [Second embodiment]

[0369] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0370] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0371] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0372] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0373] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0374] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0375] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0376] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0377] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0378] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0379] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0380] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0381] This invention is a system that efficiently generates press releases and social media posts that are well-suited to target users. This system achieves effective and safe content creation by using the following methods: information collection, data analysis, content generation, correction of inappropriate expressions, and output.

[0382] 1. System Configuration

[0383] The system mainly consists of the following components:

[0384] 1. Server:

[0385] Data Collection Module

[0386] Data Analysis Module

[0387] Content Generation Module

[0388] Expression Correction Module

[0389] Database

[0390] 2. Terminal:

[0391] User Interface

[0392] Content display and editing function

[0393] 2. Program processing (natural language explanation)

[0394] How information is collected

[0395] The server collects a large amount of information related to the target user from online sources. This process involves gathering data such as news articles, social media posts, and blog posts based on specific keywords (e.g., "environment" or "energy efficiency"). The collected data is then stored in a database.

[0396] A means of analyzing information

[0397] The server's data analysis module analyzes the collected information using natural language processing technology. For example, it extracts topics and keywords that interest the target user demographic and performs sentiment analysis to categorize positive, negative, and neutral expressions. The analysis results are then used in the next content generation module.

[0398] A means of generating content

[0399] The server's content generation module automatically generates drafts of press releases and social media posts based on the analysis results. Specifically, a generative AI model (e.g., GPT-3) creates new text from the analyzed data to effectively reach the target user demographic. For example, when creating a press release on the theme of "new product launch," it can include information about environmentally friendly design and energy efficiency.

[0400] How to fix inappropriate language

[0401] The server's expression correction module automatically detects inappropriate language or words that pose a risk of causing controversy in the generated draft and corrects them to appropriate language. For example, it replaces "language attacking competitors" with "neutral language." This process minimizes the inappropriate impact on users and society.

[0402] A means of outputting content

[0403] The device allows users to review and edit the final content. The user interface displays the generated text and includes a function that allows users to make fine adjustments as needed. For example, a marketing person may review a draft of a press release, make final adjustments, and then post it on the company's official page or social media.

[0404] Specific examples

[0405] The server collects data related to "new product launches." Next, the data analysis module analyzes the interests of users and identifies positive keywords related to "environmentally friendly new products." Based on the analysis results, the content generation module generates a draft that reads, "Our new product is designed with environmental considerations in mind and is energy efficient." The expression correction module corrects the excessive expression in this draft, replacing "excellent" with "highly acclaimed." The corrected draft is sent to the user's device, where the user performs a final review and publishes it after approval.

[0406] In this way, the present invention can automatically generate and revise effective, low-risk content, making marketing activities more efficient.

[0407] The processing flow will be explained below.

[0408] Step 1:

[0409] The server collects information related to the target user from online sources, such as news articles, social media posts, past press releases, and relevant blog posts, using web scraping and APIs. The collected data is then stored in a centralized database.

[0410] Step 2:

[0411] The data collected by the server is analyzed by the data analysis module. Specifically, natural language processing technology is used to extract topics and keywords from the collected data and identify the characteristics of the interests and concerns of the target user demographic. Sentiment analysis is also performed to classify expressions in the data as positive, negative, or neutral.

[0412] Step 3:

[0413] Based on the analysis results, the server uses a content generation module to automatically generate drafts of press releases and social media posts. Using a generative AI model (e.g., GPT-3), the server combines the analyzed and identified keywords and concepts into sentences to create content that effectively reaches the target audience.

[0414] Step 4:

[0415] The server then inspects the generated draft with a language correction module, which uses an automatic filtering algorithm to detect inappropriate language or potentially inflammatory words and replace them with appropriate language. For example, it can correct "offensive language" to "neutral language."

[0416] Step 5:

[0417] The server sends the revised draft to the user's device for review and editing. A user interface displays the generated content and allows the user to make final adjustments and fine-tuning.

[0418] Step 6:

[0419] The user performs a final review and approves the revised content. For example, a marketing person reviews the draft, makes any necessary revisions, and then finally approves it for publication.

[0420] Step 7:

[0421] The device will then post the final approved content to the company's official website or social media, making it publicly available and reaching the target audience.

[0422] Example 1

[0423] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0424] Conventional content generation systems are unable to accurately grasp the interests of target users, making it difficult to generate effective content. Furthermore, if the generated content contains inappropriate language, manual correction is required, resulting in inefficiency. Furthermore, there is no environment in place for quickly checking and editing the generated content, which means it takes a long time to reach the final output.

[0425] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0426] In this invention, the server includes means for collecting information related to target users, means for analyzing the collected information to identify user demographic characteristics, means for generating content based on the analysis results, means for correcting inappropriate language in the generated content, and means for outputting the corrected content. This makes it possible to automatically generate effective content tailored to target users and quickly correct inappropriate language. Furthermore, by including a terminal with a user interface for viewing and editing the content, the generated content can be immediately viewed, edited as necessary, and quickly published.

[0427] A "target user" is a group of users who are expected to have an interest in or demand for a particular product or service.

[0428] "Information gathering methods" are the tools and processes used to gather relevant data online based on specific keywords.

[0429] A "database" is a system that stores collected information or data so that it can be accessed and processed at a later time.

[0430] "Means for analyzing information" refers to methods for analyzing collected data using natural language processing technology, etc., to identify the characteristics and interests of user groups.

[0431] "Natural language processing technology" is a technology that enables computers to understand, generate, and manipulate human language.

[0432] "Means of generating content" refers to processes or tools that automatically generate effective text and information for target users based on analysis results.

[0433] A "generative AI model" is an artificial intelligence model that has been pre-trained with large amounts of data and has the ability to generate advanced text based on input.

[0434] A "prompt" is an explanatory or instructional text to be input into a generative AI model, indicating the theme or topic of the content to be generated.

[0435] "Profanity correction methods" are processes or tools that automatically detect profanity or risky language in generated content and correct it to appropriate language.

[0436] "Means for outputting content" refers to the functionality that displays the final revised content and allows users to access, review, and edit it.

[0437] "User interface" refers to the screens and operating means through which a user interacts with the system and checks and edits the generated content.

[0438] "Terminal" means a device that a User uses to view or edit content using the System, including, for example, a PC or tablet.

[0439] This invention is a system that efficiently generates press releases and social media posts that are well-suited to target users. This system achieves effective and safe content creation by using the following methods: information collection, data analysis, content generation, correction of inappropriate expressions, and output.

[0440] System configuration

[0441] The system mainly consists of the following components:

[0442] 1. Server:

[0443] Data Collection Module

[0444] Data Analysis Module

[0445] Content Generation Module

[0446] Expression Correction Module

[0447] Database

[0448] 2. Terminal:

[0449] User Interface

[0450] Content display and editing function

[0451] Methods of collecting information

[0452] A server collects large amounts of information relevant to a target user from online sources. This is done based on specific keywords (e.g., "environment" or "energy efficiency") and includes data such as news articles, social media posts, and blog posts. The collected data is then stored in a database. Technologies used in this process include web scraping tools (e.g., BeautifulSoup) and APIs (e.g., Twitter API).

[0453] Information analysis methods

[0454] The server's data analysis module analyzes the collected information using natural language processing techniques. Specifically, it performs topic modeling (e.g., LDA) to extract topics and keywords that interest the target user demographic. It also uses sentiment analysis tools (e.g., VADER) to classify the sentiment of the collected data into positive, negative, or neutral.

[0455] Means of content generation

[0456] The server's content generation module automatically generates drafts of press releases and social media posts based on the analysis results. This step uses a generative AI model (e.g., GPT-3). For example, the following prompt is used:

[0457] "Write a press release for a new product launch. Your target demographic is environmentally conscious and focused on energy efficiency. Highlight the product's environmentally friendly design and high energy efficiency as features."

[0458] How to correct inappropriate language

[0459] The server's expression correction module automatically detects and corrects inappropriate language and words that pose a risk of causing controversy in the generated draft. For example, it uses a text analysis tool (e.g., TextBlob) to detect and correct "language attacking competitors" and other such words to make them more neutral.

[0460] Content output method

[0461] The device allows users to review and edit the final content. The user interface displays the generated text and includes an editing function for users to make fine adjustments. Specifically, marketers can review the draft press release, make any necessary corrections, and then post it on the company's official website or social media.

[0462] Specific examples

[0463] The server collects data related to "new product launches," and the data analysis module analyzes user demographic interests and identifies positive keywords related to "environmentally friendly new products." Based on the analysis results, the content generation module generates a draft that reads, "Our new product is designed with environmental considerations in mind and is also energy efficient." The expression correction module corrects the excessive expression in this draft, replacing "excellent" with "highly acclaimed." The corrected draft is sent to the user's device, where the user performs a final review and publishes it after approval.

[0464] In this way, the present invention can automatically generate and revise effective, low-risk content, making marketing activities more efficient.

[0465] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0466] Step 1: Gather information

[0467] The server collects information relevant to the target user from the internet. As input, it is given a list of specific keywords (e.g., "environment," "energy efficiency," etc.). Specifically, it uses web scraping tools (e.g., BeautifulSoup) or APIs (e.g., Twitter API) to retrieve data such as news articles, social media posts, and blog posts. The collected data is stored in a database. As output, it obtains a large amount of unanalyzed data stored in the database.

[0468] Step 2: Information analysis

[0469] The server's data analysis module analyzes the information collected in step 1. The input is the unanalyzed data stored in the database. Specifically, natural language processing techniques (e.g., NLTK, Spacy) are used to tokenize the text of the collected data. Next, topic modeling (e.g., LDA) is performed to extract topics and keywords that interest the target user demographic. A sentiment analysis tool (e.g., VADER) is also used to classify the data into positive, negative, and neutral. The output includes the analysis results, which include data on the topics of interest to the target user demographic and sentiment classification.

[0470] Step 3: Content generation

[0471] The server's content generation module automatically generates drafts of press releases and social media posts based on the analysis results obtained in step 2. The input is data on the topics of interest and sentiment classification of the target user demographic. Specifically, the generative AI model (e.g., GPT-3) is given a prompt such as the following: "Please write a press release about the launch of a new product. The target user demographic is concerned with the environment and focuses on energy efficiency. Please emphasize the product's environmentally friendly design and high energy efficiency as features." Based on this prompt, the generative AI model generates text that will effectively reach the target users. The output is a generated draft.

[0472] Step 4: Correct inappropriate language

[0473] The server's expression correction module detects and corrects inappropriate language and words that pose a risk of causing controversy in the draft generated in step 3. The input is the generated draft. Specifically, it uses a text analysis tool (e.g., TextBlob) to detect offensive language and extreme assertions and replaces them with neutral language. For example, it corrects "language attacking competitors" to "neutral language." The output is a draft with inappropriate language corrected.

[0474] Step 5: Content Output

[0475] The user uses the terminal to check the final content corrected in step 4 and edit it as necessary. The input is the corrected draft. Specifically, the generated text is displayed on the user interface, and the user edits the necessary parts with the mouse or keyboard. For example, a marketing person makes a final check of a press release draft, makes any necessary corrections, and then posts it on the company's official website or social media. The output is the final corrected and checked text.

[0476] Through the above steps, the system efficiently generates effective, low-risk content and supports marketing activities.

[0477] (Application example 1)

[0478] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0479] In modern digital marketing, creating content that effectively appeals to target users is extremely important. However, generating effective content efficiently and disseminating information specific to food delivery services while removing inappropriate language is a time-consuming and labor-intensive task. In particular, the data collection and analysis stages from news articles, social media posts, and blog articles require a great deal of effort, which can lead to variations in the quality of the final content.

[0480] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0481] In this invention, the server includes means for automatically collecting news, social media posts, and blog articles related to food delivery, means for performing sentiment analysis on the collected data to identify positive, negative, and neutral expressions, and means for automatically generating drafts of social media posts and press releases related to food delivery services. This makes it possible to efficiently generate content that effectively appeals to target users and disseminate high-quality information by removing inappropriate expressions.

[0482] "Target users" refers to the group of customers who are the primary target of a particular marketing activity, product, or service.

[0483] "Means for collecting information" refers to functions or devices for automatically obtaining information related to the target user from various data sources.

[0484] "Means for analyzing and identifying the characteristics of the user demographic" refers to a function or device for analyzing collected information to understand characteristics such as the interests, behavioral patterns, and emotional tendencies of target users.

[0485] "Means for generating content" refers to a function or device for creating content such as text and images that effectively appeal to target users based on the analysis results.

[0486] "Means for correcting inappropriate language" refers to a function or device that automatically detects potentially misleading or inappropriate language in generated content and changes it to appropriate language.

[0487] "Means for outputting modified content" refers to the function or device that provides the user with the final content with the inappropriate language corrected, and displays it in a format that allows for review and editing.

[0488] "Food delivery" refers to a service that aims to deliver meals ordered by customers via the Internet to a specified location.

[0489] "News, social media posts, and blog posts" refers to types of online public information sources, such as news articles, posts on social networking sites, and content from blogs run by individuals or businesses, that are available to users.

[0490] "Means for sentiment analysis of data" refers to a function or device for analyzing the sentiment of collected text data and identifying its emotional tendencies, such as positive, negative, or neutral.

[0491] "Means for automatically generating drafts of social media posts and press releases" refers to a function or device for automatically creating text for new social media posts or press releases based on the analysis results.

[0492] This invention is a system that efficiently generates content related to effective food delivery services for target users. This system is mainly composed of a server and terminals, and by interoperating with each other's functions, it provides high-quality content to users.

[0493] 1. System Configuration

[0494] The system consists of the following elements:

[0495] server:

[0496] Data Collection Module: Automatically collects online news articles, social media posts, and blog posts related to food delivery. Specifically, this data is obtained using APIs.

[0497] Data Analysis Module: Analyzes collected data using natural language processing techniques to identify sentiment and interests. This process involves using the TextBlob library to perform sentiment analysis, such as positive, negative, or neutral.

[0498] Content generation module: Automatically generates drafts of new social media posts and press releases based on the analysis results. It uses generative AI models (e.g., GPT-3) to create content. For example, it can generate drafts based on a prompt such as, "Please highlight healthy and delicious dishes to introduce the new menu."

[0499] Profanity Correction Module: Automatically corrects profanity and misleading language in the generated drafts by detecting specific keywords and phrases and replacing them with appropriate ones.

[0500] Database: Stores collected data and generated content.

[0501] Device:

[0502] User Interface: Allows users to view, edit, and finalize the generated content. Implemented as a web browser or smartphone application.

[0503] Content display and editing function: Generated content can be displayed and easily edited by the user. After final confirmation, it is possible to post it to social media or the official page.

[0504] 2. Specific Examples

[0505] The system collects data about "new product launches." The data analysis module analyzes the data to identify positive keywords, such as "new healthy menu items," that interest target customers. Based on the analysis, the content generation module generates drafts for social media posts, such as:

[0506] "Our new products are designed to be environmentally friendly and provide healthy and delicious food. Give them a try!"

[0507] The expression correction module inspects this content and corrects it to a more neutral expression, such as changing "Excellent" to "Highly rated." The corrected content is sent to the device, where the user can make a final check and, if necessary, make additional corrections before posting it to social media or the official page.

[0508] The above is a specific embodiment for carrying out the invention, which makes it possible to generate content that appeals to target users effectively and quickly, and to significantly improve the efficiency of marketing activities.

[0509] Hardware and software used

[0510] Data collection: API (e.g. News API)

[0511] Data Analysis: Python's TextBlob Library

[0512] Content generation: OpenAI's GPT models (e.g., GPT-3)

[0513] Profanity fix: Custom Python scripts

[0514] Prompt Sentence Examples

[0515] "Highlight healthy and tasty dishes as an introduction to your new menu."

[0516] "Write a press release that succinctly explains the new features of your food delivery service."

[0517] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0518] Step 1:

[0519] Data collection

[0520] The server automatically collects online news articles, social media posts, and blog posts related to food delivery. The input is specific keywords (e.g., "food delivery," "food delivery," etc.), and the output is a dataset related to these keywords. It uses an API (e.g., News API) to obtain information based on the specific keywords and stores the collected data in a local database.

[0521] Step 2:

[0522] Data analysis

[0523] The data collected by the server is analyzed using natural language processing techniques. In particular, sentiment analysis is performed using Python's TextBlob library. The input is the dataset collected in step 1, and the output is the sentiment classification results (positive, negative, neutral) and keywords of interest. TextBlob is applied to the dataset to extract the sentiment score and keywords for each sentence.

[0524] Step 3:

[0525] Content Generation

[0526] The server automatically generates drafts for social media posts and press releases using a generative AI model (e.g., GPT-3) based on the sentiment analysis results and keywords of interest. The input is the sentiment analysis results and keywords obtained in step 2, as well as a prompt entered by the user (e.g., "Please emphasize healthy and delicious dishes as an introduction to the new menu item."). The output is the generated text. The prompt and analysis results are input into the generative AI model to generate new content.

[0527] Step 4:

[0528] Correction of inappropriate expressions

[0529] The server inspects the generated content and corrects profanity. The input is the text generated in step 3, and the output is the corrected text. It runs a custom script that searches for specific keywords and replaces profanity or offensive phrases with neutral ones.

[0530] Step 5:

[0531] Content Output

[0532] The server sends the final modified content to the terminal, where the user can review and edit the content. The input is the text modified in step 4, and the output is the final content for the user to edit and post. The content is displayed through a user interface, and the user can make fine adjustments as needed.

[0533] Step 6:

[0534] User Verification and Submission

[0535] The user reviews the final content, makes any necessary adjustments, and then posts it to social media or an official page. The input is the final content displayed in step 5, and the output is the published content. When the user presses the confirm button, it is automatically posted to the selected platform.

[0536] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0537] This invention is a system that efficiently generates press releases and social media posts that resonate with target users and adjusts the tone of the content by recognizing user emotions. This system achieves effective and safe content creation by using information collection, data analysis, content generation, correction of inappropriate expressions, output, and an emotion engine.

[0538] 1. System Configuration

[0539] The system mainly consists of the following components:

[0540] 1. Server:

[0541] Data Collection Module

[0542] Data Analysis Module

[0543] Content Generation Module

[0544] Expression Correction Module

[0545] Emotion Engine

[0546] Database

[0547] 2. Terminal:

[0548] User Interface

[0549] Content display and editing function

[0550] Emotion recognition function

[0551] 2. Program processing (natural language explanation)

[0552] How information is collected

[0553] The server collects large amounts of information related to the target user from online sources, such as news articles, social media posts, past press releases, and relevant blog posts, using web scraping and APIs. The collected data is then stored in a centralized database.

[0554] A means of analyzing information

[0555] The server's data analysis module analyzes the collected information using natural language processing technology. For example, it extracts topics and keywords that interest the target user demographic and performs sentiment analysis to categorize positive, negative, and neutral expressions. The analysis results are then used in the next content generation module.

[0556] A means of generating content

[0557] The server's content generation module automatically generates drafts of press releases and social media posts based on the analysis results, using a generative AI model (e.g., GPT-3) to create new text from the analyzed data and create content that effectively reaches the target audience.

[0558] How to fix inappropriate language

[0559] The server's expression correction module automatically detects inappropriate language or words that pose a risk of causing controversy in the generated draft and corrects them to appropriate language. For example, it corrects "offensive language" to "neutral language." This process minimizes the inappropriate impact on users and society.

[0560] Emotion engine that recognizes user emotions

[0561] The server is equipped with an emotion engine that analyzes real-time emotional data from users. For example, it recognizes their emotional state from facial expressions, voice, and text while they are using the system. Based on the emotion recognition results, the content generation module can adjust the tone and expression of the generated text. This enables the creation of more effective content that is tailored to the user's emotions.

[0562] A means of outputting content

[0563] The terminal allows the user to review and edit the revised final content. The user interface displays the generated text and includes functionality that allows the user to make fine adjustments as needed.

[0564] Specific examples

[0565] The server collects data on "new product launches." Next, the data analysis module analyzes user demographic interests and identifies positive keywords related to "environmentally friendly new products." Based on the analysis results, the content generation module generates a draft that reads, "Our new product is designed with environmental considerations in mind and is energy efficient." The expression correction module corrects the overstatement in this draft, replacing "excellent" with "highly acclaimed."

[0566] Furthermore, the emotion engine analyzes the user's real-time emotional data and adjusts the expression to be more positive if the user shows excitement or curiosity. The revised draft is sent to the user's device, where the user performs a final check and is published after approval.

[0567] In this way, the present invention automatically generates and modifies effective, low-risk content, and adjusts the tone to match the user's emotions, thereby streamlining marketing activities.

[0568] The processing flow will be explained below.

[0569] Step 1:

[0570] The server collects information related to the target user from online sources, specifically news articles, social media posts, past press releases, and relevant blog posts, automatically using web scraping and APIs. The collected data is then stored in a centralized database.

[0571] Step 2:

[0572] The data collected by the server is analyzed in the data analysis module. Natural language processing technology is used to extract topics and keywords that interest the target user demographic, and sentiment analysis is performed to classify expressions in the data as positive, negative, or neutral. The analysis results are used in the next content generation module.

[0573] Step 3:

[0574] The server's content generation module automatically generates drafts of press releases and social media posts based on the analysis results. Using a generative AI model (e.g., GPT-3), the analyzed and identified keywords and concepts are compiled into text. At this stage, content is created that effectively reaches the target audience.

[0575] Step 4:

[0576] The server's expression correction module inspects the generated draft. Using an automatic filtering algorithm, it detects inappropriate language or areas at high risk of flaming, and corrects them to appropriate language. For example, by replacing "offensive language" with "neutral language," the risk of misunderstanding or flaming is reduced.

[0577] Step 5:

[0578] The server's emotion engine analyzes the user's real-time emotional data, recognizing their emotional state from facial expressions, voice, and text data, and adjusting the tone of the content as needed. For example, if the user expresses joy or excitement, the tone of the content will be made more positive.

[0579] Step 6:

[0580] The device will then display the corrected and adjusted content to the user, who can then review the generated text through a user interface and make any necessary final adjustments or refinements.

[0581] Step 7:

[0582] The user will then do a final review and approve the revised content. Marketing staff will then review the draft, make any necessary revisions, and approve it for publication on the official website and social media.

[0583] Step 8:

[0584] The device then posts the final approved content to the company's official website or social media, making the generated content public and reaching the target audience.

[0585] Example 2

[0586] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0587] With conventional content generation systems, it was difficult to quickly and accurately generate content that would effectively appeal to target users. Furthermore, the generated content could contain inappropriate language, and the tone of the content was not automatically adjusted to take user emotions into account, making it difficult to maximize marketing effectiveness.

[0588] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0589] In this invention, the server includes means for collecting information related to target users, means for analyzing the collected information to identify the characteristics of the user demographic, means for generating content based on the analysis results, means for correcting inappropriate expressions in the generated content, means for recognizing the emotional state of the user and adjusting the tone and expression of the content, and means for outputting the corrected content. This makes it possible to quickly and efficiently generate content that appeals appropriately to target users and further adjust the tone to match the user's emotions.

[0590] "Target Audience" refers to the specific customer group or individual for whom a particular marketing activity or piece of content is intended.

[0591] "Information collection methods" refers to the technologies and devices used to gather online news articles, social media posts, blog posts, past press releases, etc.

[0592] "Means for analyzing information" refers to technology and software that uses natural language processing technology to analyze collected information and identify the characteristics of user demographics.

[0593] "Means for generating content" refers to software or programs that use generative AI models to create effective press releases, social media posts, and other written content based on the analysis results.

[0594] "Profanity Modification Measures" refers to technology or software that detects and modifies excessive or offensive language in generated content.

[0595] "Means for recognizing a user's emotional state" refers to technologies or devices that analyze a user's emotional data in real time and adjust the tone and expression of content to match that emotion.

[0596] "Means for outputting modified content" refers to the functionality or device that provides the final modified content to the user and allows it to be viewed and edited through a user interface.

[0597] This invention is a system that efficiently generates press releases and social media posts that resonate with target users and adjusts the tone of the content by recognizing user emotions. This system achieves effective and safe content creation by using information collection, data analysis, content generation, correction of inappropriate expressions, output, and an emotion engine.

[0598] System configuration

[0599] The system mainly consists of the following components:

[0600] 1. Server:

[0601] Data Collection Module

[0602] Data Analysis Module

[0603] Content Generation Module

[0604] Expression Correction Module

[0605] Emotion Engine

[0606] Database

[0607] 2. Terminal:

[0608] User Interface

[0609] Content display and editing function

[0610] Emotion recognition function

[0611] Program processing

[0612] How information is collected

[0613] The server collects large amounts of information related to the target user from online sources. Specifically, it uses web scraping and APIs to collect news articles, social media posts, past press releases, relevant blog posts, etc. The collected data is then stored in a centralized database. For example, articles can be extracted from news sites using the Python library "BeautifulSoup."

[0614] A means of analyzing information

[0615] The server's data analysis module analyzes the collected information using natural language processing techniques, specifically using the NLTK library to tokenize the text and extract important topics and keywords, and the Hugging Face sentiment analysis model to classify positive, negative, and neutral expressions.

[0616] A means of generating content

[0617] The server's content generation module automatically generates drafts of press releases and social media posts based on the analysis results. Specifically, it uses a generative AI model (e.g., GPT-3) to create new sentences from the analyzed data. For example, the prompt sentence could be, "Can you give me some ideas for a press release that highlights the features of our new environmentally friendly product?"

[0618] How to fix inappropriate language

[0619] The server's expression correction module automatically detects inappropriate expressions or words that may pose a risk of causing controversy in the generated draft and corrects them to appropriate expressions. For example, it converts "excellent" to "highly acclaimed." This process minimizes the inappropriate impact on users and society.

[0620] Emotion engine that recognizes user emotions

[0621] The server is equipped with an emotion engine that analyzes users' real-time emotional data. Specifically, it uses OpenCV and DeepFace to recognize the user's emotional state from facial expressions, voice, text, and other information while using the system. Based on the emotion recognition results, the content generation module adjusts the tone and expression of the generated text. This enables the creation of more effective content that matches the user's emotions.

[0622] A means of outputting content

[0623] The terminal allows the user to review and edit the final revised content. The user interface displays the generated text and includes a function that allows the user to make fine adjustments as needed. The generated content can be published on social media or as a press release.

[0624] Specific examples

[0625] The server collects data related to "new product launches." Next, the data analysis module analyzes user demographic interests and identifies positive keywords related to "environmentally friendly new products." Based on the analysis results, the content generation module generates a draft that reads, "Our new product is designed with environmental considerations in mind and is energy efficient." The expression correction module corrects the excessive expression in this draft, replacing "excellent" with "highly acclaimed."

[0626] Furthermore, the emotion engine analyzes the user's real-time emotional data and adjusts the expression to be more positive if the user shows excitement or curiosity. The revised draft is sent to the user's device, where the user performs a final check and is published after approval.

[0627] In this way, the present invention can significantly improve the efficiency of marketing activities by automatically generating and modifying effective, low-risk content and adjusting the tone to suit the user's emotions.

[0628] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0629] Step 1: Gather information

[0630] The server collects information related to the target user from online sources. Specifically, it uses Python's BeautifulSoup library to scrape articles from specific news sites and blogs. It also uses the APIs of the target sites to retrieve social media posts and past press releases. This information is then stored in a database.

[0631] Input: URL of a website or API relevant to your target users

[0632] Data processing: web scraping, API calls

[0633] Output: Collected text data

[0634] Step 2: Analyze the information

[0635] The server's data analysis module analyzes the collected information. First, it uses the NLTK library to tokenize the text and extract frequent keywords and important topics. Next, it uses Hugging Face's sentiment analysis model to classify the sentiment of the text as positive, negative, or neutral, which allows it to identify the interests of the target user demographic.

[0636] Input: Collected text data

[0637] Data processing: tokenization, keyword extraction, sentiment analysis

[0638] Output: Analyzed information (keywords, sentiment classification)

[0639] Step 3: Generate content

[0640] The server's content generation module generates prompts based on the analyzed information and uses a generative AI model (e.g., GPT-3) to draft press releases and social media posts. For example, the prompt might be, "Can you give us some ideas for a press release highlighting the features of our new eco-friendly product?"

[0641] Input: Parsed information, prompt

[0642] Data processing: Automatic text generation using AI models

[0643] Output: Generated content draft

[0644] Step 4: Fix inappropriate language

[0645] The server's correction module scans the generated draft for inappropriate or excessive language, for example, converting "excellent" to "highly acclaimed." This process makes the content more neutral and less risky.

[0646] Input: Generated content draft

[0647] Data processing: text analysis, expression correction

[0648] Output: Revised content draft

[0649] Step 5: Recognizing User Emotions

[0650] The server's emotion engine analyzes the user's real-time emotion data, for example, by using OpenCV and DeepFace to capture the user's facial expressions with a camera and recognize their emotional state. Based on this data, the content generation module adjusts the tone of the generated content.

[0651] Input: Real-time user emotion data (facial expressions, voice, text)

[0652] Data processing: facial expression analysis, emotion recognition

[0653] Output: Emotionally adjusted tone-adjusted draft content

[0654] Step 6: Outputting content

[0655] The device allows the user to review and edit the final revised content. A user interface displays the generated content and provides editing functions for the user to make fine adjustments as needed. Once final review is complete, the content is published to social media platforms and websites.

[0656] Input: Revised content draft

[0657] Data processing: Display and editing through the user interface

[0658] Output: Final content approved by the user

[0659] (Application example 2)

[0660] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0661] Conventional content generation systems have limited capabilities for automatically generating effective advertising content for target users, and are unable to adjust the tone of the advertisement based on the user's real-time emotions. As a result, the effectiveness of advertisements is often limited, and real-time adjustments to match user emotions are required.

[0662] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting information, means for analyzing the collected information to identify user demographic characteristics, means for generating content based on the analysis results, means for correcting inappropriate expressions in the generated content, means for outputting the corrected content, and means for acquiring real-time emotional data of the user and adjusting the tone of the content based on the emotional data. This makes it possible to generate advertising content that is effective for target users in real time and adjust the tone to match the user's emotions.

[0663] "Information collection means" refers to the methods and devices used to collect data related to the target user online.

[0664] "Means for analyzing collected information to identify the characteristics of the user demographic" refers to a method or device that analyzes collected data and identifies characteristics such as the interests, concerns, and behavioral patterns of target users.

[0665] "Content generation means" refers to a method or device that automatically creates content such as advertisements, press releases, and social media posts based on the characteristics of the user demographic.

[0666] "Means for correcting inappropriate language in generated content" refers to a method or device that automatically detects and corrects offensive, prejudiced, or inappropriate language in text generated by the system, such as advertisements or articles.

[0667] "Means for outputting modified content" refers to the method or device that ultimately provides users with modified or adjusted content such as advertisements, press releases, and social media posts.

[0668] "Means for acquiring real-time emotional data of a user and adjusting the tone of content based on that emotional data" refers to a method or device that detects a user's emotional state in real time from facial expressions, voice, text, etc., and dynamically adapts the tone and expression of content based on that.

[0669] MODE FOR CARRYING OUT THE INVENTION

[0670] The present invention relates to a system for efficiently and effectively generating advertising content and adjusting the tone based on a user's real-time emotions. The system includes means for information collection, data analysis, content generation, inappropriate expression correction, content output, and emotion recognition, and is capable of effectively and adaptively generating advertising content.

[0671] 1. System Configuration

[0672] server

[0673] Information Collection Module

[0674] The server uses web scraping and APIs to collect information related to the target users, including news articles, social media posts, past press releases, relevant blog posts, etc. This data is then stored in a centralized database.

[0675] Data Analysis Module

[0676] The server analyzes the collected information using natural language processing techniques (e.g., TextBlob) to identify demographic interests and characteristics and categorize positive, negative, and neutral expressions.

[0677] Content Generation Module

[0678] Based on the analysis results, a generative AI model (e.g., OpenAI's GPT-3) is used to automatically generate draft copy for advertising and social media posts. For example, a prompt such as "Generate a social media post about a new product launch with a positive tone" is generated.

[0679] Expression Correction Module

[0680] The server automatically detects inappropriate language or words that pose a risk of causing controversy in the generated draft and edits them to appropriate language, minimizing the impact on users and society.

[0681] Emotion Recognition Module

[0682] The server acquires the user's real-time emotions and adjusts the tone of the generated content accordingly. For example, it analyzes the user's facial expressions, voice, and text data when operating the system, and adjusts the tone according to their emotional state (e.g., excitement, curiosity, confusion, anxiety).

[0683] Terminal

[0684] User Interface

[0685] Users can review and edit the generated content through an on-device user interface, which displays the final content adjusted based on emotion recognition and includes the ability for users to fine-tune it as needed.

[0686] Specific examples

[0687] For example, the server collects data on "new product launches." Then, the data analysis module analyzes the interests of the user demographic and identifies positive keywords related to "new eco-friendly products." Based on the analysis results, the content generation module generates a draft of a social media post like this:

[0688] Generate social media posts about your new product launch. Keep the tone positive.

[0689] The generated content is the following sentence:

[0690] Our new products are designed to be environmentally friendly and are energy efficient! Give them a try!

[0691] The expression correction module corrects the over-expression in this draft, replacing "excellent" with "highly rated." Furthermore, the emotion recognition module analyzes the user's real-time emotional data and adjusts the expression to be more positive if the user shows excitement or curiosity. For example:

[0692] Our new product is designed to be environmentally friendly and has received rave reviews! Give it a try!

[0693] The revised draft is sent to the user's device, where the user reviews it and publishes it after approval. This process allows the invention to effectively reach the target audience and adjust the tone appropriately based on real-time emotions.

[0694] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0695] Program processing flow

[0696] Step 1: Gather information

[0697] The server collects information related to the target users from online sources, such as news articles, blog posts, social media posts, and past press releases, using web scraping or APIs. The input is keywords (e.g., "new product announcement," "environmentally friendly"), and the output is the collected text data.

[0698] Step 2: Data analysis

[0699] The server analyzes the collected information using natural language processing technology. For example, it uses TextBlob to perform sentiment analysis of text and classify it into positive, negative, and neutral expressions. The input is the collected text data, and the output is keywords related to the characteristics of the user demographic.

[0700] Step 3: Content generation

[0701] The server generates drafts of advertising copy and social media posts using a generative AI model (e.g., OpenAI's GPT-3) based on the analysis results. The input is keywords obtained from the analysis results and a prompt (e.g., "Generate a social media post about a new product launch. The tone should be positive."), and the output is the generated content.

[0702] Step 4: Correct inappropriate language

[0703] The server detects inappropriate language in the generated content and corrects it to appropriate language, e.g., changing excessive language to neutral language. The input is the generated draft, and the output is the corrected content.

[0704] Step 5: User Emotion Recognition

[0705] The server acquires the user's real-time emotional data (e.g., facial expressions, voice, text) and analyzes it with an emotion recognition library (e.g., DeepFace). The input is the user's emotional data collected in real time, and the output is the recognized emotional state.

[0706] Step 6: Adjust the tone

[0707] The server adjusts the tone of the generated content based on the recognized emotional state. For example, if the user is excited, it adjusts the tone to a more positive one, and if the user is confused, it adjusts the tone to a more calm, non-sensational one. The input is the modified content and the user's emotional state, and the output is the final adjusted content.

[0708] Step 7: Content Output

[0709] The server sends the final adjusted content to the terminal and displays it on the terminal's user interface. The input is the adjusted content, and the output is the content displayed to the user. The user can then perform further editing or final review as needed.

[0710] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0711] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0712] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0713] [Third embodiment]

[0714] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0715] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0716] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0717] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0718] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0719] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0720] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0721] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0722] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0723] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0724] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0725] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0726] This invention is a system that efficiently generates press releases and social media posts that are well-suited to target users. This system achieves effective and safe content creation by using the following methods: information collection, data analysis, content generation, correction of inappropriate expressions, and output.

[0727] 1. System Configuration

[0728] The system mainly consists of the following components:

[0729] 1. Server:

[0730] Data Collection Module

[0731] Data Analysis Module

[0732] Content Generation Module

[0733] Expression Correction Module

[0734] Database

[0735] 2. Terminal:

[0736] User Interface

[0737] Content display and editing function

[0738] 2. Program processing (natural language explanation)

[0739] How information is collected

[0740] The server collects a large amount of information related to the target user from online sources. This process involves gathering data such as news articles, social media posts, and blog posts based on specific keywords (e.g., "environment" or "energy efficiency"). The collected data is then stored in a database.

[0741] A means of analyzing information

[0742] The server's data analysis module analyzes the collected information using natural language processing technology. For example, it extracts topics and keywords that interest the target user demographic and performs sentiment analysis to categorize positive, negative, and neutral expressions. The analysis results are then used in the next content generation module.

[0743] A means of generating content

[0744] The server's content generation module automatically generates drafts of press releases and social media posts based on the analysis results. Specifically, a generative AI model (e.g., GPT-3) creates new text from the analyzed data to effectively reach the target user demographic. For example, when creating a press release on the theme of "new product launch," it can include information about environmentally friendly design and energy efficiency.

[0745] How to fix inappropriate language

[0746] The server's expression correction module automatically detects inappropriate language or words that pose a risk of causing controversy in the generated draft and corrects them to appropriate language. For example, it replaces "language attacking competitors" with "neutral language." This process minimizes the inappropriate impact on users and society.

[0747] A means of outputting content

[0748] The device allows users to review and edit the final content. The user interface displays the generated text and includes a function that allows users to make fine adjustments as needed. For example, a marketing person may review a draft of a press release, make final adjustments, and then post it on the company's official page or social media.

[0749] Specific examples

[0750] The server collects data related to "new product launches." Next, the data analysis module analyzes the interests of users and identifies positive keywords related to "environmentally friendly new products." Based on the analysis results, the content generation module generates a draft that reads, "Our new product is designed with environmental considerations in mind and is energy efficient." The expression correction module corrects the excessive expression in this draft, replacing "excellent" with "highly acclaimed." The corrected draft is sent to the user's device, where the user performs a final review and publishes it after approval.

[0751] In this way, the present invention can automatically generate and revise effective, low-risk content, making marketing activities more efficient.

[0752] The processing flow will be explained below.

[0753] Step 1:

[0754] The server collects information related to the target user from online sources, such as news articles, social media posts, past press releases, and relevant blog posts, using web scraping and APIs. The collected data is then stored in a centralized database.

[0755] Step 2:

[0756] The data collected by the server is analyzed by the data analysis module. Specifically, natural language processing technology is used to extract topics and keywords from the collected data and identify the characteristics of the interests and concerns of the target user demographic. Sentiment analysis is also performed to classify expressions in the data as positive, negative, or neutral.

[0757] Step 3:

[0758] Based on the analysis results, the server uses a content generation module to automatically generate drafts of press releases and social media posts. Using a generative AI model (e.g., GPT-3), the server combines the analyzed and identified keywords and concepts into sentences to create content that effectively reaches the target audience.

[0759] Step 4:

[0760] The server then inspects the generated draft with a language correction module, which uses an automatic filtering algorithm to detect inappropriate language or potentially inflammatory words and replace them with appropriate language. For example, it can correct "offensive language" to "neutral language."

[0761] Step 5:

[0762] The server sends the revised draft to the user's device for review and editing. A user interface displays the generated content and allows the user to make final adjustments and fine-tuning.

[0763] Step 6:

[0764] The user performs a final review and approves the revised content. For example, a marketing person reviews the draft, makes any necessary revisions, and then finally approves it for publication.

[0765] Step 7:

[0766] The device will then post the final approved content to the company's official website or social media, making it publicly available and reaching the target audience.

[0767] Example 1

[0768] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0769] Conventional content generation systems are unable to accurately grasp the interests of target users, making it difficult to generate effective content. Furthermore, if the generated content contains inappropriate language, manual correction is required, resulting in inefficiency. Furthermore, there is no environment in place for quickly checking and editing the generated content, which means it takes a long time to reach the final output.

[0770] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0771] In this invention, the server includes means for collecting information related to target users, means for analyzing the collected information to identify user demographic characteristics, means for generating content based on the analysis results, means for correcting inappropriate language in the generated content, and means for outputting the corrected content. This makes it possible to automatically generate effective content tailored to target users and quickly correct inappropriate language. Furthermore, by including a terminal with a user interface for viewing and editing the content, the generated content can be immediately viewed, edited as necessary, and quickly published.

[0772] A "target user" is a group of users who are expected to have an interest in or demand for a particular product or service.

[0773] "Information gathering methods" are the tools and processes used to gather relevant data online based on specific keywords.

[0774] A "database" is a system that stores collected information or data so that it can be accessed and processed at a later time.

[0775] "Means for analyzing information" refers to methods for analyzing collected data using natural language processing technology, etc., to identify the characteristics and interests of user groups.

[0776] "Natural language processing technology" is a technology that enables computers to understand, generate, and manipulate human language.

[0777] "Means of generating content" refers to processes or tools that automatically generate effective text and information for target users based on analysis results.

[0778] A "generative AI model" is an artificial intelligence model that has been pre-trained with large amounts of data and has the ability to generate advanced text based on input.

[0779] A "prompt" is an explanatory or instructional text to be input into a generative AI model, indicating the theme or topic of the content to be generated.

[0780] "Profanity correction methods" are processes or tools that automatically detect profanity or risky language in generated content and correct it to appropriate language.

[0781] "Means for outputting content" refers to the functionality that displays the final revised content and allows users to access, review, and edit it.

[0782] "User interface" refers to the screens and operating means through which a user interacts with the system and checks and edits the generated content.

[0783] "Terminal" means a device that a User uses to view or edit content using the System, including, for example, a PC or tablet.

[0784] This invention is a system that efficiently generates press releases and social media posts that are well-suited to target users. This system achieves effective and safe content creation by using the following methods: information collection, data analysis, content generation, correction of inappropriate expressions, and output.

[0785] System configuration

[0786] The system mainly consists of the following components:

[0787] 1. Server:

[0788] Data Collection Module

[0789] Data Analysis Module

[0790] Content Generation Module

[0791] Expression Correction Module

[0792] Database

[0793] 2. Terminal:

[0794] User Interface

[0795] Content display and editing function

[0796] Methods of collecting information

[0797] A server collects large amounts of information relevant to a target user from online sources. This is done based on specific keywords (e.g., "environment" or "energy efficiency") and includes data such as news articles, social media posts, and blog posts. The collected data is then stored in a database. Technologies used in this process include web scraping tools (e.g., BeautifulSoup) and APIs (e.g., Twitter API).

[0798] Information analysis methods

[0799] The server's data analysis module analyzes the collected information using natural language processing techniques. Specifically, it performs topic modeling (e.g., LDA) to extract topics and keywords that interest the target user demographic. It also uses sentiment analysis tools (e.g., VADER) to classify the sentiment of the collected data into positive, negative, or neutral.

[0800] Means of content generation

[0801] The server's content generation module automatically generates drafts of press releases and social media posts based on the analysis results. This step uses a generative AI model (e.g., GPT-3). For example, the following prompt is used:

[0802] "Write a press release for a new product launch. Your target demographic is environmentally conscious and focused on energy efficiency. Highlight the product's environmentally friendly design and high energy efficiency as features."

[0803] How to correct inappropriate language

[0804] The server's expression correction module automatically detects and corrects inappropriate language and words that pose a risk of causing controversy in the generated draft. For example, it uses a text analysis tool (e.g., TextBlob) to detect and correct "language attacking competitors" and other such words to make them more neutral.

[0805] Content output method

[0806] The device allows users to review and edit the final content. The user interface displays the generated text and includes an editing function for users to make fine adjustments. Specifically, marketers can review the draft press release, make any necessary corrections, and then post it on the company's official website or social media.

[0807] Specific examples

[0808] The server collects data related to "new product launches," and the data analysis module analyzes user demographic interests and identifies positive keywords related to "environmentally friendly new products." Based on the analysis results, the content generation module generates a draft that reads, "Our new product is designed with environmental considerations in mind and is also energy efficient." The expression correction module corrects the excessive expression in this draft, replacing "excellent" with "highly acclaimed." The corrected draft is sent to the user's device, where the user performs a final review and publishes it after approval.

[0809] In this way, the present invention can automatically generate and revise effective, low-risk content, making marketing activities more efficient.

[0810] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0811] Step 1: Gather information

[0812] The server collects information relevant to the target user from the internet. As input, it is given a list of specific keywords (e.g., "environment," "energy efficiency," etc.). Specifically, it uses web scraping tools (e.g., BeautifulSoup) or APIs (e.g., Twitter API) to retrieve data such as news articles, social media posts, and blog posts. The collected data is stored in a database. As output, it obtains a large amount of unanalyzed data stored in the database.

[0813] Step 2: Information analysis

[0814] The server's data analysis module analyzes the information collected in step 1. The input is the unanalyzed data stored in the database. Specifically, natural language processing techniques (e.g., NLTK, Spacy) are used to tokenize the text of the collected data. Next, topic modeling (e.g., LDA) is performed to extract topics and keywords that interest the target user demographic. A sentiment analysis tool (e.g., VADER) is also used to classify the data into positive, negative, and neutral. The output includes the analysis results, which include data on the topics of interest to the target user demographic and sentiment classification.

[0815] Step 3: Content generation

[0816] The server's content generation module automatically generates drafts of press releases and social media posts based on the analysis results obtained in step 2. The input is data on the topics of interest and sentiment classification of the target user demographic. Specifically, the generative AI model (e.g., GPT-3) is given a prompt such as the following: "Please write a press release about the launch of a new product. The target user demographic is concerned with the environment and focuses on energy efficiency. Please emphasize the product's environmentally friendly design and high energy efficiency as features." Based on this prompt, the generative AI model generates text that will effectively reach the target users. The output is a generated draft.

[0817] Step 4: Correct inappropriate language

[0818] The server's expression correction module detects and corrects inappropriate language and words that pose a risk of causing controversy in the draft generated in step 3. The input is the generated draft. Specifically, it uses a text analysis tool (e.g., TextBlob) to detect offensive language and extreme assertions and replaces them with neutral language. For example, it corrects "language attacking competitors" to "neutral language." The output is a draft with inappropriate language corrected.

[0819] Step 5: Content Output

[0820] The user uses the terminal to check the final content corrected in step 4 and edit it as necessary. The input is the corrected draft. Specifically, the generated text is displayed on the user interface, and the user edits the necessary parts with the mouse or keyboard. For example, a marketing person makes a final check of a press release draft, makes any necessary corrections, and then posts it on the company's official website or social media. The output is the final corrected and checked text.

[0821] Through the above steps, the system efficiently generates effective, low-risk content and supports marketing activities.

[0822] (Application example 1)

[0823] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0824] In modern digital marketing, creating content that effectively appeals to target users is extremely important. However, generating effective content efficiently and disseminating information specific to food delivery services while removing inappropriate language is a time-consuming and labor-intensive task. In particular, the data collection and analysis stages from news articles, social media posts, and blog articles require a great deal of effort, which can lead to variations in the quality of the final content.

[0825] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0826] In this invention, the server includes means for automatically collecting news, social media posts, and blog articles related to food delivery, means for performing sentiment analysis on the collected data to identify positive, negative, and neutral expressions, and means for automatically generating drafts of social media posts and press releases related to food delivery services. This makes it possible to efficiently generate content that effectively appeals to target users and disseminate high-quality information by removing inappropriate expressions.

[0827] "Target users" refers to the group of customers who are the primary target of a particular marketing activity, product, or service.

[0828] "Means for collecting information" refers to functions or devices for automatically obtaining information related to the target user from various data sources.

[0829] "Means for analyzing and identifying the characteristics of the user demographic" refers to a function or device for analyzing collected information to understand characteristics such as the interests, behavioral patterns, and emotional tendencies of target users.

[0830] "Means for generating content" refers to a function or device for creating content such as text and images that effectively appeal to target users based on the analysis results.

[0831] "Means for correcting inappropriate language" refers to a function or device that automatically detects potentially misleading or inappropriate language in generated content and changes it to appropriate language.

[0832] "Means for outputting modified content" refers to the function or device that provides the user with the final content with the inappropriate language corrected, and displays it in a format that allows for review and editing.

[0833] "Food delivery" refers to a service that aims to deliver meals ordered by customers via the Internet to a specified location.

[0834] "News, social media posts, and blog posts" refers to types of online public information sources, such as news articles, posts on social networking sites, and content from blogs run by individuals or businesses, that are available to users.

[0835] "Means for sentiment analysis of data" refers to a function or device for analyzing the sentiment of collected text data and identifying its emotional tendencies, such as positive, negative, or neutral.

[0836] "Means for automatically generating drafts of social media posts and press releases" refers to a function or device for automatically creating text for new social media posts or press releases based on the analysis results.

[0837] This invention is a system that efficiently generates content related to effective food delivery services for target users. This system is mainly composed of a server and terminals, and by interoperating with each other's functions, it provides high-quality content to users.

[0838] 1. System Configuration

[0839] The system consists of the following elements:

[0840] server:

[0841] Data Collection Module: Automatically collects online news articles, social media posts, and blog posts related to food delivery. Specifically, this data is obtained using APIs.

[0842] Data Analysis Module: Analyzes collected data using natural language processing techniques to identify sentiment and interests. This process involves using the TextBlob library to perform sentiment analysis, such as positive, negative, or neutral.

[0843] Content generation module: Automatically generates drafts of new social media posts and press releases based on the analysis results. It uses generative AI models (e.g., GPT-3) to create content. For example, it can generate drafts based on a prompt such as, "Please highlight healthy and delicious dishes to introduce the new menu."

[0844] Profanity Correction Module: Automatically corrects profanity and misleading language in the generated drafts by detecting specific keywords and phrases and replacing them with appropriate ones.

[0845] Database: Stores collected data and generated content.

[0846] Device:

[0847] User Interface: Allows users to view, edit, and finalize the generated content. Implemented as a web browser or smartphone application.

[0848] Content display and editing function: Generated content can be displayed and easily edited by the user. After final confirmation, it is possible to post it to social media or the official page.

[0849] 2. Specific Examples

[0850] The system collects data about "new product launches." The data analysis module analyzes the data to identify positive keywords, such as "new healthy menu items," that interest target customers. Based on the analysis, the content generation module generates drafts for social media posts, such as:

[0851] "Our new products are designed to be environmentally friendly and provide healthy and delicious food. Give them a try!"

[0852] The expression correction module inspects this content and corrects it to a more neutral expression, such as changing "Excellent" to "Highly rated." The corrected content is sent to the device, where the user can make a final check and, if necessary, make additional corrections before posting it to social media or the official page.

[0853] The above is a specific embodiment for carrying out the invention, which makes it possible to generate content that appeals to target users effectively and quickly, and to significantly improve the efficiency of marketing activities.

[0854] Hardware and software used

[0855] Data collection: API (e.g. News API)

[0856] Data Analysis: Python's TextBlob Library

[0857] Content generation: OpenAI's GPT models (e.g., GPT-3)

[0858] Profanity fix: Custom Python scripts

[0859] Prompt Sentence Examples

[0860] "Highlight healthy and tasty dishes as an introduction to your new menu."

[0861] "Write a press release that succinctly explains the new features of your food delivery service."

[0862] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0863] Step 1:

[0864] Data collection

[0865] The server automatically collects online news articles, social media posts, and blog posts related to food delivery. The input is specific keywords (e.g., "food delivery," "food delivery," etc.), and the output is a dataset related to these keywords. It uses an API (e.g., News API) to obtain information based on the specific keywords and stores the collected data in a local database.

[0866] Step 2:

[0867] Data analysis

[0868] The data collected by the server is analyzed using natural language processing techniques. In particular, sentiment analysis is performed using Python's TextBlob library. The input is the dataset collected in step 1, and the output is the sentiment classification results (positive, negative, neutral) and keywords of interest. TextBlob is applied to the dataset to extract the sentiment score and keywords for each sentence.

[0869] Step 3:

[0870] Content Generation

[0871] The server automatically generates drafts for social media posts and press releases using a generative AI model (e.g., GPT-3) based on the sentiment analysis results and keywords of interest. The input is the sentiment analysis results and keywords obtained in step 2, as well as a prompt entered by the user (e.g., "Please emphasize healthy and delicious dishes as an introduction to the new menu item."). The output is the generated text. The prompt and analysis results are input into the generative AI model to generate new content.

[0872] Step 4:

[0873] Correction of inappropriate expressions

[0874] The server inspects the generated content and corrects profanity. The input is the text generated in step 3, and the output is the corrected text. It runs a custom script that searches for specific keywords and replaces profanity or offensive phrases with neutral ones.

[0875] Step 5:

[0876] Content Output

[0877] The server sends the final modified content to the terminal, where the user can review and edit the content. The input is the text modified in step 4, and the output is the final content for the user to edit and post. The content is displayed through a user interface, and the user can make fine adjustments as needed.

[0878] Step 6:

[0879] User Verification and Submission

[0880] The user reviews the final content, makes any necessary adjustments, and then posts it to social media or an official page. The input is the final content displayed in step 5, and the output is the published content. When the user presses the confirm button, it is automatically posted to the selected platform.

[0881] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0882] This invention is a system that efficiently generates press releases and social media posts that resonate with target users and adjusts the tone of the content by recognizing user emotions. This system achieves effective and safe content creation by using information collection, data analysis, content generation, correction of inappropriate expressions, output, and an emotion engine.

[0883] 1. System Configuration

[0884] The system mainly consists of the following components:

[0885] 1. Server:

[0886] Data Collection Module

[0887] Data Analysis Module

[0888] Content Generation Module

[0889] Expression Correction Module

[0890] Emotion Engine

[0891] Database

[0892] 2. Terminal:

[0893] User Interface

[0894] Content display and editing function

[0895] Emotion recognition function

[0896] 2. Program processing (natural language explanation)

[0897] How information is collected

[0898] The server collects large amounts of information related to the target user from online sources, such as news articles, social media posts, past press releases, and relevant blog posts, using web scraping and APIs. The collected data is then stored in a centralized database.

[0899] A means of analyzing information

[0900] The server's data analysis module analyzes the collected information using natural language processing technology. For example, it extracts topics and keywords that interest the target user demographic and performs sentiment analysis to categorize positive, negative, and neutral expressions. The analysis results are then used in the next content generation module.

[0901] A means of generating content

[0902] The server's content generation module automatically generates drafts of press releases and social media posts based on the analysis results, using a generative AI model (e.g., GPT-3) to create new text from the analyzed data and create content that effectively reaches the target audience.

[0903] How to fix inappropriate language

[0904] The server's expression correction module automatically detects inappropriate language or words that pose a risk of causing controversy in the generated draft and corrects them to appropriate language. For example, it corrects "offensive language" to "neutral language." This process minimizes the inappropriate impact on users and society.

[0905] Emotion engine that recognizes user emotions

[0906] The server is equipped with an emotion engine that analyzes real-time emotional data from users. For example, it recognizes their emotional state from facial expressions, voice, and text while they are using the system. Based on the emotion recognition results, the content generation module can adjust the tone and expression of the generated text. This enables the creation of more effective content that is tailored to the user's emotions.

[0907] A means of outputting content

[0908] The terminal allows the user to review and edit the revised final content. The user interface displays the generated text and includes functionality that allows the user to make fine adjustments as needed.

[0909] Specific examples

[0910] The server collects data on "new product launches." Next, the data analysis module analyzes user demographic interests and identifies positive keywords related to "environmentally friendly new products." Based on the analysis results, the content generation module generates a draft that reads, "Our new product is designed with environmental considerations in mind and is energy efficient." The expression correction module corrects the overstatement in this draft, replacing "excellent" with "highly acclaimed."

[0911] Furthermore, the emotion engine analyzes the user's real-time emotional data and adjusts the expression to be more positive if the user shows excitement or curiosity. The revised draft is sent to the user's device, where the user performs a final check and is published after approval.

[0912] In this way, the present invention automatically generates and modifies effective, low-risk content, and adjusts the tone to match the user's emotions, thereby streamlining marketing activities.

[0913] The processing flow will be explained below.

[0914] Step 1:

[0915] The server collects information related to the target user from online sources, specifically news articles, social media posts, past press releases, and relevant blog posts, automatically using web scraping and APIs. The collected data is then stored in a centralized database.

[0916] Step 2:

[0917] The data collected by the server is analyzed in the data analysis module. Natural language processing technology is used to extract topics and keywords that interest the target user demographic, and sentiment analysis is performed to classify expressions in the data as positive, negative, or neutral. The analysis results are used in the next content generation module.

[0918] Step 3:

[0919] The server's content generation module automatically generates drafts of press releases and social media posts based on the analysis results. Using a generative AI model (e.g., GPT-3), the analyzed and identified keywords and concepts are compiled into text. At this stage, content is created that effectively reaches the target audience.

[0920] Step 4:

[0921] The server's expression correction module inspects the generated draft. Using an automatic filtering algorithm, it detects inappropriate language or areas at high risk of flaming, and corrects them to appropriate language. For example, by replacing "offensive language" with "neutral language," the risk of misunderstanding or flaming is reduced.

[0922] Step 5:

[0923] The server's emotion engine analyzes the user's real-time emotional data, recognizing their emotional state from facial expressions, voice, and text data, and adjusting the tone of the content as needed. For example, if the user expresses joy or excitement, the tone of the content will be made more positive.

[0924] Step 6:

[0925] The device will then display the corrected and adjusted content to the user, who can then review the generated text through a user interface and make any necessary final adjustments or refinements.

[0926] Step 7:

[0927] The user will then do a final review and approve the revised content. Marketing staff will then review the draft, make any necessary revisions, and approve it for publication on the official website and social media.

[0928] Step 8:

[0929] The device then posts the final approved content to the company's official website or social media, making the generated content public and reaching the target audience.

[0930] Example 2

[0931] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0932] With conventional content generation systems, it was difficult to quickly and accurately generate content that would effectively appeal to target users. Furthermore, the generated content could contain inappropriate language, and the tone of the content was not automatically adjusted to take user emotions into account, making it difficult to maximize marketing effectiveness.

[0933] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0934] In this invention, the server includes means for collecting information related to target users, means for analyzing the collected information to identify the characteristics of the user demographic, means for generating content based on the analysis results, means for correcting inappropriate expressions in the generated content, means for recognizing the emotional state of the user and adjusting the tone and expression of the content, and means for outputting the corrected content. This makes it possible to quickly and efficiently generate content that appeals appropriately to target users and further adjust the tone to match the user's emotions.

[0935] "Target Audience" refers to the specific customer group or individual for whom a particular marketing activity or piece of content is intended.

[0936] "Information collection methods" refers to the technologies and devices used to gather online news articles, social media posts, blog posts, past press releases, etc.

[0937] "Means for analyzing information" refers to technology and software that uses natural language processing technology to analyze collected information and identify the characteristics of user demographics.

[0938] "Means for generating content" refers to software or programs that use generative AI models to create effective press releases, social media posts, and other written content based on the analysis results.

[0939] "Profanity Modification Measures" refers to technology or software that detects and modifies excessive or offensive language in generated content.

[0940] "Means for recognizing a user's emotional state" refers to technologies or devices that analyze a user's emotional data in real time and adjust the tone and expression of content to match that emotion.

[0941] "Means for outputting modified content" refers to the functionality or device that provides the final modified content to the user and allows it to be viewed and edited through a user interface.

[0942] This invention is a system that efficiently generates press releases and social media posts that resonate with target users and adjusts the tone of the content by recognizing user emotions. This system achieves effective and safe content creation by using information collection, data analysis, content generation, correction of inappropriate expressions, output, and an emotion engine.

[0943] System configuration

[0944] The system mainly consists of the following components:

[0945] 1. Server:

[0946] Data Collection Module

[0947] Data Analysis Module

[0948] Content Generation Module

[0949] Expression Correction Module

[0950] Emotion Engine

[0951] Database

[0952] 2. Terminal:

[0953] User Interface

[0954] Content display and editing function

[0955] Emotion recognition function

[0956] Program processing

[0957] How information is collected

[0958] The server collects large amounts of information related to the target user from online sources. Specifically, it uses web scraping and APIs to collect news articles, social media posts, past press releases, and related blog posts. The collected data is then stored in a centralized database. For example, articles can be extracted from news sites using the Python library "BeautifulSoup."

[0959] A means of analyzing information

[0960] The server's data analysis module analyzes the collected information using natural language processing techniques, specifically using the NLTK library to tokenize the text and extract important topics and keywords, and the Hugging Face sentiment analysis model to classify positive, negative, and neutral expressions.

[0961] A means of generating content

[0962] The server's content generation module automatically generates drafts of press releases and social media posts based on the analysis results. Specifically, it uses a generative AI model (e.g., GPT-3) to create new sentences from the analyzed data. For example, the prompt sentence could be, "Can you give me some ideas for a press release that highlights the features of our new environmentally friendly product?"

[0963] How to fix inappropriate language

[0964] The server's expression correction module automatically detects inappropriate expressions or words that may pose a risk of causing controversy in the generated draft and corrects them to appropriate expressions. For example, it converts "excellent" to "highly acclaimed." This process minimizes the inappropriate impact on users and society.

[0965] Emotion engine that recognizes user emotions

[0966] The server is equipped with an emotion engine that analyzes users' real-time emotional data. Specifically, it uses OpenCV and DeepFace to recognize the user's emotional state from facial expressions, voice, text, and other information while using the system. Based on the emotion recognition results, the content generation module adjusts the tone and expression of the generated text. This enables the creation of more effective content that matches the user's emotions.

[0967] A means of outputting content

[0968] The terminal allows the user to review and edit the final revised content. The user interface displays the generated text and includes a function that allows the user to make fine adjustments as needed. The generated content can be published on social media or as a press release.

[0969] Specific examples

[0970] The server collects data related to "new product launches." Next, the data analysis module analyzes user demographic interests and identifies positive keywords related to "environmentally friendly new products." Based on the analysis results, the content generation module generates a draft that reads, "Our new product is designed with environmental considerations in mind and is energy efficient." The expression correction module corrects the excessive expression in this draft, replacing "excellent" with "highly acclaimed."

[0971] Furthermore, the emotion engine analyzes the user's real-time emotional data and adjusts the expression to be more positive if the user shows excitement or curiosity. The revised draft is sent to the user's device, where the user performs a final check and is published after approval.

[0972] In this way, the present invention can significantly improve the efficiency of marketing activities by automatically generating and modifying effective, low-risk content and adjusting the tone to suit the user's emotions.

[0973] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0974] Step 1: Gather information

[0975] The server collects information related to the target user from online sources. Specifically, it uses Python's BeautifulSoup library to scrape articles from specific news sites and blogs. It also uses the APIs of the target sites to retrieve social media posts and past press releases. This information is then stored in a database.

[0976] Input: URL of a website or API relevant to your target users

[0977] Data processing: web scraping, API calls

[0978] Output: Collected text data

[0979] Step 2: Analyze the information

[0980] The server's data analysis module analyzes the collected information. First, it uses the NLTK library to tokenize the text and extract frequent keywords and important topics. Next, it uses Hugging Face's sentiment analysis model to classify the sentiment of the text as positive, negative, or neutral, which allows it to identify the interests of the target user demographic.

[0981] Input: Collected text data

[0982] Data processing: tokenization, keyword extraction, sentiment analysis

[0983] Output: Analyzed information (keywords, sentiment classification)

[0984] Step 3: Generate content

[0985] The server's content generation module generates prompts based on the analyzed information and uses a generative AI model (e.g., GPT-3) to draft press releases and social media posts. For example, the prompt might be, "Can you give us some ideas for a press release highlighting the features of our new eco-friendly product?"

[0986] Input: Parsed information, prompt

[0987] Data processing: Automatic text generation using AI models

[0988] Output: Generated content draft

[0989] Step 4: Fix inappropriate language

[0990] The server's correction module scans the generated draft for inappropriate or excessive language, for example, converting "excellent" to "highly acclaimed." This process makes the content more neutral and less risky.

[0991] Input: Generated content draft

[0992] Data processing: text analysis, expression correction

[0993] Output: Revised content draft

[0994] Step 5: Recognizing User Emotions

[0995] The server's emotion engine analyzes the user's real-time emotion data, for example, by using OpenCV and DeepFace to capture the user's facial expressions with a camera and recognize their emotional state. Based on this data, the content generation module adjusts the tone of the generated content.

[0996] Input: Real-time user emotion data (facial expressions, voice, text)

[0997] Data processing: facial expression analysis, emotion recognition

[0998] Output: Emotionally adjusted tone-adjusted draft content

[0999] Step 6: Outputting content

[1000] The device allows the user to review and edit the final revised content. A user interface displays the generated content and provides editing functions for the user to make fine adjustments as needed. Once final review is complete, the content is published to social media platforms and websites.

[1001] Input: Revised content draft

[1002] Data processing: Display and editing through the user interface

[1003] Output: Final content approved by the user

[1004] (Application example 2)

[1005] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1006] Conventional content generation systems have limited capabilities for automatically generating effective advertising content for target users, and are unable to adjust the tone of the advertisement based on the user's real-time emotions. As a result, the effectiveness of advertisements is often limited, and real-time adjustments to match user emotions are required.

[1007] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting information, means for analyzing the collected information to identify user demographic characteristics, means for generating content based on the analysis results, means for correcting inappropriate expressions in the generated content, means for outputting the corrected content, and means for acquiring real-time emotional data of the user and adjusting the tone of the content based on the emotional data. This makes it possible to generate advertising content that is effective for target users in real time and adjust the tone to match the user's emotions.

[1008] "Information collection means" refers to the methods and devices used to collect data related to the target user online.

[1009] "Means for analyzing collected information to identify the characteristics of the user demographic" refers to a method or device that analyzes collected data and identifies characteristics such as the interests, concerns, and behavioral patterns of target users.

[1010] "Content generation means" refers to a method or device that automatically creates content such as advertisements, press releases, and social media posts based on the characteristics of the user demographic.

[1011] "Means for correcting inappropriate language in generated content" refers to a method or device that automatically detects and corrects offensive, prejudiced, or inappropriate language in text generated by the system, such as advertisements or articles.

[1012] "Means for outputting modified content" refers to the method or device that ultimately provides users with modified or adjusted content such as advertisements, press releases, and social media posts.

[1013] "Means for acquiring real-time emotional data of a user and adjusting the tone of content based on that emotional data" refers to a method or device that detects a user's emotional state in real time from facial expressions, voice, text, etc., and dynamically adapts the tone and expression of content based on that.

[1014] MODE FOR CARRYING OUT THE INVENTION

[1015] The present invention relates to a system for efficiently and effectively generating advertising content and adjusting the tone based on a user's real-time emotions. The system includes means for information collection, data analysis, content generation, inappropriate expression correction, content output, and emotion recognition, and is capable of effectively and adaptively generating advertising content.

[1016] 1. System Configuration

[1017] server

[1018] Information Collection Module

[1019] The server uses web scraping and APIs to collect information related to the target users, including news articles, social media posts, past press releases, relevant blog posts, etc. This data is then stored in a centralized database.

[1020] Data Analysis Module

[1021] The server analyzes the collected information using natural language processing techniques (e.g., TextBlob) to identify demographic interests and characteristics and categorize positive, negative, and neutral expressions.

[1022] Content Generation Module

[1023] Based on the analysis results, a generative AI model (e.g., OpenAI's GPT-3) is used to automatically generate draft copy for advertising and social media posts. For example, a prompt such as "Generate a social media post about a new product launch with a positive tone" is generated.

[1024] Expression Correction Module

[1025] The server automatically detects inappropriate language or words that pose a risk of causing controversy in the generated draft and edits them to appropriate language, minimizing the impact on users and society.

[1026] Emotion Recognition Module

[1027] The server acquires the user's real-time emotions and adjusts the tone of the generated content accordingly. For example, it analyzes the user's facial expressions, voice, and text data when operating the system, and adjusts the tone according to their emotional state (e.g., excitement, curiosity, confusion, anxiety).

[1028] Terminal

[1029] User Interface

[1030] Users can review and edit the generated content through an on-device user interface, which displays the final content adjusted based on emotion recognition and includes the ability for users to fine-tune it as needed.

[1031] Specific examples

[1032] For example, the server collects data on "new product launches." Then, the data analysis module analyzes the interests of the user demographic and identifies positive keywords related to "new eco-friendly products." Based on the analysis results, the content generation module generates a draft of a social media post like this:

[1033] Generate social media posts about your new product launch. Keep the tone positive.

[1034] The generated content is the following sentence:

[1035] Our new products are designed to be environmentally friendly and are energy efficient! Give them a try!

[1036] The expression correction module corrects the over-expression in this draft, replacing "excellent" with "highly rated." Furthermore, the emotion recognition module analyzes the user's real-time emotional data and adjusts the expression to be more positive if the user shows excitement or curiosity. For example:

[1037] Our new product is designed to be environmentally friendly and has received rave reviews! Give it a try!

[1038] The revised draft is sent to the user's device, where the user reviews it and publishes it after approval. This process allows the invention to effectively reach the target audience and adjust the tone appropriately based on real-time emotions.

[1039] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1040] Program processing flow

[1041] Step 1: Gather information

[1042] The server collects information related to the target users from online sources, such as news articles, blog posts, social media posts, and past press releases, using web scraping or APIs. The input is keywords (e.g., "new product announcement," "environmentally friendly"), and the output is the collected text data.

[1043] Step 2: Data analysis

[1044] The server analyzes the collected information using natural language processing technology. For example, it uses TextBlob to perform sentiment analysis of text and classify it into positive, negative, and neutral expressions. The input is the collected text data, and the output is keywords related to the characteristics of the user demographic.

[1045] Step 3: Content generation

[1046] The server generates drafts of advertising copy and social media posts using a generative AI model (e.g., OpenAI's GPT-3) based on the analysis results. The input is keywords obtained from the analysis results and a prompt (e.g., "Generate a social media post about a new product launch. The tone should be positive."), and the output is the generated content.

[1047] Step 4: Correct inappropriate language

[1048] The server detects inappropriate language in the generated content and corrects it to appropriate language, e.g., changing excessive language to neutral language. The input is the generated draft, and the output is the corrected content.

[1049] Step 5: User Emotion Recognition

[1050] The server acquires the user's real-time emotional data (e.g., facial expressions, voice, text) and analyzes it with an emotion recognition library (e.g., DeepFace). The input is the user's emotional data collected in real time, and the output is the recognized emotional state.

[1051] Step 6: Adjust the tone

[1052] The server adjusts the tone of the generated content based on the recognized emotional state. For example, if the user is excited, it adjusts the tone to a more positive one, and if the user is confused, it adjusts the tone to a more calm, non-sensational one. The input is the modified content and the user's emotional state, and the output is the final adjusted content.

[1053] Step 7: Content Output

[1054] The server sends the final adjusted content to the terminal and displays it on the terminal's user interface. The input is the adjusted content, and the output is the content displayed to the user. The user can then perform further editing or final review as needed.

[1055] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1056] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1057] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1058] [Fourth embodiment]

[1059] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1060] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1061] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1062] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1063] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1064] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1065] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1066] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1067] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1068] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1069] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1070] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1071] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1072] This invention is a system that efficiently generates press releases and social media posts that are well-suited to target users. This system achieves effective and safe content creation by using the following methods: information collection, data analysis, content generation, correction of inappropriate expressions, and output.

[1073] 1. System Configuration

[1074] The system mainly consists of the following components:

[1075] 1. Server:

[1076] Data Collection Module

[1077] Data Analysis Module

[1078] Content Generation Module

[1079] Expression Correction Module

[1080] Database

[1081] 2. Terminal:

[1082] User Interface

[1083] Content display and editing function

[1084] 2. Program processing (natural language explanation)

[1085] How information is collected

[1086] The server collects a large amount of information related to the target user from online sources. This process involves gathering data such as news articles, social media posts, and blog posts based on specific keywords (e.g., "environment" or "energy efficiency"). The collected data is then stored in a database.

[1087] A means of analyzing information

[1088] The server's data analysis module analyzes the collected information using natural language processing technology. For example, it extracts topics and keywords that interest the target user demographic and performs sentiment analysis to categorize positive, negative, and neutral expressions. The analysis results are then used in the next content generation module.

[1089] A means of generating content

[1090] The server's content generation module automatically generates drafts of press releases and social media posts based on the analysis results. Specifically, a generative AI model (e.g., GPT-3) creates new text from the analyzed data to effectively reach the target user demographic. For example, when creating a press release on the theme of "new product launch," it can include information about environmentally friendly design and energy efficiency.

[1091] How to fix inappropriate language

[1092] The server's expression correction module automatically detects inappropriate language or words that pose a risk of causing controversy in the generated draft and corrects them to appropriate language. For example, it replaces "language attacking competitors" with "neutral language." This process minimizes the inappropriate impact on users and society.

[1093] A means of outputting content

[1094] The device allows users to review and edit the final content. The user interface displays the generated text and includes a function that allows users to make fine adjustments as needed. For example, a marketing person may review a draft of a press release, make final adjustments, and then post it on the company's official page or social media.

[1095] Specific examples

[1096] The server collects data related to "new product launches." Next, the data analysis module analyzes the interests of users and identifies positive keywords related to "environmentally friendly new products." Based on the analysis results, the content generation module generates a draft that reads, "Our new product is designed with environmental considerations in mind and is energy efficient." The expression correction module corrects the excessive expression in this draft, replacing "excellent" with "highly acclaimed." The corrected draft is sent to the user's device, where the user performs a final review and publishes it after approval.

[1097] In this way, the present invention can automatically generate and revise effective, low-risk content, making marketing activities more efficient.

[1098] The processing flow will be explained below.

[1099] Step 1:

[1100] The server collects information related to the target user from online sources, such as news articles, social media posts, past press releases, and relevant blog posts, using web scraping and APIs. The collected data is then stored in a centralized database.

[1101] Step 2:

[1102] The data collected by the server is analyzed by the data analysis module. Specifically, natural language processing technology is used to extract topics and keywords from the collected data and identify the characteristics of the interests and concerns of the target user demographic. Sentiment analysis is also performed to classify expressions in the data as positive, negative, or neutral.

[1103] Step 3:

[1104] Based on the analysis results, the server uses a content generation module to automatically generate drafts of press releases and social media posts. Using a generative AI model (e.g., GPT-3), the server combines the analyzed and identified keywords and concepts into sentences to create content that effectively reaches the target audience.

[1105] Step 4:

[1106] The server then inspects the generated draft with a language correction module, which uses an automatic filtering algorithm to detect inappropriate language or potentially inflammatory words and replace them with appropriate language. For example, it can correct "offensive language" to "neutral language."

[1107] Step 5:

[1108] The server sends the revised draft to the user's device for review and editing. A user interface displays the generated content and allows the user to make final adjustments and fine-tuning.

[1109] Step 6:

[1110] The user performs a final review and approves the revised content. For example, a marketing person reviews the draft, makes any necessary revisions, and then finally approves it for publication.

[1111] Step 7:

[1112] The device will then post the final approved content to the company's official website or social media, making it publicly available and reaching the target audience.

[1113] Example 1

[1114] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1115] Conventional content generation systems are unable to accurately grasp the interests of target users, making it difficult to generate effective content. Furthermore, if the generated content contains inappropriate language, manual correction is required, resulting in inefficiency. Furthermore, there is no environment in place for quickly checking and editing the generated content, which means it takes a long time to reach the final output.

[1116] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1117] In this invention, the server includes means for collecting information related to target users, means for analyzing the collected information to identify user demographic characteristics, means for generating content based on the analysis results, means for correcting inappropriate language in the generated content, and means for outputting the corrected content. This makes it possible to automatically generate effective content tailored to target users and quickly correct inappropriate language. Furthermore, by including a terminal with a user interface for viewing and editing the content, the generated content can be immediately viewed, edited as necessary, and quickly published.

[1118] A "target user" is a group of users who are expected to have an interest in or demand for a particular product or service.

[1119] "Information gathering methods" are the tools and processes used to gather relevant data online based on specific keywords.

[1120] A "database" is a system that stores collected information or data so that it can be accessed and processed at a later time.

[1121] "Means for analyzing information" refers to methods for analyzing collected data using natural language processing technology, etc., to identify the characteristics and interests of user groups.

[1122] "Natural language processing technology" is a technology that enables computers to understand, generate, and manipulate human language.

[1123] "Means of generating content" refers to processes or tools that automatically generate effective text and information for target users based on analysis results.

[1124] A "generative AI model" is an artificial intelligence model that has been pre-trained with large amounts of data and has the ability to generate advanced text based on input.

[1125] A "prompt" is an explanatory or instructional text to be input into a generative AI model, indicating the theme or topic of the content to be generated.

[1126] "Profanity correction methods" are processes or tools that automatically detect profanity or risky language in generated content and correct it to appropriate language.

[1127] "Means for outputting content" refers to the functionality that displays the final revised content and allows users to access, review, and edit it.

[1128] "User interface" refers to the screens and operating means through which a user interacts with the system and checks and edits the generated content.

[1129] "Terminal" means a device that a User uses to view or edit content using the System, including, for example, a PC or tablet.

[1130] This invention is a system that efficiently generates press releases and social media posts that are well-suited to target users. This system achieves effective and safe content creation by using the following methods: information collection, data analysis, content generation, correction of inappropriate expressions, and output.

[1131] System configuration

[1132] The system mainly consists of the following components:

[1133] 1. Server:

[1134] Data Collection Module

[1135] Data Analysis Module

[1136] Content Generation Module

[1137] Expression Correction Module

[1138] Database

[1139] 2. Terminal:

[1140] User Interface

[1141] Content display and editing function

[1142] Methods of collecting information

[1143] A server collects large amounts of information relevant to a target user from online sources. This is done based on specific keywords (e.g., "environment" or "energy efficiency") and includes data such as news articles, social media posts, and blog posts. The collected data is then stored in a database. Technologies used in this process include web scraping tools (e.g., BeautifulSoup) and APIs (e.g., Twitter API).

[1144] Information analysis methods

[1145] The server's data analysis module analyzes the collected information using natural language processing techniques. Specifically, it performs topic modeling (e.g., LDA) to extract topics and keywords that interest the target user demographic. It also uses sentiment analysis tools (e.g., VADER) to classify the sentiment of the collected data into positive, negative, or neutral.

[1146] Means of content generation

[1147] The server's content generation module automatically generates drafts of press releases and social media posts based on the analysis results. This step uses a generative AI model (e.g., GPT-3). For example, the following prompt is used:

[1148] "Write a press release for a new product launch. Your target demographic is environmentally conscious and focused on energy efficiency. Highlight the product's environmentally friendly design and high energy efficiency as features."

[1149] How to correct inappropriate language

[1150] The server's expression correction module automatically detects and corrects inappropriate language and words that pose a risk of causing controversy in the generated draft. For example, it uses a text analysis tool (e.g., TextBlob) to detect and correct "language attacking competitors" and other such words to make them more neutral.

[1151] Content output method

[1152] The device allows users to review and edit the final content. The user interface displays the generated text and includes an editing function for users to make fine adjustments. Specifically, marketers can review the draft press release, make any necessary corrections, and then post it on the company's official website or social media.

[1153] Specific examples

[1154] The server collects data related to "new product launches," and the data analysis module analyzes user demographic interests and identifies positive keywords related to "environmentally friendly new products." Based on the analysis results, the content generation module generates a draft that reads, "Our new product is designed with environmental considerations in mind and is also energy efficient." The expression correction module corrects the excessive expression in this draft, replacing "excellent" with "highly acclaimed." The corrected draft is sent to the user's device, where the user performs a final review and publishes it after approval.

[1155] In this way, the present invention can automatically generate and revise effective, low-risk content, making marketing activities more efficient.

[1156] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1157] Step 1: Gather information

[1158] The server collects information relevant to the target user from the internet. As input, it is given a list of specific keywords (e.g., "environment," "energy efficiency," etc.). Specifically, it uses web scraping tools (e.g., BeautifulSoup) or APIs (e.g., Twitter API) to retrieve data such as news articles, social media posts, and blog posts. The collected data is stored in a database. As output, it obtains a large amount of unanalyzed data stored in the database.

[1159] Step 2: Information analysis

[1160] The server's data analysis module analyzes the information collected in step 1. The input is the unanalyzed data stored in the database. Specifically, natural language processing techniques (e.g., NLTK, Spacy) are used to tokenize the text of the collected data. Next, topic modeling (e.g., LDA) is performed to extract topics and keywords that interest the target user demographic. A sentiment analysis tool (e.g., VADER) is also used to classify the data into positive, negative, and neutral. The output includes the analysis results, which include data on the topics of interest to the target user demographic and sentiment classification.

[1161] Step 3: Content generation

[1162] The server's content generation module automatically generates drafts of press releases and social media posts based on the analysis results obtained in step 2. The input is data on the topics of interest and sentiment classification of the target user demographic. Specifically, the generative AI model (e.g., GPT-3) is given a prompt such as the following: "Please write a press release about the launch of a new product. The target user demographic is concerned with the environment and focuses on energy efficiency. Please emphasize the product's environmentally friendly design and high energy efficiency as features." Based on this prompt, the generative AI model generates text that will effectively reach the target users. The output is a generated draft.

[1163] Step 4: Correct inappropriate language

[1164] The server's expression correction module detects and corrects inappropriate language and words that pose a risk of causing controversy in the draft generated in step 3. The input is the generated draft. Specifically, it uses a text analysis tool (e.g., TextBlob) to detect offensive language and extreme assertions and replaces them with neutral language. For example, it corrects "language attacking competitors" to "neutral language." The output is a draft with inappropriate language corrected.

[1165] Step 5: Content Output

[1166] The user uses the terminal to check the final content corrected in step 4 and edit it as necessary. The input is the corrected draft. Specifically, the generated text is displayed on the user interface, and the user edits the necessary parts with the mouse or keyboard. For example, a marketing person makes a final check of a press release draft, makes any necessary corrections, and then posts it on the company's official website or social media. The output is the final corrected and checked text.

[1167] Through the above steps, the system efficiently generates effective, low-risk content and supports marketing activities.

[1168] (Application example 1)

[1169] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1170] In modern digital marketing, creating content that effectively appeals to target users is extremely important. However, generating effective content efficiently and disseminating information specific to food delivery services while removing inappropriate language is a time-consuming and labor-intensive task. In particular, the data collection and analysis stages from news articles, social media posts, and blog articles require a great deal of effort, which can lead to variations in the quality of the final content.

[1171] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1172] In this invention, the server includes means for automatically collecting news, social media posts, and blog articles related to food delivery, means for performing sentiment analysis on the collected data to identify positive, negative, and neutral expressions, and means for automatically generating drafts of social media posts and press releases related to food delivery services. This makes it possible to efficiently generate content that effectively appeals to target users and disseminate high-quality information by removing inappropriate expressions.

[1173] "Target users" refers to the group of customers who are the primary target of a particular marketing activity, product, or service.

[1174] "Means for collecting information" refers to functions or devices for automatically obtaining information related to the target user from various data sources.

[1175] "Means for analyzing and identifying the characteristics of the user demographic" refers to a function or device for analyzing collected information to understand characteristics such as the interests, behavioral patterns, and emotional tendencies of target users.

[1176] "Means for generating content" refers to a function or device for creating content such as text and images that effectively appeal to target users based on the analysis results.

[1177] "Means for correcting inappropriate language" refers to a function or device that automatically detects potentially misleading or inappropriate language in generated content and changes it to appropriate language.

[1178] "Means for outputting modified content" refers to the function or device that provides the user with the final content with the inappropriate language corrected, and displays it in a format that allows for review and editing.

[1179] "Food delivery" refers to a service that aims to deliver meals ordered by customers via the Internet to a specified location.

[1180] "News, social media posts, and blog posts" refers to types of online public information sources, such as news articles, posts on social networking sites, and content from blogs run by individuals or businesses, that are available to users.

[1181] "Means for sentiment analysis of data" refers to a function or device for analyzing the sentiment of collected text data and identifying its emotional tendencies, such as positive, negative, or neutral.

[1182] "Means for automatically generating drafts of social media posts and press releases" refers to a function or device for automatically creating text for new social media posts or press releases based on the analysis results.

[1183] This invention is a system that efficiently generates content related to effective food delivery services for target users. This system is mainly composed of a server and terminals, and by interoperating with each other's functions, it provides high-quality content to users.

[1184] 1. System Configuration

[1185] The system consists of the following elements:

[1186] server:

[1187] Data Collection Module: Automatically collects online news articles, social media posts, and blog posts related to food delivery. Specifically, this data is obtained using APIs.

[1188] Data Analysis Module: Analyzes collected data using natural language processing techniques to identify sentiment and interests. This process involves using the TextBlob library to perform sentiment analysis, such as positive, negative, or neutral.

[1189] Content generation module: Automatically generates drafts of new social media posts and press releases based on the analysis results. It uses generative AI models (e.g., GPT-3) to create content. For example, it can generate drafts based on a prompt such as, "Please highlight healthy and delicious dishes to introduce the new menu."

[1190] Profanity Correction Module: Automatically corrects profanity and misleading language in the generated drafts by detecting specific keywords and phrases and replacing them with appropriate ones.

[1191] Database: Stores collected data and generated content.

[1192] Device:

[1193] User Interface: Allows users to view, edit, and finalize the generated content. Implemented as a web browser or smartphone application.

[1194] Content display and editing function: Generated content can be displayed and easily edited by the user. After final confirmation, it is possible to post it to social media or the official page.

[1195] 2. Specific Examples

[1196] The system collects data about "new product launches." The data analysis module analyzes the data to identify positive keywords, such as "new healthy menu items," that interest target customers. Based on the analysis, the content generation module generates drafts for social media posts, such as:

[1197] "Our new products are designed to be environmentally friendly and provide healthy and delicious food. Give them a try!"

[1198] The expression correction module inspects this content and corrects it to a more neutral expression, such as changing "Excellent" to "Highly rated." The corrected content is sent to the device, where the user can make a final check and, if necessary, make additional corrections before posting it to social media or the official page.

[1199] The above is a specific embodiment for carrying out the invention, which makes it possible to generate content that appeals to target users effectively and quickly, and to significantly improve the efficiency of marketing activities.

[1200] Hardware and software used

[1201] Data collection: API (e.g. News API)

[1202] Data Analysis: Python's TextBlob Library

[1203] Content generation: OpenAI's GPT models (e.g., GPT-3)

[1204] Profanity fix: Custom Python scripts

[1205] Prompt Sentence Examples

[1206] "Highlight healthy and tasty dishes as an introduction to your new menu."

[1207] "Write a press release that succinctly explains the new features of your food delivery service."

[1208] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1209] Step 1:

[1210] Data collection

[1211] The server automatically collects online news articles, social media posts, and blog posts related to food delivery. The input is specific keywords (e.g., "food delivery," "food delivery," etc.), and the output is a dataset related to these keywords. It uses an API (e.g., News API) to obtain information based on the specific keywords and stores the collected data in a local database.

[1212] Step 2:

[1213] Data analysis

[1214] The data collected by the server is analyzed using natural language processing techniques. In particular, sentiment analysis is performed using Python's TextBlob library. The input is the dataset collected in step 1, and the output is the sentiment classification results (positive, negative, neutral) and keywords of interest. TextBlob is applied to the dataset to extract the sentiment score and keywords for each sentence.

[1215] Step 3:

[1216] Content Generation

[1217] The server automatically generates drafts for social media posts and press releases using a generative AI model (e.g., GPT-3) based on the sentiment analysis results and keywords of interest. The input is the sentiment analysis results and keywords obtained in step 2, as well as a prompt entered by the user (e.g., "Please emphasize healthy and delicious dishes as an introduction to the new menu item."). The output is the generated text. The prompt and analysis results are input into the generative AI model to generate new content.

[1218] Step 4:

[1219] Correction of inappropriate expressions

[1220] The server inspects the generated content and corrects profanity. The input is the text generated in step 3, and the output is the corrected text. It runs a custom script that searches for specific keywords and replaces profanity or offensive phrases with neutral ones.

[1221] Step 5:

[1222] Content Output

[1223] The server sends the final modified content to the terminal, where the user can review and edit the content. The input is the text modified in step 4, and the output is the final content for the user to edit and post. The content is displayed through a user interface, and the user can make fine adjustments as needed.

[1224] Step 6:

[1225] User Verification and Submission

[1226] The user reviews the final content, makes any necessary adjustments, and then posts it to social media or an official page. The input is the final content displayed in step 5, and the output is the published content. When the user presses the confirm button, it is automatically posted to the selected platform.

[1227] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1228] This invention is a system that efficiently generates press releases and social media posts that resonate with target users and adjusts the tone of the content by recognizing user emotions. This system achieves effective and safe content creation by using information collection, data analysis, content generation, correction of inappropriate expressions, output, and an emotion engine.

[1229] 1. System Configuration

[1230] The system mainly consists of the following components:

[1231] 1. Server:

[1232] Data Collection Module

[1233] Data Analysis Module

[1234] Content Generation Module

[1235] Expression Correction Module

[1236] Emotion Engine

[1237] Database

[1238] 2. Terminal:

[1239] User Interface

[1240] Content display and editing function

[1241] Emotion recognition function

[1242] 2. Program processing (natural language explanation)

[1243] How information is collected

[1244] The server collects large amounts of information related to the target user from online sources, such as news articles, social media posts, past press releases, and relevant blog posts, using web scraping and APIs. The collected data is then stored in a centralized database.

[1245] A means of analyzing information

[1246] The server's data analysis module analyzes the collected information using natural language processing technology. For example, it extracts topics and keywords that interest the target user demographic and performs sentiment analysis to categorize positive, negative, and neutral expressions. The analysis results are then used in the next content generation module.

[1247] A means of generating content

[1248] The server's content generation module automatically generates drafts of press releases and social media posts based on the analysis results, using a generative AI model (e.g., GPT-3) to create new text from the analyzed data and create content that effectively reaches the target audience.

[1249] How to fix inappropriate language

[1250] The server's expression correction module automatically detects inappropriate language or words that pose a risk of causing controversy in the generated draft and corrects them to appropriate language. For example, it corrects "offensive language" to "neutral language." This process minimizes the inappropriate impact on users and society.

[1251] Emotion engine that recognizes user emotions

[1252] The server is equipped with an emotion engine that analyzes real-time emotional data from users. For example, it recognizes their emotional state from facial expressions, voice, and text while they are using the system. Based on the emotion recognition results, the content generation module can adjust the tone and expression of the generated text. This enables the creation of more effective content that is tailored to the user's emotions.

[1253] A means of outputting content

[1254] The terminal allows the user to review and edit the revised final content. The user interface displays the generated text and includes functionality that allows the user to make fine adjustments as needed.

[1255] Specific examples

[1256] The server collects data on "new product launches." Next, the data analysis module analyzes user demographic interests and identifies positive keywords related to "environmentally friendly new products." Based on the analysis results, the content generation module generates a draft that reads, "Our new product is designed with environmental considerations in mind and is energy efficient." The expression correction module corrects the overstatement in this draft, replacing "excellent" with "highly acclaimed."

[1257] Furthermore, the emotion engine analyzes the user's real-time emotional data and adjusts the expression to be more positive if the user shows excitement or curiosity. The revised draft is sent to the user's device, where the user performs a final check and is published after approval.

[1258] In this way, the present invention automatically generates and modifies effective, low-risk content, and adjusts the tone to match the user's emotions, thereby streamlining marketing activities.

[1259] The processing flow will be explained below.

[1260] Step 1:

[1261] The server collects information related to the target user from online sources, specifically news articles, social media posts, past press releases, and relevant blog posts, automatically using web scraping and APIs. The collected data is then stored in a centralized database.

[1262] Step 2:

[1263] The data collected by the server is analyzed in the data analysis module. Natural language processing technology is used to extract topics and keywords that interest the target user demographic, and sentiment analysis is performed to classify expressions in the data as positive, negative, or neutral. The analysis results are used in the next content generation module.

[1264] Step 3:

[1265] The server's content generation module automatically generates drafts of press releases and social media posts based on the analysis results. Using a generative AI model (e.g., GPT-3), the analyzed and identified keywords and concepts are compiled into text. At this stage, content is created that effectively reaches the target audience.

[1266] Step 4:

[1267] The server's expression correction module inspects the generated draft. Using an automatic filtering algorithm, it detects inappropriate language or areas at high risk of flaming, and corrects them to appropriate language. For example, by replacing "offensive language" with "neutral language," the risk of misunderstanding or flaming is reduced.

[1268] Step 5:

[1269] The server's emotion engine analyzes the user's real-time emotional data, recognizing their emotional state from facial expressions, voice, and text data, and adjusting the tone of the content as needed. For example, if the user expresses joy or excitement, the tone of the content will be made more positive.

[1270] Step 6:

[1271] The device will then display the corrected and adjusted content to the user, who can then review the generated text through a user interface and make any necessary final adjustments or refinements.

[1272] Step 7:

[1273] The user will then do a final review and approve the revised content. Marketing staff will then review the draft, make any necessary revisions, and approve it for publication on the official website and social media.

[1274] Step 8:

[1275] The device then posts the final approved content to the company's official website or social media, making the generated content public and reaching the target audience.

[1276] Example 2

[1277] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1278] With conventional content generation systems, it was difficult to quickly and accurately generate content that would effectively appeal to target users. Furthermore, the generated content could contain inappropriate language, and the tone of the content was not automatically adjusted to take user emotions into account, making it difficult to maximize marketing effectiveness.

[1279] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1280] In this invention, the server includes means for collecting information related to target users, means for analyzing the collected information to identify the characteristics of the user demographic, means for generating content based on the analysis results, means for correcting inappropriate expressions in the generated content, means for recognizing the emotional state of the user and adjusting the tone and expression of the content, and means for outputting the corrected content. This makes it possible to quickly and efficiently generate content that appeals appropriately to target users and further adjust the tone to match the user's emotions.

[1281] "Target Audience" refers to the specific customer group or individual for whom a particular marketing activity or piece of content is intended.

[1282] "Information collection methods" refers to the technologies and devices used to gather online news articles, social media posts, blog posts, past press releases, etc.

[1283] "Means for analyzing information" refers to technology and software that uses natural language processing technology to analyze collected information and identify the characteristics of user demographics.

[1284] "Means for generating content" refers to software or programs that use generative AI models to create effective press releases, social media posts, and other written content based on the analysis results.

[1285] "Profanity Modification Measures" refers to technology or software that detects and modifies excessive or offensive language in generated content.

[1286] "Means for recognizing a user's emotional state" refers to technologies or devices that analyze a user's emotional data in real time and adjust the tone and expression of content to match that emotion.

[1287] "Means for outputting modified content" refers to the functionality or device that provides the final modified content to the user and allows it to be viewed and edited through a user interface.

[1288] This invention is a system that efficiently generates press releases and social media posts that resonate with target users and adjusts the tone of the content by recognizing user emotions. This system achieves effective and safe content creation by using information collection, data analysis, content generation, correction of inappropriate expressions, output, and an emotion engine.

[1289] System configuration

[1290] The system mainly consists of the following components:

[1291] 1. Server:

[1292] Data Collection Module

[1293] Data Analysis Module

[1294] Content Generation Module

[1295] Expression Correction Module

[1296] Emotion Engine

[1297] Database

[1298] 2. Terminal:

[1299] User Interface

[1300] Content display and editing function

[1301] Emotion recognition function

[1302] Program processing

[1303] How information is collected

[1304] The server collects large amounts of information related to the target user from online sources. Specifically, it uses web scraping and APIs to collect news articles, social media posts, past press releases, relevant blog posts, etc. The collected data is then stored in a centralized database. For example, articles can be extracted from news sites using the Python library "BeautifulSoup."

[1305] A means of analyzing information

[1306] The server's data analysis module analyzes the collected information using natural language processing techniques, specifically using the NLTK library to tokenize the text and extract important topics and keywords, and the Hugging Face sentiment analysis model to classify positive, negative, and neutral expressions.

[1307] A means of generating content

[1308] The server's content generation module automatically generates drafts of press releases and social media posts based on the analysis results. Specifically, it uses a generative AI model (e.g., GPT-3) to create new sentences from the analyzed data. For example, the prompt sentence could be, "Can you give me some ideas for a press release that highlights the features of our new environmentally friendly product?"

[1309] How to fix inappropriate language

[1310] The server's expression correction module automatically detects inappropriate expressions or words that may pose a risk of causing controversy in the generated draft and corrects them to appropriate expressions. For example, it converts "excellent" to "highly acclaimed." This process minimizes the inappropriate impact on users and society.

[1311] Emotion engine that recognizes user emotions

[1312] The server is equipped with an emotion engine that analyzes users' real-time emotional data. Specifically, it uses OpenCV and DeepFace to recognize the user's emotional state from facial expressions, voice, text, and other information while using the system. Based on the emotion recognition results, the content generation module adjusts the tone and expression of the generated text. This enables the creation of more effective content that matches the user's emotions.

[1313] A means of outputting content

[1314] The terminal allows the user to review and edit the final revised content. The user interface displays the generated text and includes a function that allows the user to make fine adjustments as needed. The generated content can be published on social media or as a press release.

[1315] Specific examples

[1316] The server collects data related to "new product launches." Next, the data analysis module analyzes user demographic interests and identifies positive keywords related to "environmentally friendly new products." Based on the analysis results, the content generation module generates a draft that reads, "Our new product is designed with environmental considerations in mind and is energy efficient." The expression correction module corrects the excessive expression in this draft, replacing "excellent" with "highly acclaimed."

[1317] Furthermore, the emotion engine analyzes the user's real-time emotional data and adjusts the expression to be more positive if the user shows excitement or curiosity. The revised draft is sent to the user's device, where the user performs a final check and is published after approval.

[1318] In this way, the present invention can significantly improve the efficiency of marketing activities by automatically generating and modifying effective, low-risk content and adjusting the tone to suit the user's emotions.

[1319] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1320] Step 1: Gather information

[1321] The server collects information related to the target user from online sources. Specifically, it uses Python's BeautifulSoup library to scrape articles from specific news sites and blogs. It also uses the APIs of the target sites to retrieve social media posts and past press releases. This information is then stored in a database.

[1322] Input: URL of a website or API relevant to your target users

[1323] Data processing: web scraping, API calls

[1324] Output: Collected text data

[1325] Step 2: Analyze the information

[1326] The server's data analysis module analyzes the collected information. First, it uses the NLTK library to tokenize the text and extract frequent keywords and important topics. Next, it uses Hugging Face's sentiment analysis model to classify the sentiment of the text as positive, negative, or neutral, which allows it to identify the interests of the target user demographic.

[1327] Input: Collected text data

[1328] Data processing: tokenization, keyword extraction, sentiment analysis

[1329] Output: Analyzed information (keywords, sentiment classification)

[1330] Step 3: Generate content

[1331] The server's content generation module generates prompts based on the analyzed information and uses a generative AI model (e.g., GPT-3) to draft press releases and social media posts. For example, the prompt might be, "Can you give us some ideas for a press release highlighting the features of our new eco-friendly product?"

[1332] Input: Parsed information, prompt

[1333] Data processing: Automatic text generation using AI models

[1334] Output: Generated content draft

[1335] Step 4: Fix inappropriate language

[1336] The server's correction module scans the generated draft for inappropriate or excessive language, for example, converting "excellent" to "highly acclaimed." This process makes the content more neutral and less risky.

[1337] Input: Generated content draft

[1338] Data processing: text analysis, expression correction

[1339] Output: Revised content draft

[1340] Step 5: Recognizing User Emotions

[1341] The server's emotion engine analyzes the user's real-time emotion data, for example, by using OpenCV and DeepFace to capture the user's facial expressions with a camera and recognize their emotional state. Based on this data, the content generation module adjusts the tone of the generated content.

[1342] Input: Real-time user emotion data (facial expressions, voice, text)

[1343] Data processing: facial expression analysis, emotion recognition

[1344] Output: Emotionally adjusted tone-adjusted draft content

[1345] Step 6: Outputting content

[1346] The device allows the user to review and edit the final revised content. A user interface displays the generated content and provides editing functions for the user to make fine adjustments as needed. Once final review is complete, the content is published to social media platforms and websites.

[1347] Input: Revised content draft

[1348] Data processing: Display and editing through the user interface

[1349] Output: Final content approved by the user

[1350] (Application example 2)

[1351] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1352] Conventional content generation systems have limited capabilities for automatically generating effective advertising content for target users, and are unable to adjust the tone of the advertisement based on the user's real-time emotions. As a result, the effectiveness of advertisements is often limited, and real-time adjustments to match user emotions are required.

[1353] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting information, means for analyzing the collected information to identify user demographic characteristics, means for generating content based on the analysis results, means for correcting inappropriate expressions in the generated content, means for outputting the corrected content, and means for acquiring real-time emotional data of the user and adjusting the tone of the content based on the emotional data. This makes it possible to generate advertising content that is effective for target users in real time and adjust the tone to match the user's emotions.

[1354] "Information collection means" refers to the methods and devices used to collect data related to the target user online.

[1355] "Means for analyzing collected information to identify the characteristics of the user demographic" refers to a method or device that analyzes collected data and identifies characteristics such as the interests, concerns, and behavioral patterns of target users.

[1356] "Content generation means" refers to a method or device that automatically creates content such as advertisements, press releases, and social media posts based on the characteristics of the user demographic.

[1357] "Means for correcting inappropriate language in generated content" refers to a method or device that automatically detects and corrects offensive, prejudiced, or inappropriate language in text generated by the system, such as advertisements or articles.

[1358] "Means for outputting modified content" refers to the method or device that ultimately provides users with modified or adjusted content such as advertisements, press releases, and social media posts.

[1359] "Means for acquiring real-time emotional data of a user and adjusting the tone of content based on that emotional data" refers to a method or device that detects a user's emotional state in real time from facial expressions, voice, text, etc., and dynamically adapts the tone and expression of content based on that.

[1360] MODE FOR CARRYING OUT THE INVENTION

[1361] The present invention relates to a system for efficiently and effectively generating advertising content and adjusting the tone based on a user's real-time emotions. The system includes means for information collection, data analysis, content generation, inappropriate expression correction, content output, and emotion recognition, and is capable of effectively and adaptively generating advertising content.

[1362] 1. System Configuration

[1363] server

[1364] Information Collection Module

[1365] The server uses web scraping and APIs to collect information related to the target users, including news articles, social media posts, past press releases, relevant blog posts, etc. This data is then stored in a centralized database.

[1366] Data Analysis Module

[1367] The server analyzes the collected information using natural language processing techniques (e.g., TextBlob) to identify demographic interests and characteristics and categorize positive, negative, and neutral expressions.

[1368] Content Generation Module

[1369] Based on the analysis results, a generative AI model (e.g., OpenAI's GPT-3) is used to automatically generate draft copy for advertising and social media posts. For example, a prompt such as "Generate a social media post about a new product launch with a positive tone" is generated.

[1370] Expression Correction Module

[1371] The server automatically detects inappropriate language or words that pose a risk of causing controversy in the generated draft and edits them to appropriate language, minimizing the impact on users and society.

[1372] Emotion Recognition Module

[1373] The server acquires the user's real-time emotions and adjusts the tone of the generated content accordingly. For example, it analyzes the user's facial expressions, voice, and text data when operating the system, and adjusts the tone according to their emotional state (e.g., excitement, curiosity, confusion, anxiety).

[1374] Terminal

[1375] User Interface

[1376] Users can review and edit the generated content through an on-device user interface, which displays the final content adjusted based on emotion recognition and includes the ability for users to fine-tune it as needed.

[1377] Specific examples

[1378] For example, the server collects data on "new product launches." Then, the data analysis module analyzes the interests of the user demographic and identifies positive keywords related to "new eco-friendly products." Based on the analysis results, the content generation module generates a draft of a social media post like this:

[1379] Generate social media posts about your new product launch. Keep the tone positive.

[1380] The generated content is the following sentence:

[1381] Our new products are designed to be environmentally friendly and are energy efficient! Give them a try!

[1382] The expression correction module corrects the over-expression in this draft, replacing "excellent" with "highly rated." Furthermore, the emotion recognition module analyzes the user's real-time emotional data and adjusts the expression to be more positive if the user shows excitement or curiosity. For example:

[1383] Our new product is designed to be environmentally friendly and has received rave reviews! Give it a try!

[1384] The revised draft is sent to the user's device, where the user reviews it and publishes it after approval. This process allows the invention to effectively reach the target audience and adjust the tone appropriately based on real-time emotions.

[1385] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1386] Program processing flow

[1387] Step 1: Gather information

[1388] The server collects information related to the target users from online sources, such as news articles, blog posts, social media posts, and past press releases, using web scraping or APIs. The input is keywords (e.g., "new product announcement," "environmentally friendly"), and the output is the collected text data.

[1389] Step 2: Data analysis

[1390] The server analyzes the collected information using natural language processing technology. For example, it uses TextBlob to perform sentiment analysis of text and classify it into positive, negative, and neutral expressions. The input is the collected text data, and the output is keywords related to the characteristics of the user demographic.

[1391] Step 3: Content generation

[1392] The server generates drafts of advertising copy and social media posts using a generative AI model (e.g., OpenAI's GPT-3) based on the analysis results. The input is keywords obtained from the analysis results and a prompt (e.g., "Generate a social media post about a new product launch. The tone should be positive."), and the output is the generated content.

[1393] Step 4: Correct inappropriate language

[1394] The server detects inappropriate language in the generated content and corrects it to appropriate language, e.g., changing excessive language to neutral language. The input is the generated draft, and the output is the corrected content.

[1395] Step 5: User Emotion Recognition

[1396] The server acquires the user's real-time emotional data (e.g., facial expressions, voice, text) and analyzes it with an emotion recognition library (e.g., DeepFace). The input is the user's emotional data collected in real time, and the output is the recognized emotional state.

[1397] Step 6: Adjust the tone

[1398] The server adjusts the tone of the generated content based on the recognized emotional state. For example, if the user is excited, it adjusts the tone to a more positive one, and if the user is confused, it adjusts the tone to a more calm, non-sensational one. The input is the modified content and the user's emotional state, and the output is the final adjusted content.

[1399] Step 7: Content Output

[1400] The server sends the final adjusted content to the terminal and displays it on the terminal's user interface. The input is the adjusted content, and the output is the content displayed to the user. The user can then perform further editing or final review as needed.

[1401] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1402] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1403] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1404] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1405] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1406] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1407] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1408] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1409] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1410] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1411] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1412] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1413] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1414] 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.

[1415] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1416] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1417] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1418] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1419] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1420] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1421] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1422] The following is further disclosed regarding the above embodiment.

[1423] (Claim 1)

[1424] A means of collecting information relevant to your target users;

[1425] A means of analyzing the collected information to identify the characteristics of the user demographic;

[1426] a means for generating content based on the analysis results;

[1427] a means for modifying profanity in the generated content; and

[1428] means for outputting the modified content;

[1429] A system including:

[1430] (Claim 2)

[1431] 10. The system of claim 1, wherein the analyzing means analyzes the information using natural language processing techniques.

[1432] (Claim 3)

[1433] 10. The system of claim 1, wherein the content generation means utilizes a generative AI model to generate the content.

[1434] "Example 1"

[1435] (Claim 1)

[1436] A means of collecting information relevant to your target users;

[1437] A means of analyzing the collected information to identify the characteristics of the user demographic;

[1438] a means for generating content based on the analysis results;

[1439] a means for modifying profanity in the generated content; and

[1440] means for outputting the modified content; and

[1441] means for storing the collected information in a database;

[1442] A means of gathering information online based on keywords relevant to your target users;

[1443] means for analyzing the information using natural language processing techniques; and

[1444] a means for generating content using a generative AI model;

[1445] a terminal with a user interface for displaying and editing the generated content;

[1446] system.

[1447] (Claim 2)

[1448] 10. The system of claim 1, wherein the analyzing means analyzes the information using natural language processing techniques.

[1449] (Claim 3)

[1450] 10. The system of claim 1, wherein the content generation means utilizes a generative AI model to generate the content.

[1451] "Application Example 1"

[1452] (Claim 1)

[1453] A means of collecting information relevant to your target users;

[1454] A means of analyzing the collected information to identify the characteristics of the user demographic;

[1455] a means for generating content based on the analysis results;

[1456] a means for modifying profanity in the generated content; and

[1457] means for outputting the modified content;

[1458] A means to automatically collect news, social media posts, and blog articles related to food delivery;

[1459] A means of sentiment analysis of the collected data to identify positive, negative, and neutral expressions;

[1460] A means to automatically generate drafts of social media posts and press releases related to food delivery services,

[1461] A system including:

[1462] (Claim 2)

[1463] 10. The system of claim 1, wherein the analyzing means analyzes the information using natural language processing techniques.

[1464] (Claim 3)

[1465] 10. The system of claim 1, wherein the content generation means utilizes a generative AI model to generate the content.

[1466] "Example 2: Combining Emotion Engines"

[1467] (Claim 1)

[1468] A means of collecting information relevant to your target users;

[1469] A means of analyzing the collected information to identify the characteristics of the user demographic;

[1470] a means for generating content based on the analysis results;

[1471] a means for modifying profanity in the generated content; and

[1472] a means for recognizing a user's emotional state and adjusting the tone and expression of the content;

[1473] means for outputting the modified content;

[1474] A system including:

[1475] (Claim 2)

[1476] 10. The system of claim 1, wherein the analyzing means analyzes the information using natural language processing techniques.

[1477] (Claim 3)

[1478] 10. The system of claim 1, wherein the content generation means utilizes a generative AI model to generate the content.

[1479] "Application example 2 when combining emotion engines"

[1480] (Claim 1)

[1481] A means of collecting information relevant to your target users;

[1482] A means of analyzing the collected information to identify the characteristics of the user demographic;

[1483] a means for generating content based on the analysis results;

[1484] a means for modifying profanity in the generated content; and

[1485] means for outputting the modified content;

[1486] a means for obtaining real-time emotional data of a user and adjusting the tone of the content based on the emotional data;

[1487] A system including:

[1488] (Claim 2)

[1489] 10. The system of claim 1, wherein the analyzing means analyzes the information using natural language processing techniques.

[1490] (Claim 3)

[1491] 10. The system of claim 1, wherein the content generation means utilizes a generative AI model to generate the content. [Explanation of symbols]

[1492] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of collecting information relevant to your target users; A means of analyzing the collected information to identify the characteristics of the user demographic; a means for generating content based on the analysis results; a means for modifying profanity in the generated content; and means for outputting the modified content; A system including:

2. The system of claim 1 , wherein the analyzing means analyzes the information using natural language processing techniques.

3. The system of claim 1 , wherein the content generation means generates the content using a generative AI model.

Images