system

Abstract

A system is provided.SOLUTION: The system includes a means for collecting a past transmission history, a means for analyzing the collected transmission history and extracting the transmission style of a user, a means for generating a new transmission sentence on the basis of the extracted transmission style, a means for presenting the generated transmission sentence to the user and making the user confirm and correct it, and a means for finally contributing the confirmed and corrected transmission sentence to an SNS.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] As the number of SNS users increases, maintaining regular daily posts is crucial for maintaining and expanding followers. However, manually posting this regular content requires a great deal of effort, which often leads to SNS fatigue. Furthermore, the content of everyday posts is often just greetings or small incidents, so there is little need for them to be handwritten. There is a need to solve this issue and provide an environment where users can focus on posting truly important messages. [Means for solving the problem]

[0005] This system collects and analyzes past communication history, extracts the user's communication style, and then provides a means to automatically generate new messages. It also includes a means to present the generated messages to the user, allowing them to confirm and edit them. This method allows users to easily send messages on a regular basis. It also supports messages that include image information, and automatically generates text that matches the image content, further reducing the burden on the user. It also incorporates a method that uses machine learning models to analyze the tone, theme, and word choice of individual messages from the communication history, enabling messages to be more "personal."

[0006] "Past call history" refers to text data of calls made by a user in the past, obtained from the SNS account.

[0007] "Means of collection" refers to APIs and programs that use authentication information from social media accounts to collect past communication history.

[0008] "Means of analysis" refers to algorithms or machine learning models used to analyze a user's communication style based on collected communication history.

[0009] "Communication style" refers to the tendency of a user to communicate, including characteristics such as the specific language, sentence structure, and tone that they normally use.

[0010] "Extraction means" refers to the techniques used to extract specific features or patterns from the analyzed data.

[0011] "Means for generation" refers to algorithms or programs that create new messages based on the extracted user's communication style.

[0012] "Means for review and correction" refers to an interface or program that presents the generated message to the user and allows the user to review it and make corrections as necessary.

[0013] "Means of posting" refers to the API or program that allows the user to post the message that they have confirmed and edited to the SNS.

[0014] "Image information" refers to image data that a user adds when sending a message.

[0015] "Copyright" refers to the text generated in association with the image information.

[0016] A "machine learning model" is a type of algorithm used in data analysis, which learns from past communication history and is used to analyze a user's communication style.

[0017] "Tone" refers to the expression of emotion and attitude in a message, including the mood and tone of the text.

[0018] A "theme" is the central topic or subject of a message. [Brief explanation of the drawings]

[0019] [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

[0020] 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.

[0021] First, the terms used in the following description will be explained.

[0022] 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).

[0023] 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.

[0024] 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.

[0025] 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.

[0026] 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."

[0027] [First embodiment]

[0028] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0029] 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.

[0030] 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).

[0031] 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.

[0032] 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.

[0033] 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.

[0034] 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.

[0035] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0036] 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.

[0037] 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.

[0038] 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.

[0039] 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."

[0040] The embodiments of the present invention will be specifically described below.

[0041] Data collection

[0042] The server collects past call history using authentication information for the SNS account provided by the user. This authentication information includes an API key and a secret token, and uses this to obtain the latest call history from the SNS platform.

[0043] Style Analysis

[0044] The collected call history is analyzed by the server. This analysis includes the tone, theme, wording, sentence structure, etc. of the text. Specifically, natural language processing technology and machine learning algorithms are used to extract features from the call history.

[0045] Creating a new message

[0046] The server generates new messages based on the analysis results. At this stage, generative AI (e.g., GPT-2 model) is used to create messages that match the user's style. The generated messages match the user's past communication style, allowing them to communicate in a way that is "true to their personality."

[0047] Check and correct the message

[0048] The generated message is presented to the user via the terminal. The user can review the message and make corrections as necessary. This process prevents accidental and incorrect messages from being sent, while significantly reducing the amount of manual input required by the user.

[0049] Final call

[0050] Once the user has confirmed and corrected the message, it is finally posted to the SNS platform by the server, which also posts the message via API.

[0051] Call response with images

[0052] This system also supports messages that include image information. When a user uploads an image, it automatically generates text that matches the image content. For example, when uploading a landscape photo, text that matches the atmosphere of the photo is generated. This makes it easy to send messages that include visual elements.

[0053] Utilizing machine learning models

[0054] The server uses a machine learning model to analyze the collected call history, learning the topics, tone, and word choice of the user from their call history and incorporating these characteristics into the generation of new messages.

[0055] Specific examples

[0056] For example, if a user has previously sent a message such as "It's a nice day today. I want to go for a walk!", the system will recognize that themes such as "weather" and "walks" are part of the user's style. Next, when generating a new message, the system will automatically generate a message such as "It's a nice, sunny day today. I'm looking forward to going outside!"

[0057] This invention allows users to efficiently send out regular messages every day. Furthermore, if there is a message that users really want to convey, they can send it manually each time, which helps to reduce SNS fatigue and maintain and increase the number of followers.

[0058] The processing flow will be explained below.

[0059] Step 1:

[0060] A user enters their social networking account authentication information (API key, API secret, access token, access token secret) into the system, which allows the system to access the specific social networking account.

[0061] Step 2:

[0062] The server accesses the SNS account using the authentication information provided by the user and collects past call history. Specifically, it uses the SNS API to obtain the latest call history. At this time, it is set to collect a certain amount of past call history (for example, 200 items).

[0063] Step 3:

[0064] The server analyzes the collected call history. The analysis uses natural language processing and machine learning techniques, applying algorithms that identify frequently occurring words and topics in texts to extract features such as tone, theme, wording, and sentence structure.

[0065] Step 4:

[0066] The server extracts the user's communication style. This extraction step identifies specific patterns and styles from past communication history. For example, frequently used phrases and specific tones (friendly, professional, etc.) are analyzed.

[0067] Step 5:

[0068] The server uses a generative AI to generate new sentences based on the extracted user's speech style. The generative AI (e.g., a GPT-2 model) creates natural sentences that reflect the user's style. In this generation step, a specific prompt is input and a sentence is generated accordingly.

[0069] Step 6:

[0070] The generated message is displayed to the user through the terminal. The terminal provides an interface for the user to review the new message and modify it if necessary. This interface includes a function that displays the generated message in a text box and allows the user to edit its content.

[0071] Step 7:

[0072] The user checks the generated message and makes any corrections. When the user has completed the corrections and is ready to send the message, they press the Confirm button. By pressing this button, the final message is confirmed.

[0073] Step 8:

[0074] The server receives the final message that the user has confirmed and corrected, and posts it to the SNS platform. This posting is also done via API, and the final message is published to the SNS from the user's account.

[0075] Step 9:

[0076] When a user uploads image information, the server automatically generates text that matches the image content. For this purpose, an image analysis algorithm is used to generate relevant text based on the image content.

[0077] Step 10:

[0078] Similarly, text with images is presented to the user via the device, and they are given the opportunity to check and correct it. After going through a series of checking and correcting processes, the final message with images is posted to the SNS.

[0079] These are the specific steps for automatically generating new messages based on past message history, and then posting them to SNS after the user has confirmed and edited them. This process allows users to efficiently send regular daily messages and saves time and effort in maintaining and expanding their followers.

[0080] Example 1

[0081] 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."

[0082] Modern social media users often feel burdened by frequent posting. Users' posts can also be inconsistent, making it difficult to keep their followers engaged. Furthermore, as posts increasingly include visual elements, quickly generating appropriate copy for images can be time-consuming. A new system is needed to solve these problems.

[0083] 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.

[0084] In this invention, the server includes means for collecting past call history using authentication information provided by the user, means for analyzing the collected call history and extracting the user's call style, and means for generating new messages using a generative AI model based on the extracted call style. This enables users to efficiently send messages with consistency. The present invention also includes means for presenting the generated messages to the user via a terminal for confirmation and correction, and means for posting the final confirmed and corrected messages to an SNS via an API. This enables users to send messages quickly and accurately. Furthermore, the present invention includes means for including image information in the generated messages and automatically generating drafts based on the image content, making it easy to send messages that include visual elements.

[0085] "Authentication information" refers to information such as an API key or secret token that a user needs to access their SNS account.

[0086] "Outgoing communication history" refers to a record of all posts and comments that a user has made on a social networking platform in the past.

[0087] A "server" is a computer system that processes, stores, and controls access to data over a network.

[0088] "Communication style" refers to characteristics such as tone, theme, wording, and sentence structure of text extracted from a user's past communication history.

[0089] A "generative AI model" is an artificial intelligence model that uses natural language processing and machine learning algorithms to generate new text or information from given data.

[0090] A "terminal" is a device such as a computer or smartphone that is directly operated by a user.

[0091] "API" stands for Application Programming Interface, and refers to an interface for exchanging data and functions between different software programs.

[0092] "Image information" refers to image data used when a user posts to an SNS.

[0093] A "machine learning model" is a set of algorithms that learn from data and perform tasks such as prediction, classification, and generation.

[0094] The embodiments of the present invention will be specifically described below.

[0095] Data collection and analysis

[0096] 1. The server collects authentication information for the SNS account provided by the user. This authentication information includes an API key and secret token, and uses this information to retrieve the user's past call history from the SNS platform. If authentication is successful, the server calls the SNS API, collects the past call history, and stores it in a database.

[0097] 2. The server analyzes the collected communication history and extracts the user's communication style. This analysis uses natural language processing techniques (e.g., NLTK, spaCy) and machine learning algorithms (e.g., BERT, GPT-2). The server tokenizes each piece of text data and performs sentiment analysis, topic modeling, and contextual analysis. As a result, the user's communication style is generated as a digital profile.

[0098] Creating a new message

[0099] 3. The server generates a new message based on the analysis results. Here, a generative AI model (e.g., GPT-3) is used. The server inputs the analysis data as a prompt into the generative AI model to generate a new sentence.

[0100] Check and correct the message

[0101] 4. The generated message is presented to the user via the device. The user can review the message and make corrections as necessary. These corrections are reflected in real time on the device screen, preventing errors and significantly reducing the amount of manual input required by the user.

[0102] Final call

[0103] 5. After the user has confirmed and corrected the message, it is finally posted to the SNS platform by the server. This posting is also automated using APIs. When the user presses the "Post" button on their device, the server sends the corrected message to the SNS API and executes the posting.

[0104] Support for sending messages with images

[0105] 6. When a user uploads an image, the server analyzes the image and automatically generates a sentence that corresponds to the content. For example, when a user uploads a landscape photo from their device, the server recognizes and analyzes the image and generates a sentence that corresponds to the atmosphere of the landscape photo.

[0106] Specific examples

[0107] For example, if a user has previously sent a message such as "It's a nice day today. I want to go for a walk!", the system will recognize that themes such as "weather" and "walks" are part of the user's communication style. When generating a new message, the system will automatically generate a sentence such as "It's a nice, sunny day today. I'm looking forward to going outside!"

[0108] Prompt Sentence Examples

[0109] Example prompt: "Generate new posts based on the user's past posting style. The post might look something like, 'It's a nice day today. I want to go for a walk!' Generate new posts based on this."

[0110] This invention allows users to efficiently post on SNS every day. It also contributes to maintaining and expanding the number of followers because users continue to post consistently. It also makes it easy to post content that includes visual elements, reducing SNS fatigue and enabling effective communication.

[0111] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0112] Step 1:

[0113] A user uses a device to enter authentication information for their social networking account. This authentication information includes an API key and a secret token. The entered authentication information is sent to the server. The server receives and securely stores the authentication information. The input of this process is the authentication information entered by the user, and the output is the securely stored authentication information.

[0114] Step 2:

[0115] The server uses the stored authentication information to call the API of the social media platform and collect the user's past call history. During this process, real-time data is obtained via the API and stored in a database. The input is the authentication information and the call history data obtained as a result of the API call, and the output is the call history stored in the database.

[0116] How it works: The server connects to the social networking platform and fetches the user's past posts. This data is stored in a database in chronological order.

[0117] Step 3:

[0118] The server analyzes the collected call history. This analysis uses natural language processing technology and machine learning algorithms to extract the user's call style. Specifically, the tone, theme, vocabulary, and sentence structure of the text are analyzed. The input is the call history data, and the output is a digital profile that shows the user's call style.

[0119] Specific operation: The server uses analysis tools (e.g., NLTK, spaCy) to analyze the text data and saves the results in a database as a digital profile.

[0120] Step 4:

[0121] The server uses a generative AI model based on the analysis results to generate a new message. At this time, the analysis data is input to the generative AI model as a prompt. The input is a digital profile that indicates the user's communication style and a prompt, and the output is a new message.

[0122] Specific operation: The server inputs the prompt sentence into a generative AI model (e.g., GPT-3) to generate a new message, which is then stored in a database.

[0123] Step 5:

[0124] The generated message is presented to the user via the terminal. The user can check this message and modify it as necessary. The input is the generated message, and the output is the message that has been checked and modified by the user.

[0125] Specific operation: The generated message is displayed on the device screen, and the user can edit the text using the keyboard or touch screen.

[0126] Step 6:

[0127] The message that the user has confirmed and modified is finally posted to the SNS platform by the server. This posting is also automated using APIs. The input is the message that the user modified, and the output is a new post on the SNS platform.

[0128] Specific operation: When the user presses the "Post" button, the server sends the revised message to the SNS API and executes the post.

[0129] Step 7:

[0130] The server analyzes the image uploaded by the user and generates a sentence based on its content. After analyzing the content of the image using image recognition technology, the server generates a sentence using a generative AI model. The input is the image data uploaded by the user, and the output is a sentence that matches the image.

[0131] Specific operation: When a user uploads an image from their device, the server recognizes the image and generates an appropriate draft based on the analysis results. The draft is then presented to the user, who can review and edit it.

[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] Previous message generation systems on social media platforms were limited to generating new messages based on a user's past message styles, and were unable to efficiently generate and post ad copy. This meant that advertising agencies and marketing teams had to spend a great deal of time and effort creating ad copy that matched their clients' styles.

[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 collecting past communication history, means for analyzing the collected communication history and extracting the user's communication style, means for generating new messages based on the extracted communication style, means for presenting the generated messages to the user for confirmation and revision, means for posting the confirmed and revised messages to an SNS, means for automatically generating advertising copy, means for presenting the generated advertising copy to the user for confirmation and revision, and means for posting the revised advertising copy to a distribution platform. This enables efficient generation and posting of advertising copy based on the user's past communication style.

[0137] "Outgoing message history" refers to a record of text, images, etc. that a user has posted on a social media platform in the past.

[0138] "Communication style" refers to the combined characteristics of the tone, theme, language, and writing style of the content a user posts.

[0139] "Copyright" means text created to promote a particular product or service.

[0140] "Means of collection" refers to the methods and systems for obtaining a user's past posting history from a social media platform.

[0141] "Means of analysis" refers to a method of analyzing collected communication history and extracting the user's communication style from it.

[0142] The "means of generation" refers to a method for creating new messages or advertising copy based on the extracted communication style.

[0143] "Means for confirmation and correction" refers to a method in which the generated message or advertising copy is presented to the user, allowing the user to check the content themselves and make corrections as necessary.

[0144] "Method of posting" refers to the method of posting the final confirmed and revised message or advertising copy on social media or distribution platforms.

[0145] A "machine learning model" is an algorithm or system that learns patterns and features from data and makes predictions and classifications for new data.

[0146] The system for implementing this invention is made up of a server, a terminal, and a user. Each component and its function will be specifically described below.

[0147] server

[0148] The server has the function of collecting past posting history. It retrieves past posting history from the social media platform using the social media account authentication information (API key or secret token) provided by the user. The collected posting history is analyzed using natural language processing technology and machine learning algorithms. Through the analysis, features such as the tone, theme, wording, and sentence structure of the text are extracted. The server then generates new postings based on the extracted style. This generation uses a generative AI model such as OpenAI's GPT-3 model.

[0149] Terminal

[0150] The device provides a user interface, allowing the user to review and edit the generated message or ad copy. The new message or ad copy generated by the server is sent to the device and presented to the user. The user reviews the presented copy and makes any necessary edits. Once the user has reviewed and edited the message or ad copy, it is sent back to the server.

[0151] User

[0152] Users provide their social media account credentials and authorize the collection of past messaging history. They are then asked to review the generated messaging and ad copy and make any necessary revisions. Users then post the final messaging and ad copy to the social media platform.

[0153] Hardware and software used

[0154] Hardware: Smartphones, servers

[0155] Software: requests library, TextBlob library, OpenAI library, SNS platform API

[0156] Specific examples

[0157] For example, an advertising agency might generate new ad copy based on a particular client's social media accounts. If the client's past posts tend to be positive and emotional, the resulting ad copy will also have a similarly positive tone. An example of a prompt might look like this:

[0158] Generate ad copy based on your client's social media posting style. Consider the following style characteristics:

[0159] Polarity: 0.8, Subjectivity: 0.6

[0160] This invention allows advertising agencies and marketing teams to efficiently generate ad copy that matches their client's style, then review and revise it before posting it on the target platform, improving the efficiency and effectiveness of ad copy creation and enhancing the quality of marketing activities.

[0161] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0162] Step 1:

[0163] The server collects past call history using the authentication information of the SNS account provided by the user. The input is the user's authentication information, and the output is the call history obtained from the SNS platform. In this process, the API of the SNS platform is used to obtain the latest call history.

[0164] Step 2:

[0165] The server analyzes the collected communication history and extracts the user's communication style. The input is the collected communication history, and the output is the characteristics of the user's communication style (tone, theme, wording, sentence structure). This analysis uses natural language processing technology and the TextBlob library to perform sentiment analysis of the text and word frequency analysis.

[0166] Step 3:

[0167] The server generates a new message or ad copy based on the extracted style. The input is the user's message style characteristics, and the output is the generated new message or ad copy. The copy is generated using a generative AI model such as OpenAI's GPT-3 model. The prompt sentence is provided as input to the model.

[0168] Step 4:

[0169] The generated message or copy draft is sent from the server to the terminal and presented to the user. The input is the generated message or copy draft, and the output is a screen display presented to the user for confirmation and correction, provided via the terminal's user interface.

[0170] Step 5:

[0171] The user checks the message or advertising copy presented to them and makes corrections as necessary. The input is the message or advertising copy presented to the user, and the output is the copy that has been checked and corrected by the user. The user makes corrections through their terminal and sends the corrections to the server.

[0172] Step 6:

[0173] The server finally posts the message or advertising copy that has been confirmed and revised by the user to the SNS platform. The input is the confirmed and revised copy, and the output is the final copy posted to the SNS platform. In this process, the posting is made via the SNS platform's API.

[0174] Step 7:

[0175] When image information is included in the generated message, the server provides a means to automatically generate a message that matches the image content. The input is an image uploaded by the user and a prompt message based on its content, and the output is a new message that matches the image. A generative AI model is used to generate a message that matches the image.

[0176] This process allows users to efficiently create new messages and copy drafts, review and revise them, and finally post them to social media platforms, allowing advertising agencies and marketing teams to generate effective copy drafts that fit their clients' communication styles.

[0177] 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.

[0178] The embodiments of the present invention will be specifically described below.

[0179] Data collection

[0180] A user enters the authentication information of a social networking account (API key, API secret, access token, access token secret) into the system, which allows the system to access the specific social networking account.

[0181] Collection of past call history

[0182] The server accesses the SNS account using the authentication information provided by the user and collects past call history. This collection is done by using the SNS API to obtain the latest call history. The server is set to collect a certain amount of past call history (for example, 200 items).

[0183] Style and Sentiment Analysis

[0184] The collected call history is analyzed by the server. First, natural language processing of the text data is used to extract features such as tone, theme, wording, and sentence structure. At the same time, an emotion engine is used to detect the user's emotional state from the call history. The emotion engine identifies emotional categories such as positive, negative, and neutral, and labels the data accordingly.

[0185] Extracting the delivery style

[0186] The server extracts the user's communication style based on the analysis results. At this stage, the analysis also includes the relationship between the user's communication style and their emotional state. For example, it distinguishes between messages sent when the user was in a positive emotional state and messages sent when the user was in a negative emotional state.

[0187] Creating a new message

[0188] The server generates new messages based on the extracted user's communication style. In this process, a generative AI (e.g., GPT-2 model) is used to create messages that reflect the user's emotional state. The generation prompts include topics and tones based on the user's current emotional state.

[0189] Emotion recognition and correction of messages

[0190] The generated message is displayed to the user through the device. Based on the user's feedback and corrections, the generated message is checked for appropriate sentiment. The interface includes a function where the generated message is displayed in a text box and the user can enter corrections.

[0191] Final call

[0192] The final message, after being confirmed and corrected by the user, is posted to the SNS platform by the server. This posting is also done via API, so the final message is published to the SNS from the user's account.

[0193] Image information support

[0194] The system also supports messages containing image information. When a user uploads an image, the server automatically generates appropriate text based on the image content. This includes a process that uses an image analysis algorithm to generate text that corresponds to the image content.

[0195] Utilizing machine learning models

[0196] The server uses a machine learning model to analyze the collected message history and emotional data. The model learns from the user's message history and emotional state and uses the information to generate future messages.

[0197] Specific examples

[0198] For example, suppose a user previously made a positive statement such as, "Today was such a fun day! I discovered a new place!" If the system recognizes the user's current emotional state as positive, it will generate a new statement such as, "The weather was great today, and I went to a new cafe. It was so relaxing!" Conversely, if the emotion engine detects a negative emotion, it will generate a statement such as, "Today was a stressful day, but I'm home and relaxing."

[0199] This invention allows users to efficiently post regular messages on SNS every day, and enables them to post more naturally, reflecting their own emotional state. Also, if there is a message that users really want to convey, they can post it manually each time, which helps to reduce SNS fatigue and maintain and increase the number of followers.

[0200] The processing flow will be explained below.

[0201] Step 1:

[0202] A user enters their social networking account authentication information (API key, API secret, access token, access token secret) into the system, which allows the system to access the specific social networking account.

[0203] Step 2:

[0204] The server accesses the SNS account using the authentication information provided by the user and collects the past call history. To do this, it uses the SNS API to retrieve the most recent call history. The past call history covers 200 calls.

[0205] Step 3:

[0206] The server analyzes the collected call history using natural language processing technology to extract features such as the tone, theme, vocabulary, and sentence structure of the text.

[0207] Step 4:

[0208] The server uses an emotion engine to analyze the user's emotional state from the collected call history, which involves identifying emotional categories such as positive, negative, and neutral.

[0209] Step 5:

[0210] The server extracts the user's communication style based on the analyzed data. This extraction process identifies distinctive patterns and styles from communication history and distinguishes styles according to emotional states.

[0211] Step 6:

[0212] The server generates new messages based on the extracted user's communication style and current emotional state. Here, generative AI (e.g., GPT-2 model) is used to create natural-sounding sentences that reflect the user's emotional state.

[0213] Step 7:

[0214] The generated message is displayed to the user through the terminal. The terminal provides an interface for the user to review the new message and modify it if necessary. This interface includes a function that displays the generated message in a text box and allows the user to edit its content.

[0215] Step 8:

[0216] The user can check and modify the generated message. Once the user has completed the modifications and is ready to send the message, the user presses the confirmation button to finalize the message.

[0217] Step 9:

[0218] The server posts the final message that the user has confirmed and corrected to the SNS platform. This posting is also done via API, and the message is published to the SNS from the user's account.

[0219] Step 10:

[0220] When a user uploads image information, the server automatically generates appropriate copy based on the image content, including generating text based on the image content using image analysis algorithms.

[0221] Step 11:

[0222] Messages with images are also displayed to the user via the device, and the user is given the opportunity to review and edit them. Once the user has reviewed and edited the message, it is finally posted to the SNS.

[0223] These are the specific processing steps of the invention that combines the emotion engine. This process allows users to efficiently post natural messages that correspond to their emotional state. The combination of the emotion engine and generative AI enables users to post messages that are more in line with their own personalities, reducing social media fatigue and enabling users to maintain and expand their following.

[0224] Example 2

[0225] 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."

[0226] On existing social media platforms, users spend a lot of time and effort on daily posting. Furthermore, the content of posts tends to be monotonous, often resulting in low engagement with followers. Furthermore, it is difficult to post natural, rich content that reflects the user's emotional state. In such situations, users may become exhausted by social media activity and ultimately abandon or delete their accounts. There is a need for a system that can solve these issues and enable users to post efficiently and appropriately.

[0227] 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.

[0228] In this invention, the server includes means for a user to input authentication information for the SNS service, means for collecting past message history, means for analyzing the collected message history and extracting the user's message style and emotional state, means using a generative AI model to generate new messages based on the extracted message style and emotional state, means for presenting the generated messages to the user and allowing them to confirm and revise, means for finally posting the confirmed and revised messages to the SNS service, means for analyzing collected image information and generating appropriate drafts based on the contents, and means for using a machine learning model to learn the collected message history and emotional state. This enables users to efficiently post to SNS every day and to post natural and diverse messages that reflect their own emotions.

[0229] "User" refers to an individual or organization that uses the SNS service.

[0230] "Authentication Information" refers to information such as API keys, API secrets, access tokens, and access token secrets required to access SNS services.

[0231] "SNS services" refers to social networking services, such as Twitter and Facebook.

[0232] "Server" refers to a computer system on a network that accesses SNS services and collects, analyzes, generates, and posts data.

[0233] "Outgoing message history" refers to a record of posts and comments a user has made on social media in the past.

[0234] "Analysis" refers to data processing performed to extract specific characteristics from the collected call history.

[0235] "Communication style" refers to the patterns of language, tone, and themes used by users on social media.

[0236] "Emotional state" refers to the user's emotional state, such as positive, negative, or neutral, extracted from the call history.

[0237] A "generative AI model" refers to an artificial intelligence model that uses machine learning techniques to generate new text.

[0238] A "prompt" refers to the input data or instructions that a generative AI model uses to generate new text.

[0239] "Confirmation and correction" refers to the act of the user confirming the generated message and correcting the content as necessary.

[0240] "Posting" refers to the act of publishing a message that has been finally confirmed and corrected from a user's SNS account.

[0241] "Image information" refers to the image data used by users when posting to SNS.

[0242] "Draft" refers to the text content generated based on image information.

[0243] A "machine learning model" refers to an algorithm or system used to learn from data and accomplish a specific task.

[0244] The present invention relates to a system that improves the efficiency of users' SNS activities and enables natural posting that reflects their emotions. Specific embodiments of the present invention will be described below.

[0245] First, the user enters the authentication information (API key, API secret, access token, access token secret) of the SNS service into the system, which allows the system to access the specific SNS account.

[0246] Next, the server uses the authentication information to access the API of the SNS service and retrieve the user's past call history. Specifically, the server is configured to collect the most recent 200 call histories. The collected data is stored in a database.

[0247] The server performs text analysis based on the collected call history. This analysis uses natural language processing libraries (e.g., NLTK and spaCy) and emotion engines (e.g., Azure Sentiment Analysis). This extracts features such as tone, theme, wording, and sentence structure from the call history, and labels the emotional state (e.g., positive, negative, neutral, etc.).

[0248] Based on the analysis results, the server extracts the user's communication style. Here, it distinguishes between messages sent by the user in a positive emotional state and messages sent by the user in a negative emotional state. For example, if a user's communication style is dominated by "fun" and "joy," it is classified as a positive style.

[0249] The server then uses a generative AI model (e.g., OpenAI's GPT-3) to generate a new message. The generation prompt is populated with the user's current emotional state and topic based on the analysis results. For example, a prompt might be, "Generate a new message based on past positive messages." The generated message is then sent from the server to the device and presented to the user.

[0250] The terminal displays the generated message received from the server to the user, allowing the user to confirm and edit the message using the text box. When the user enters the edits, the information is sent back to the server.

[0251] The final message that has been confirmed and corrected by the user is posted to the SNS from the user's account via the SNS service's API by the server.

[0252] The system also supports users uploading images. The server uses an image analysis algorithm (e.g., Google Cloud Vision) to analyze the image content and generate appropriate text based on the content. For example, if a "landscape image" is recognized, the system generates the text "I found a beautiful landscape!"

[0253] Furthermore, the server uses a machine learning model (e.g., Scikit-learn) to learn from the collected message history and emotional state. This model learns from the user's message history and emotional data and uses it to generate future messages.

[0254] This will enable users to post on social media every day efficiently and enable natural and diverse communication that appropriately reflects their emotions, which is expected to improve engagement and reduce social media fatigue.

[0255] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0256] Program processing flow

[0257] Step 1:

[0258] The user enters the authentication information (API key, API secret, access token, access token secret) for the social networking service into a web form and clicks a button to submit it.

[0259] The data entered is authentication information, and based on this information, information that allows access to the user's SNS account is obtained.

[0260] Step 2:

[0261] The server receives the authentication information sent by the user and accesses the API of the SNS service to obtain the past call history.

[0262] Specifically, the server sends an API request to the SNS endpoint and saves the most recent 200 outgoing calls in a database.

[0263] The input is the SNS authentication information and request, and the output is the call history data.

[0264] Step 3:

[0265] The server analyzes the collected call history.

[0266] A natural language processing library (e.g., NLTK or spaCy) is used to analyze the tone, theme, wording, sentence structure, etc. of the text data. In addition, a sentiment engine (e.g., Azure Sentiment Analysis) is used to label the sentiment state (positive, negative, neutral) of the call history.

[0267] The input is the call history data, and the output is the analysis results (tone, theme, wording, sentence structure, emotional state).

[0268] Step 4:

[0269] The server extracts the user's communication style based on the analysis results.

[0270] Using the analysis results, we distinguish between messages sent in a positive emotional state and those sent in a negative emotional state, and clarify the user's communication style. For example, sentences containing many words like "fun" and "joy" are classified as a positive style.

[0271] The input is the analysis result, and the output is the transmission style.

[0272] Step 5:

[0273] The server generates a new message based on the extracted message style.

[0274] A prompt (e.g., "Generate a new message based on past positive messages") is input into a generative AI model (e.g., OpenAI's GPT-3) and the generated text is received.

[0275] The input is the communication style and the prompt sentence, and the output is the generated communication sentence.

[0276] Step 6:

[0277] The terminal displays the generated message to the user.

[0278] The generated message is displayed in a text box, providing an interface that allows the user to check and edit it.

[0279] The input is the generated message, and the output is the user's confirmation and correction.

[0280] Step 7:

[0281] The user checks the generated message and enters corrections as necessary.

[0282] Check the message displayed in the text box and correct any inappropriate parts. The corrections will be sent to the server again.

[0283] The input is the generated message and the user's suggestions for revision, and the output is the revised message.

[0284] Step 8:

[0285] The server posts the final, corrected message to the SNS service.

[0286] The revised message will be published from the user's account via the SNS API.

[0287] The input is the modified message and the output is a post on a social networking site.

[0288] Step 9:

[0289] The server analyzes the image information uploaded by the user and generates appropriate text.

[0290] It uses image analysis algorithms (e.g., Google Cloud Vision) to analyze the image content and generate text based on that content. For example, if it recognizes an image as a landscape, it will generate the sentence "I found a beautiful landscape!"

[0291] The input is image information, and the output is a generated draft based on the image.

[0292] Step 10:

[0293] The server uses a machine learning model to learn from the collected call history and emotional state.

[0294] The collected message history and emotion labels are used to train a Scikit-learn model, which will provide learning results that will be useful for generating future messages.

[0295] The input is the call history and emotion labels, and the output is a trained machine learning model.

[0296] (Application example 2)

[0297] 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."

[0298] In modern society, many users use social media to share information, but the content of their posts is subjective and inconsistent, making it difficult to reflect their own emotional state or style. It is also difficult to automatically and efficiently create effective ad copy that matches a user's emotions and style when generating ads. To solve these problems, a system is needed that analyzes a user's past posting history and generates new posts and ad copy that reflect that user's style and emotional state.

[0299] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0300] In this invention, the server includes means for collecting past call history, means for analyzing the collected call history and extracting the user's call style, and means for generating new messages based on the extracted call style. This makes it possible to automatically generate messages and advertising copy that reflect the user's past call history and emotional state, present them to the user, and post them to SNS or advertising media after final confirmation.

[0301] "Past communication history" refers to all of the content posted or transmitted by a user in the past on social media platforms and other platforms.

[0302] "Means of collection" refers to the technical processes and functions that the system uses to obtain past communication history using the user's social media account authentication information.

[0303] "Means of analyzing and extracting a user's communication style" refers to a means of analyzing collected communication history using natural language processing and machine learning technology to identify characteristics of the user's writing, such as tone, theme, and wording.

[0304] "Means for generating new messages" refers to technology for creating new messages using AI models, etc., based on the extracted user's communication style and emotional state.

[0305] "Means for presenting the generated message to the user and allowing them to review and correct it" refers to an interface or process that displays the generated message to the user, allows the user to review its contents, and makes corrections as necessary.

[0306] "Means for posting the final verified and corrected message on a social networking site" refers to the technical process by which the final message that has been verified and corrected by the user is automatically posted on a platform such as a social networking site.

[0307] "Means for generating advertising copy and advertising visuals based on the extracted communication style and emotional state" refers to technology for automatically creating advertising copy and visuals for commercial purposes, taking into account the user's communication style and emotional state.

[0308] "Means for presenting the generated ad copy to the user and allowing the user to modify and finalize it" refers to an interface or function that displays the generated ad copy to the user, allowing the user to review it and modify it as necessary.

[0309] "Means for posting to designated media" refers to the technical process for automatically distributing and posting the final, verified and corrected advertising copy to the designated advertising media or platform.

[0310] The following is a specific description of an embodiment of the present invention. The entire system is composed of a server, a terminal, and a user. The programs that are part of this system are realized as follows.

[0311] Data collection methods

[0312] The server uses the authentication information for the SNS account provided by the user to collect past call history via the SNS API. Specifically, it accesses the SNS account using authentication information such as the API key, API secret, access token, and access token secret, and retrieves the past call history (e.g., 200 calls). At this stage, data is collected from the SNS API and stored on the server.

[0313] Analysis of communication style and emotional state

[0314] The collected call history is analyzed within the server. Natural language processing technology is used to extract features such as tone, theme, and wording of the call. An emotion engine is also used to detect the user's emotional state (e.g., positive, negative, neutral) from the call history. This analysis uses machine learning models (e.g., BERT) and sentiment analysis models.

[0315] Generate new messaging and ad copy

[0316] The server then uses a generative AI model (e.g., GPT-2) to generate new messages based on the extracted user's communication style and emotional state. The generation prompts are populated with topics and tones based on the user's emotional state, and new messages are created. Similarly, ad copy and visuals are automatically generated to match the user's communication style and emotional state.

[0317] For example, the following prompt might be used:

[0318] Prompt: "Create an ad for a new cafe with a positive tone that matches the user's communication style."

[0319] Feedback and Corrections

[0320] The generated message and advertisement copy are presented to the user via the device. The user can review these copies and make corrections as necessary. This procedure is an important step to ensure that the automatically generated copies match the user's expectations and intentions. An interface is provided that allows the user to provide feedback and make corrections.

[0321] Final Post

[0322] The final message and advertisement copy that has been checked and corrected is posted by the server to the social media platform or designated advertising medium. This posting is also done via API and is published from the user's account.

[0323] In this way, the present invention can efficiently generate consistent messages and advertisements that reflect the user's past message history and emotional state, and finally publish them after confirmation and correction.

[0324] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0325] Step 1: Data collection

[0326] The server receives the authentication information (API key, API secret, access token, access token secret) of the user's SNS account as input. Using this authentication information, it accesses the SNS API and retrieves the user's past call history. At this time, the most recent 200 call history entries are collected and stored on the server. The output is the collected call history data.

[0327] Step 2: Analyze communication style and emotional state

[0328] The server analyzes the collected call history data as input. First, it uses natural language processing technology to extract features such as tone, theme, and wording from the call history. At the same time, it uses an emotion engine to detect the emotional state (positive, negative, neutral) of each message. This identifies the user's call style and emotional state. The output is the analyzed call style and emotional state data.

[0329] Step 3: Generate a new message

[0330] The server generates a new message using the analyzed message style and emotional state data. It uses a generative AI model (GPT-2) to generate the message. The following prompt is input:

[0331] Prompt: "Create an ad for a new cafe with a positive tone that matches the user's communication style."

[0332] Based on this prompt, a new message is generated. The output is the generated message.

[0333] Step 4: Generate ad copy and visuals

[0334] The server generates ad copy and visuals using the analyzed customer communication style and emotional state data as input. Using a generative AI model, ad copy that matches the user's emotional state is created based on the prompt text. The output is the generated ad copy and visuals.

[0335] Step 5: Feedback and revisions

[0336] The terminal presents the generated message and advertisement copy to the user. The user checks the contents and makes any necessary corrections. At this time, the terminal sends the user's corrections to the server, which reflects the suggested corrections. The input is the generated message and advertisement copy, and the output is the final version that has been checked and corrected by the user.

[0337] Step 6: Final submission

[0338] The server receives as input the final message and advertisement copy that has been reviewed and revised by the user. It posts this to the social media platform or designated advertising medium. The posting is again done using the social media API, and is made public through the user's account. The output is the publication of the final message and advertisement copy.

[0339] 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.

[0340] 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.

[0341] 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.

[0342] [Second embodiment]

[0343] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0344] 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.

[0345] 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).

[0346] 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.

[0347] 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.

[0348] 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).

[0349] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0350] 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.

[0351] 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.

[0352] 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.

[0353] 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.

[0354] 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."

[0355] The embodiments of the present invention will be specifically described below.

[0356] Data collection

[0357] The server collects past call history using authentication information for the SNS account provided by the user. This authentication information includes an API key and a secret token, and uses this to obtain the latest call history from the SNS platform.

[0358] Style Analysis

[0359] The collected call history is analyzed by the server. This analysis includes the tone, theme, wording, sentence structure, etc. of the text. Specifically, natural language processing technology and machine learning algorithms are used to extract features from the call history.

[0360] Creating a new message

[0361] The server generates new messages based on the analysis results. At this stage, generative AI (e.g., GPT-2 model) is used to create messages that match the user's style. The generated messages match the user's past communication style, allowing them to communicate in a way that is "true to their personality."

[0362] Check and correct the message

[0363] The generated message is presented to the user via the terminal. The user can review the message and make corrections as necessary. This process prevents accidental and incorrect messages from being sent, while significantly reducing the amount of manual input required by the user.

[0364] Final call

[0365] Once the user has confirmed and corrected the message, it is finally posted to the SNS platform by the server, which also posts the message via API.

[0366] Call response with images

[0367] This system also supports messages that include image information. When a user uploads an image, it automatically generates text that matches the image content. For example, when uploading a landscape photo, text that matches the atmosphere of the photo is generated. This makes it easy to send messages that include visual elements.

[0368] Utilizing machine learning models

[0369] The server uses a machine learning model to analyze the collected call history, learning the topics, tone, and word choice of the user from their call history and incorporating these characteristics into the generation of new messages.

[0370] Specific examples

[0371] For example, if a user has previously sent a message such as "It's a nice day today. I want to go for a walk!", the system will recognize that themes such as "weather" and "walks" are part of the user's style. Next, when generating a new message, the system will automatically generate a message such as "It's a nice, sunny day today. I'm looking forward to going outside!"

[0372] This invention allows users to efficiently send out regular messages every day. Furthermore, if there is a message that users really want to convey, they can send it manually each time, which helps to reduce SNS fatigue and maintain and increase the number of followers.

[0373] The processing flow will be explained below.

[0374] Step 1:

[0375] A user enters their social networking account authentication information (API key, API secret, access token, access token secret) into the system, which allows the system to access the specific social networking account.

[0376] Step 2:

[0377] The server accesses the SNS account using the authentication information provided by the user and collects past call history. Specifically, it uses the SNS API to obtain the latest call history. At this time, it is set to collect a certain amount of past call history (for example, 200 items).

[0378] Step 3:

[0379] The server analyzes the collected call history. The analysis uses natural language processing and machine learning techniques, applying algorithms that identify frequently occurring words and topics in texts to extract features such as tone, theme, wording, and sentence structure.

[0380] Step 4:

[0381] The server extracts the user's communication style. This extraction step identifies specific patterns and styles from past communication history. For example, frequently used phrases and specific tones (friendly, professional, etc.) are analyzed.

[0382] Step 5:

[0383] The server uses a generative AI to generate new sentences based on the extracted user's speech style. The generative AI (e.g., a GPT-2 model) creates natural sentences that reflect the user's style. In this generation step, a specific prompt is input and a sentence is generated accordingly.

[0384] Step 6:

[0385] The generated message is displayed to the user through the terminal. The terminal provides an interface for the user to review the new message and modify it if necessary. This interface includes a function that displays the generated message in a text box and allows the user to edit its content.

[0386] Step 7:

[0387] The user checks the generated message and makes any corrections. When the user has completed the corrections and is ready to send the message, they press the Confirm button. By pressing this button, the final message is confirmed.

[0388] Step 8:

[0389] The server receives the final message that the user has confirmed and corrected, and posts it to the SNS platform. This posting is also done via API, and the final message is published to the SNS from the user's account.

[0390] Step 9:

[0391] When a user uploads image information, the server automatically generates text that matches the image content. For this purpose, an image analysis algorithm is used to generate relevant text based on the image content.

[0392] Step 10:

[0393] Similarly, text with images is presented to the user via the device, and they are given the opportunity to check and correct it. After going through a series of checking and correcting processes, the final message with images is posted to the SNS.

[0394] These are the specific steps for automatically generating new messages based on past message history, and then posting them to SNS after the user has confirmed and edited them. This process allows users to efficiently send regular daily messages and saves time and effort in maintaining and expanding their followers.

[0395] Example 1

[0396] 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."

[0397] Modern social media users often feel burdened by frequent posting. Users' posts can also be inconsistent, making it difficult to keep their followers engaged. Furthermore, as posts increasingly include visual elements, quickly generating appropriate copy for images can be time-consuming. A new system is needed to solve these problems.

[0398] 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.

[0399] In this invention, the server includes means for collecting past call history using authentication information provided by the user, means for analyzing the collected call history and extracting the user's call style, and means for generating new messages using a generative AI model based on the extracted call style. This enables users to efficiently send messages with consistency. The present invention also includes means for presenting the generated messages to the user via a terminal for confirmation and correction, and means for posting the final confirmed and corrected messages to an SNS via an API. This enables users to send messages quickly and accurately. Furthermore, the present invention includes means for including image information in the generated messages and automatically generating drafts based on the image content, making it easy to send messages that include visual elements.

[0400] "Authentication information" refers to information such as an API key or secret token that a user needs to access their SNS account.

[0401] "Outgoing communication history" refers to a record of all posts and comments that a user has made on a social networking platform in the past.

[0402] A "server" is a computer system that processes, stores, and controls access to data over a network.

[0403] "Communication style" refers to characteristics such as tone, theme, wording, and sentence structure of text extracted from a user's past communication history.

[0404] A "generative AI model" is an artificial intelligence model that uses natural language processing and machine learning algorithms to generate new text or information from given data.

[0405] A "terminal" is a device such as a computer or smartphone that is directly operated by a user.

[0406] "API" stands for Application Programming Interface, and refers to an interface for exchanging data and functions between different software programs.

[0407] "Image information" refers to image data used when a user posts to an SNS.

[0408] A "machine learning model" is a set of algorithms that learn from data and perform tasks such as prediction, classification, and generation.

[0409] The embodiments of the present invention will be specifically described below.

[0410] Data collection and analysis

[0411] 1. The server collects authentication information for the SNS account provided by the user. This authentication information includes an API key and secret token, and uses this information to retrieve the user's past call history from the SNS platform. If authentication is successful, the server calls the SNS API, collects the past call history, and stores it in a database.

[0412] 2. The server analyzes the collected communication history and extracts the user's communication style. This analysis uses natural language processing techniques (e.g., NLTK, spaCy) and machine learning algorithms (e.g., BERT, GPT-2). The server tokenizes each piece of text data and performs sentiment analysis, topic modeling, and contextual analysis. As a result, the user's communication style is generated as a digital profile.

[0413] Creating a new message

[0414] 3. The server generates a new message based on the analysis results. Here, a generative AI model (e.g., GPT-3) is used. The server inputs the analysis data as a prompt into the generative AI model to generate a new sentence.

[0415] Check and correct the message

[0416] 4. The generated message is presented to the user via the device. The user can review the message and make corrections as necessary. These corrections are reflected in real time on the device screen, preventing errors and significantly reducing the amount of manual input required by the user.

[0417] Final call

[0418] 5. After the user has confirmed and corrected the message, it is finally posted to the SNS platform by the server. This posting is also automated using APIs. When the user presses the "Post" button on their device, the server sends the corrected message to the SNS API and executes the posting.

[0419] Support for sending messages with images

[0420] 6. When a user uploads an image, the server analyzes the image and automatically generates a sentence that corresponds to the content. For example, when a user uploads a landscape photo from their device, the server recognizes and analyzes the image and generates a sentence that corresponds to the atmosphere of the landscape photo.

[0421] Specific examples

[0422] For example, if a user has previously sent a message such as "It's a nice day today. I want to go for a walk!", the system will recognize that themes such as "weather" and "walks" are part of the user's communication style. When generating a new message, the system will automatically generate a sentence such as "It's a nice, sunny day today. I'm looking forward to going outside!"

[0423] Prompt Sentence Examples

[0424] Example prompt: "Generate new posts based on the user's past posting style. The post might look something like, 'It's a nice day today. I want to go for a walk!' Generate new posts based on this."

[0425] This invention allows users to efficiently post on SNS every day. It also contributes to maintaining and expanding the number of followers because users continue to post consistently. It also makes it easy to post content that includes visual elements, reducing SNS fatigue and enabling effective communication.

[0426] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0427] Step 1:

[0428] A user uses a device to enter authentication information for their social networking account. This authentication information includes an API key and a secret token. The entered authentication information is sent to the server. The server receives and securely stores the authentication information. The input of this process is the authentication information entered by the user, and the output is the securely stored authentication information.

[0429] Step 2:

[0430] The server uses the stored authentication information to call the API of the social media platform and collect the user's past call history. During this process, real-time data is obtained via the API and stored in a database. The input is the authentication information and the call history data obtained as a result of the API call, and the output is the call history stored in the database.

[0431] How it works: The server connects to the social networking platform and fetches the user's past posts. This data is stored in a database in chronological order.

[0432] Step 3:

[0433] The server analyzes the collected call history. This analysis uses natural language processing technology and machine learning algorithms to extract the user's call style. Specifically, the tone, theme, vocabulary, and sentence structure of the text are analyzed. The input is the call history data, and the output is a digital profile that shows the user's call style.

[0434] Specific operation: The server uses analysis tools (e.g., NLTK, spaCy) to analyze the text data and saves the results in a database as a digital profile.

[0435] Step 4:

[0436] The server uses a generative AI model based on the analysis results to generate a new message. At this time, the analysis data is input to the generative AI model as a prompt. The input is a digital profile that indicates the user's communication style and a prompt, and the output is a new message.

[0437] Specific operation: The server inputs the prompt sentence into a generative AI model (e.g., GPT-3) to generate a new message, which is then stored in a database.

[0438] Step 5:

[0439] The generated message is presented to the user via the terminal. The user can check this message and modify it as necessary. The input is the generated message, and the output is the message that has been checked and modified by the user.

[0440] Specific operation: The generated message is displayed on the device screen, and the user can edit the text using the keyboard or touch screen.

[0441] Step 6:

[0442] The message that the user has confirmed and modified is finally posted to the SNS platform by the server. This posting is also automated using APIs. The input is the message that the user modified, and the output is a new post on the SNS platform.

[0443] Specific operation: When the user presses the "Post" button, the server sends the revised message to the SNS API and executes the post.

[0444] Step 7:

[0445] The server analyzes the image uploaded by the user and generates a sentence based on its content. After analyzing the content of the image using image recognition technology, the server generates a sentence using a generative AI model. The input is the image data uploaded by the user, and the output is a sentence that matches the image.

[0446] Specific operation: When a user uploads an image from their device, the server recognizes the image and generates an appropriate draft based on the analysis results. The draft is then presented to the user, who can review and edit it.

[0447] (Application example 1)

[0448] 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."

[0449] Previous message generation systems on social media platforms were limited to generating new messages based on a user's past message styles, and were unable to efficiently generate and post ad copy. This meant that advertising agencies and marketing teams had to spend a great deal of time and effort creating ad copy that matched their clients' styles.

[0450] 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.

[0451] In this invention, the server includes means for collecting past communication history, means for analyzing the collected communication history and extracting the user's communication style, means for generating new messages based on the extracted communication style, means for presenting the generated messages to the user for confirmation and revision, means for posting the confirmed and revised messages to an SNS, means for automatically generating advertising copy, means for presenting the generated advertising copy to the user for confirmation and revision, and means for posting the revised advertising copy to a distribution platform. This enables efficient generation and posting of advertising copy based on the user's past communication style.

[0452] "Outgoing message history" refers to a record of text, images, etc. that a user has posted on a social media platform in the past.

[0453] "Communication style" refers to the combined characteristics of the tone, theme, language, and writing style of the content a user posts.

[0454] "Copyright" means text created to promote a particular product or service.

[0455] "Means of collection" refers to the methods and systems for obtaining a user's past posting history from a social media platform.

[0456] "Means of analysis" refers to a method of analyzing collected communication history and extracting the user's communication style from it.

[0457] The "means of generation" refers to a method for creating new messages or advertising copy based on the extracted communication style.

[0458] "Means for confirmation and correction" refers to a method in which the generated message or advertising copy is presented to the user, allowing the user to check the content themselves and make corrections as necessary.

[0459] "Method of posting" refers to the method of posting the final confirmed and revised message or advertising copy on social media or distribution platforms.

[0460] A "machine learning model" is an algorithm or system that learns patterns and features from data and makes predictions and classifications for new data.

[0461] The system for implementing this invention is made up of a server, a terminal, and a user. Each component and its function will be specifically described below.

[0462] server

[0463] The server has the function of collecting past posting history. It retrieves past posting history from the social media platform using the social media account authentication information (API key or secret token) provided by the user. The collected posting history is analyzed using natural language processing technology and machine learning algorithms. Through the analysis, features such as the tone, theme, wording, and sentence structure of the text are extracted. The server then generates new postings based on the extracted style. This generation uses a generative AI model such as OpenAI's GPT-3 model.

[0464] Terminal

[0465] The device provides a user interface, allowing the user to review and edit the generated message or ad copy. The new message or ad copy generated by the server is sent to the device and presented to the user. The user reviews the presented copy and makes any necessary edits. Once the user has reviewed and edited the message or ad copy, it is sent back to the server.

[0466] User

[0467] Users provide their social media account credentials and authorize the collection of past messaging history. They are then asked to review the generated messaging and ad copy and make any necessary revisions. Users then post the final messaging and ad copy to the social media platform.

[0468] Hardware and software used

[0469] Hardware: Smartphones, servers

[0470] Software: requests library, TextBlob library, OpenAI library, SNS platform API

[0471] Specific examples

[0472] For example, an advertising agency might generate new ad copy based on a particular client's social media accounts. If the client's past posts tend to be positive and emotional, the resulting ad copy will also have a similarly positive tone. An example of a prompt might look like this:

[0473] Generate ad copy based on your client's social media posting style. Consider the following style characteristics:

[0474] Polarity: 0.8, Subjectivity: 0.6

[0475] This invention allows advertising agencies and marketing teams to efficiently generate ad copy that matches their client's style, then review and revise it before posting it on the target platform, improving the efficiency and effectiveness of ad copy creation and enhancing the quality of marketing activities.

[0476] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0477] Step 1:

[0478] The server collects past call history using the authentication information of the SNS account provided by the user. The input is the user's authentication information, and the output is the call history obtained from the SNS platform. In this process, the API of the SNS platform is used to obtain the latest call history.

[0479] Step 2:

[0480] The server analyzes the collected communication history and extracts the user's communication style. The input is the collected communication history, and the output is the characteristics of the user's communication style (tone, theme, wording, sentence structure). This analysis uses natural language processing technology and the TextBlob library to perform sentiment analysis of the text and word frequency analysis.

[0481] Step 3:

[0482] The server generates a new message or ad copy based on the extracted style. The input is the user's message style characteristics, and the output is the generated new message or ad copy. The copy is generated using a generative AI model such as OpenAI's GPT-3 model. The prompt sentence is provided as input to the model.

[0483] Step 4:

[0484] The generated message or copy draft is sent from the server to the terminal and presented to the user. The input is the generated message or copy draft, and the output is a screen display presented to the user for confirmation and correction, provided via the terminal's user interface.

[0485] Step 5:

[0486] The user checks the message or advertising copy presented to them and makes corrections as necessary. The input is the message or advertising copy presented to the user, and the output is the copy that has been checked and corrected by the user. The user makes corrections through their terminal and sends the corrections to the server.

[0487] Step 6:

[0488] The server finally posts the message or advertising copy that has been confirmed and revised by the user to the SNS platform. The input is the confirmed and revised copy, and the output is the final copy posted to the SNS platform. In this process, the posting is made via the SNS platform's API.

[0489] Step 7:

[0490] When image information is included in the generated message, the server provides a means to automatically generate a message that matches the image content. The input is an image uploaded by the user and a prompt message based on its content, and the output is a new message that matches the image. A generative AI model is used to generate a message that matches the image.

[0491] This process allows users to efficiently create new messages and copy drafts, review and revise them, and finally post them to social media platforms, allowing advertising agencies and marketing teams to generate effective copy drafts that fit their clients' communication styles.

[0492] 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.

[0493] The embodiments of the present invention will be specifically described below.

[0494] Data collection

[0495] A user enters the authentication information of a social networking account (API key, API secret, access token, access token secret) into the system, which allows the system to access the specific social networking account.

[0496] Collection of past call history

[0497] The server accesses the SNS account using the authentication information provided by the user and collects past call history. This collection is done by using the SNS API to obtain the latest call history. The server is set to collect a certain amount of past call history (for example, 200 items).

[0498] Style and Sentiment Analysis

[0499] The collected call history is analyzed by the server. First, natural language processing of the text data is used to extract features such as tone, theme, wording, and sentence structure. At the same time, an emotion engine is used to detect the user's emotional state from the call history. The emotion engine identifies emotional categories such as positive, negative, and neutral, and labels the data accordingly.

[0500] Extracting the delivery style

[0501] The server extracts the user's communication style based on the analysis results. At this stage, the analysis also includes the relationship between the user's communication style and their emotional state. For example, it distinguishes between messages sent when the user was in a positive emotional state and messages sent when the user was in a negative emotional state.

[0502] Creating a new message

[0503] The server generates new messages based on the extracted user's communication style. In this process, a generative AI (e.g., GPT-2 model) is used to create messages that reflect the user's emotional state. The generation prompts include topics and tones based on the user's current emotional state.

[0504] Emotion recognition and correction of messages

[0505] The generated message is displayed to the user through the device. Based on the user's feedback and corrections, the generated message is checked for appropriate sentiment. The interface includes a function where the generated message is displayed in a text box and the user can enter corrections.

[0506] Final call

[0507] The final message, after being confirmed and corrected by the user, is posted to the SNS platform by the server. This posting is also done via API, so the final message is published to the SNS from the user's account.

[0508] Image information support

[0509] The system also supports messages containing image information. When a user uploads an image, the server automatically generates appropriate text based on the image content. This includes a process that uses an image analysis algorithm to generate text that corresponds to the image content.

[0510] Utilizing machine learning models

[0511] The server uses a machine learning model to analyze the collected message history and emotional data. The model learns from the user's message history and emotional state and uses the information to generate future messages.

[0512] Specific examples

[0513] For example, suppose a user previously made a positive statement such as, "Today was such a fun day! I discovered a new place!" If the system recognizes the user's current emotional state as positive, it will generate a new statement such as, "The weather was great today, and I went to a new cafe. It was so relaxing!" Conversely, if the emotion engine detects a negative emotion, it will generate a statement such as, "Today was a stressful day, but I'm home and relaxing."

[0514] This invention allows users to efficiently post regular messages on SNS every day, and enables them to post more naturally, reflecting their own emotional state. Also, if there is a message that users really want to convey, they can post it manually each time, which helps to reduce SNS fatigue and maintain and increase the number of followers.

[0515] The processing flow will be explained below.

[0516] Step 1:

[0517] A user enters their social networking account authentication information (API key, API secret, access token, access token secret) into the system, which allows the system to access the specific social networking account.

[0518] Step 2:

[0519] The server accesses the SNS account using the authentication information provided by the user and collects the past call history. To do this, it uses the SNS API to retrieve the most recent call history. The past call history covers 200 calls.

[0520] Step 3:

[0521] The server analyzes the collected call history using natural language processing technology to extract features such as the tone, theme, vocabulary, and sentence structure of the text.

[0522] Step 4:

[0523] The server uses an emotion engine to analyze the user's emotional state from the collected call history, which involves identifying emotional categories such as positive, negative, and neutral.

[0524] Step 5:

[0525] The server extracts the user's communication style based on the analyzed data. This extraction process identifies distinctive patterns and styles from communication history and distinguishes styles according to emotional states.

[0526] Step 6:

[0527] The server generates new messages based on the extracted user's communication style and current emotional state. Here, generative AI (e.g., GPT-2 model) is used to create natural-sounding sentences that reflect the user's emotional state.

[0528] Step 7:

[0529] The generated message is displayed to the user through the terminal. The terminal provides an interface for the user to review the new message and modify it if necessary. This interface includes a function that displays the generated message in a text box and allows the user to edit its content.

[0530] Step 8:

[0531] The user can check and modify the generated message. Once the user has completed the modifications and is ready to send the message, the user presses the confirmation button to finalize the message.

[0532] Step 9:

[0533] The server posts the final message that the user has confirmed and corrected to the SNS platform. This posting is also done via API, and the message is published to the SNS from the user's account.

[0534] Step 10:

[0535] When a user uploads image information, the server automatically generates appropriate copy based on the image content, including generating text based on the image content using image analysis algorithms.

[0536] Step 11:

[0537] Messages with images are also displayed to the user via the device, and the user is given the opportunity to review and edit them. Once the user has reviewed and edited the message, it is finally posted to the SNS.

[0538] These are the specific processing steps of the invention that combines the emotion engine. This process allows users to efficiently post natural messages that correspond to their emotional state. The combination of the emotion engine and generative AI enables users to post messages that are more in line with their own personalities, reducing social media fatigue and enabling users to maintain and expand their following.

[0539] Example 2

[0540] 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."

[0541] On existing social media platforms, users spend a lot of time and effort on daily posting. Furthermore, the content of posts tends to be monotonous, often resulting in low engagement with followers. Furthermore, it is difficult to post natural, rich content that reflects the user's emotional state. In such situations, users may become exhausted by social media activity and ultimately abandon or delete their accounts. There is a need for a system that can solve these issues and enable users to post efficiently and appropriately.

[0542] 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.

[0543] In this invention, the server includes means for a user to input authentication information for the SNS service, means for collecting past message history, means for analyzing the collected message history and extracting the user's message style and emotional state, means using a generative AI model to generate new messages based on the extracted message style and emotional state, means for presenting the generated messages to the user and allowing them to confirm and revise, means for finally posting the confirmed and revised messages to the SNS service, means for analyzing collected image information and generating appropriate drafts based on the contents, and means for using a machine learning model to learn the collected message history and emotional state. This enables users to efficiently post to SNS every day and to post natural and diverse messages that reflect their own emotions.

[0544] "User" refers to an individual or organization that uses the SNS service.

[0545] "Authentication Information" refers to information such as API keys, API secrets, access tokens, and access token secrets required to access SNS services.

[0546] "SNS services" refers to social networking services, such as Twitter and Facebook.

[0547] "Server" refers to a computer system on a network that accesses SNS services and collects, analyzes, generates, and posts data.

[0548] "Outgoing message history" refers to a record of posts and comments a user has made on social media in the past.

[0549] "Analysis" refers to data processing performed to extract specific characteristics from the collected call history.

[0550] "Communication style" refers to the patterns of language, tone, and themes used by users on social media.

[0551] "Emotional state" refers to the user's emotional state, such as positive, negative, or neutral, extracted from the call history.

[0552] A "generative AI model" refers to an artificial intelligence model that uses machine learning techniques to generate new text.

[0553] A "prompt" refers to the input data or instructions that a generative AI model uses to generate new text.

[0554] "Confirmation and correction" refers to the act of the user confirming the generated message and correcting the content as necessary.

[0555] "Posting" refers to the act of publishing a message that has been finally confirmed and corrected from a user's SNS account.

[0556] "Image information" refers to the image data used by users when posting to SNS.

[0557] "Draft" refers to the text content generated based on image information.

[0558] A "machine learning model" refers to an algorithm or system used to learn from data and accomplish a specific task.

[0559] The present invention relates to a system that improves the efficiency of users' SNS activities and enables natural posting that reflects their emotions. Specific embodiments of the present invention will be described below.

[0560] First, the user enters the authentication information (API key, API secret, access token, access token secret) of the SNS service into the system, which allows the system to access the specific SNS account.

[0561] Next, the server uses the authentication information to access the API of the SNS service and retrieve the user's past call history. Specifically, the server is configured to collect the most recent 200 call histories. The collected data is stored in a database.

[0562] The server performs text analysis based on the collected call history. This analysis uses natural language processing libraries (e.g., NLTK and spaCy) and emotion engines (e.g., Azure Sentiment Analysis). This extracts features such as tone, theme, wording, and sentence structure from the call history, and labels the emotional state (e.g., positive, negative, neutral, etc.).

[0563] Based on the analysis results, the server extracts the user's communication style. Here, it distinguishes between messages sent by the user in a positive emotional state and messages sent by the user in a negative emotional state. For example, if a user's communication style is dominated by "fun" and "joy," it is classified as a positive style.

[0564] The server then uses a generative AI model (e.g., OpenAI's GPT-3) to generate a new message. The generation prompt is populated with the user's current emotional state and topic based on the analysis results. For example, a prompt might be, "Generate a new message based on past positive messages." The generated message is then sent from the server to the device and presented to the user.

[0565] The terminal displays the generated message received from the server to the user, allowing the user to confirm and edit the message using the text box. When the user enters the edits, the information is sent back to the server.

[0566] The final message that has been confirmed and corrected by the user is posted to the SNS from the user's account via the SNS service's API by the server.

[0567] The system also supports users uploading images. The server uses an image analysis algorithm (e.g., Google Cloud Vision) to analyze the image content and generate appropriate text based on the content. For example, if a "landscape image" is recognized, the system generates the text "I found a beautiful landscape!"

[0568] Furthermore, the server uses a machine learning model (e.g., Scikit-learn) to learn from the collected message history and emotional state. This model learns from the user's message history and emotional data and uses it to generate future messages.

[0569] This will enable users to post on social media every day efficiently and enable natural and diverse communication that appropriately reflects their emotions, which is expected to improve engagement and reduce social media fatigue.

[0570] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0571] Program processing flow

[0572] Step 1:

[0573] The user enters the authentication information (API key, API secret, access token, access token secret) for the social networking service into a web form and clicks a button to submit it.

[0574] The data entered is authentication information, and based on this information, information that allows access to the user's SNS account is obtained.

[0575] Step 2:

[0576] The server receives the authentication information sent by the user and accesses the API of the SNS service to obtain the past call history.

[0577] Specifically, the server sends an API request to the SNS endpoint and saves the most recent 200 outgoing calls in a database.

[0578] The input is the SNS authentication information and request, and the output is the call history data.

[0579] Step 3:

[0580] The server analyzes the collected call history.

[0581] A natural language processing library (e.g., NLTK or spaCy) is used to analyze the tone, theme, wording, sentence structure, etc. of the text data. In addition, a sentiment engine (e.g., Azure Sentiment Analysis) is used to label the sentiment state (positive, negative, neutral) of the call history.

[0582] The input is the call history data, and the output is the analysis results (tone, theme, wording, sentence structure, emotional state).

[0583] Step 4:

[0584] The server extracts the user's communication style based on the analysis results.

[0585] Using the analysis results, we distinguish between messages sent in a positive emotional state and those sent in a negative emotional state, and clarify the user's communication style. For example, sentences containing many words like "fun" and "joy" are classified as a positive style.

[0586] The input is the analysis result, and the output is the transmission style.

[0587] Step 5:

[0588] The server generates a new message based on the extracted message style.

[0589] A prompt (e.g., "Generate a new message based on past positive messages") is input into a generative AI model (e.g., OpenAI's GPT-3) and the generated text is received.

[0590] The input is the communication style and the prompt sentence, and the output is the generated communication sentence.

[0591] Step 6:

[0592] The terminal displays the generated message to the user.

[0593] The generated message is displayed in a text box, providing an interface that allows the user to check and edit it.

[0594] The input is the generated message, and the output is the user's confirmation and correction.

[0595] Step 7:

[0596] The user checks the generated message and enters corrections as necessary.

[0597] Check the message displayed in the text box and correct any inappropriate parts. The corrections will be sent to the server again.

[0598] The input is the generated message and the user's suggestions for revision, and the output is the revised message.

[0599] Step 8:

[0600] The server posts the final, corrected message to the SNS service.

[0601] The revised message will be published from the user's account via the SNS API.

[0602] The input is the modified message and the output is a post on a social networking site.

[0603] Step 9:

[0604] The server analyzes the image information uploaded by the user and generates appropriate text.

[0605] It uses image analysis algorithms (e.g., Google Cloud Vision) to analyze the image content and generate text based on that content. For example, if it recognizes an image as a landscape, it will generate the sentence "I found a beautiful landscape!"

[0606] The input is image information, and the output is a generated draft based on the image.

[0607] Step 10:

[0608] The server uses a machine learning model to learn from the collected call history and emotional state.

[0609] The collected message history and emotion labels are used to train a Scikit-learn model, which will provide learning results that will be useful for generating future messages.

[0610] The input is the call history and emotion labels, and the output is a trained machine learning model.

[0611] (Application example 2)

[0612] 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."

[0613] In modern society, many users use social media to share information, but the content of their posts is subjective and inconsistent, making it difficult to reflect their own emotional state or style. It is also difficult to automatically and efficiently create effective ad copy that matches a user's emotions and style when generating ads. To solve these problems, a system is needed that analyzes a user's past posting history and generates new posts and ad copy that reflect that user's style and emotional state.

[0614] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0615] In this invention, the server includes means for collecting past call history, means for analyzing the collected call history and extracting the user's call style, and means for generating new messages based on the extracted call style. This makes it possible to automatically generate messages and advertising copy that reflect the user's past call history and emotional state, present them to the user, and post them to SNS or advertising media after final confirmation.

[0616] "Past communication history" refers to all of the content posted or transmitted by a user in the past on social media platforms and other platforms.

[0617] "Means of collection" refers to the technical processes and functions that the system uses to obtain past communication history using the user's social media account authentication information.

[0618] "Means of analyzing and extracting a user's communication style" refers to a means of analyzing collected communication history using natural language processing and machine learning technology to identify characteristics of the user's writing, such as tone, theme, and wording.

[0619] "Means for generating new messages" refers to technology for creating new messages using AI models, etc., based on the extracted user's communication style and emotional state.

[0620] "Means for presenting the generated message to the user and allowing them to review and correct it" refers to an interface or process that displays the generated message to the user, allows the user to review its contents, and makes corrections as necessary.

[0621] "Means for posting the final verified and corrected message on a social networking site" refers to the technical process by which the final message that has been verified and corrected by the user is automatically posted on a platform such as a social networking site.

[0622] "Means for generating advertising copy and advertising visuals based on the extracted communication style and emotional state" refers to technology for automatically creating advertising copy and visuals for commercial purposes, taking into account the user's communication style and emotional state.

[0623] "Means for presenting the generated ad copy to the user and allowing the user to modify and finalize it" refers to an interface or function that displays the generated ad copy to the user, allowing the user to review it and modify it as necessary.

[0624] "Means for posting to designated media" refers to the technical process for automatically distributing and posting the final, verified and corrected advertising copy to the designated advertising media or platform.

[0625] The following is a specific description of an embodiment of the present invention. The entire system is composed of a server, a terminal, and a user. The programs that are part of this system are realized as follows.

[0626] Data collection methods

[0627] The server uses the authentication information for the SNS account provided by the user to collect past call history via the SNS API. Specifically, it accesses the SNS account using authentication information such as the API key, API secret, access token, and access token secret, and retrieves the past call history (e.g., 200 calls). At this stage, data is collected from the SNS API and stored on the server.

[0628] Analysis of communication style and emotional state

[0629] The collected call history is analyzed within the server. Natural language processing technology is used to extract features such as tone, theme, and wording of the call. An emotion engine is also used to detect the user's emotional state (e.g., positive, negative, neutral) from the call history. This analysis uses machine learning models (e.g., BERT) and sentiment analysis models.

[0630] Generate new messaging and ad copy

[0631] The server then uses a generative AI model (e.g., GPT-2) to generate new messages based on the extracted user's communication style and emotional state. The generation prompts are populated with topics and tones based on the user's emotional state, and new messages are created. Similarly, ad copy and visuals are automatically generated to match the user's communication style and emotional state.

[0632] For example, the following prompt might be used:

[0633] Prompt: "Create an ad for a new cafe with a positive tone that matches the user's communication style."

[0634] Feedback and Corrections

[0635] The generated message and advertisement copy are presented to the user via the device. The user can review these copies and make corrections as necessary. This procedure is an important step to ensure that the automatically generated copies match the user's expectations and intentions. An interface is provided that allows the user to provide feedback and make corrections.

[0636] Final Post

[0637] The final message and advertisement copy that has been checked and corrected is posted by the server to the social media platform or designated advertising medium. This posting is also done via API and is published from the user's account.

[0638] In this way, the present invention can efficiently generate consistent messages and advertisements that reflect the user's past message history and emotional state, and finally publish them after confirmation and correction.

[0639] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0640] Step 1: Data collection

[0641] The server receives the authentication information (API key, API secret, access token, access token secret) of the user's SNS account as input. Using this authentication information, it accesses the SNS API and retrieves the user's past call history. At this time, the most recent 200 call history entries are collected and stored on the server. The output is the collected call history data.

[0642] Step 2: Analyze communication style and emotional state

[0643] The server analyzes the collected call history data as input. First, it uses natural language processing technology to extract features such as tone, theme, and wording from the call history. At the same time, it uses an emotion engine to detect the emotional state (positive, negative, neutral) of each message. This identifies the user's call style and emotional state. The output is the analyzed call style and emotional state data.

[0644] Step 3: Generate a new message

[0645] The server generates a new message using the analyzed message style and emotional state data. It uses a generative AI model (GPT-2) to generate the message. The following prompt is input:

[0646] Prompt: "Create an ad for a new cafe with a positive tone that matches the user's communication style."

[0647] Based on this prompt, a new message is generated. The output is the generated message.

[0648] Step 4: Generate ad copy and visuals

[0649] The server generates ad copy and visuals using the analyzed customer communication style and emotional state data as input. Using a generative AI model, ad copy that matches the user's emotional state is created based on the prompt text. The output is the generated ad copy and visuals.

[0650] Step 5: Feedback and revisions

[0651] The terminal presents the generated message and advertisement copy to the user. The user checks the contents and makes any necessary corrections. At this time, the terminal sends the user's corrections to the server, which reflects the suggested corrections. The input is the generated message and advertisement copy, and the output is the final version that has been checked and corrected by the user.

[0652] Step 6: Final submission

[0653] The server receives as input the final message and advertisement copy that has been reviewed and revised by the user. It posts this to the social media platform or designated advertising medium. The posting is again done using the social media API, and is made public through the user's account. The output is the publication of the final message and advertisement copy.

[0654] 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.

[0655] 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.

[0656] 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.

[0657] [Third embodiment]

[0658] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0659] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0660] 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).

[0661] 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.

[0662] 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.

[0663] 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).

[0664] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0665] 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.

[0666] 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.

[0667] 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.

[0668] 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.

[0669] 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."

[0670] The embodiments of the present invention will be specifically described below.

[0671] Data collection

[0672] The server collects past call history using authentication information for the SNS account provided by the user. This authentication information includes an API key and a secret token, and uses this to obtain the latest call history from the SNS platform.

[0673] Style Analysis

[0674] The collected call history is analyzed by the server. This analysis includes the tone, theme, wording, sentence structure, etc. of the text. Specifically, natural language processing technology and machine learning algorithms are used to extract features from the call history.

[0675] Creating a new message

[0676] The server generates new messages based on the analysis results. At this stage, generative AI (e.g., GPT-2 model) is used to create messages that match the user's style. The generated messages match the user's past communication style, allowing them to communicate in a way that is "true to their personality."

[0677] Check and correct the message

[0678] The generated message is presented to the user via the terminal. The user can review the message and make corrections as necessary. This process prevents accidental and incorrect messages from being sent, while significantly reducing the amount of manual input required by the user.

[0679] Final call

[0680] Once the user has confirmed and corrected the message, it is finally posted to the SNS platform by the server, which also posts the message via API.

[0681] Call response with images

[0682] This system also supports messages that include image information. When a user uploads an image, it automatically generates text that matches the image content. For example, when uploading a landscape photo, text that matches the atmosphere of the photo is generated. This makes it easy to send messages that include visual elements.

[0683] Utilizing machine learning models

[0684] The server uses a machine learning model to analyze the collected call history, learning the topics, tone, and word choice of the user from their call history and incorporating these characteristics into the generation of new messages.

[0685] Specific examples

[0686] For example, if a user has previously sent a message such as "It's a nice day today. I want to go for a walk!", the system will recognize that themes such as "weather" and "walks" are part of the user's style. Next, when generating a new message, the system will automatically generate a message such as "It's a nice, sunny day today. I'm looking forward to going outside!"

[0687] This invention allows users to efficiently send out regular messages every day. Furthermore, if there is a message that users really want to convey, they can send it manually each time, which helps to reduce SNS fatigue and maintain and increase the number of followers.

[0688] The processing flow will be explained below.

[0689] Step 1:

[0690] A user enters their social networking account authentication information (API key, API secret, access token, access token secret) into the system, which allows the system to access the specific social networking account.

[0691] Step 2:

[0692] The server accesses the SNS account using the authentication information provided by the user and collects past call history. Specifically, it uses the SNS API to obtain the latest call history. At this time, it is set to collect a certain amount of past call history (for example, 200 items).

[0693] Step 3:

[0694] The server analyzes the collected call history. The analysis uses natural language processing and machine learning techniques, applying algorithms that identify frequently occurring words and topics in texts to extract features such as tone, theme, wording, and sentence structure.

[0695] Step 4:

[0696] The server extracts the user's communication style. This extraction step identifies specific patterns and styles from past communication history. For example, frequently used phrases and specific tones (friendly, professional, etc.) are analyzed.

[0697] Step 5:

[0698] The server uses a generative AI to generate new sentences based on the extracted user's speech style. The generative AI (e.g., a GPT-2 model) creates natural sentences that reflect the user's style. In this generation step, a specific prompt is input and a sentence is generated accordingly.

[0699] Step 6:

[0700] The generated message is displayed to the user through the terminal. The terminal provides an interface for the user to review the new message and modify it if necessary. This interface includes a function that displays the generated message in a text box and allows the user to edit its content.

[0701] Step 7:

[0702] The user checks the generated message and makes any corrections. When the user has completed the corrections and is ready to send the message, they press the Confirm button. By pressing this button, the final message is confirmed.

[0703] Step 8:

[0704] The server receives the final message that the user has confirmed and corrected, and posts it to the SNS platform. This posting is also done via API, and the final message is published to the SNS from the user's account.

[0705] Step 9:

[0706] When a user uploads image information, the server automatically generates text that matches the image content. For this purpose, an image analysis algorithm is used to generate relevant text based on the image content.

[0707] Step 10:

[0708] Similarly, text with images is presented to the user via the device, and they are given the opportunity to check and correct it. After going through a series of checking and correcting processes, the final message with images is posted to the SNS.

[0709] These are the specific steps for automatically generating new messages based on past message history, and then posting them to SNS after the user has confirmed and edited them. This process allows users to efficiently send regular daily messages and saves time and effort in maintaining and expanding their followers.

[0710] Example 1

[0711] 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."

[0712] Modern social media users often feel burdened by frequent posting. Users' posts can also be inconsistent, making it difficult to keep their followers engaged. Furthermore, as posts increasingly include visual elements, quickly generating appropriate copy for images can be time-consuming. A new system is needed to solve these problems.

[0713] 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.

[0714] In this invention, the server includes means for collecting past call history using authentication information provided by the user, means for analyzing the collected call history and extracting the user's call style, and means for generating new messages using a generative AI model based on the extracted call style. This enables users to efficiently send messages with consistency. The present invention also includes means for presenting the generated messages to the user via a terminal for confirmation and correction, and means for posting the final confirmed and corrected messages to an SNS via an API. This enables users to send messages quickly and accurately. Furthermore, the present invention includes means for including image information in the generated messages and automatically generating drafts based on the image content, making it easy to send messages that include visual elements.

[0715] "Authentication information" refers to information such as an API key or secret token that a user needs to access their SNS account.

[0716] "Outgoing communication history" refers to a record of all posts and comments that a user has made on a social networking platform in the past.

[0717] A "server" is a computer system that processes, stores, and controls access to data over a network.

[0718] "Communication style" refers to characteristics such as tone, theme, wording, and sentence structure of text extracted from a user's past communication history.

[0719] A "generative AI model" is an artificial intelligence model that uses natural language processing and machine learning algorithms to generate new text or information from given data.

[0720] A "terminal" is a device such as a computer or smartphone that is directly operated by a user.

[0721] "API" stands for Application Programming Interface, and refers to an interface for exchanging data and functions between different software programs.

[0722] "Image information" refers to image data used when a user posts to an SNS.

[0723] A "machine learning model" is a set of algorithms that learn from data and perform tasks such as prediction, classification, and generation.

[0724] The embodiments of the present invention will be specifically described below.

[0725] Data collection and analysis

[0726] 1. The server collects authentication information for the SNS account provided by the user. This authentication information includes an API key and secret token, and uses this information to retrieve the user's past call history from the SNS platform. If authentication is successful, the server calls the SNS API, collects the past call history, and stores it in a database.

[0727] 2. The server analyzes the collected communication history and extracts the user's communication style. This analysis uses natural language processing techniques (e.g., NLTK, spaCy) and machine learning algorithms (e.g., BERT, GPT-2). The server tokenizes each piece of text data and performs sentiment analysis, topic modeling, and contextual analysis. As a result, the user's communication style is generated as a digital profile.

[0728] Creating a new message

[0729] 3. The server generates a new message based on the analysis results. Here, a generative AI model (e.g., GPT-3) is used. The server inputs the analysis data as a prompt into the generative AI model to generate a new sentence.

[0730] Check and correct the message

[0731] 4. The generated message is presented to the user via the device. The user can review the message and make corrections as necessary. These corrections are reflected in real time on the device screen, preventing errors and significantly reducing the amount of manual input required by the user.

[0732] Final call

[0733] 5. After the user has confirmed and corrected the message, it is finally posted to the SNS platform by the server. This posting is also automated using APIs. When the user presses the "Post" button on their device, the server sends the corrected message to the SNS API and executes the posting.

[0734] Support for sending messages with images

[0735] 6. When a user uploads an image, the server analyzes the image and automatically generates a sentence that corresponds to the content. For example, when a user uploads a landscape photo from their device, the server recognizes and analyzes the image and generates a sentence that corresponds to the atmosphere of the landscape photo.

[0736] Specific examples

[0737] For example, if a user has previously sent a message such as "It's a nice day today. I want to go for a walk!", the system will recognize that themes such as "weather" and "walks" are part of the user's communication style. When generating a new message, the system will automatically generate a sentence such as "It's a nice, sunny day today. I'm looking forward to going outside!"

[0738] Prompt Sentence Examples

[0739] Example prompt: "Generate new posts based on the user's past posting style. The post might look something like, 'It's a nice day today. I want to go for a walk!' Generate new posts based on this."

[0740] This invention allows users to efficiently post on SNS every day. It also contributes to maintaining and expanding the number of followers because users continue to post consistently. It also makes it easy to post content that includes visual elements, reducing SNS fatigue and enabling effective communication.

[0741] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0742] Step 1:

[0743] A user uses a device to enter authentication information for their social networking account. This authentication information includes an API key and a secret token. The entered authentication information is sent to the server. The server receives and securely stores the authentication information. The input of this process is the authentication information entered by the user, and the output is the securely stored authentication information.

[0744] Step 2:

[0745] The server uses the stored authentication information to call the API of the social media platform and collect the user's past call history. During this process, real-time data is obtained via the API and stored in a database. The input is the authentication information and the call history data obtained as a result of the API call, and the output is the call history stored in the database.

[0746] How it works: The server connects to the social networking platform and fetches the user's past posts. This data is stored in a database in chronological order.

[0747] Step 3:

[0748] The server analyzes the collected call history. This analysis uses natural language processing technology and machine learning algorithms to extract the user's call style. Specifically, the tone, theme, vocabulary, and sentence structure of the text are analyzed. The input is the call history data, and the output is a digital profile that shows the user's call style.

[0749] Specific operation: The server uses analysis tools (e.g., NLTK, spaCy) to analyze the text data and saves the results in a database as a digital profile.

[0750] Step 4:

[0751] The server uses a generative AI model based on the analysis results to generate a new message. At this time, the analysis data is input to the generative AI model as a prompt. The input is a digital profile that indicates the user's communication style and a prompt, and the output is a new message.

[0752] Specific operation: The server inputs the prompt sentence into a generative AI model (e.g., GPT-3) to generate a new message, which is then stored in a database.

[0753] Step 5:

[0754] The generated message is presented to the user via the terminal. The user can check this message and modify it as necessary. The input is the generated message, and the output is the message that has been checked and modified by the user.

[0755] Specific operation: The generated message is displayed on the device screen, and the user can edit the text using the keyboard or touch screen.

[0756] Step 6:

[0757] The message that the user has confirmed and modified is finally posted to the SNS platform by the server. This posting is also automated using APIs. The input is the message that the user modified, and the output is a new post on the SNS platform.

[0758] Specific operation: When the user presses the "Post" button, the server sends the revised message to the SNS API and executes the post.

[0759] Step 7:

[0760] The server analyzes the image uploaded by the user and generates a sentence based on its content. After analyzing the content of the image using image recognition technology, the server generates a sentence using a generative AI model. The input is the image data uploaded by the user, and the output is a sentence that matches the image.

[0761] Specific operation: When a user uploads an image from their device, the server recognizes the image and generates an appropriate draft based on the analysis results. The draft is then presented to the user, who can review and edit it.

[0762] (Application example 1)

[0763] 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."

[0764] Previous message generation systems on social media platforms were limited to generating new messages based on a user's past message styles, and were unable to efficiently generate and post ad copy. This meant that advertising agencies and marketing teams had to spend a great deal of time and effort creating ad copy that matched their clients' styles.

[0765] 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.

[0766] In this invention, the server includes means for collecting past communication history, means for analyzing the collected communication history and extracting the user's communication style, means for generating new messages based on the extracted communication style, means for presenting the generated messages to the user for confirmation and revision, means for posting the confirmed and revised messages to an SNS, means for automatically generating advertising copy, means for presenting the generated advertising copy to the user for confirmation and revision, and means for posting the revised advertising copy to a distribution platform. This enables efficient generation and posting of advertising copy based on the user's past communication style.

[0767] "Outgoing message history" refers to a record of text, images, etc. that a user has posted on a social media platform in the past.

[0768] "Communication style" refers to the combined characteristics of the tone, theme, language, and writing style of the content a user posts.

[0769] "Copyright" means text created to promote a particular product or service.

[0770] "Means of collection" refers to the methods and systems for obtaining a user's past posting history from a social media platform.

[0771] "Means of analysis" refers to a method of analyzing collected communication history and extracting the user's communication style from it.

[0772] The "means of generation" refers to a method for creating new messages or advertising copy based on the extracted communication style.

[0773] "Means for confirmation and correction" refers to a method in which the generated message or advertising copy is presented to the user, allowing the user to check the content themselves and make corrections as necessary.

[0774] "Method of posting" refers to the method of posting the final confirmed and revised message or advertising copy on social media or distribution platforms.

[0775] A "machine learning model" is an algorithm or system that learns patterns and features from data and makes predictions and classifications for new data.

[0776] The system for implementing this invention is made up of a server, a terminal, and a user. Each component and its function will be specifically described below.

[0777] server

[0778] The server has the function of collecting past posting history. It retrieves past posting history from the social media platform using the social media account authentication information (API key or secret token) provided by the user. The collected posting history is analyzed using natural language processing technology and machine learning algorithms. Through the analysis, features such as the tone, theme, wording, and sentence structure of the text are extracted. The server then generates new postings based on the extracted style. This generation uses a generative AI model such as OpenAI's GPT-3 model.

[0779] Terminal

[0780] The device provides a user interface, allowing the user to review and edit the generated message or ad copy. The new message or ad copy generated by the server is sent to the device and presented to the user. The user reviews the presented copy and makes any necessary edits. Once the user has reviewed and edited the message or ad copy, it is sent back to the server.

[0781] User

[0782] Users provide their social media account credentials and authorize the collection of past messaging history. They are then asked to review the generated messaging and ad copy and make any necessary revisions. Users then post the final messaging and ad copy to the social media platform.

[0783] Hardware and software used

[0784] Hardware: Smartphones, servers

[0785] Software: requests library, TextBlob library, OpenAI library, SNS platform API

[0786] Specific examples

[0787] For example, an advertising agency might generate new ad copy based on a particular client's social media accounts. If the client's past posts tend to be positive and emotional, the resulting ad copy will also have a similarly positive tone. An example of a prompt might look like this:

[0788] Generate ad copy based on your client's social media posting style. Consider the following style characteristics:

[0789] Polarity: 0.8, Subjectivity: 0.6

[0790] This invention allows advertising agencies and marketing teams to efficiently generate ad copy that matches their client's style, then review and revise it before posting it on the target platform, improving the efficiency and effectiveness of ad copy creation and enhancing the quality of marketing activities.

[0791] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0792] Step 1:

[0793] The server collects past call history using the authentication information of the SNS account provided by the user. The input is the user's authentication information, and the output is the call history obtained from the SNS platform. In this process, the API of the SNS platform is used to obtain the latest call history.

[0794] Step 2:

[0795] The server analyzes the collected communication history and extracts the user's communication style. The input is the collected communication history, and the output is the characteristics of the user's communication style (tone, theme, wording, sentence structure). This analysis uses natural language processing technology and the TextBlob library to perform sentiment analysis of the text and word frequency analysis.

[0796] Step 3:

[0797] The server generates a new message or ad copy based on the extracted style. The input is the user's message style characteristics, and the output is the generated new message or ad copy. The copy is generated using a generative AI model such as OpenAI's GPT-3 model. The prompt sentence is provided as input to the model.

[0798] Step 4:

[0799] The generated message or copy draft is sent from the server to the terminal and presented to the user. The input is the generated message or copy draft, and the output is a screen display presented to the user for confirmation and correction, provided via the terminal's user interface.

[0800] Step 5:

[0801] The user checks the message or advertising copy presented to them and makes corrections as necessary. The input is the message or advertising copy presented to the user, and the output is the copy that has been checked and corrected by the user. The user makes corrections through their terminal and sends the corrections to the server.

[0802] Step 6:

[0803] The server finally posts the message or advertising copy that has been confirmed and revised by the user to the SNS platform. The input is the confirmed and revised copy, and the output is the final copy posted to the SNS platform. In this process, the posting is made via the SNS platform's API.

[0804] Step 7:

[0805] When image information is included in the generated message, the server provides a means to automatically generate a message that matches the image content. The input is an image uploaded by the user and a prompt message based on its content, and the output is a new message that matches the image. A generative AI model is used to generate a message that matches the image.

[0806] This process allows users to efficiently create new messages and copy drafts, review and revise them, and finally post them to social media platforms, allowing advertising agencies and marketing teams to generate effective copy drafts that fit their clients' communication styles.

[0807] 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.

[0808] The embodiments of the present invention will be specifically described below.

[0809] Data collection

[0810] A user enters the authentication information of a social networking account (API key, API secret, access token, access token secret) into the system, which allows the system to access the specific social networking account.

[0811] Collection of past call history

[0812] The server accesses the SNS account using the authentication information provided by the user and collects past call history. This collection is done by using the SNS API to obtain the latest call history. The server is set to collect a certain amount of past call history (for example, 200 items).

[0813] Style and Sentiment Analysis

[0814] The collected call history is analyzed by the server. First, natural language processing of the text data is used to extract features such as tone, theme, wording, and sentence structure. At the same time, an emotion engine is used to detect the user's emotional state from the call history. The emotion engine identifies emotional categories such as positive, negative, and neutral, and labels the data accordingly.

[0815] Extracting the delivery style

[0816] The server extracts the user's communication style based on the analysis results. At this stage, the analysis also includes the relationship between the user's communication style and their emotional state. For example, it distinguishes between messages sent when the user was in a positive emotional state and messages sent when the user was in a negative emotional state.

[0817] Creating a new message

[0818] The server generates new messages based on the extracted user's communication style. In this process, a generative AI (e.g., GPT-2 model) is used to create messages that reflect the user's emotional state. The generation prompts include topics and tones based on the user's current emotional state.

[0819] Emotion recognition and correction of messages

[0820] The generated message is displayed to the user through the device. Based on the user's feedback and corrections, the generated message is checked for appropriate sentiment. The interface includes a function where the generated message is displayed in a text box and the user can enter corrections.

[0821] Final call

[0822] The final message, after being confirmed and corrected by the user, is posted to the SNS platform by the server. This posting is also done via API, so the final message is published to the SNS from the user's account.

[0823] Image information support

[0824] The system also supports messages containing image information. When a user uploads an image, the server automatically generates appropriate text based on the image content. This includes a process that uses an image analysis algorithm to generate text that corresponds to the image content.

[0825] Utilizing machine learning models

[0826] The server uses a machine learning model to analyze the collected message history and emotional data. The model learns from the user's message history and emotional state and uses the information to generate future messages.

[0827] Specific examples

[0828] For example, suppose a user previously made a positive statement such as, "Today was such a fun day! I discovered a new place!" If the system recognizes the user's current emotional state as positive, it will generate a new statement such as, "The weather was great today, and I went to a new cafe. It was so relaxing!" Conversely, if the emotion engine detects a negative emotion, it will generate a statement such as, "Today was a stressful day, but I'm home and relaxing."

[0829] This invention allows users to efficiently post regular messages on SNS every day, and enables them to post more naturally, reflecting their own emotional state. Also, if there is a message that users really want to convey, they can post it manually each time, which helps to reduce SNS fatigue and maintain and increase the number of followers.

[0830] The processing flow will be explained below.

[0831] Step 1:

[0832] A user enters their social networking account authentication information (API key, API secret, access token, access token secret) into the system, which allows the system to access the specific social networking account.

[0833] Step 2:

[0834] The server accesses the SNS account using the authentication information provided by the user and collects the past call history. To do this, it uses the SNS API to retrieve the most recent call history. The past call history covers 200 calls.

[0835] Step 3:

[0836] The server analyzes the collected call history using natural language processing technology to extract features such as the tone, theme, vocabulary, and sentence structure of the text.

[0837] Step 4:

[0838] The server uses an emotion engine to analyze the user's emotional state from the collected call history, which involves identifying emotional categories such as positive, negative, and neutral.

[0839] Step 5:

[0840] The server extracts the user's communication style based on the analyzed data. This extraction process identifies distinctive patterns and styles from communication history and distinguishes styles according to emotional states.

[0841] Step 6:

[0842] The server generates new messages based on the extracted user's communication style and current emotional state. Here, generative AI (e.g., GPT-2 model) is used to create natural-sounding sentences that reflect the user's emotional state.

[0843] Step 7:

[0844] The generated message is displayed to the user through the terminal. The terminal provides an interface for the user to review the new message and modify it if necessary. This interface includes a function that displays the generated message in a text box and allows the user to edit its content.

[0845] Step 8:

[0846] The user can check and modify the generated message. Once the user has completed the modifications and is ready to send the message, the user presses the confirmation button to finalize the message.

[0847] Step 9:

[0848] The server posts the final message that the user has confirmed and corrected to the SNS platform. This posting is also done via API, and the message is published to the SNS from the user's account.

[0849] Step 10:

[0850] When a user uploads image information, the server automatically generates appropriate copy based on the image content, including generating text based on the image content using image analysis algorithms.

[0851] Step 11:

[0852] Messages with images are also displayed to the user via the device, and the user is given the opportunity to review and edit them. Once the user has reviewed and edited the message, it is finally posted to the SNS.

[0853] These are the specific processing steps of the invention that combines the emotion engine. This process allows users to efficiently post natural messages that correspond to their emotional state. The combination of the emotion engine and generative AI enables users to post messages that are more in line with their own personalities, reducing social media fatigue and enabling users to maintain and expand their following.

[0854] Example 2

[0855] 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."

[0856] On existing social media platforms, users spend a lot of time and effort on daily posting. Furthermore, the content of posts tends to be monotonous, often resulting in low engagement with followers. Furthermore, it is difficult to post natural, rich content that reflects the user's emotional state. In such situations, users may become exhausted by social media activity and ultimately abandon or delete their accounts. There is a need for a system that can solve these issues and enable users to post efficiently and appropriately.

[0857] 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.

[0858] In this invention, the server includes means for a user to input authentication information for the SNS service, means for collecting past message history, means for analyzing the collected message history and extracting the user's message style and emotional state, means using a generative AI model to generate new messages based on the extracted message style and emotional state, means for presenting the generated messages to the user and allowing them to confirm and revise, means for finally posting the confirmed and revised messages to the SNS service, means for analyzing collected image information and generating appropriate drafts based on the contents, and means for using a machine learning model to learn the collected message history and emotional state. This enables users to efficiently post to SNS every day and to post natural and diverse messages that reflect their own emotions.

[0859] "User" refers to an individual or organization that uses the SNS service.

[0860] "Authentication Information" refers to information such as API keys, API secrets, access tokens, and access token secrets required to access SNS services.

[0861] "SNS services" refers to social networking services, such as Twitter and Facebook.

[0862] "Server" refers to a computer system on a network that accesses SNS services and collects, analyzes, generates, and posts data.

[0863] "Outgoing message history" refers to a record of posts and comments a user has made on social media in the past.

[0864] "Analysis" refers to data processing performed to extract specific characteristics from the collected call history.

[0865] "Communication style" refers to the patterns of language, tone, and themes used by users on social media.

[0866] "Emotional state" refers to the user's emotional state, such as positive, negative, or neutral, extracted from the call history.

[0867] A "generative AI model" refers to an artificial intelligence model that uses machine learning techniques to generate new text.

[0868] A "prompt" refers to the input data or instructions that a generative AI model uses to generate new text.

[0869] "Confirmation and correction" refers to the act of the user confirming the generated message and correcting the content as necessary.

[0870] "Posting" refers to the act of publishing a message that has been finally confirmed and corrected from a user's SNS account.

[0871] "Image information" refers to the image data used by users when posting to SNS.

[0872] "Draft" refers to the text content generated based on image information.

[0873] A "machine learning model" refers to an algorithm or system used to learn from data and accomplish a specific task.

[0874] The present invention relates to a system that improves the efficiency of users' SNS activities and enables natural posting that reflects their emotions. Specific embodiments of the present invention will be described below.

[0875] First, the user enters the authentication information (API key, API secret, access token, access token secret) of the SNS service into the system, which allows the system to access the specific SNS account.

[0876] Next, the server uses the authentication information to access the API of the SNS service and retrieve the user's past call history. Specifically, the server is configured to collect the most recent 200 call histories. The collected data is stored in a database.

[0877] The server performs text analysis based on the collected call history. This analysis uses natural language processing libraries (e.g., NLTK and spaCy) and emotion engines (e.g., Azure Sentiment Analysis). This extracts features such as tone, theme, wording, and sentence structure from the call history, and labels the emotional state (e.g., positive, negative, neutral, etc.).

[0878] Based on the analysis results, the server extracts the user's communication style. Here, it distinguishes between messages sent by the user in a positive emotional state and messages sent by the user in a negative emotional state. For example, if a user's communication style is dominated by "fun" and "joy," it is classified as a positive style.

[0879] The server then uses a generative AI model (e.g., OpenAI's GPT-3) to generate a new message. The generation prompt is populated with the user's current emotional state and topic based on the analysis results. For example, a prompt might be, "Generate a new message based on past positive messages." The generated message is then sent from the server to the device and presented to the user.

[0880] The terminal displays the generated message received from the server to the user, allowing the user to confirm and edit the message using the text box. When the user enters the edits, the information is sent back to the server.

[0881] The final message that has been confirmed and corrected by the user is posted to the SNS from the user's account via the SNS service's API by the server.

[0882] The system also supports users uploading images. The server uses an image analysis algorithm (e.g., Google Cloud Vision) to analyze the image content and generate appropriate text based on the content. For example, if a "landscape image" is recognized, the system generates the text "I found a beautiful landscape!"

[0883] Furthermore, the server uses a machine learning model (e.g., Scikit-learn) to learn from the collected message history and emotional state. This model learns from the user's message history and emotional data and uses it to generate future messages.

[0884] This will enable users to post on social media every day efficiently and enable natural and diverse communication that appropriately reflects their emotions, which is expected to improve engagement and reduce social media fatigue.

[0885] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0886] Program processing flow

[0887] Step 1:

[0888] The user enters the authentication information (API key, API secret, access token, access token secret) for the social networking service into a web form and clicks a button to submit it.

[0889] The data entered is authentication information, and based on this information, information that allows access to the user's SNS account is obtained.

[0890] Step 2:

[0891] The server receives the authentication information sent by the user and accesses the API of the SNS service to obtain the past call history.

[0892] Specifically, the server sends an API request to the SNS endpoint and saves the most recent 200 outgoing calls in a database.

[0893] The input is the SNS authentication information and request, and the output is the call history data.

[0894] Step 3:

[0895] The server analyzes the collected call history.

[0896] A natural language processing library (e.g., NLTK or spaCy) is used to analyze the tone, theme, wording, sentence structure, etc. of the text data. In addition, a sentiment engine (e.g., Azure Sentiment Analysis) is used to label the sentiment state (positive, negative, neutral) of the call history.

[0897] The input is the call history data, and the output is the analysis results (tone, theme, wording, sentence structure, emotional state).

[0898] Step 4:

[0899] The server extracts the user's communication style based on the analysis results.

[0900] Using the analysis results, we distinguish between messages sent in a positive emotional state and those sent in a negative emotional state, and clarify the user's communication style. For example, sentences containing many words like "fun" and "joy" are classified as a positive style.

[0901] The input is the analysis result, and the output is the transmission style.

[0902] Step 5:

[0903] The server generates a new message based on the extracted message style.

[0904] A prompt (e.g., "Generate a new message based on past positive messages") is input into a generative AI model (e.g., OpenAI's GPT-3) and the generated text is received.

[0905] The input is the communication style and the prompt sentence, and the output is the generated communication sentence.

[0906] Step 6:

[0907] The terminal displays the generated message to the user.

[0908] The generated message is displayed in a text box, providing an interface that allows the user to check and edit it.

[0909] The input is the generated message, and the output is the user's confirmation and correction.

[0910] Step 7:

[0911] The user checks the generated message and enters corrections as necessary.

[0912] Check the message displayed in the text box and correct any inappropriate parts. The corrections will be sent to the server again.

[0913] The input is the generated message and the user's suggestions for revision, and the output is the revised message.

[0914] Step 8:

[0915] The server posts the final, corrected message to the SNS service.

[0916] The revised message will be published from the user's account via the SNS API.

[0917] The input is the modified message and the output is a post on a social networking site.

[0918] Step 9:

[0919] The server analyzes the image information uploaded by the user and generates appropriate text.

[0920] It uses image analysis algorithms (e.g., Google Cloud Vision) to analyze the image content and generate text based on that content. For example, if it recognizes an image as a landscape, it will generate the sentence "I found a beautiful landscape!"

[0921] The input is image information, and the output is a generated draft based on the image.

[0922] Step 10:

[0923] The server uses a machine learning model to learn from the collected call history and emotional state.

[0924] The collected message history and emotion labels are used to train a Scikit-learn model, which will provide learning results that will be useful for generating future messages.

[0925] The input is the call history and emotion labels, and the output is a trained machine learning model.

[0926] (Application example 2)

[0927] 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."

[0928] In modern society, many users use social media to share information, but the content of their posts is subjective and inconsistent, making it difficult to reflect their own emotional state or style. It is also difficult to automatically and efficiently create effective ad copy that matches a user's emotions and style when generating ads. To solve these problems, a system is needed that analyzes a user's past posting history and generates new posts and ad copy that reflect that user's style and emotional state.

[0929] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0930] In this invention, the server includes means for collecting past call history, means for analyzing the collected call history and extracting the user's call style, and means for generating new messages based on the extracted call style. This makes it possible to automatically generate messages and advertising copy that reflect the user's past call history and emotional state, present them to the user, and post them to SNS or advertising media after final confirmation.

[0931] "Past communication history" refers to all of the content posted or transmitted by a user in the past on social media platforms and other platforms.

[0932] "Means of collection" refers to the technical processes and functions that the system uses to obtain past communication history using the user's social media account authentication information.

[0933] "Means of analyzing and extracting a user's communication style" refers to a means of analyzing collected communication history using natural language processing and machine learning technology to identify characteristics of the user's writing, such as tone, theme, and wording.

[0934] "Means for generating new messages" refers to technology for creating new messages using AI models, etc., based on the extracted user's communication style and emotional state.

[0935] "Means for presenting the generated message to the user and allowing them to review and correct it" refers to an interface or process that displays the generated message to the user, allows the user to review its contents, and makes corrections as necessary.

[0936] "Means for posting the final verified and corrected message on a social networking site" refers to the technical process by which the final message that has been verified and corrected by the user is automatically posted on a platform such as a social networking site.

[0937] "Means for generating advertising copy and advertising visuals based on the extracted communication style and emotional state" refers to technology for automatically creating advertising copy and visuals for commercial purposes, taking into account the user's communication style and emotional state.

[0938] "Means for presenting the generated ad copy to the user and allowing the user to modify and finalize it" refers to an interface or function that displays the generated ad copy to the user, allowing the user to review it and modify it as necessary.

[0939] "Means for posting to designated media" refers to the technical process for automatically distributing and posting the final, verified and corrected advertising copy to the designated advertising media or platform.

[0940] The following is a specific description of an embodiment of the present invention. The entire system is composed of a server, a terminal, and a user. The programs that are part of this system are realized as follows.

[0941] Data collection methods

[0942] The server uses the authentication information for the SNS account provided by the user to collect past call history via the SNS API. Specifically, it accesses the SNS account using authentication information such as the API key, API secret, access token, and access token secret, and retrieves the past call history (e.g., 200 calls). At this stage, data is collected from the SNS API and stored on the server.

[0943] Analysis of communication style and emotional state

[0944] The collected call history is analyzed within the server. Natural language processing technology is used to extract features such as tone, theme, and wording of the call. An emotion engine is also used to detect the user's emotional state (e.g., positive, negative, neutral) from the call history. This analysis uses machine learning models (e.g., BERT) and sentiment analysis models.

[0945] Generate new messaging and ad copy

[0946] The server then uses a generative AI model (e.g., GPT-2) to generate new messages based on the extracted user's communication style and emotional state. The generation prompts are populated with topics and tones based on the user's emotional state, and new messages are created. Similarly, ad copy and visuals are automatically generated to match the user's communication style and emotional state.

[0947] For example, the following prompt might be used:

[0948] Prompt: "Create an ad for a new cafe with a positive tone that matches the user's communication style."

[0949] Feedback and Corrections

[0950] The generated message and advertisement copy are presented to the user via the device. The user can review these copies and make corrections as necessary. This procedure is an important step to ensure that the automatically generated copies match the user's expectations and intentions. An interface is provided that allows the user to provide feedback and make corrections.

[0951] Final Post

[0952] The final message and advertisement copy that has been checked and corrected is posted by the server to the social media platform or designated advertising medium. This posting is also done via API and is published from the user's account.

[0953] In this way, the present invention can efficiently generate consistent messages and advertisements that reflect the user's past message history and emotional state, and finally publish them after confirmation and correction.

[0954] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0955] Step 1: Data collection

[0956] The server receives the authentication information (API key, API secret, access token, access token secret) of the user's SNS account as input. Using this authentication information, it accesses the SNS API and retrieves the user's past call history. At this time, the most recent 200 call history entries are collected and stored on the server. The output is the collected call history data.

[0957] Step 2: Analyze communication style and emotional state

[0958] The server analyzes the collected call history data as input. First, it uses natural language processing technology to extract features such as tone, theme, and wording from the call history. At the same time, it uses an emotion engine to detect the emotional state (positive, negative, neutral) of each message. This identifies the user's call style and emotional state. The output is the analyzed call style and emotional state data.

[0959] Step 3: Generate a new message

[0960] The server generates a new message using the analyzed message style and emotional state data. It uses a generative AI model (GPT-2) to generate the message. The following prompt is input:

[0961] Prompt: "Create an ad for a new cafe with a positive tone that matches the user's communication style."

[0962] Based on this prompt, a new message is generated. The output is the generated message.

[0963] Step 4: Generate ad copy and visuals

[0964] The server generates ad copy and visuals using the analyzed customer communication style and emotional state data as input. Using a generative AI model, ad copy that matches the user's emotional state is created based on the prompt text. The output is the generated ad copy and visuals.

[0965] Step 5: Feedback and revisions

[0966] The terminal presents the generated message and advertisement copy to the user. The user checks the contents and makes any necessary corrections. At this time, the terminal sends the user's corrections to the server, which reflects the suggested corrections. The input is the generated message and advertisement copy, and the output is the final version that has been checked and corrected by the user.

[0967] Step 6: Final submission

[0968] The server receives as input the final message and advertisement copy that has been reviewed and revised by the user. It posts this to the social media platform or designated advertising medium. The posting is again done using the social media API, and is made public through the user's account. The output is the publication of the final message and advertisement copy.

[0969] 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.

[0970] 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.

[0971] 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.

[0972] [Fourth embodiment]

[0973] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0974] 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.

[0975] 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).

[0976] 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.

[0977] 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.

[0978] 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).

[0979] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0980] 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.

[0981] 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.

[0982] 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.

[0983] 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.

[0984] 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.

[0985] 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."

[0986] The embodiments of the present invention will be specifically described below.

[0987] Data collection

[0988] The server collects past call history using authentication information for the SNS account provided by the user. This authentication information includes an API key and a secret token, and uses this to obtain the latest call history from the SNS platform.

[0989] Style Analysis

[0990] The collected call history is analyzed by the server. This analysis includes the tone, theme, wording, sentence structure, etc. of the text. Specifically, natural language processing technology and machine learning algorithms are used to extract features from the call history.

[0991] Creating a new message

[0992] The server generates new messages based on the analysis results. At this stage, generative AI (e.g., GPT-2 model) is used to create messages that match the user's style. The generated messages match the user's past communication style, allowing them to communicate in a way that is "true to their personality."

[0993] Check and correct the message

[0994] The generated message is presented to the user via the terminal. The user can review the message and make corrections as necessary. This process prevents accidental and incorrect messages from being sent, while significantly reducing the amount of manual input required by the user.

[0995] Final call

[0996] Once the user has confirmed and corrected the message, it is finally posted to the SNS platform by the server, which also posts the message via API.

[0997] Call response with images

[0998] This system also supports messages that include image information. When a user uploads an image, it automatically generates text that matches the image content. For example, when uploading a landscape photo, text that matches the atmosphere of the photo is generated. This makes it easy to send messages that include visual elements.

[0999] Utilizing machine learning models

[1000] The server uses a machine learning model to analyze the collected call history, learning the topics, tone, and word choice of the user from their call history and incorporating these characteristics into the generation of new messages.

[1001] Specific examples

[1002] For example, if a user has previously sent a message such as "It's a nice day today. I want to go for a walk!", the system will recognize that themes such as "weather" and "walks" are part of the user's style. Next, when generating a new message, the system will automatically generate a message such as "It's a nice, sunny day today. I'm looking forward to going outside!"

[1003] This invention allows users to efficiently send out regular messages every day. Furthermore, if there is a message that users really want to convey, they can send it manually each time, which helps to reduce SNS fatigue and maintain and increase the number of followers.

[1004] The processing flow will be explained below.

[1005] Step 1:

[1006] A user enters their social networking account authentication information (API key, API secret, access token, access token secret) into the system, which allows the system to access the specific social networking account.

[1007] Step 2:

[1008] The server accesses the SNS account using the authentication information provided by the user and collects past call history. Specifically, it uses the SNS API to obtain the latest call history. At this time, it is set to collect a certain amount of past call history (for example, 200 items).

[1009] Step 3:

[1010] The server analyzes the collected call history. The analysis uses natural language processing and machine learning techniques, applying algorithms that identify frequently occurring words and topics in texts to extract features such as tone, theme, wording, and sentence structure.

[1011] Step 4:

[1012] The server extracts the user's communication style. This extraction step identifies specific patterns and styles from past communication history. For example, frequently used phrases and specific tones (friendly, professional, etc.) are analyzed.

[1013] Step 5:

[1014] The server uses a generative AI to generate new sentences based on the extracted user's speech style. The generative AI (e.g., a GPT-2 model) creates natural sentences that reflect the user's style. In this generation step, a specific prompt is input and a sentence is generated accordingly.

[1015] Step 6:

[1016] The generated message is displayed to the user through the terminal. The terminal provides an interface for the user to review the new message and modify it if necessary. This interface includes a function that displays the generated message in a text box and allows the user to edit its content.

[1017] Step 7:

[1018] The user checks the generated message and makes any corrections. When the user has completed the corrections and is ready to send the message, they press the Confirm button. By pressing this button, the final message is confirmed.

[1019] Step 8:

[1020] The server receives the final message that the user has confirmed and corrected, and posts it to the SNS platform. This posting is also done via API, and the final message is published to the SNS from the user's account.

[1021] Step 9:

[1022] When a user uploads image information, the server automatically generates text that matches the image content. For this purpose, an image analysis algorithm is used to generate relevant text based on the image content.

[1023] Step 10:

[1024] Similarly, text with images is presented to the user via the device, and they are given the opportunity to check and correct it. After going through a series of checking and correcting processes, the final message with images is posted to the SNS.

[1025] These are the specific steps for automatically generating new messages based on past message history, and then posting them to SNS after the user has confirmed and edited them. This process allows users to efficiently send regular daily messages and saves time and effort in maintaining and expanding their followers.

[1026] Example 1

[1027] 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."

[1028] Modern social media users often feel burdened by frequent posting. Users' posts can also be inconsistent, making it difficult to keep their followers engaged. Furthermore, as posts increasingly include visual elements, quickly generating appropriate copy for images can be time-consuming. A new system is needed to solve these problems.

[1029] 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.

[1030] In this invention, the server includes means for collecting past call history using authentication information provided by the user, means for analyzing the collected call history and extracting the user's call style, and means for generating new messages using a generative AI model based on the extracted call style. This enables users to efficiently send messages with consistency. The present invention also includes means for presenting the generated messages to the user via a terminal for confirmation and correction, and means for posting the final confirmed and corrected messages to an SNS via an API. This enables users to send messages quickly and accurately. Furthermore, the present invention includes means for including image information in the generated messages and automatically generating drafts based on the image content, making it easy to send messages that include visual elements.

[1031] "Authentication information" refers to information such as an API key or secret token that a user needs to access their SNS account.

[1032] "Outgoing communication history" refers to a record of all posts and comments that a user has made on a social networking platform in the past.

[1033] A "server" is a computer system that processes, stores, and controls access to data over a network.

[1034] "Communication style" refers to characteristics such as tone, theme, wording, and sentence structure of text extracted from a user's past communication history.

[1035] A "generative AI model" is an artificial intelligence model that uses natural language processing and machine learning algorithms to generate new text or information from given data.

[1036] A "terminal" is a device such as a computer or smartphone that is directly operated by a user.

[1037] "API" stands for Application Programming Interface, and refers to an interface for exchanging data and functions between different software programs.

[1038] "Image information" refers to image data used when a user posts to an SNS.

[1039] A "machine learning model" is a set of algorithms that learn from data and perform tasks such as prediction, classification, and generation.

[1040] The embodiments of the present invention will be specifically described below.

[1041] Data collection and analysis

[1042] 1. The server collects authentication information for the SNS account provided by the user. This authentication information includes an API key and secret token, and uses this information to retrieve the user's past call history from the SNS platform. If authentication is successful, the server calls the SNS API, collects the past call history, and stores it in a database.

[1043] 2. The server analyzes the collected communication history and extracts the user's communication style. This analysis uses natural language processing techniques (e.g., NLTK, spaCy) and machine learning algorithms (e.g., BERT, GPT-2). The server tokenizes each piece of text data and performs sentiment analysis, topic modeling, and contextual analysis. As a result, the user's communication style is generated as a digital profile.

[1044] Creating a new message

[1045] 3. The server generates a new message based on the analysis results. Here, a generative AI model (e.g., GPT-3) is used. The server inputs the analysis data as a prompt into the generative AI model to generate a new sentence.

[1046] Check and correct the message

[1047] 4. The generated message is presented to the user via the device. The user can review the message and make corrections as necessary. These corrections are reflected in real time on the device screen, preventing errors and significantly reducing the amount of manual input required by the user.

[1048] Final call

[1049] 5. After the user has confirmed and corrected the message, it is finally posted to the SNS platform by the server. This posting is also automated using APIs. When the user presses the "Post" button on their device, the server sends the corrected message to the SNS API and executes the posting.

[1050] Support for sending messages with images

[1051] 6. When a user uploads an image, the server analyzes the image and automatically generates a sentence that corresponds to the content. For example, when a user uploads a landscape photo from their device, the server recognizes and analyzes the image and generates a sentence that corresponds to the atmosphere of the landscape photo.

[1052] Specific examples

[1053] For example, if a user has previously sent a message such as "It's a nice day today. I want to go for a walk!", the system will recognize that themes such as "weather" and "walks" are part of the user's communication style. When generating a new message, the system will automatically generate a sentence such as "It's a nice, sunny day today. I'm looking forward to going outside!"

[1054] Prompt Sentence Examples

[1055] Example prompt: "Generate new posts based on the user's past posting style. The post might look something like, 'It's a nice day today. I want to go for a walk!' Generate new posts based on this."

[1056] This invention allows users to efficiently post on SNS every day. It also contributes to maintaining and expanding the number of followers because users continue to post consistently. It also makes it easy to post content that includes visual elements, reducing SNS fatigue and enabling effective communication.

[1057] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1058] Step 1:

[1059] A user uses a device to enter authentication information for their social networking account. This authentication information includes an API key and a secret token. The entered authentication information is sent to the server. The server receives and securely stores the authentication information. The input of this process is the authentication information entered by the user, and the output is the securely stored authentication information.

[1060] Step 2:

[1061] The server uses the stored authentication information to call the API of the social media platform and collect the user's past call history. During this process, real-time data is obtained via the API and stored in a database. The input is the authentication information and the call history data obtained as a result of the API call, and the output is the call history stored in the database.

[1062] How it works: The server connects to the social networking platform and fetches the user's past posts. This data is stored in a database in chronological order.

[1063] Step 3:

[1064] The server analyzes the collected call history. This analysis uses natural language processing technology and machine learning algorithms to extract the user's call style. Specifically, the tone, theme, vocabulary, and sentence structure of the text are analyzed. The input is the call history data, and the output is a digital profile that shows the user's call style.

[1065] Specific operation: The server uses analysis tools (e.g., NLTK, spaCy) to analyze the text data and saves the results in a database as a digital profile.

[1066] Step 4:

[1067] The server uses a generative AI model based on the analysis results to generate a new message. At this time, the analysis data is input to the generative AI model as a prompt. The input is a digital profile that indicates the user's communication style and a prompt, and the output is a new message.

[1068] Specific operation: The server inputs the prompt sentence into a generative AI model (e.g., GPT-3) to generate a new message, which is then stored in a database.

[1069] Step 5:

[1070] The generated message is presented to the user via the terminal. The user can check this message and modify it as necessary. The input is the generated message, and the output is the message that has been checked and modified by the user.

[1071] Specific operation: The generated message is displayed on the device screen, and the user can edit the text using the keyboard or touch screen.

[1072] Step 6:

[1073] The message that the user has confirmed and modified is finally posted to the SNS platform by the server. This posting is also automated using APIs. The input is the message that the user modified, and the output is a new post on the SNS platform.

[1074] Specific operation: When the user presses the "Post" button, the server sends the revised message to the SNS API and executes the post.

[1075] Step 7:

[1076] The server analyzes the image uploaded by the user and generates a sentence based on its content. After analyzing the content of the image using image recognition technology, the server generates a sentence using a generative AI model. The input is the image data uploaded by the user, and the output is a sentence that matches the image.

[1077] Specific operation: When a user uploads an image from their device, the server recognizes the image and generates an appropriate draft based on the analysis results. The draft is then presented to the user, who can review and edit it.

[1078] (Application example 1)

[1079] 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."

[1080] Previous message generation systems on social media platforms were limited to generating new messages based on a user's past message styles, and were unable to efficiently generate and post ad copy. This meant that advertising agencies and marketing teams had to spend a great deal of time and effort creating ad copy that matched their clients' styles.

[1081] 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.

[1082] In this invention, the server includes means for collecting past communication history, means for analyzing the collected communication history and extracting the user's communication style, means for generating new messages based on the extracted communication style, means for presenting the generated messages to the user for confirmation and revision, means for posting the confirmed and revised messages to an SNS, means for automatically generating advertising copy, means for presenting the generated advertising copy to the user for confirmation and revision, and means for posting the revised advertising copy to a distribution platform. This enables efficient generation and posting of advertising copy based on the user's past communication style.

[1083] "Outgoing message history" refers to a record of text, images, etc. that a user has posted on a social media platform in the past.

[1084] "Communication style" refers to the combined characteristics of the tone, theme, language, and writing style of the content a user posts.

[1085] "Copyright" means text created to promote a particular product or service.

[1086] "Means of collection" refers to the methods and systems for obtaining a user's past posting history from a social media platform.

[1087] "Means of analysis" refers to a method of analyzing collected communication history and extracting the user's communication style from it.

[1088] The "means of generation" refers to a method for creating new messages or advertising copy based on the extracted communication style.

[1089] "Means for confirmation and correction" refers to a method in which the generated message or advertising copy is presented to the user, allowing the user to check the content themselves and make corrections as necessary.

[1090] "Method of posting" refers to the method of posting the final confirmed and revised message or advertising copy on social media or distribution platforms.

[1091] A "machine learning model" is an algorithm or system that learns patterns and features from data and makes predictions and classifications for new data.

[1092] The system for implementing this invention is made up of a server, a terminal, and a user. Each component and its function will be specifically described below.

[1093] server

[1094] The server has the function of collecting past posting history. It retrieves past posting history from the social media platform using the social media account authentication information (API key or secret token) provided by the user. The collected posting history is analyzed using natural language processing technology and machine learning algorithms. Through the analysis, features such as the tone, theme, wording, and sentence structure of the text are extracted. The server then generates new postings based on the extracted style. This generation uses a generative AI model such as OpenAI's GPT-3 model.

[1095] Terminal

[1096] The device provides a user interface, allowing the user to review and edit the generated message or ad copy. The new message or ad copy generated by the server is sent to the device and presented to the user. The user reviews the presented copy and makes any necessary edits. Once the user has reviewed and edited the message or ad copy, it is sent back to the server.

[1097] User

[1098] Users provide their social media account credentials and authorize the collection of past messaging history. They are then asked to review the generated messaging and ad copy and make any necessary revisions. Users then post the final messaging and ad copy to the social media platform.

[1099] Hardware and software used

[1100] Hardware: Smartphones, servers

[1101] Software: requests library, TextBlob library, OpenAI library, SNS platform API

[1102] Specific examples

[1103] For example, an advertising agency might generate new ad copy based on a particular client's social media accounts. If the client's past posts tend to be positive and emotional, the resulting ad copy will also have a similarly positive tone. An example of a prompt might look like this:

[1104] Generate ad copy based on your client's social media posting style. Consider the following style characteristics:

[1105] Polarity: 0.8, Subjectivity: 0.6

[1106] This invention allows advertising agencies and marketing teams to efficiently generate ad copy that matches their client's style, then review and revise it before posting it on the target platform, improving the efficiency and effectiveness of ad copy creation and enhancing the quality of marketing activities.

[1107] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1108] Step 1:

[1109] The server collects past call history using the authentication information of the SNS account provided by the user. The input is the user's authentication information, and the output is the call history obtained from the SNS platform. In this process, the API of the SNS platform is used to obtain the latest call history.

[1110] Step 2:

[1111] The server analyzes the collected communication history and extracts the user's communication style. The input is the collected communication history, and the output is the characteristics of the user's communication style (tone, theme, wording, sentence structure). This analysis uses natural language processing technology and the TextBlob library to perform sentiment analysis of the text and word frequency analysis.

[1112] Step 3:

[1113] The server generates a new message or ad copy based on the extracted style. The input is the user's message style characteristics, and the output is the generated new message or ad copy. The copy is generated using a generative AI model such as OpenAI's GPT-3 model. The prompt sentence is provided as input to the model.

[1114] Step 4:

[1115] The generated message or copy draft is sent from the server to the terminal and presented to the user. The input is the generated message or copy draft, and the output is a screen display presented to the user for confirmation and correction, provided via the terminal's user interface.

[1116] Step 5:

[1117] The user checks the message or advertising copy presented to them and makes corrections as necessary. The input is the message or advertising copy presented to the user, and the output is the copy that has been checked and corrected by the user. The user makes corrections through their terminal and sends the corrections to the server.

[1118] Step 6:

[1119] The server finally posts the message or advertising copy that has been confirmed and revised by the user to the SNS platform. The input is the confirmed and revised copy, and the output is the final copy posted to the SNS platform. In this process, the posting is made via the SNS platform's API.

[1120] Step 7:

[1121] When image information is included in the generated message, the server provides a means to automatically generate a message that matches the image content. The input is an image uploaded by the user and a prompt message based on its content, and the output is a new message that matches the image. A generative AI model is used to generate a message that matches the image.

[1122] This process allows users to efficiently create new messages and copy drafts, review and revise them, and finally post them to social media platforms, allowing advertising agencies and marketing teams to generate effective copy drafts that fit their clients' communication styles.

[1123] 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.

[1124] The embodiments of the present invention will be specifically described below.

[1125] Data collection

[1126] A user enters the authentication information of a social networking account (API key, API secret, access token, access token secret) into the system, which allows the system to access the specific social networking account.

[1127] Collection of past call history

[1128] The server accesses the SNS account using the authentication information provided by the user and collects past call history. This collection is done by using the SNS API to obtain the latest call history. The server is set to collect a certain amount of past call history (for example, 200 items).

[1129] Style and Sentiment Analysis

[1130] The collected call history is analyzed by the server. First, natural language processing of the text data is used to extract features such as tone, theme, wording, and sentence structure. At the same time, an emotion engine is used to detect the user's emotional state from the call history. The emotion engine identifies emotional categories such as positive, negative, and neutral, and labels the data accordingly.

[1131] Extracting the delivery style

[1132] The server extracts the user's communication style based on the analysis results. At this stage, the analysis also includes the relationship between the user's communication style and their emotional state. For example, it distinguishes between messages sent when the user was in a positive emotional state and messages sent when the user was in a negative emotional state.

[1133] Creating a new message

[1134] The server generates new messages based on the extracted user's communication style. In this process, a generative AI (e.g., GPT-2 model) is used to create messages that reflect the user's emotional state. The generation prompts include topics and tones based on the user's current emotional state.

[1135] Emotion recognition and correction of messages

[1136] The generated message is displayed to the user through the device. Based on the user's feedback and corrections, the generated message is checked for appropriate sentiment. The interface includes a function where the generated message is displayed in a text box and the user can enter corrections.

[1137] Final call

[1138] The final message, after being confirmed and corrected by the user, is posted to the SNS platform by the server. This posting is also done via API, so the final message is published to the SNS from the user's account.

[1139] Image information support

[1140] The system also supports messages containing image information. When a user uploads an image, the server automatically generates appropriate text based on the image content. This includes a process that uses an image analysis algorithm to generate text that corresponds to the image content.

[1141] Utilizing machine learning models

[1142] The server uses a machine learning model to analyze the collected message history and emotional data. The model learns from the user's message history and emotional state and uses the information to generate future messages.

[1143] Specific examples

[1144] For example, suppose a user previously made a positive statement such as, "Today was such a fun day! I discovered a new place!" If the system recognizes the user's current emotional state as positive, it will generate a new statement such as, "The weather was great today, and I went to a new cafe. It was so relaxing!" Conversely, if the emotion engine detects a negative emotion, it will generate a statement such as, "Today was a stressful day, but I'm home and relaxing."

[1145] This invention allows users to efficiently post regular messages on SNS every day, and enables them to post more naturally, reflecting their own emotional state. Also, if there is a message that users really want to convey, they can post it manually each time, which helps to reduce SNS fatigue and maintain and increase the number of followers.

[1146] The processing flow will be explained below.

[1147] Step 1:

[1148] A user enters their social networking account authentication information (API key, API secret, access token, access token secret) into the system, which allows the system to access the specific social networking account.

[1149] Step 2:

[1150] The server accesses the SNS account using the authentication information provided by the user and collects the past call history. To do this, it uses the SNS API to retrieve the most recent call history. The past call history covers 200 calls.

[1151] Step 3:

[1152] The server analyzes the collected call history using natural language processing technology to extract features such as the tone, theme, vocabulary, and sentence structure of the text.

[1153] Step 4:

[1154] The server uses an emotion engine to analyze the user's emotional state from the collected call history, which involves identifying emotional categories such as positive, negative, and neutral.

[1155] Step 5:

[1156] The server extracts the user's communication style based on the analyzed data. This extraction process identifies distinctive patterns and styles from communication history and distinguishes styles according to emotional states.

[1157] Step 6:

[1158] The server generates new messages based on the extracted user's communication style and current emotional state. Here, generative AI (e.g., GPT-2 model) is used to create natural-sounding sentences that reflect the user's emotional state.

[1159] Step 7:

[1160] The generated message is displayed to the user through the terminal. The terminal provides an interface for the user to review the new message and modify it if necessary. This interface includes a function that displays the generated message in a text box and allows the user to edit its content.

[1161] Step 8:

[1162] The user can check and modify the generated message. Once the user has completed the modifications and is ready to send the message, the user presses the confirmation button to finalize the message.

[1163] Step 9:

[1164] The server posts the final message that the user has confirmed and corrected to the SNS platform. This posting is also done via API, and the message is published to the SNS from the user's account.

[1165] Step 10:

[1166] When a user uploads image information, the server automatically generates appropriate copy based on the image content, including generating text based on the image content using image analysis algorithms.

[1167] Step 11:

[1168] Messages with images are also displayed to the user via the device, and the user is given the opportunity to review and edit them. Once the user has reviewed and edited the message, it is finally posted to the SNS.

[1169] These are the specific processing steps of the invention that combines the emotion engine. This process allows users to efficiently post natural messages that correspond to their emotional state. The combination of the emotion engine and generative AI enables users to post messages that are more in line with their own personalities, reducing social media fatigue and enabling users to maintain and expand their following.

[1170] Example 2

[1171] 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."

[1172] On existing social media platforms, users spend a lot of time and effort on daily posting. Furthermore, the content of posts tends to be monotonous, often resulting in low engagement with followers. Furthermore, it is difficult to post natural, rich content that reflects the user's emotional state. In such situations, users may become exhausted by social media activity and ultimately abandon or delete their accounts. There is a need for a system that can solve these issues and enable users to post efficiently and appropriately.

[1173] 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.

[1174] In this invention, the server includes means for a user to input authentication information for the SNS service, means for collecting past message history, means for analyzing the collected message history and extracting the user's message style and emotional state, means using a generative AI model to generate new messages based on the extracted message style and emotional state, means for presenting the generated messages to the user and allowing them to confirm and revise, means for finally posting the confirmed and revised messages to the SNS service, means for analyzing collected image information and generating appropriate drafts based on the contents, and means for using a machine learning model to learn the collected message history and emotional state. This enables users to efficiently post to SNS every day and to post natural and diverse messages that reflect their own emotions.

[1175] "User" refers to an individual or organization that uses the SNS service.

[1176] "Authentication Information" refers to information such as API keys, API secrets, access tokens, and access token secrets required to access SNS services.

[1177] "SNS services" refers to social networking services, such as Twitter and Facebook.

[1178] "Server" refers to a computer system on a network that accesses SNS services and collects, analyzes, generates, and posts data.

[1179] "Outgoing message history" refers to a record of posts and comments a user has made on social media in the past.

[1180] "Analysis" refers to data processing performed to extract specific characteristics from the collected call history.

[1181] "Communication style" refers to the patterns of language, tone, and themes used by users on social media.

[1182] "Emotional state" refers to the user's emotional state, such as positive, negative, or neutral, extracted from the call history.

[1183] A "generative AI model" refers to an artificial intelligence model that uses machine learning techniques to generate new text.

[1184] A "prompt" refers to the input data or instructions that a generative AI model uses to generate new text.

[1185] "Confirmation and correction" refers to the act of the user confirming the generated message and correcting the content as necessary.

[1186] "Posting" refers to the act of publishing a message that has been finally confirmed and corrected from a user's SNS account.

[1187] "Image information" refers to the image data used by users when posting to SNS.

[1188] "Draft" refers to the text content generated based on image information.

[1189] A "machine learning model" refers to an algorithm or system used to learn from data and accomplish a specific task.

[1190] The present invention relates to a system that improves the efficiency of users' SNS activities and enables natural posting that reflects their emotions. Specific embodiments of the present invention will be described below.

[1191] First, the user enters the authentication information (API key, API secret, access token, access token secret) of the SNS service into the system, which allows the system to access the specific SNS account.

[1192] Next, the server uses the authentication information to access the API of the SNS service and retrieve the user's past call history. Specifically, the server is configured to collect the most recent 200 call histories. The collected data is stored in a database.

[1193] The server performs text analysis based on the collected call history. This analysis uses natural language processing libraries (e.g., NLTK and spaCy) and emotion engines (e.g., Azure Sentiment Analysis). This extracts features such as tone, theme, wording, and sentence structure from the call history, and labels the emotional state (e.g., positive, negative, neutral, etc.).

[1194] Based on the analysis results, the server extracts the user's communication style. Here, it distinguishes between messages sent by the user in a positive emotional state and messages sent by the user in a negative emotional state. For example, if a user's communication style is dominated by "fun" and "joy," it is classified as a positive style.

[1195] The server then uses a generative AI model (e.g., OpenAI's GPT-3) to generate a new message. The generation prompt is populated with the user's current emotional state and topic based on the analysis results. For example, a prompt might be, "Generate a new message based on past positive messages." The generated message is then sent from the server to the device and presented to the user.

[1196] The terminal displays the generated message received from the server to the user, allowing the user to confirm and edit the message using the text box. When the user enters the edits, the information is sent back to the server.

[1197] The final message that has been confirmed and corrected by the user is posted to the SNS from the user's account via the SNS service's API by the server.

[1198] The system also supports users uploading images. The server uses an image analysis algorithm (e.g., Google Cloud Vision) to analyze the image content and generate appropriate text based on the content. For example, if a "landscape image" is recognized, the system generates the text "I found a beautiful landscape!"

[1199] Furthermore, the server uses a machine learning model (e.g., Scikit-learn) to learn from the collected message history and emotional state. This model learns from the user's message history and emotional data and uses it to generate future messages.

[1200] This will enable users to post on social media every day efficiently and enable natural and diverse communication that appropriately reflects their emotions, which is expected to improve engagement and reduce social media fatigue.

[1201] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1202] Program processing flow

[1203] Step 1:

[1204] The user enters the authentication information (API key, API secret, access token, access token secret) for the social networking service into a web form and clicks a button to submit it.

[1205] The data entered is authentication information, and based on this information, information that allows access to the user's SNS account is obtained.

[1206] Step 2:

[1207] The server receives the authentication information sent by the user and accesses the API of the SNS service to obtain the past call history.

[1208] Specifically, the server sends an API request to the SNS endpoint and saves the most recent 200 outgoing calls in a database.

[1209] The input is the SNS authentication information and request, and the output is the call history data.

[1210] Step 3:

[1211] The server analyzes the collected call history.

[1212] A natural language processing library (e.g., NLTK or spaCy) is used to analyze the tone, theme, wording, sentence structure, etc. of the text data. In addition, a sentiment engine (e.g., Azure Sentiment Analysis) is used to label the sentiment state (positive, negative, neutral) of the call history.

[1213] The input is the call history data, and the output is the analysis results (tone, theme, wording, sentence structure, emotional state).

[1214] Step 4:

[1215] The server extracts the user's communication style based on the analysis results.

[1216] Using the analysis results, we distinguish between messages sent in a positive emotional state and those sent in a negative emotional state, and clarify the user's communication style. For example, sentences containing many words like "fun" and "joy" are classified as a positive style.

[1217] The input is the analysis result, and the output is the transmission style.

[1218] Step 5:

[1219] The server generates a new message based on the extracted message style.

[1220] A prompt (e.g., "Generate a new message based on past positive messages") is input into a generative AI model (e.g., OpenAI's GPT-3) and the generated text is received.

[1221] The input is the communication style and the prompt sentence, and the output is the generated communication sentence.

[1222] Step 6:

[1223] The terminal displays the generated message to the user.

[1224] The generated message is displayed in a text box, providing an interface that allows the user to check and edit it.

[1225] The input is the generated message, and the output is the user's confirmation and correction.

[1226] Step 7:

[1227] The user checks the generated message and enters corrections as necessary.

[1228] Check the message displayed in the text box and correct any inappropriate parts. The corrections will be sent to the server again.

[1229] The input is the generated message and the user's suggestions for revision, and the output is the revised message.

[1230] Step 8:

[1231] The server posts the final, corrected message to the SNS service.

[1232] The revised message will be published from the user's account via the SNS API.

[1233] The input is the modified message and the output is a post on a social networking site.

[1234] Step 9:

[1235] The server analyzes the image information uploaded by the user and generates appropriate text.

[1236] It uses image analysis algorithms (e.g., Google Cloud Vision) to analyze the image content and generate text based on that content. For example, if it recognizes an image as a landscape, it will generate the sentence "I found a beautiful landscape!"

[1237] The input is image information, and the output is a generated draft based on the image.

[1238] Step 10:

[1239] The server uses a machine learning model to learn from the collected call history and emotional state.

[1240] The collected message history and emotion labels are used to train a Scikit-learn model, which will provide learning results that will be useful for generating future messages.

[1241] The input is the call history and emotion labels, and the output is a trained machine learning model.

[1242] (Application example 2)

[1243] 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."

[1244] In modern society, many users use social media to share information, but the content of their posts is subjective and inconsistent, making it difficult to reflect their own emotional state or style. It is also difficult to automatically and efficiently create effective ad copy that matches a user's emotions and style when generating ads. To solve these problems, a system is needed that analyzes a user's past posting history and generates new posts and ad copy that reflect that user's style and emotional state.

[1245] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1246] In this invention, the server includes means for collecting past call history, means for analyzing the collected call history and extracting the user's call style, and means for generating new messages based on the extracted call style. This makes it possible to automatically generate messages and advertising copy that reflect the user's past call history and emotional state, present them to the user, and post them to SNS or advertising media after final confirmation.

[1247] "Past communication history" refers to all of the content posted or transmitted by a user in the past on social media platforms and other platforms.

[1248] "Means of collection" refers to the technical processes and functions that the system uses to obtain past communication history using the user's social media account authentication information.

[1249] "Means of analyzing and extracting a user's communication style" refers to a means of analyzing collected communication history using natural language processing and machine learning technology to identify characteristics of the user's writing, such as tone, theme, and wording.

[1250] "Means for generating new messages" refers to technology for creating new messages using AI models, etc., based on the extracted user's communication style and emotional state.

[1251] "Means for presenting the generated message to the user and allowing them to review and correct it" refers to an interface or process that displays the generated message to the user, allows the user to review its contents, and makes corrections as necessary.

[1252] "Means for posting the final verified and corrected message on a social networking site" refers to the technical process by which the final message that has been verified and corrected by the user is automatically posted on a platform such as a social networking site.

[1253] "Means for generating advertising copy and advertising visuals based on the extracted communication style and emotional state" refers to technology for automatically creating advertising copy and visuals for commercial purposes, taking into account the user's communication style and emotional state.

[1254] "Means for presenting the generated ad copy to the user and allowing the user to modify and finalize it" refers to an interface or function that displays the generated ad copy to the user, allowing the user to review it and modify it as necessary.

[1255] "Means for posting to designated media" refers to the technical process for automatically distributing and posting the final, verified and corrected advertising copy to the designated advertising media or platform.

[1256] The following is a specific description of an embodiment of the present invention. The entire system is composed of a server, a terminal, and a user. The programs that are part of this system are realized as follows.

[1257] Data collection methods

[1258] The server uses the authentication information for the SNS account provided by the user to collect past call history via the SNS API. Specifically, it accesses the SNS account using authentication information such as the API key, API secret, access token, and access token secret, and retrieves the past call history (e.g., 200 calls). At this stage, data is collected from the SNS API and stored on the server.

[1259] Analysis of communication style and emotional state

[1260] The collected call history is analyzed within the server. Natural language processing technology is used to extract features such as tone, theme, and wording of the call. An emotion engine is also used to detect the user's emotional state (e.g., positive, negative, neutral) from the call history. This analysis uses machine learning models (e.g., BERT) and sentiment analysis models.

[1261] Generate new messaging and ad copy

[1262] The server then uses a generative AI model (e.g., GPT-2) to generate new messages based on the extracted user's communication style and emotional state. The generation prompts are populated with topics and tones based on the user's emotional state, and new messages are created. Similarly, ad copy and visuals are automatically generated to match the user's communication style and emotional state.

[1263] For example, the following prompt might be used:

[1264] Prompt: "Create an ad for a new cafe with a positive tone that matches the user's communication style."

[1265] Feedback and Corrections

[1266] The generated message and advertisement copy are presented to the user via the device. The user can review these copies and make corrections as necessary. This procedure is an important step to ensure that the automatically generated copies match the user's expectations and intentions. An interface is provided that allows the user to provide feedback and make corrections.

[1267] Final Post

[1268] The final message and advertisement copy that has been checked and corrected is posted by the server to the social media platform or designated advertising medium. This posting is also done via API and is published from the user's account.

[1269] In this way, the present invention can efficiently generate consistent messages and advertisements that reflect the user's past message history and emotional state, and finally publish them after confirmation and correction.

[1270] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1271] Step 1: Data collection

[1272] The server receives the authentication information (API key, API secret, access token, access token secret) of the user's SNS account as input. Using this authentication information, it accesses the SNS API and retrieves the user's past call history. At this time, the most recent 200 call history entries are collected and stored on the server. The output is the collected call history data.

[1273] Step 2: Analyze communication style and emotional state

[1274] The server analyzes the collected call history data as input. First, it uses natural language processing technology to extract features such as tone, theme, and wording from the call history. At the same time, it uses an emotion engine to detect the emotional state (positive, negative, neutral) of each message. This identifies the user's call style and emotional state. The output is the analyzed call style and emotional state data.

[1275] Step 3: Generate a new message

[1276] The server generates a new message using the analyzed message style and emotional state data. It uses a generative AI model (GPT-2) to generate the message. The following prompt is input:

[1277] Prompt: "Create an ad for a new cafe with a positive tone that matches the user's communication style."

[1278] Based on this prompt, a new message is generated. The output is the generated message.

[1279] Step 4: Generate ad copy and visuals

[1280] The server generates ad copy and visuals using the analyzed customer communication style and emotional state data as input. Using a generative AI model, ad copy that matches the user's emotional state is created based on the prompt text. The output is the generated ad copy and visuals.

[1281] Step 5: Feedback and revisions

[1282] The terminal presents the generated message and advertisement copy to the user. The user checks the contents and makes any necessary corrections. At this time, the terminal sends the user's corrections to the server, which reflects the suggested corrections. The input is the generated message and advertisement copy, and the output is the final version that has been checked and corrected by the user.

[1283] Step 6: Final submission

[1284] The server receives as input the final message and advertisement copy that has been reviewed and revised by the user. It posts this to the social media platform or designated advertising medium. The posting is again done using the social media API, and is made public through the user's account. The output is the publication of the final message and advertisement copy.

[1285] 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.

[1286] 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.

[1287] 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.

[1288] 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.

[1289] 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.

[1290] 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.

[1291] 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).

[1292] 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.

[1293] 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."

[1294] 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.

[1295] 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).

[1296] 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.

[1297] 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.

[1298] 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.

[1299] 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.

[1300] 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.

[1301] 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.

[1302] 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.

[1303] 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.

[1304] 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.

[1305] 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.

[1306] The following is further disclosed regarding the above embodiment.

[1307] (Claim 1)

[1308] A means for collecting past call history;

[1309] A means for analyzing the collected call history and extracting the user's call style;

[1310] a means for generating a new message based on the extracted message style;

[1311] means for presenting the generated message to the user for confirmation and correction;

[1312] The system includes a means for posting the final confirmed and corrected message to SNS.

[1313] (Claim 2)

[1314] 2. The system according to claim 1, further comprising means for including image information in the generated message and automatically generating a draft message that matches the image content.

[1315] (Claim 3)

[1316] The system of claim 1, including a machine learning model for analyzing the tone, theme, and word choice of individual messages from the collected message history.

[1317] "Example 1"

[1318] (Claim 1)

[1319] means for collecting past call history using authentication information provided by a user;

[1320] A means for analyzing the collected call history and extracting the user's call style;

[1321] A means for generating new speech using a generative AI model based on the extracted speech style;

[1322] means for presenting the generated message to a user via a terminal for confirmation and correction;

[1323] The system includes a means to post the final confirmed and corrected message to SNS via API.

[1324] (Claim 2)

[1325] 2. The system according to claim 1, further comprising means for including image information in the generated message and automatically generating a draft message in accordance with the image content.

[1326] (Claim 3)

[1327] The system of claim 1, including a machine learning model for analyzing the tone, theme, and word choice of individual messages from the collected message history.

[1328] "Application Example 1"

[1329] (Claim 1)

[1330] A means for collecting past call history;

[1331] A means for analyzing the collected call history and extracting the user's call style;

[1332] a means for generating a new message based on the extracted message style;

[1333] means for presenting the generated message to the user for confirmation and correction;

[1334] A way to post the final confirmed and corrected message on social media,

[1335] A means for automatically generating advertising copy;

[1336] A means for presenting the generated ad copy to the user for confirmation and correction;

[1337] The system includes a means for posting the revised copy to a distribution platform.

[1338] (Claim 2)

[1339] 2. The system according to claim 1, further comprising means for including image information in the generated message and automatically generating a draft message that matches the image content.

[1340] (Claim 3)

[1341] The system of claim 1, including a machine learning model for analyzing the tone, theme, and word choice of individual messages from the collected message history.

[1342] "Example 2: Combining Emotion Engines"

[1343] (Claim 1)

[1344] A means for a user to input authentication information for a social networking service;

[1345] A means for collecting past call history;

[1346] A means for analyzing the collected call history and extracting the user's call style and emotional state;

[1347] A means using a generative AI model to generate new messages based on the extracted message style and emotional state;

[1348] means for presenting the generated message to the user for confirmation and correction;

[1349] A means for posting the final confirmed and corrected message to an SNS service,

[1350] A means for analyzing the collected image information and generating appropriate copy based on the content thereof;

[1351] The system includes means for using a machine learning model to learn the collected call history and emotional state.

[1352] (Claim 2)

[1353] 2. The system according to claim 1, further comprising means for including image information in the generated message and automatically generating a draft message that matches the image content.

[1354] (Claim 3)

[1355] The system of claim 1, further comprising technology for analyzing the tone, theme, and word choice of individual messages from the collected message history.

[1356] "Application example 2 when combining emotion engines"

[1357] (Claim 1)

[1358] A means for collecting past call history;

[1359] A means for analyzing the collected call history and extracting the user's call style;

[1360] a means for generating a new message based on the extracted message style;

[1361] means for presenting the generated message to the user for confirmation and correction;

[1362] A way to post the final confirmed and corrected message on social media,

[1363] A means for generating advertising copy and advertising visuals based on the extracted communication style and emotional state;

[1364] A means for presenting the generated ad copy to the user, allowing the user to make corrections and final confirmation;

[1365] A system including a means for posting the confirmed and corrected advertising copy to a predetermined medium.

[1366] (Claim 2)

[1367] 2. The system according to claim 1, further comprising means for including image information in the generated message and automatically generating a draft message that matches the image content.

[1368] (Claim 3)

[1369] The system of claim 1, including a machine learning model for analyzing the tone, theme, and word choice of individual messages from the collected message history. [Explanation of symbols]

[1370] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for collecting past call history; A means for analyzing the collected call history and extracting the user's call style; a means for generating a new message based on the extracted message style; means for presenting the generated message to the user for confirmation and correction; A system that includes a means for posting the final confirmed and corrected message to a social networking site.

2. 2. The system according to claim 1, further comprising means for including image information in the generated message and automatically generating a copy of the message that matches the image content.

3. The system of claim 1, further comprising a machine learning model for analyzing the tone, theme, and word choice of individual messages from the collected message history.

Images