system

Abstract

We provide the system. [Solution] A generation method for generating the first draft of a report using natural language generation technology, A means of accumulating and analyzing past revision history and feedback, An interaction mechanism that presents correction suggestions to users and collects feedback from them, A learning method that improves the accuracy of proposed generation methods using feedback, A system that includes this.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, and includes steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation 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

Summary of the Invention

Problems to be Solved by the Invention

[0004] Simultaneously achieving efficiency and quality improvement in report creation is an important issue in many industries. The conventional report creation process requires a lot of time and human resources, and the burden is particularly large in the stage of correcting content consistency and typos. There is a need for a method to effectively utilize feedback while solving such problems and guarantee the quality of the report.

Means for Solving the Problems

[0005] This invention provides a system comprising a generation means for automatically generating the first draft of a report using natural language generation technology. It also includes an interaction means for providing and collecting accurate revision suggestions to the user through an accumulation means for accumulating and analyzing past revision history and feedback. Furthermore, it includes a learning means for improving the accuracy of suggestions using the collected feedback, thereby enabling efficient and high-quality report creation.

[0006] "Natural language generation technology" is artificial intelligence technology that automatically generates sentences and texts that mimic human speech.

[0007] A "first draft" is a preliminary version of a report or document.

[0008] "Generating means" refers to a technology or device for automatically creating information for a specific purpose.

[0009] "Storage methods" refer to methods and systems for collecting and storing data and information.

[0010] "Analysis" is a method of breaking down data and information, understanding its constituent elements, and finding meaning in them.

[0011] "Interaction means" refers to methods or devices for a system and a user to exchange information with each other.

[0012] "Suggestion accuracy" is a measure that indicates the accuracy and appropriateness of the suggestions and advice provided by the system.

[0013] "Learning methods" refer to technologies and processes that enable a system to automatically improve its performance based on new information and data. [Brief explanation of the drawing]

[0014] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] It is a conceptual diagram showing an example of the main functions of a data processing device and a smart device according to the first embodiment. [Figure 3] It is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] It is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] It is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] It is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] It is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] It is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] It shows an emotion map to which a plurality of emotions are mapped. [Figure 10] It shows an emotion map to which a plurality of emotions are mapped. [[ID=2​​​​​​​​​​​​​​​​​​​​​First, the terms used in the following description will be explained.

[0017] In the following embodiments, the numbered processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0018] In the following embodiments, the numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0019] In the following embodiments, the numbered storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, and the like.

[0020] In the following embodiments, the numbered communication I / F (Interface) is an interface including a communication processor and 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), or Bluetooth (registered trademark), and the like.

[0021] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0022] [First Embodiment]

[0023] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0024] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0025] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0026] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0027] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0028] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0029] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0030] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0031] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0032] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0033] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0034] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0035] This system consists of servers, terminals, and users, and each entity works together to efficiently and accurately produce reports. The details are described below.

[0036] server

[0037] The server first utilizes natural language generation technology to generate an initial draft of a report based on requests received from users via their devices. This generation process leverages a pre-trained AI model to automatically create text based on themes and guidelines specified by the user. The server also manages past user revision history and feedback, storing it in a database. This data later becomes a crucial resource for improving the AI ​​model's performance.

[0038] terminal

[0039] The terminal provides the user interface, allowing users to easily interact with the system. All requests for initial draft generation and revisions are made through this interface. Information entered by the user and results returned from the server are displayed on the terminal to assist in user decision-making. In addition, the terminal also functions as a relay for sending feedback information to the server.

[0040] User

[0041] The user requests the generation of the initial draft of the report via their terminal and receives the automatically generated document from the server. They review the received initial draft and, if necessary, request revision suggestions from the server. After considering the suggested revisions, the user makes the necessary corrections and finalizes the version. The user also submits feedback on the provided revision suggestions, and this information is used to improve the accuracy of future suggestions.

[0042] Specific example

[0043] For example, if a user wants to create a "quarterly performance report," they specify a theme via their terminal and request the server to generate an initial draft. The server automatically generates a draft of the report using natural language generation technology and sends it to the user. The user reviews the draft and requests revisions via their terminal for any areas that need correction. The server analyzes past feedback, generates the most suitable revisions, and sends them back to the terminal. Finally, the user finalizes the report based on the suggestions, provides feedback, and completes the process.

[0044] Thus, the present invention improves the quality and efficiency of the report creation process while automating it by incorporating natural language generation technology and user feedback.

[0045] The following describes the processing flow.

[0046] Step 1:

[0047] The user starts up their device and opens the report creation application. Next, the user enters the report's theme and necessary information, and requests the generation of the first draft.

[0048] Step 2:

[0049] The terminal sends the information entered by the user to the server as data packets. This includes details and specific requirements for the report to be generated.

[0050] Step 3:

[0051] The server analyzes the received data packets and generates a preliminary draft of the report using natural language generation technology. The generated preliminary draft is based on a pre-configured algorithm and data.

[0052] Step 4:

[0053] The server sends the generated initial draft to the terminal. The terminal displays it in a format that is easy for the user to understand.

[0054] Step 5:

[0055] Users review the initial draft on their devices and add comments and revision suggestions for areas that need correction. This information will be used in the next step.

[0056] Step 6:

[0057] The terminal sends correction suggestions from the user to the server. The server analyzes these suggestions and generates the optimal suggestions by referring to the accumulated correction history and feedback data.

[0058] Step 7:

[0059] The server generates suggested fixes based on the analysis results and sends them back to the terminal. The terminal presents the suggestions to the user and prompts them to apply the necessary fixes.

[0060] Step 8:

[0061] The user makes revisions based on the server's suggestions and completes the final report. After completion, they provide feedback on the effectiveness of the suggestions.

[0062] Step 9:

[0063] The device sends user feedback to the server. The server stores the received feedback in a database and uses it for future improvements.

[0064] Through this process, the system will be continuously improved, resulting in more accurate suggestions being provided in subsequent report generation.

[0065] (Example 1)

[0066] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0067] In document creation, such as reports, a challenge is enabling users to efficiently produce high-quality initial drafts. Conventional systems often require extensive manual revisions due to insufficient accuracy and efficiency in document generation. Furthermore, it is difficult to improve the accuracy of the generation model by fully utilizing user feedback.

[0068] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0069] In this invention, the server includes a generation means for generating an initial draft of a document using natural language, a storage means for recording and analyzing past revision history and user evaluations, and an interaction means for presenting revision proposals to the user and collecting user evaluations. This makes it possible to provide highly accurate document generation and efficient revision suggestions, and to significantly reduce manual work for the user.

[0070] "Natural language" refers to the language that humans use on a daily basis, and it is analyzed in order to enable information processing by computers.

[0071] "Generation means" refers to technologies and devices that automatically generate the initial draft of a document using natural language processing techniques based on specified guidelines.

[0072] "Storage methods" refer to technologies and devices for recording past revision history and user feedback, and for analyzing this data.

[0073] "Interaction means" refers to interfaces and technologies for presenting revised versions to users and collecting feedback from them.

[0074] "Learning methods" refer to technologies and devices that use collected feedback to adjust and enhance AI models in order to improve the accuracy and quality of the documents they generate.

[0075] "Display means" refers to technologies including screens and devices used to present generated initial drafts, revised versions, etc., to users.

[0076] "Input means" refers to interfaces or devices that allow users to input requests or modification suggestions into the system.

[0077] This invention is a document generation system in which a server, a terminal, and a user work together. The server uses a pre-trained generative AI model to generate a first draft of a document based on themes and guidelines received from the user. For example, a specific AI framework or API can be used as the model based on natural language generation technology. The generated first draft is presented to the user via the terminal.

[0078] When generating this initial draft, the server accumulates past revision history and user feedback, which is used to adjust and improve the generation model. This data is stored in a database on the server, and the model is retrained and optimized as needed through learning mechanisms.

[0079] The terminal provides a user interface and a means for the user to input themes and guidelines for reports. It also displays the initial draft sent from the server to the user, allowing them to input revision suggestions and feedback. For example, if a user wants to create a "Quarterly Performance Report," they could input the theme into the terminal and send a prompt message to the server requesting the generation of the initial draft.

[0080] The user requests the server to generate an initial draft via their device and reviews the suggested revisions and feedback from the server. Based on this feedback, the user revises the report and finalizes the document on their device. The device collects the user's feedback and sends it to the server, contributing to improving the accuracy of the AI ​​model in subsequent iterations.

[0081] As an example of a prompt, you can send the following to the server: "Generate the first draft of the quarterly performance report based on the following themes: Sales trends and market analysis." This allows the system to automatically generate documents that match the user's requests, streamlining manual work.

[0082] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0083] Step 1:

[0084] The terminal accepts the report's theme and guidelines as input from the user. The user enters a specific theme, such as "Quarterly Performance Report." The terminal receives this input and formats it as a prompt. The formatted prompt is then prepared as a request to the server.

[0085] Step 2:

[0086] The terminal sends a formatted prompt message to the server. This prompt message becomes the input, and the server uses it to begin the document generation process. In this process, the terminal's role is to send data to the server reliably and quickly.

[0087] Step 3:

[0088] The server uses a pre-trained generative AI model to generate a first draft of a document, taking the received prompt text as input. During this process, the AI ​​model uses algorithms to construct the text based on the content of the prompt text. The generated first draft is temporarily stored on the server.

[0089] Step 4:

[0090] The server sends the generated initial draft to the terminal. The server's output is the initial draft, which is then returned to the terminal. The core of the server's operation is an accurate and efficient generation process using a model.

[0091] Step 5:

[0092] The terminal presents the initial draft sent from the server to the user. The user receives the initial draft in a visually verifiable format. The terminal formats the initial draft appropriately for easy viewing.

[0093] Step 6:

[0094] The user reviews the initial draft and inputs any necessary corrections or additional instructions into the terminal. The user's feedback and suggested revisions are then used as input for the next data processing step.

[0095] Step 7:

[0096] The terminal sends correction requests from the user to the server. This data is analyzed by the server and used to generate further correction suggestions. The terminal's function is to accurately transmit the data to the server.

[0097] Step 8:

[0098] The server receives user feedback and compares it with past data to generate optimal correction suggestions. At this stage, the server runs an optimization algorithm based on the correction history and feedback data. The generated correction suggestions become the new output.

[0099] Step 9:

[0100] The server sends the generated proposed corrections to the terminal. The server's role is to utilize its accumulated knowledge to return useful information to the user.

[0101] Step 10:

[0102] Users review the proposed revisions via their devices and incorporate them into the final document. The final version is then finalized and saved on the device as a completed report. It is crucial that the user's intentions are accurately reflected.

[0103] (Application Example 1)

[0104] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0105] In creating business reports and inventory management reports, there is a need to improve both efficiency and accuracy simultaneously. In particular, in the retail industry, it is essential to grasp sales and inventory status in real time and quickly create reports. However, manual report creation is time-consuming, labor-intensive, and prone to errors. Therefore, technology is needed to achieve a high-quality and efficient report creation process.

[0106] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0107] In this invention, the server includes a generation means for generating an initial draft of a report using natural language generation technology, an storage means for accumulating and analyzing past revision history and feedback, and an interaction means for presenting revision suggestions to the user and collecting feedback from the user. This enables efficient processing of sales information and inventory information and accurate report creation.

[0108] "Natural language generation technology" is a technology that allows computers to understand human language and generate sentences in a natural-sounding form.

[0109] "Generation means" refers to a device or process that has the function of automatically creating the first draft of a report using natural language generation technology.

[0110] A "storage device" refers to a device or process equipped with a mechanism for collecting and storing past revision history and feedback information, and for utilizing it for subsequent analysis and improvement.

[0111] "Interaction means" refers to technologies that provide an interface for a system and a user to exchange information and have the function of collecting feedback from the user.

[0112] "Business information processing means" refers to a system or method for preparing data necessary for report creation by inputting sales information and inventory information in digital format and processing it quickly and accurately.

[0113] A "learning method" is a process or device for training an AI model to improve the accuracy of the generation method's suggestions based on feedback collected from users.

[0114] The system for realizing this invention involves the coordinated operation of a server, a terminal, and a user. The server is responsible for generating the initial draft of the report using natural language generation technology. Specifically, it uses a pre-trained generative AI model to generate prompt sentences for the report based on sales and inventory information provided by the user, and then presents the results to the user.

[0115] The terminal functions as a user interface, allowing users to input sales and inventory information. It also displays the initial draft of reports returned from the server, providing an interface for users to input corrections and improvement suggestions. This allows users to review the report content, make necessary revisions, and finalize the version.

[0116] Users can easily input daily work information using their smartphones and review reports generated based on that information. The feedback entered by users is sent to the server and used to improve the system in the future.

[0117] The primary hardware used is a smartphone, and the software implements a GPT model using the OpenAI® API. The server processes this information and efficiently manages sales and inventory data. As a concrete example, assuming a situation where sales of 500,000 yen and inventory of 20 units are managed on a given day, an example of a prompt message would be as follows:

[0118] Example prompt: "Today's sales were 500,000 yen, and inventory is 20 units. Please create a business report based on this information."

[0119] The reports generated in this way are used as an important source of information for planning the next day's work and placing orders.

[0120] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0121] Step 1:

[0122] Users input sales and inventory information using their smartphones. The entered data is sent to the server as basic information for creating daily business reports. Specifically, users enter sales figures and inventory status into the input fields on the screen and press the "Submit" button to send the data to the server.

[0123] Step 2:

[0124] The server generates a prompt message based on the received sales and inventory information. This prompt message clarifies the information needed to create the initial draft of the report and prepares it for the AI ​​generation model. Specifically, it converts the input numerical data into natural language sentences and formats them to something like, "Today's sales were ¥XX, and inventory is XX units. Please create a report."

[0125] Step 3:

[0126] The server feeds the generated prompt sentences into a generative AI model to generate a first draft of the report. This process uses the OpenAI API to input the prompt sentences into the model and create the necessary report text data. The output is a draft of the report in natural language obtained from the generative AI model.

[0127] Step 4:

[0128] The initial draft of the generated report is sent from the server to the terminal and displayed to the user. The user can review it and, if necessary, make corrections or comments on the terminal screen. User feedback and correction instructions are sent to the server via the interaction mechanism.

[0129] Step 5:

[0130] The server receives feedback from users and stores it in a storage system along with past revision history. This data is used to improve the performance of the AI ​​model when creating the next report. This process continuously improves the quality of the reports and the efficiency of the generation process.

[0131] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0132] This invention optimizes the report generation process by integrating natural language generation technology, user feedback, and emotion recognition technology. The system is primarily implemented using three elements: a server, a terminal, and a user.

[0133] server

[0134] The server hosts natural language generation technology to generate the initial draft of reports and automatically generates reports based on generation requests from users. The server also features an emotion engine that analyzes the user's emotional state and adjusts the tone and style of the initial draft based on that data. The server also has a learning mechanism to continuously improve its generation algorithm by accumulating feedback and emotion data.

[0135] terminal

[0136] The terminal provides an interface, creating an environment where users can easily interact with the system. In addition to displaying user input and responses from the server, it has the ability to send user emotions to the server in real time.

[0137] User

[0138] Users input their report creation requests into the system via their terminal. Once the initial draft is sent from the server, the user reviews its contents and requests revisions as needed. User feedback and sentiment data are sent to the server and used to improve the accuracy of future proposals.

[0139] Specific example

[0140] For example, if a user wants to create an "annual report," they input the theme and goals into the server via their terminal. The server generates a first draft using natural language generation technology, while simultaneously analyzing the user's emotional state with an emotion engine and adjusting the document's tone accordingly. This process generates a more natural report that reflects the user's intentions and emotions. Users can request revisions, and in doing so, they also send feedback, including emotional data, from their terminal to the server, allowing the system to further improve the accuracy of future reports.

[0141] Thus, the present invention provides a method that simultaneously improves the efficiency and quality of report creation by taking user emotions into consideration.

[0142] The following describes the processing flow.

[0143] Step 1:

[0144] The user starts up their device and accesses the report creation application. Next, they enter the theme and specific requirements of the report they want to create and request the generation of the initial draft.

[0145] Step 2:

[0146] The terminal converts the information provided by the user into data packets and sends them to the server. These data packets contain detailed information about the theme and the user's requests.

[0147] Step 3:

[0148] The server receives data packets and generates a first draft of the report using natural language generation technology. This generation process uses a pre-configured algorithm and constructs the content as needed.

[0149] Step 4:

[0150] Simultaneously, the server uses an emotion engine to analyze emotional data sent from the user's device. The results of this analysis are reflected in the tone and style of the report.

[0151] Step 5:

[0152] The generated initial draft and the adjusted tone information are sent from the server to the terminal. The terminal receives this information and displays it visually to the user.

[0153] Step 6:

[0154] Users review the initial draft on their device and provide feedback and revision suggestions for areas they feel need correction. During this process, the device also records the user's real-time emotions.

[0155] Step 7:

[0156] The device sends user feedback and sentiment data to the server. The server analyzes this data, referring to the revision history and accumulated feedback, to generate optimal revision suggestions.

[0157] Step 8:

[0158] The server sends the generated correction suggestions to the terminal. The terminal then presents them to the user, allowing them to freely apply the corrections.

[0159] Step 9:

[0160] The user finalizes the report based on the provided revision suggestions. After completion, they provide feedback on the quality of the suggestions and the report via their device.

[0161] Step 10:

[0162] The terminal sends the final feedback to the server. The server stores this feedback in a database and uses it to improve the accuracy of suggestions in the future.

[0163] Through this series of processes, the system can effectively utilize user emotions and feedback to improve the quality and efficiency of report generation.

[0164] (Example 2)

[0165] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0166] In report creation, traditional systems have made it difficult to efficiently produce documents that fully reflect the user's emotions and intentions. There is a need to quickly produce high-quality reports by appropriately adjusting the tone and style of the report while taking user feedback and emotional states into consideration.

[0167] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0168] In this invention, the server includes a generation means for generating a first draft of a report using natural language generation technology, an analysis means for analyzing the user's emotional state, and an adjustment means for adjusting the tone and style using feedback and emotional data. This makes it possible to efficiently generate high-quality reports that reflect the user's intentions and emotions.

[0169] "Natural language generation technology" refers to the technology that enables computer systems to automatically generate human language.

[0170] The "generation method" refers to the part that has the function of automatically creating the first draft of the report using natural language generation technology.

[0171] "Storage methods" refer to the part of the system that stores past revision history and user feedback in a database and has the function to analyze them.

[0172] "Interaction mechanisms" refer to the parts that have functions for presenting modification suggestions to the user and collecting feedback from the user.

[0173] "Adjustment mechanisms" refer to the part of the report that uses feedback and sentiment data to adjust the tone and style of the report.

[0174] The "analysis means" refers to the part of the system that analyzes user emotional data and applies that information to other functions of the system.

[0175] The "learning method" refers to the part that uses accumulated feedback and sentiment data to improve the accuracy and proposal capabilities of the initial draft generation algorithm.

[0176] "Revision means" refers to the part that has the function of revising the generated initial draft based on the user's requests and feelings.

[0177] "Optimization means" refers to the part that has the function of retraining or adjusting the generation technology based on user feedback and sentiment data to improve performance.

[0178] This invention is a system that integrates natural language generation technology, user feedback, and emotion recognition technology to optimize the report creation process. This system mainly consists of three elements: a server, a terminal, and a user.

[0179] server

[0180] The server hosts a natural language generation AI model and generates a draft of the report based on user prompts. For this purpose, the server utilizes a natural language processing framework, such as the GPT model. The generated draft is temporarily stored in storage. In addition, the server has an emotion engine that analyzes the user's emotional state using, for example, an emotion analysis API. Based on the results, the server adjusts the tone and style of the report according to the user's emotions.

[0181] terminal

[0182] The terminal provides a user interface and sets up an environment for the user to interact with the system. The terminal sends the prompt text entered by the user to the server and displays the initial draft of the generated report. Furthermore, the terminal has an emotion recognition function, which uses sensors to analyze the user's emotions in real time and sends this information to the server.

[0183] User

[0184] Users submit report creation requests through their terminals. For example, they might enter a prompt such as, "Please create an annual report on the company's performance for fiscal year 2023. The tone should be formal yet positive." Once the initial draft is sent from the server to the terminal, the user can review its contents and provide specific revision requests or emotional feedback.

[0185] As a concrete example, consider a scenario where a user wants to create an "annual report." The user enters prompt text into the terminal, and the server generates a first draft of the report based on those instructions. During the generation process, the tone is adjusted to take into account the user's emotional state, resulting in a more natural report that aligns with the user's intentions. This allows the user to efficiently create a high-quality report.

[0186] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0187] Step 1:

[0188] The user inputs the information necessary for report creation as prompt messages into the terminal. These prompt messages include specific themes, tones, and goals. The entered prompt messages are then sent from the terminal to the server.

[0189] Step 2:

[0190] The server automatically generates a first draft of the report using a natural language generation AI model based on the received prompt text. In this process, it analyzes the prompt text provided as input and outputs a logically structured document according to its content.

[0191] Step 3:

[0192] The server uses an emotion engine to analyze the user's emotional data. It receives emotional data sent from the terminal as input and uses an emotion analysis API to evaluate the user's current emotional state. This evaluation result is used to adjust the tone and style of the report.

[0193] Step 4:

[0194] The server adjusts the tone and style of the report based on the sentiment analysis results. This adjustment generates an appropriate report that matches the user's emotions, and the adjusted content is sent to the terminal as the final output.

[0195] Step 5:

[0196] The user reviews the initial draft of the report sent from the server on their device. The user then inputs revision requests or sentiment feedback for specific parts of the report, as needed. This feedback is then sent back to the server for further adjustments and improvements.

[0197] Step 6:

[0198] The server readjusts the report based on the received feedback and revision requests. Specifically, it analyzes the feedback, applies it again to the natural language generation AI model, and generates a revised report. The improved report is then sent to the terminal as the final deliverable.

[0199] Step 7:

[0200] The server uses accumulated feedback and sentiment data to train its generative AI model. This improves the accuracy and adaptability of subsequent report generation. By using feedback and sentiment analysis results as input and executing the learning process, the quality of suggestions is continuously improved.

[0201] (Application Example 2)

[0202] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0203] In modern society, generating personalized content that responds to the diverse feelings and needs of users is a crucial challenge. In particular, content distribution services require more natural and effective communication by adjusting content to reflect the user's emotional state. However, current systems struggle to generate content that appropriately reflects user emotions, making it difficult to improve user satisfaction. Therefore, it is necessary to enable the adjustment of tone and style according to the user's emotional state and to efficiently generate personalized content.

[0204] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0205] In this invention, the server includes a generation means for generating an initial draft of a document using natural language generation technology, an storage means for accumulating and analyzing past revision history and user feedback, and an adjustment means for recognizing the user's emotional state and adjusting the tone and style of the document based on that. This makes it possible to generate personalized content that reflects the user's emotional state.

[0206] "Natural language generation technology" is a technology that allows computers to automatically generate text using human language.

[0207] The term "first draft of a document" refers to the initial, basic form of the document, before any further revisions or adjustments are made.

[0208] "Generation means" refers to a device or software that has the function of creating a first draft of a document using natural language generation technology.

[0209] A "storage method" refers to a device or software that has the function of storing past revision history and user feedback in a database, etc., and analyzing this data.

[0210] "Opinions" refer to feedback and suggestions for corrections that users provide to the system.

[0211] A "dialogue means" is a device or software that interactively exchanges information with the user and presents suggestions for correction.

[0212] A "learning tool" is a device or software that has the function of adjusting the system to improve the accuracy of natural language generation based on collected opinions.

[0213] "Emotional state" refers to the user's current feelings and psychological state, and recognizing this is an element that allows for adjustment of the tone and style of the content.

[0214] "Adjustment means" refers to a device or software that has the function of appropriately changing the tone and style of a generated document based on the user's emotional state.

[0215] To realize this invention, a system is constructed by appropriately combining three elements: a server, a terminal, and a user.

[0216] The server has the functionality to generate a first draft of a document using natural language generation technology. Specifically, it can use a natural language generation API (e.g., OpenAI GPT) to automatically generate a document in response to a user's request, using prompt text as input data. The server also analyzes the user's emotional state using emotion recognition technology and uses the emotion data to adjust the tone and style of the document. The server uses technologies such as TENSORFLOW to build an emotion recognition model. Furthermore, the server accumulates feedback and emotion data, and uses these to improve the generation method through learning mechanisms.

[0217] The terminal provides an interface with the user and enables communication with the server. Using smart glasses or a dedicated device, users can input and send emotional data to the server in real time. The terminal easily collects user feedback, sends it to the server, and uses it to improve future content creation.

[0218] Users use the system to send document creation requests to the server via their terminal. For example, while relaxing in a park, they can request the creation of a document in a relaxed tone that matches their mood. The feedback and mood data provided by the user greatly contribute to the subsequent content generation on the server.

[0219] For example, if a user requests a "relaxing article," one could send a prompt message to the AI ​​model saying, "This user is relaxed. Please generate an article with relaxing content that matches this state."

[0220] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0221] Step 1:

[0222] The user enters a prompt message via smart glasses or a device, requesting the server to generate specific content. The entered prompt message is sent to the server as initial data to reflect the content and tone of the document desired by the user.

[0223] Step 2:

[0224] The device uses built-in sensors and cameras to analyze the user's real-time emotional state and transmits that data to a server. This emotional data is then used to adjust the tone and style of documents. Input data includes the user's facial expressions and voice tone, while output is numerical or categorical data corresponding to that emotion.

[0225] Step 3:

[0226] The server uses a generative AI model to generate a first draft of a document based on the received prompt text and sentiment data. This step uses natural language generation technology to optimize the tone to match the sentiment. The input is the prompt text and sentiment data, and the output is the first draft of the generated document.

[0227] Step 4:

[0228] The server sends the generated initial draft to the terminal and requests the user to review its contents. The user reviews the initial draft on the terminal and provides feedback. This feedback includes requests for document revisions as needed. The input consists of the user's feedback comments and revision instructions.

[0229] Step 5:

[0230] The server analyzes user feedback and modifies the document as needed. This process allows the server to learn and improve the accuracy of its suggestions. The input at this stage is feedback data, and the output is the modified document.

[0231] Step 6:

[0232] Finally, the server sends the revised document to the terminal and requests the user's final confirmation. The user reviews the final version of the document and approves or requests further revisions. The output at this step is the completed document.

[0233] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0234] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0235] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0236] [Second Embodiment]

[0237] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0238] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0239] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0240] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0241] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0242] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0243] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0244] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0245] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0246] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0247] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0248] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0249] This system consists of servers, terminals, and users, and each entity works together to efficiently and accurately produce reports. The details are described below.

[0250] server

[0251] The server first utilizes natural language generation technology to generate an initial draft of a report based on requests received from users via their devices. This generation process leverages a pre-trained AI model to automatically create text based on themes and guidelines specified by the user. The server also manages past user revision history and feedback, storing it in a database. This data later becomes a crucial resource for improving the AI ​​model's performance.

[0252] terminal

[0253] The terminal provides the user interface, allowing users to easily interact with the system. All requests for initial draft generation and revisions are made through this interface. Information entered by the user and results returned from the server are displayed on the terminal to assist in user decision-making. In addition, the terminal also functions as a relay for sending feedback information to the server.

[0254] User

[0255] The user requests the generation of the initial draft of the report via their terminal and receives the automatically generated document from the server. They review the received initial draft and, if necessary, request revision suggestions from the server. After considering the suggested revisions, the user makes the necessary corrections and finalizes the version. The user also submits feedback on the provided revision suggestions, and this information is used to improve the accuracy of future suggestions.

[0256] Specific example

[0257] For example, if a user wants to create a "quarterly performance report," they specify a theme via their terminal and request the server to generate an initial draft. The server automatically generates a draft of the report using natural language generation technology and sends it to the user. The user reviews the draft and requests revisions via their terminal for any areas that need correction. The server analyzes past feedback, generates the most suitable revisions, and sends them back to the terminal. Finally, the user finalizes the report based on the suggestions, provides feedback, and completes the process.

[0258] Thus, the present invention improves the quality and efficiency of the report creation process while automating it by incorporating natural language generation technology and user feedback.

[0259] The following describes the processing flow.

[0260] Step 1:

[0261] The user starts up their device and opens the report creation application. Next, the user enters the report's theme and necessary information, and requests the generation of the first draft.

[0262] Step 2:

[0263] The terminal sends the information entered by the user to the server as data packets. This includes details and specific requirements for the report to be generated.

[0264] Step 3:

[0265] The server analyzes the received data packets and generates a preliminary draft of the report using natural language generation technology. The generated preliminary draft is based on a pre-configured algorithm and data.

[0266] Step 4:

[0267] The server sends the generated initial draft to the terminal. The terminal displays it in a format that is easy for the user to understand.

[0268] Step 5:

[0269] Users review the initial draft on their devices and add comments and revision suggestions for areas that need correction. This information will be used in the next step.

[0270] Step 6:

[0271] The terminal sends correction suggestions from the user to the server. The server analyzes these suggestions and generates the optimal suggestions by referring to the accumulated correction history and feedback data.

[0272] Step 7:

[0273] The server generates suggested fixes based on the analysis results and sends them back to the terminal. The terminal presents the suggestions to the user and prompts them to apply the necessary fixes.

[0274] Step 8:

[0275] The user makes revisions based on the server's suggestions and completes the final report. After completion, they provide feedback on the effectiveness of the suggestions.

[0276] Step 9:

[0277] The device sends user feedback to the server. The server stores the received feedback in a database and uses it for future improvements.

[0278] Through this process, the system will be continuously improved, resulting in more accurate suggestions being provided in subsequent report generation.

[0279] (Example 1)

[0280] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0281] In the creation of documents such as reports, it is an issue to enable users to efficiently create high-quality first drafts. In conventional systems, since the accuracy and efficiency of document generation are not sufficient, a great deal of manual correction work is often required. In addition, it is difficult to sufficiently utilize feedback from users to improve the accuracy of the generation model.

[0282] The specific processing by the specific processing unit 290 of the data processing apparatus 12 in the first embodiment is realized by the following respective means.

[0283] In this invention, the server includes a generation means for generating a first draft of a document using natural language, a storage means for recording and analyzing past correction histories and evaluations from users, and an interaction means for presenting amendment proposals to the users and collecting evaluations from the users. Thereby, it becomes possible to provide highly accurate document generation and efficient amendment proposals, and significantly reduce the manual work of users.

[0284] "Natural language" refers to the language that humans commonly use in daily life and is analyzed to enable information processing by a computer.

[0285] "Generation means" refers to the technology and apparatus for automatically generating a first draft of a document using natural language processing technology based on specified guidelines.

[0286] "Storage means" refers to the technology and apparatus for recording past correction histories and evaluations from users and analyzing them as data.

[0287] "Interaction means" refers to the interface and technology for presenting amendment proposals to users and collecting feedback from users.

[0288] "Learning means" refers to the technology and apparatus for adjusting and strengthening an AI model in order to improve the accuracy and quality of the generated document by utilizing the collected feedback.

[0289] "Display means" refers to technologies including screens and devices used to present generated initial drafts, revised versions, etc., to users.

[0290] "Input means" refers to interfaces or devices that allow users to input requests or modification suggestions into the system.

[0291] This invention is a document generation system in which a server, a terminal, and a user work together. The server uses a pre-trained generative AI model to generate a first draft of a document based on themes and guidelines received from the user. For example, a specific AI framework or API can be used as the model based on natural language generation technology. The generated first draft is presented to the user via the terminal.

[0292] When generating this initial draft, the server accumulates past revision history and user feedback, which is used to adjust and improve the generation model. This data is stored in a database on the server, and the model is retrained and optimized as needed through learning mechanisms.

[0293] The terminal provides a user interface and a means for the user to input themes and guidelines for reports. It also displays the initial draft sent from the server to the user, allowing them to input revision suggestions and feedback. For example, if a user wants to create a "Quarterly Performance Report," they could input the theme into the terminal and send a prompt message to the server requesting the generation of the initial draft.

[0294] The user requests the server to generate an initial draft via their device and reviews the suggested revisions and feedback from the server. Based on this feedback, the user revises the report and finalizes the document on their device. The device collects the user's feedback and sends it to the server, contributing to improving the accuracy of the AI ​​model in subsequent iterations.

[0295] As an example of a prompt, you can send the following to the server: "Generate the first draft of the quarterly performance report based on the following themes: Sales trends and market analysis." This allows the system to automatically generate documents that match the user's requests, streamlining manual work.

[0296] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0297] Step 1:

[0298] The terminal accepts the report's theme and guidelines as input from the user. The user enters a specific theme, such as "Quarterly Performance Report." The terminal receives this input and formats it as a prompt. The formatted prompt is then prepared as a request to the server.

[0299] Step 2:

[0300] The terminal sends a formatted prompt message to the server. This prompt message becomes the input, and the server uses it to begin the document generation process. In this process, the terminal's role is to send data to the server reliably and quickly.

[0301] Step 3:

[0302] The server uses a pre-trained generative AI model to generate a first draft of a document, taking the received prompt text as input. During this process, the AI ​​model uses algorithms to construct the text based on the content of the prompt text. The generated first draft is temporarily stored on the server.

[0303] Step 4:

[0304] The server sends the generated initial draft to the terminal. The server's output is the initial draft, which is then returned to the terminal. The core of the server's operation is an accurate and efficient generation process using a model.

[0305] Step 5:

[0306] The terminal presents the first draft sent from the server to the user. The user receives the first draft in a visually confirmable form. When displaying the first draft, the terminal makes it easy to view in an appropriate format.

[0307] Step 6:

[0308] The user checks the presented first draft and enters necessary corrections and additional instructions into the terminal. The user's feedback and proposed corrections serve as the input for the next data processing.

[0309] Step 7:

[0310] The terminal sends the correction request from the user to the server. This data is analyzed by the server and used for generating further amendments. The terminal functions to accurately send the data to the server.

[0311] Step 8:

[0312] The server receives the feedback from the user, collates it with past data, and generates an optimal amendment. At this stage, the server executes an optimization algorithm based on the correction history and feedback data. The generated amendment becomes the new output.

[0313] Step 9:

[0314] The server sends the generated amendment to the terminal. The role of the server is to utilize the accumulated knowledge and return useful information to the user.

[0315] Step 10:

[0316] The user checks the amendment through the terminal and reflects it in the final document. The final version is finalized and saved as a completed report on the terminal. It is important that the user's intention is accurately reflected.

[0317] (Application Example 1)

[0318] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0319] In creating business reports and inventory management reports, there is a need to improve both efficiency and accuracy simultaneously. In particular, in the retail industry, it is essential to grasp sales and inventory status in real time and quickly create reports. However, manual report creation is time-consuming, labor-intensive, and prone to errors. Therefore, technology is needed to achieve a high-quality and efficient report creation process.

[0320] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0321] In this invention, the server includes a generation means for generating an initial draft of a report using natural language generation technology, an storage means for accumulating and analyzing past revision history and feedback, and an interaction means for presenting revision suggestions to the user and collecting feedback from the user. This enables efficient processing of sales information and inventory information and accurate report creation.

[0322] "Natural language generation technology" is a technology that allows computers to understand human language and generate sentences in a natural-sounding form.

[0323] "Generation means" refers to a device or process that has the function of automatically creating the first draft of a report using natural language generation technology.

[0324] A "storage device" refers to a device or process equipped with a mechanism for collecting and storing past revision history and feedback information, and for utilizing it for subsequent analysis and improvement.

[0325] "Interaction means" refers to technologies that provide an interface for a system and a user to exchange information and have the function of collecting feedback from the user.

[0326] "Business information processing means" refers to a system or method for preparing data necessary for report creation by inputting sales information and inventory information in digital format and processing it quickly and accurately.

[0327] A "learning method" is a process or device for training an AI model to improve the accuracy of the generation method's suggestions based on feedback collected from users.

[0328] The system for realizing this invention involves the coordinated operation of a server, a terminal, and a user. The server is responsible for generating the initial draft of the report using natural language generation technology. Specifically, it uses a pre-trained generative AI model to generate prompt sentences for the report based on sales and inventory information provided by the user, and then presents the results to the user.

[0329] The terminal functions as a user interface, allowing users to input sales and inventory information. It also displays the initial draft of reports returned from the server, providing an interface for users to input corrections and improvement suggestions. This allows users to review the report content, make necessary revisions, and finalize the version.

[0330] Users can easily input daily work information using their smartphones and review reports generated based on that information. The feedback entered by users is sent to the server and used to improve the system in the future.

[0331] The primary hardware used is a smartphone, and the software implements a GPT model using the OpenAI API. The server processes this information and efficiently manages sales and inventory data. As a concrete example, assuming a situation where sales of 500,000 yen and inventory of 20 units are managed on a given day, an example of a prompt message would be as follows:

[0332] Example prompt: "Today's sales were 500,000 yen, and inventory is 20 units. Please create a business report based on this information."

[0333] The reports generated in this way are used as an important source of information for planning the next day's work and placing orders.

[0334] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0335] Step 1:

[0336] Users input sales and inventory information using their smartphones. The entered data is sent to the server as basic information for creating daily business reports. Specifically, users enter sales figures and inventory status into the input fields on the screen and press the "Submit" button to send the data to the server.

[0337] Step 2:

[0338] The server generates a prompt message based on the received sales and inventory information. This prompt message clarifies the information needed to create the initial draft of the report and prepares it for the AI ​​generation model. Specifically, it converts the input numerical data into natural language sentences and formats them to something like, "Today's sales were ¥XX, and inventory is XX units. Please create a report."

[0339] Step 3:

[0340] The server feeds the generated prompt sentences into a generative AI model to generate a first draft of the report. This process uses the OpenAI API to input the prompt sentences into the model and create the necessary report text data. The output is a draft of the report in natural language obtained from the generative AI model.

[0341] Step 4:

[0342] The initial draft of the generated report is sent from the server to the terminal and displayed to the user. The user can review it and, if necessary, make corrections or comments on the terminal screen. User feedback and correction instructions are sent to the server via the interaction mechanism.

[0343] Step 5:

[0344] The server receives feedback from users and stores it in a storage system along with past revision history. This data is used to improve the performance of the AI ​​model when creating the next report. This process continuously improves the quality of the reports and the efficiency of the generation process.

[0345] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0346] This invention optimizes the report generation process by integrating natural language generation technology, user feedback, and emotion recognition technology. The system is primarily implemented using three elements: a server, a terminal, and a user.

[0347] server

[0348] The server hosts natural language generation technology to generate the initial draft of reports and automatically generates reports based on generation requests from users. The server also features an emotion engine that analyzes the user's emotional state and adjusts the tone and style of the initial draft based on that data. The server also has a learning mechanism to continuously improve its generation algorithm by accumulating feedback and emotion data.

[0349] terminal

[0350] The terminal provides an interface, creating an environment where users can easily interact with the system. In addition to displaying user input and responses from the server, it has the ability to send user emotions to the server in real time.

[0351] User

[0352] Users input their report creation requests into the system via their terminal. Once the initial draft is sent from the server, the user reviews its contents and requests revisions as needed. User feedback and sentiment data are sent to the server and used to improve the accuracy of future proposals.

[0353] Specific example

[0354] For example, if a user wants to create an "annual report," they input the theme and goals into the server via their terminal. The server generates a first draft using natural language generation technology, while simultaneously analyzing the user's emotional state with an emotion engine and adjusting the document's tone accordingly. This process generates a more natural report that reflects the user's intentions and emotions. Users can request revisions, and in doing so, they also send feedback, including emotional data, from their terminal to the server, allowing the system to further improve the accuracy of future reports.

[0355] Thus, the present invention provides a method that simultaneously improves the efficiency and quality of report creation by taking user emotions into consideration.

[0356] The following describes the processing flow.

[0357] Step 1:

[0358] The user starts up their device and accesses the report creation application. Next, they enter the theme and specific requirements of the report they want to create and request the generation of the initial draft.

[0359] Step 2:

[0360] The terminal converts the information provided by the user into data packets and sends them to the server. These data packets contain detailed information about the theme and the user's requests.

[0361] Step 3:

[0362] The server receives data packets and generates a first draft of the report using natural language generation technology. This generation process uses a pre-configured algorithm and constructs the content as needed.

[0363] Step 4:

[0364] Simultaneously, the server uses an emotion engine to analyze emotional data sent from the user's device. The results of this analysis are reflected in the tone and style of the report.

[0365] Step 5:

[0366] The generated initial draft and the adjusted tone information are sent from the server to the terminal. The terminal receives this information and displays it visually to the user.

[0367] Step 6:

[0368] Users review the initial draft on their device and provide feedback and revision suggestions for areas they feel need correction. During this process, the device also records the user's real-time emotions.

[0369] Step 7:

[0370] The device sends user feedback and sentiment data to the server. The server analyzes this data, referring to the revision history and accumulated feedback, to generate optimal revision suggestions.

[0371] Step 8:

[0372] The server sends the generated correction suggestions to the terminal. The terminal then presents them to the user, allowing them to freely apply the corrections.

[0373] Step 9:

[0374] The user finalizes the report based on the provided revision suggestions. After completion, they provide feedback on the quality of the suggestions and the report via their device.

[0375] Step 10:

[0376] The terminal sends the final feedback to the server. The server stores this feedback in a database and uses it to improve the accuracy of suggestions in the future.

[0377] Through this series of processes, the system can effectively utilize user emotions and feedback to improve the quality and efficiency of report generation.

[0378] (Example 2)

[0379] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0380] In report creation, traditional systems have made it difficult to efficiently produce documents that fully reflect the user's emotions and intentions. There is a need to quickly produce high-quality reports by appropriately adjusting the tone and style of the report while taking user feedback and emotional states into consideration.

[0381] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0382] In this invention, the server includes a generation means for generating a first draft of a report using natural language generation technology, an analysis means for analyzing the user's emotional state, and an adjustment means for adjusting the tone and style using feedback and emotional data. This makes it possible to efficiently generate high-quality reports that reflect the user's intentions and emotions.

[0383] "Natural language generation technology" refers to the technology that enables computer systems to automatically generate human language.

[0384] The "generation method" refers to the part that has the function of automatically creating the first draft of the report using natural language generation technology.

[0385] "Storage methods" refer to the part of the system that stores past revision history and user feedback in a database and has the function to analyze them.

[0386] "Interaction mechanisms" refer to the parts that have functions for presenting modification suggestions to the user and collecting feedback from the user.

[0387] "Adjustment mechanisms" refer to the part of the report that uses feedback and sentiment data to adjust the tone and style of the report.

[0388] The "analysis means" refers to the part of the system that analyzes user emotional data and applies that information to other functions of the system.

[0389] The "learning method" refers to the part that uses accumulated feedback and sentiment data to improve the accuracy and proposal capabilities of the initial draft generation algorithm.

[0390] "Revision means" refers to the part that has the function of revising the generated initial draft based on the user's requests and feelings.

[0391] "Optimization means" refers to the part that has the function of retraining or adjusting the generation technology based on user feedback and sentiment data to improve performance.

[0392] This invention is a system that integrates natural language generation technology, user feedback, and emotion recognition technology to optimize the report creation process. This system mainly consists of three elements: a server, a terminal, and a user.

[0393] server

[0394] The server hosts a natural language generation AI model and generates a draft of the report based on user prompts. For this purpose, the server utilizes a natural language processing framework, such as the GPT model. The generated draft is temporarily stored in storage. In addition, the server has an emotion engine that analyzes the user's emotional state using, for example, an emotion analysis API. Based on the results, the server adjusts the tone and style of the report according to the user's emotions.

[0395] terminal

[0396] The terminal provides a user interface and sets up an environment for the user to interact with the system. The terminal sends the prompt text entered by the user to the server and displays the initial draft of the generated report. Furthermore, the terminal has an emotion recognition function, which uses sensors to analyze the user's emotions in real time and sends this information to the server.

[0397] User

[0398] Users submit report creation requests through their terminals. For example, they might enter a prompt such as, "Please create an annual report on the company's performance for fiscal year 2023. The tone should be formal yet positive." Once the initial draft is sent from the server to the terminal, the user can review its contents and provide specific revision requests or emotional feedback.

[0399] As a concrete example, consider a scenario where a user wants to create an "annual report." The user enters prompt text into the terminal, and the server generates a first draft of the report based on those instructions. During the generation process, the tone is adjusted to take into account the user's emotional state, resulting in a more natural report that aligns with the user's intentions. This allows the user to efficiently create a high-quality report.

[0400] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0401] Step 1:

[0402] The user inputs the information necessary for report creation as prompt messages into the terminal. These prompt messages include specific themes, tones, and goals. The entered prompt messages are then sent from the terminal to the server.

[0403] Step 2:

[0404] The server automatically generates a first draft of the report using a natural language generation AI model based on the received prompt text. In this process, it analyzes the prompt text provided as input and outputs a logically structured document according to its content.

[0405] Step 3:

[0406] The server uses an emotion engine to analyze the user's emotional data. It receives emotional data sent from the terminal as input and uses an emotion analysis API to evaluate the user's current emotional state. This evaluation result is used to adjust the tone and style of the report.

[0407] Step 4:

[0408] The server adjusts the tone and style of the report based on the sentiment analysis results. This adjustment generates an appropriate report that matches the user's emotions, and the adjusted content is sent to the terminal as the final output.

[0409] Step 5:

[0410] The user reviews the initial draft of the report sent from the server on their device. The user then inputs revision requests or sentiment feedback for specific parts of the report, as needed. This feedback is then sent back to the server for further adjustments and improvements.

[0411] Step 6:

[0412] The server readjusts the report based on the received feedback and revision requests. Specifically, it analyzes the feedback, applies it again to the natural language generation AI model, and generates a revised report. The improved report is then sent to the terminal as the final deliverable.

[0413] Step 7:

[0414] The server uses accumulated feedback and sentiment data to train its generative AI model. This improves the accuracy and adaptability of subsequent report generation. By using feedback and sentiment analysis results as input and executing the learning process, the quality of suggestions is continuously improved.

[0415] (Application Example 2)

[0416] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0417] In modern society, generating personalized content that responds to the diverse feelings and needs of users is a crucial challenge. In particular, content distribution services require more natural and effective communication by adjusting content to reflect the user's emotional state. However, current systems struggle to generate content that appropriately reflects user emotions, making it difficult to improve user satisfaction. Therefore, it is necessary to enable the adjustment of tone and style according to the user's emotional state and to efficiently generate personalized content.

[0418] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0419] In this invention, the server includes a generation means for generating an initial draft of a document using natural language generation technology, an storage means for accumulating and analyzing past revision history and user feedback, and an adjustment means for recognizing the user's emotional state and adjusting the tone and style of the document based on that. This makes it possible to generate personalized content that reflects the user's emotional state.

[0420] "Natural language generation technology" is a technology that allows computers to automatically generate text using human language.

[0421] The term "first draft of a document" refers to the initial, basic form of the document, before any further revisions or adjustments are made.

[0422] "Generation means" refers to a device or software that has the function of creating a first draft of a document using natural language generation technology.

[0423] A "storage method" refers to a device or software that has the function of storing past revision history and user feedback in a database, etc., and analyzing this data.

[0424] "Opinions" refer to feedback and suggestions for corrections that users provide to the system.

[0425] A "dialogue means" is a device or software that interactively exchanges information with the user and presents suggestions for correction.

[0426] A "learning tool" is a device or software that has the function of adjusting the system to improve the accuracy of natural language generation based on collected opinions.

[0427] "Emotional state" refers to the user's current feelings and psychological state, and recognizing this is an element that allows for adjustment of the tone and style of the content.

[0428] "Adjustment means" refers to a device or software that has the function of appropriately changing the tone and style of a generated document based on the user's emotional state.

[0429] To realize this invention, a system is constructed by appropriately combining three elements: a server, a terminal, and a user.

[0430] The server has the functionality to generate a first draft of a document using natural language generation technology. Specifically, it can use a natural language generation API (e.g., OpenAI GPT) to automatically generate a document in response to a user's request, using prompt text as input data. The server also analyzes the user's emotional state using emotion recognition technology and uses the emotion data to adjust the tone and style of the document. The server uses TensorFlow or similar tools to build an emotion recognition model. Furthermore, the server accumulates feedback and emotion data, and uses this information to improve the generation method through learning mechanisms.

[0431] The terminal provides an interface with the user and enables communication with the server. Using smart glasses or a dedicated device, users can input and send emotional data to the server in real time. The terminal easily collects user feedback, sends it to the server, and uses it to improve future content creation.

[0432] Users use the system to send document creation requests to the server via their terminal. For example, while relaxing in a park, they can request the creation of a document in a relaxed tone that matches their mood. The feedback and mood data provided by the user greatly contribute to the subsequent content generation on the server.

[0433] For example, if a user requests a "relaxing article," one could send a prompt message to the AI ​​model saying, "This user is relaxed. Please generate an article with relaxing content that matches this state."

[0434] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0435] Step 1:

[0436] The user enters a prompt message via smart glasses or a device, requesting the server to generate specific content. The entered prompt message is sent to the server as initial data to reflect the content and tone of the document desired by the user.

[0437] Step 2:

[0438] The device uses built-in sensors and cameras to analyze the user's real-time emotional state and transmits that data to a server. This emotional data is then used to adjust the tone and style of documents. Input data includes the user's facial expressions and voice tone, while output is numerical or categorical data corresponding to that emotion.

[0439] Step 3:

[0440] The server uses a generative AI model to generate a first draft of a document based on the received prompt text and sentiment data. This step uses natural language generation technology to optimize the tone to match the sentiment. The input is the prompt text and sentiment data, and the output is the first draft of the generated document.

[0441] Step 4:

[0442] The server sends the generated initial draft to the terminal and requests the user to review its contents. The user reviews the initial draft on the terminal and provides feedback. This feedback includes requests for document revisions as needed. The input consists of the user's feedback comments and revision instructions.

[0443] Step 5:

[0444] The server analyzes user feedback and modifies the document as needed. This process allows the server to learn and improve the accuracy of its suggestions. The input at this stage is feedback data, and the output is the modified document.

[0445] Step 6:

[0446] Finally, the server sends the revised document to the terminal and requests the user's final confirmation. The user reviews the final version of the document and approves or requests further revisions. The output at this step is the completed document.

[0447] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0448] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0449] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0450] [Third Embodiment]

[0451] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0452] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0453] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0454] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0455] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0456] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0457] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0458] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0459] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0460] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0461] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0462] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0463] This system consists of servers, terminals, and users, and each entity works together to efficiently and accurately produce reports. The details are described below.

[0464] server

[0465] The server first utilizes natural language generation technology to generate an initial draft of a report based on requests received from users via their devices. This generation process leverages a pre-trained AI model to automatically create text based on themes and guidelines specified by the user. The server also manages past user revision history and feedback, storing it in a database. This data later becomes a crucial resource for improving the AI ​​model's performance.

[0466] terminal

[0467] The terminal provides the user interface, allowing users to easily interact with the system. All requests for initial draft generation and revisions are made through this interface. Information entered by the user and results returned from the server are displayed on the terminal to assist in user decision-making. In addition, the terminal also functions as a relay for sending feedback information to the server.

[0468] User

[0469] The user requests the generation of the initial draft of the report via their terminal and receives the automatically generated document from the server. They review the received initial draft and, if necessary, request revision suggestions from the server. After considering the suggested revisions, the user makes the necessary corrections and finalizes the version. The user also submits feedback on the provided revision suggestions, and this information is used to improve the accuracy of future suggestions.

[0470] Specific example

[0471] For example, if a user wants to create a "quarterly performance report," they specify a theme via their terminal and request the server to generate an initial draft. The server automatically generates a draft of the report using natural language generation technology and sends it to the user. The user reviews the draft and requests revisions via their terminal for any areas that need correction. The server analyzes past feedback, generates the most suitable revisions, and sends them back to the terminal. Finally, the user finalizes the report based on the suggestions, provides feedback, and completes the process.

[0472] Thus, the present invention improves the quality and efficiency of the report creation process while automating it by incorporating natural language generation technology and user feedback.

[0473] The following describes the processing flow.

[0474] Step 1:

[0475] The user starts up their device and opens the report creation application. Next, the user enters the report's theme and necessary information, and requests the generation of the first draft.

[0476] Step 2:

[0477] The terminal sends the information entered by the user to the server as data packets. This includes details and specific requirements for the report to be generated.

[0478] Step 3:

[0479] The server analyzes the received data packets and generates a preliminary draft of the report using natural language generation technology. The generated preliminary draft is based on a pre-configured algorithm and data.

[0480] Step 4:

[0481] The server sends the generated initial draft to the terminal. The terminal displays it in a format that is easy for the user to understand.

[0482] Step 5:

[0483] Users review the initial draft on their devices and add comments and revision suggestions for areas that need correction. This information will be used in the next step.

[0484] Step 6:

[0485] The terminal sends correction suggestions from the user to the server. The server analyzes these suggestions and generates the optimal suggestions by referring to the accumulated correction history and feedback data.

[0486] Step 7:

[0487] The server generates suggested fixes based on the analysis results and sends them back to the terminal. The terminal presents the suggestions to the user and prompts them to apply the necessary fixes.

[0488] Step 8:

[0489] The user makes revisions based on the server's suggestions and completes the final report. After completion, they provide feedback on the effectiveness of the suggestions.

[0490] Step 9:

[0491] The device sends user feedback to the server. The server stores the received feedback in a database and uses it for future improvements.

[0492] Through this process, the system will be continuously improved, resulting in more accurate suggestions being provided in subsequent report generation.

[0493] (Example 1)

[0494] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0495] In document creation, such as reports, a challenge is enabling users to efficiently produce high-quality initial drafts. Conventional systems often require extensive manual revisions due to insufficient accuracy and efficiency in document generation. Furthermore, it is difficult to improve the accuracy of the generation model by fully utilizing user feedback.

[0496] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0497] In this invention, the server includes a generation means for generating an initial draft of a document using natural language, a storage means for recording and analyzing past revision history and user evaluations, and an interaction means for presenting revision proposals to the user and collecting user evaluations. This makes it possible to provide highly accurate document generation and efficient revision suggestions, and to significantly reduce manual work for the user.

[0498] "Natural language" refers to the language that humans use on a daily basis, and it is analyzed in order to enable information processing by computers.

[0499] "Generation means" refers to technologies and devices that automatically generate the initial draft of a document using natural language processing techniques based on specified guidelines.

[0500] "Storage methods" refer to technologies and devices for recording past revision history and user feedback, and for analyzing this data.

[0501] "Interaction means" refers to interfaces and technologies for presenting revised versions to users and collecting feedback from them.

[0502] "Learning methods" refer to technologies and devices that use collected feedback to adjust and enhance AI models in order to improve the accuracy and quality of the documents they generate.

[0503] "Display means" refers to technologies including screens and devices used to present generated initial drafts, revised versions, etc., to users.

[0504] "Input means" refers to interfaces or devices that allow users to input requests or modification suggestions into the system.

[0505] This invention is a document generation system in which a server, a terminal, and a user work together. The server uses a pre-trained generative AI model to generate a first draft of a document based on themes and guidelines received from the user. For example, a specific AI framework or API can be used as the model based on natural language generation technology. The generated first draft is presented to the user via the terminal.

[0506] When generating this initial draft, the server accumulates past revision history and user feedback, which is used to adjust and improve the generation model. This data is stored in a database on the server, and the model is retrained and optimized as needed through learning mechanisms.

[0507] The terminal provides a user interface and a means for the user to input themes and guidelines for reports. It also displays the initial draft sent from the server to the user, allowing them to input revision suggestions and feedback. For example, if a user wants to create a "Quarterly Performance Report," they could input the theme into the terminal and send a prompt message to the server requesting the generation of the initial draft.

[0508] The user requests the server to generate an initial draft via their device and reviews the suggested revisions and feedback from the server. Based on this feedback, the user revises the report and finalizes the document on their device. The device collects the user's feedback and sends it to the server, contributing to improving the accuracy of the AI ​​model in subsequent iterations.

[0509] As an example of a prompt, you can send the following to the server: "Generate the first draft of the quarterly performance report based on the following themes: Sales trends and market analysis." This allows the system to automatically generate documents that match the user's requests, streamlining manual work.

[0510] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0511] Step 1:

[0512] The terminal accepts the report's theme and guidelines as input from the user. The user enters a specific theme, such as "Quarterly Performance Report." The terminal receives this input and formats it as a prompt. The formatted prompt is then prepared as a request to the server.

[0513] Step 2:

[0514] The terminal sends a formatted prompt message to the server. This prompt message becomes the input, and the server uses it to begin the document generation process. In this process, the terminal's role is to send data to the server reliably and quickly.

[0515] Step 3:

[0516] The server uses a pre-trained generative AI model to generate a first draft of a document, taking the received prompt text as input. During this process, the AI ​​model uses algorithms to construct the text based on the content of the prompt text. The generated first draft is temporarily stored on the server.

[0517] Step 4:

[0518] The server sends the generated initial draft to the terminal. The server's output is the initial draft, which is then returned to the terminal. The core of the server's operation is an accurate and efficient generation process using a model.

[0519] Step 5:

[0520] The terminal presents the initial draft sent from the server to the user. The user receives the initial draft in a visually verifiable format. The terminal formats the initial draft appropriately for easy viewing.

[0521] Step 6:

[0522] The user reviews the initial draft and inputs any necessary corrections or additional instructions into the terminal. The user's feedback and suggested revisions are then used as input for the next data processing step.

[0523] Step 7:

[0524] The terminal sends correction requests from the user to the server. This data is analyzed by the server and used to generate further correction suggestions. The terminal's function is to accurately transmit the data to the server.

[0525] Step 8:

[0526] The server receives user feedback and compares it with past data to generate optimal correction suggestions. At this stage, the server runs an optimization algorithm based on the correction history and feedback data. The generated correction suggestions become the new output.

[0527] Step 9:

[0528] The server sends the generated proposed corrections to the terminal. The server's role is to utilize its accumulated knowledge to return useful information to the user.

[0529] Step 10:

[0530] Users review the proposed revisions via their devices and incorporate them into the final document. The final version is then finalized and saved on the device as a completed report. It is crucial that the user's intentions are accurately reflected.

[0531] (Application Example 1)

[0532] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0533] In creating business reports and inventory management reports, there is a need to improve both efficiency and accuracy simultaneously. In particular, in the retail industry, it is essential to grasp sales and inventory status in real time and quickly create reports. However, manual report creation is time-consuming, labor-intensive, and prone to errors. Therefore, technology is needed to achieve a high-quality and efficient report creation process.

[0534] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0535] In this invention, the server includes a generation means for generating an initial draft of a report using natural language generation technology, an storage means for accumulating and analyzing past revision history and feedback, and an interaction means for presenting revision suggestions to the user and collecting feedback from the user. This enables efficient processing of sales information and inventory information and accurate report creation.

[0536] "Natural language generation technology" is a technology that allows computers to understand human language and generate sentences in a natural-sounding form.

[0537] "Generation means" refers to a device or process that has the function of automatically creating the first draft of a report using natural language generation technology.

[0538] A "storage device" refers to a device or process equipped with a mechanism for collecting and storing past revision history and feedback information, and for utilizing it for subsequent analysis and improvement.

[0539] "Interaction means" refers to technologies that provide an interface for a system and a user to exchange information and have the function of collecting feedback from the user.

[0540] "Business information processing means" refers to a system or method for preparing data necessary for report creation by inputting sales information and inventory information in digital format and processing it quickly and accurately.

[0541] A "learning method" is a process or device for training an AI model to improve the accuracy of the generation method's suggestions based on feedback collected from users.

[0542] The system for realizing this invention involves the coordinated operation of a server, a terminal, and a user. The server is responsible for generating the initial draft of the report using natural language generation technology. Specifically, it uses a pre-trained generative AI model to generate prompt sentences for the report based on sales and inventory information provided by the user, and then presents the results to the user.

[0543] The terminal functions as a user interface, allowing users to input sales and inventory information. It also displays the initial draft of reports returned from the server, providing an interface for users to input corrections and improvement suggestions. This allows users to review the report content, make necessary revisions, and finalize the version.

[0544] Users can easily input daily work information using their smartphones and review reports generated based on that information. The feedback entered by users is sent to the server and used to improve the system in the future.

[0545] The primary hardware used is a smartphone, and the software implements a GPT model using the OpenAI API. The server processes this information and efficiently manages sales and inventory data. As a concrete example, assuming a situation where sales of 500,000 yen and inventory of 20 units are managed on a given day, an example of a prompt message would be as follows:

[0546] Example prompt: "Today's sales were 500,000 yen, and inventory is 20 units. Please create a business report based on this information."

[0547] The reports generated in this way are used as an important source of information for planning the next day's work and placing orders.

[0548] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0549] Step 1:

[0550] Users input sales and inventory information using their smartphones. The entered data is sent to the server as basic information for creating daily business reports. Specifically, users enter sales figures and inventory status into the input fields on the screen and press the "Submit" button to send the data to the server.

[0551] Step 2:

[0552] The server generates a prompt message based on the received sales and inventory information. This prompt message clarifies the information needed to create the initial draft of the report and prepares it for the AI ​​generation model. Specifically, it converts the input numerical data into natural language sentences and formats them to something like, "Today's sales were ¥XX, and inventory is XX units. Please create a report."

[0553] Step 3:

[0554] The server feeds the generated prompt sentences into a generative AI model to generate a first draft of the report. This process uses the OpenAI API to input the prompt sentences into the model and create the necessary report text data. The output is a draft of the report in natural language obtained from the generative AI model.

[0555] Step 4:

[0556] The initial draft of the generated report is sent from the server to the terminal and displayed to the user. The user can review it and, if necessary, make corrections or comments on the terminal screen. User feedback and correction instructions are sent to the server via the interaction mechanism.

[0557] Step 5:

[0558] The server receives feedback from users and stores it in a storage system along with past revision history. This data is used to improve the performance of the AI ​​model when creating the next report. This process continuously improves the quality of the reports and the efficiency of the generation process.

[0559] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0560] This invention optimizes the report generation process by integrating natural language generation technology, user feedback, and emotion recognition technology. The system is primarily implemented using three elements: a server, a terminal, and a user.

[0561] server

[0562] The server hosts natural language generation technology to generate the initial draft of reports and automatically generates reports based on generation requests from users. The server also features an emotion engine that analyzes the user's emotional state and adjusts the tone and style of the initial draft based on that data. The server also has a learning mechanism to continuously improve its generation algorithm by accumulating feedback and emotion data.

[0563] terminal

[0564] The terminal provides an interface, creating an environment where users can easily interact with the system. In addition to displaying user input and responses from the server, it has the ability to send user emotions to the server in real time.

[0565] User

[0566] Users input their report creation requests into the system via their terminal. Once the initial draft is sent from the server, the user reviews its contents and requests revisions as needed. User feedback and sentiment data are sent to the server and used to improve the accuracy of future proposals.

[0567] Specific example

[0568] For example, if a user wants to create an "annual report," they input the theme and goals into the server via their terminal. The server generates a first draft using natural language generation technology, while simultaneously analyzing the user's emotional state with an emotion engine and adjusting the document's tone accordingly. This process generates a more natural report that reflects the user's intentions and emotions. Users can request revisions, and in doing so, they also send feedback, including emotional data, from their terminal to the server, allowing the system to further improve the accuracy of future reports.

[0569] Thus, the present invention provides a method that simultaneously improves the efficiency and quality of report creation by taking user emotions into consideration.

[0570] The following describes the processing flow.

[0571] Step 1:

[0572] The user starts up their device and accesses the report creation application. Next, they enter the theme and specific requirements of the report they want to create and request the generation of the initial draft.

[0573] Step 2:

[0574] The terminal converts the information provided by the user into data packets and sends them to the server. These data packets contain detailed information about the theme and the user's requests.

[0575] Step 3:

[0576] The server receives data packets and generates a first draft of the report using natural language generation technology. This generation process uses a pre-configured algorithm and constructs the content as needed.

[0577] Step 4:

[0578] Simultaneously, the server uses an emotion engine to analyze emotional data sent from the user's device. The results of this analysis are reflected in the tone and style of the report.

[0579] Step 5:

[0580] The generated initial draft and the adjusted tone information are sent from the server to the terminal. The terminal receives this information and displays it visually to the user.

[0581] Step 6:

[0582] Users review the initial draft on their device and provide feedback and revision suggestions for areas they feel need correction. During this process, the device also records the user's real-time emotions.

[0583] Step 7:

[0584] The device sends user feedback and sentiment data to the server. The server analyzes this data, referring to the revision history and accumulated feedback, to generate optimal revision suggestions.

[0585] Step 8:

[0586] The server sends the generated correction suggestions to the terminal. The terminal then presents them to the user, allowing them to freely apply the corrections.

[0587] Step 9:

[0588] The user finalizes the report based on the provided revision suggestions. After completion, they provide feedback on the quality of the suggestions and the report via their device.

[0589] Step 10:

[0590] The terminal sends the final feedback to the server. The server stores this feedback in a database and uses it to improve the accuracy of suggestions in the future.

[0591] Through this series of processes, the system can effectively utilize user emotions and feedback to improve the quality and efficiency of report generation.

[0592] (Example 2)

[0593] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0594] In report creation, traditional systems have made it difficult to efficiently produce documents that fully reflect the user's emotions and intentions. There is a need to quickly produce high-quality reports by appropriately adjusting the tone and style of the report while taking user feedback and emotional states into consideration.

[0595] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0596] In this invention, the server includes a generation means for generating a first draft of a report using natural language generation technology, an analysis means for analyzing the user's emotional state, and an adjustment means for adjusting the tone and style using feedback and emotional data. This makes it possible to efficiently generate high-quality reports that reflect the user's intentions and emotions.

[0597] "Natural language generation technology" refers to the technology that enables computer systems to automatically generate human language.

[0598] The "generation method" refers to the part that has the function of automatically creating the first draft of the report using natural language generation technology.

[0599] "Storage methods" refer to the part of the system that stores past revision history and user feedback in a database and has the function to analyze them.

[0600] "Interaction mechanisms" refer to the parts that have functions for presenting modification suggestions to the user and collecting feedback from the user.

[0601] "Adjustment mechanisms" refer to the part of the report that uses feedback and sentiment data to adjust the tone and style of the report.

[0602] The "analysis means" refers to the part of the system that analyzes user emotional data and applies that information to other functions of the system.

[0603] The "learning method" refers to the part that uses accumulated feedback and sentiment data to improve the accuracy and proposal capabilities of the initial draft generation algorithm.

[0604] "Revision means" refers to the part that has the function of revising the generated initial draft based on the user's requests and feelings.

[0605] "Optimization means" refers to the part that has the function of retraining or adjusting the generation technology based on user feedback and sentiment data to improve performance.

[0606] This invention is a system that integrates natural language generation technology, user feedback, and emotion recognition technology to optimize the report creation process. This system mainly consists of three elements: a server, a terminal, and a user.

[0607] server

[0608] The server hosts a natural language generation AI model and generates a draft of the report based on user prompts. For this purpose, the server utilizes a natural language processing framework, such as the GPT model. The generated draft is temporarily stored in storage. In addition, the server has an emotion engine that analyzes the user's emotional state using, for example, an emotion analysis API. Based on the results, the server adjusts the tone and style of the report according to the user's emotions.

[0609] terminal

[0610] The terminal provides a user interface and sets up an environment for the user to interact with the system. The terminal sends the prompt text entered by the user to the server and displays the initial draft of the generated report. Furthermore, the terminal has an emotion recognition function, which uses sensors to analyze the user's emotions in real time and sends this information to the server.

[0611] User

[0612] Users submit report creation requests through their terminals. For example, they might enter a prompt such as, "Please create an annual report on the company's performance for fiscal year 2023. The tone should be formal yet positive." Once the initial draft is sent from the server to the terminal, the user can review its contents and provide specific revision requests or emotional feedback.

[0613] As a concrete example, consider a scenario where a user wants to create an "annual report." The user enters prompt text into the terminal, and the server generates a first draft of the report based on those instructions. During the generation process, the tone is adjusted to take into account the user's emotional state, resulting in a more natural report that aligns with the user's intentions. This allows the user to efficiently create a high-quality report.

[0614] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0615] Step 1:

[0616] The user inputs the information necessary for report creation as prompt messages into the terminal. These prompt messages include specific themes, tones, and goals. The entered prompt messages are then sent from the terminal to the server.

[0617] Step 2:

[0618] The server automatically generates a first draft of the report using a natural language generation AI model based on the received prompt text. In this process, it analyzes the prompt text provided as input and outputs a logically structured document according to its content.

[0619] Step 3:

[0620] The server uses an emotion engine to analyze the user's emotional data. It receives emotional data sent from the terminal as input and uses an emotion analysis API to evaluate the user's current emotional state. This evaluation result is used to adjust the tone and style of the report.

[0621] Step 4:

[0622] The server adjusts the tone and style of the report based on the sentiment analysis results. This adjustment generates an appropriate report that matches the user's emotions, and the adjusted content is sent to the terminal as the final output.

[0623] Step 5:

[0624] The user reviews the initial draft of the report sent from the server on their device. The user then inputs revision requests or sentiment feedback for specific parts of the report, as needed. This feedback is then sent back to the server for further adjustments and improvements.

[0625] Step 6:

[0626] The server readjusts the report based on the received feedback and revision requests. Specifically, it analyzes the feedback, applies it again to the natural language generation AI model, and generates a revised report. The improved report is then sent to the terminal as the final deliverable.

[0627] Step 7:

[0628] The server uses accumulated feedback and sentiment data to train its generative AI model. This improves the accuracy and adaptability of subsequent report generation. By using feedback and sentiment analysis results as input and executing the learning process, the quality of suggestions is continuously improved.

[0629] (Application Example 2)

[0630] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0631] In modern society, generating personalized content that responds to the diverse feelings and needs of users is a crucial challenge. In particular, content distribution services require more natural and effective communication by adjusting content to reflect the user's emotional state. However, current systems struggle to generate content that appropriately reflects user emotions, making it difficult to improve user satisfaction. Therefore, it is necessary to enable the adjustment of tone and style according to the user's emotional state and to efficiently generate personalized content.

[0632] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0633] In this invention, the server includes a generation means for generating an initial draft of a document using natural language generation technology, an storage means for accumulating and analyzing past revision history and user feedback, and an adjustment means for recognizing the user's emotional state and adjusting the tone and style of the document based on that. This makes it possible to generate personalized content that reflects the user's emotional state.

[0634] "Natural language generation technology" is a technology that allows computers to automatically generate text using human language.

[0635] The term "first draft of a document" refers to the initial, basic form of the document, before any further revisions or adjustments are made.

[0636] "Generation means" refers to a device or software that has the function of creating a first draft of a document using natural language generation technology.

[0637] A "storage method" refers to a device or software that has the function of storing past revision history and user feedback in a database, etc., and analyzing this data.

[0638] "Opinions" refer to feedback and suggestions for corrections that users provide to the system.

[0639] A "dialogue means" is a device or software that interactively exchanges information with the user and presents suggestions for correction.

[0640] A "learning tool" is a device or software that has the function of adjusting the system to improve the accuracy of natural language generation based on collected opinions.

[0641] "Emotional state" refers to the user's current feelings and psychological state, and recognizing this is an element that allows for adjustment of the tone and style of the content.

[0642] "Adjustment means" refers to a device or software that has the function of appropriately changing the tone and style of a generated document based on the user's emotional state.

[0643] To realize this invention, a system is constructed by appropriately combining three elements: a server, a terminal, and a user.

[0644] The server has the functionality to generate a first draft of a document using natural language generation technology. Specifically, it can use a natural language generation API (e.g., OpenAI GPT) to automatically generate a document in response to a user's request, using prompt text as input data. The server also analyzes the user's emotional state using emotion recognition technology and uses the emotion data to adjust the tone and style of the document. The server uses TensorFlow or similar tools to build an emotion recognition model. Furthermore, the server accumulates feedback and emotion data, and uses this information to improve the generation method through learning mechanisms.

[0645] The terminal provides an interface with the user and enables communication with the server. Using smart glasses or a dedicated device, users can input and send emotional data to the server in real time. The terminal easily collects user feedback, sends it to the server, and uses it to improve future content creation.

[0646] Users use the system to send document creation requests to the server via their terminal. For example, while relaxing in a park, they can request the creation of a document in a relaxed tone that matches their mood. The feedback and mood data provided by the user greatly contribute to the subsequent content generation on the server.

[0647] For example, if a user requests a "relaxing article," one could send a prompt message to the AI ​​model saying, "This user is relaxed. Please generate an article with relaxing content that matches this state."

[0648] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0649] Step 1:

[0650] The user enters a prompt message via smart glasses or a device, requesting the server to generate specific content. The entered prompt message is sent to the server as initial data to reflect the content and tone of the document desired by the user.

[0651] Step 2:

[0652] The device uses built-in sensors and cameras to analyze the user's real-time emotional state and transmits that data to a server. This emotional data is then used to adjust the tone and style of documents. Input data includes the user's facial expressions and voice tone, while output is numerical or categorical data corresponding to that emotion.

[0653] Step 3:

[0654] The server uses a generative AI model to generate a first draft of a document based on the received prompt text and sentiment data. This step uses natural language generation technology to optimize the tone to match the sentiment. The input is the prompt text and sentiment data, and the output is the first draft of the generated document.

[0655] Step 4:

[0656] The server sends the generated initial draft to the terminal and requests the user to review its contents. The user reviews the initial draft on the terminal and provides feedback. This feedback includes requests for document revisions as needed. The input consists of the user's feedback comments and revision instructions.

[0657] Step 5:

[0658] The server analyzes user feedback and modifies the document as needed. This process allows the server to learn and improve the accuracy of its suggestions. The input at this stage is feedback data, and the output is the modified document.

[0659] Step 6:

[0660] Finally, the server sends the revised document to the terminal and requests the user's final confirmation. The user reviews the final version of the document and approves or requests further revisions. The output at this step is the completed document.

[0661] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0662] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0663] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0664] [Fourth Embodiment]

[0665] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0666] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0667] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0668] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0669] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0670] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0671] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0672] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0673] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0674] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0675] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0676] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0677] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0678] This system consists of servers, terminals, and users, and each entity works together to efficiently and accurately produce reports. The details are described below.

[0679] server

[0680] The server first utilizes natural language generation technology to generate an initial draft of a report based on requests received from users via their devices. This generation process leverages a pre-trained AI model to automatically create text based on themes and guidelines specified by the user. The server also manages past user revision history and feedback, storing it in a database. This data later becomes a crucial resource for improving the AI ​​model's performance.

[0681] terminal

[0682] The terminal provides the user interface, allowing users to easily interact with the system. All requests for initial draft generation and revisions are made through this interface. Information entered by the user and results returned from the server are displayed on the terminal to assist in user decision-making. In addition, the terminal also functions as a relay for sending feedback information to the server.

[0683] User

[0684] The user requests the generation of the initial draft of the report via their terminal and receives the automatically generated document from the server. They review the received initial draft and, if necessary, request revision suggestions from the server. After considering the suggested revisions, the user makes the necessary corrections and finalizes the version. The user also submits feedback on the provided revision suggestions, and this information is used to improve the accuracy of future suggestions.

[0685] Specific example

[0686] For example, if a user wants to create a "quarterly performance report," they specify a theme via their terminal and request the server to generate an initial draft. The server automatically generates a draft of the report using natural language generation technology and sends it to the user. The user reviews the draft and requests revisions via their terminal for any areas that need correction. The server analyzes past feedback, generates the most suitable revisions, and sends them back to the terminal. Finally, the user finalizes the report based on the suggestions, provides feedback, and completes the process.

[0687] Thus, the present invention improves the quality and efficiency of the report creation process while automating it by incorporating natural language generation technology and user feedback.

[0688] The following describes the processing flow.

[0689] Step 1:

[0690] The user starts up their device and opens the report creation application. Next, the user enters the report's theme and necessary information, and requests the generation of the first draft.

[0691] Step 2:

[0692] The terminal sends the information entered by the user to the server as data packets. This includes details and specific requirements for the report to be generated.

[0693] Step 3:

[0694] The server analyzes the received data packets and generates a preliminary draft of the report using natural language generation technology. The generated preliminary draft is based on a pre-configured algorithm and data.

[0695] Step 4:

[0696] The server sends the generated initial draft to the terminal. The terminal displays it in a format that is easy for the user to understand.

[0697] Step 5:

[0698] Users review the initial draft on their devices and add comments and revision suggestions for areas that need correction. This information will be used in the next step.

[0699] Step 6:

[0700] The terminal sends correction suggestions from the user to the server. The server analyzes these suggestions and generates the optimal suggestions by referring to the accumulated correction history and feedback data.

[0701] Step 7:

[0702] The server generates suggested fixes based on the analysis results and sends them back to the terminal. The terminal presents the suggestions to the user and prompts them to apply the necessary fixes.

[0703] Step 8:

[0704] The user makes revisions based on the server's suggestions and completes the final report. After completion, they provide feedback on the effectiveness of the suggestions.

[0705] Step 9:

[0706] The device sends user feedback to the server. The server stores the received feedback in a database and uses it for future improvements.

[0707] Through this process, the system will be continuously improved, resulting in more accurate suggestions being provided in subsequent report generation.

[0708] (Example 1)

[0709] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0710] In document creation, such as reports, a challenge is enabling users to efficiently produce high-quality initial drafts. Conventional systems often require extensive manual revisions due to insufficient accuracy and efficiency in document generation. Furthermore, it is difficult to improve the accuracy of the generation model by fully utilizing user feedback.

[0711] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0712] In this invention, the server includes a generation means for generating an initial draft of a document using natural language, a storage means for recording and analyzing past revision history and user evaluations, and an interaction means for presenting revision proposals to the user and collecting user evaluations. This makes it possible to provide highly accurate document generation and efficient revision suggestions, and to significantly reduce manual work for the user.

[0713] "Natural language" refers to the language that humans use on a daily basis, and it is analyzed in order to enable information processing by computers.

[0714] "Generation means" refers to technologies and devices that automatically generate the initial draft of a document using natural language processing techniques based on specified guidelines.

[0715] "Storage methods" refer to technologies and devices for recording past revision history and user feedback, and for analyzing this data.

[0716] "Interaction means" refers to interfaces and technologies for presenting revised versions to users and collecting feedback from them.

[0717] "Learning methods" refer to technologies and devices that use collected feedback to adjust and enhance AI models in order to improve the accuracy and quality of the documents they generate.

[0718] "Display means" refers to technologies including screens and devices used to present generated initial drafts, revised versions, etc., to users.

[0719] "Input means" refers to interfaces or devices that allow users to input requests or modification suggestions into the system.

[0720] This invention is a document generation system in which a server, a terminal, and a user work together. The server uses a pre-trained generative AI model to generate a first draft of a document based on themes and guidelines received from the user. For example, a specific AI framework or API can be used as the model based on natural language generation technology. The generated first draft is presented to the user via the terminal.

[0721] When generating this initial draft, the server accumulates past revision history and user feedback, which is used to adjust and improve the generation model. This data is stored in a database on the server, and the model is retrained and optimized as needed through learning mechanisms.

[0722] The terminal provides a user interface and a means for the user to input themes and guidelines for reports. It also displays the initial draft sent from the server to the user, allowing them to input revision suggestions and feedback. For example, if a user wants to create a "Quarterly Performance Report," they could input the theme into the terminal and send a prompt message to the server requesting the generation of the initial draft.

[0723] The user requests the server to generate an initial draft via their device and reviews the suggested revisions and feedback from the server. Based on this feedback, the user revises the report and finalizes the document on their device. The device collects the user's feedback and sends it to the server, contributing to improving the accuracy of the AI ​​model in subsequent iterations.

[0724] As an example of a prompt, you can send the following to the server: "Generate the first draft of the quarterly performance report based on the following themes: Sales trends and market analysis." This allows the system to automatically generate documents that match the user's requests, streamlining manual work.

[0725] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0726] Step 1:

[0727] The terminal accepts the report's theme and guidelines as input from the user. The user enters a specific theme, such as "Quarterly Performance Report." The terminal receives this input and formats it as a prompt. The formatted prompt is then prepared as a request to the server.

[0728] Step 2:

[0729] The terminal sends a formatted prompt message to the server. This prompt message becomes the input, and the server uses it to begin the document generation process. In this process, the terminal's role is to send data to the server reliably and quickly.

[0730] Step 3:

[0731] The server uses a pre-trained generative AI model to generate a first draft of a document, taking the received prompt text as input. During this process, the AI ​​model uses algorithms to construct the text based on the content of the prompt text. The generated first draft is temporarily stored on the server.

[0732] Step 4:

[0733] The server sends the generated initial draft to the terminal. The server's output is the initial draft, which is then returned to the terminal. The core of the server's operation is an accurate and efficient generation process using a model.

[0734] Step 5:

[0735] The terminal presents the initial draft sent from the server to the user. The user receives the initial draft in a visually verifiable format. The terminal formats the initial draft appropriately for easy viewing.

[0736] Step 6:

[0737] The user reviews the initial draft and inputs any necessary corrections or additional instructions into the terminal. The user's feedback and suggested revisions are then used as input for the next data processing step.

[0738] Step 7:

[0739] The terminal sends correction requests from the user to the server. This data is analyzed by the server and used to generate further correction suggestions. The terminal's function is to accurately transmit the data to the server.

[0740] Step 8:

[0741] The server receives user feedback and compares it with past data to generate optimal correction suggestions. At this stage, the server runs an optimization algorithm based on the correction history and feedback data. The generated correction suggestions become the new output.

[0742] Step 9:

[0743] The server sends the generated proposed corrections to the terminal. The server's role is to utilize its accumulated knowledge to return useful information to the user.

[0744] Step 10:

[0745] Users review the proposed revisions via their devices and incorporate them into the final document. The final version is then finalized and saved on the device as a completed report. It is crucial that the user's intentions are accurately reflected.

[0746] (Application Example 1)

[0747] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0748] In creating business reports and inventory management reports, there is a need to improve both efficiency and accuracy simultaneously. In particular, in the retail industry, it is essential to grasp sales and inventory status in real time and quickly create reports. However, manual report creation is time-consuming, labor-intensive, and prone to errors. Therefore, technology is needed to achieve a high-quality and efficient report creation process.

[0749] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0750] In this invention, the server includes a generation means for generating an initial draft of a report using natural language generation technology, an storage means for accumulating and analyzing past revision history and feedback, and an interaction means for presenting revision suggestions to the user and collecting feedback from the user. This enables efficient processing of sales information and inventory information and accurate report creation.

[0751] "Natural language generation technology" is a technology that allows computers to understand human language and generate sentences in a natural-sounding form.

[0752] "Generation means" refers to a device or process that has the function of automatically creating the first draft of a report using natural language generation technology.

[0753] A "storage device" refers to a device or process equipped with a mechanism for collecting and storing past revision history and feedback information, and for utilizing it for subsequent analysis and improvement.

[0754] "Interaction means" refers to technologies that provide an interface for a system and a user to exchange information and have the function of collecting feedback from the user.

[0755] "Business information processing means" refers to a system or method for preparing data necessary for report creation by inputting sales information and inventory information in digital format and processing it quickly and accurately.

[0756] A "learning method" is a process or device for training an AI model to improve the accuracy of the generation method's suggestions based on feedback collected from users.

[0757] The system for realizing this invention involves the coordinated operation of a server, a terminal, and a user. The server is responsible for generating the initial draft of the report using natural language generation technology. Specifically, it uses a pre-trained generative AI model to generate prompt sentences for the report based on sales and inventory information provided by the user, and then presents the results to the user.

[0758] The terminal functions as a user interface, allowing users to input sales and inventory information. It also displays the initial draft of reports returned from the server, providing an interface for users to input corrections and improvement suggestions. This allows users to review the report content, make necessary revisions, and finalize the version.

[0759] Users can easily input daily work information using their smartphones and review reports generated based on that information. The feedback entered by users is sent to the server and used to improve the system in the future.

[0760] The primary hardware used is a smartphone, and the software implements a GPT model using the OpenAI API. The server processes this information and efficiently manages sales and inventory data. As a concrete example, assuming a situation where sales of 500,000 yen and inventory of 20 units are managed on a given day, an example of a prompt message would be as follows:

[0761] Example prompt: "Today's sales were 500,000 yen, and inventory is 20 units. Please create a business report based on this information."

[0762] The reports generated in this way are used as an important source of information for planning the next day's work and placing orders.

[0763] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0764] Step 1:

[0765] Users input sales and inventory information using their smartphones. The entered data is sent to the server as basic information for creating daily business reports. Specifically, users enter sales figures and inventory status into the input fields on the screen and press the "Submit" button to send the data to the server.

[0766] Step 2:

[0767] The server generates a prompt message based on the received sales and inventory information. This prompt message clarifies the information needed to create the initial draft of the report and prepares it for the AI ​​generation model. Specifically, it converts the input numerical data into natural language sentences and formats them to something like, "Today's sales were ¥XX, and inventory is XX units. Please create a report."

[0768] Step 3:

[0769] The server feeds the generated prompt sentences into a generative AI model to generate a first draft of the report. This process uses the OpenAI API to input the prompt sentences into the model and create the necessary report text data. The output is a draft of the report in natural language obtained from the generative AI model.

[0770] Step 4:

[0771] The initial draft of the generated report is sent from the server to the terminal and displayed to the user. The user can review it and, if necessary, make corrections or comments on the terminal screen. User feedback and correction instructions are sent to the server via the interaction mechanism.

[0772] Step 5:

[0773] The server receives feedback from users and stores it in a storage system along with past revision history. This data is used to improve the performance of the AI ​​model when creating the next report. This process continuously improves the quality of the reports and the efficiency of the generation process.

[0774] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0775] This invention optimizes the report generation process by integrating natural language generation technology, user feedback, and emotion recognition technology. The system is primarily implemented using three elements: a server, a terminal, and a user.

[0776] server

[0777] The server hosts natural language generation technology to generate the initial draft of reports and automatically generates reports based on generation requests from users. The server also features an emotion engine that analyzes the user's emotional state and adjusts the tone and style of the initial draft based on that data. The server also has a learning mechanism to continuously improve its generation algorithm by accumulating feedback and emotion data.

[0778] terminal

[0779] The terminal provides an interface, creating an environment where users can easily interact with the system. In addition to displaying user input and responses from the server, it has the ability to send user emotions to the server in real time.

[0780] User

[0781] Users input their report creation requests into the system via their terminal. Once the initial draft is sent from the server, the user reviews its contents and requests revisions as needed. User feedback and sentiment data are sent to the server and used to improve the accuracy of future proposals.

[0782] Specific example

[0783] For example, if a user wants to create an "annual report," they input the theme and goals into the server via their terminal. The server generates a first draft using natural language generation technology, while simultaneously analyzing the user's emotional state with an emotion engine and adjusting the document's tone accordingly. This process generates a more natural report that reflects the user's intentions and emotions. Users can request revisions, and in doing so, they also send feedback, including emotional data, from their terminal to the server, allowing the system to further improve the accuracy of future reports.

[0784] Thus, the present invention provides a method that simultaneously improves the efficiency and quality of report creation by taking user emotions into consideration.

[0785] The following describes the processing flow.

[0786] Step 1:

[0787] The user starts up their device and accesses the report creation application. Next, they enter the theme and specific requirements of the report they want to create and request the generation of the initial draft.

[0788] Step 2:

[0789] The terminal converts the information provided by the user into data packets and sends them to the server. These data packets contain detailed information about the theme and the user's requests.

[0790] Step 3:

[0791] The server receives data packets and generates a first draft of the report using natural language generation technology. This generation process uses a pre-configured algorithm and constructs the content as needed.

[0792] Step 4:

[0793] Simultaneously, the server uses an emotion engine to analyze emotional data sent from the user's device. The results of this analysis are reflected in the tone and style of the report.

[0794] Step 5:

[0795] The generated initial draft and the adjusted tone information are sent from the server to the terminal. The terminal receives this information and displays it visually to the user.

[0796] Step 6:

[0797] Users review the initial draft on their device and provide feedback and revision suggestions for areas they feel need correction. During this process, the device also records the user's real-time emotions.

[0798] Step 7:

[0799] The device sends user feedback and sentiment data to the server. The server analyzes this data, referring to the revision history and accumulated feedback, to generate optimal revision suggestions.

[0800] Step 8:

[0801] The server sends the generated correction suggestions to the terminal. The terminal then presents them to the user, allowing them to freely apply the corrections.

[0802] Step 9:

[0803] The user finalizes the report based on the provided revision suggestions. After completion, they provide feedback on the quality of the suggestions and the report via their device.

[0804] Step 10:

[0805] The terminal sends the final feedback to the server. The server stores this feedback in a database and uses it to improve the accuracy of suggestions in the future.

[0806] Through this series of processes, the system can effectively utilize user emotions and feedback to improve the quality and efficiency of report generation.

[0807] (Example 2)

[0808] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0809] In report creation, traditional systems have made it difficult to efficiently produce documents that fully reflect the user's emotions and intentions. There is a need to quickly produce high-quality reports by appropriately adjusting the tone and style of the report while taking user feedback and emotional states into consideration.

[0810] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0811] In this invention, the server includes a generation means for generating a first draft of a report using natural language generation technology, an analysis means for analyzing the user's emotional state, and an adjustment means for adjusting the tone and style using feedback and emotional data. This makes it possible to efficiently generate high-quality reports that reflect the user's intentions and emotions.

[0812] "Natural language generation technology" refers to the technology that enables computer systems to automatically generate human language.

[0813] The "generation method" refers to the part that has the function of automatically creating the first draft of the report using natural language generation technology.

[0814] "Storage methods" refer to the part of the system that stores past revision history and user feedback in a database and has the function to analyze them.

[0815] "Interaction mechanisms" refer to the parts that have functions for presenting modification suggestions to the user and collecting feedback from the user.

[0816] "Adjustment mechanisms" refer to the part of the report that uses feedback and sentiment data to adjust the tone and style of the report.

[0817] The "analysis means" refers to the part of the system that analyzes user emotional data and applies that information to other functions of the system.

[0818] The "learning method" refers to the part that uses accumulated feedback and sentiment data to improve the accuracy and proposal capabilities of the initial draft generation algorithm.

[0819] "Revision means" refers to the part that has the function of revising the generated initial draft based on the user's requests and feelings.

[0820] "Optimization means" refers to the part that has the function of retraining or adjusting the generation technology based on user feedback and sentiment data to improve performance.

[0821] This invention is a system that integrates natural language generation technology, user feedback, and emotion recognition technology to optimize the report creation process. This system mainly consists of three elements: a server, a terminal, and a user.

[0822] server

[0823] The server hosts a natural language generation AI model and generates a draft of the report based on user prompts. For this purpose, the server utilizes a natural language processing framework, such as the GPT model. The generated draft is temporarily stored in storage. In addition, the server has an emotion engine that analyzes the user's emotional state using, for example, an emotion analysis API. Based on the results, the server adjusts the tone and style of the report according to the user's emotions.

[0824] terminal

[0825] The terminal provides a user interface and sets up an environment for the user to interact with the system. The terminal sends the prompt text entered by the user to the server and displays the initial draft of the generated report. Furthermore, the terminal has an emotion recognition function, which uses sensors to analyze the user's emotions in real time and sends this information to the server.

[0826] User

[0827] Users submit report creation requests through their terminals. For example, they might enter a prompt such as, "Please create an annual report on the company's performance for fiscal year 2023. The tone should be formal yet positive." Once the initial draft is sent from the server to the terminal, the user can review its contents and provide specific revision requests or emotional feedback.

[0828] As a concrete example, consider a scenario where a user wants to create an "annual report." The user enters prompt text into the terminal, and the server generates a first draft of the report based on those instructions. During the generation process, the tone is adjusted to take into account the user's emotional state, resulting in a more natural report that aligns with the user's intentions. This allows the user to efficiently create a high-quality report.

[0829] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0830] Step 1:

[0831] The user inputs the information necessary for report creation as prompt messages into the terminal. These prompt messages include specific themes, tones, and goals. The entered prompt messages are then sent from the terminal to the server.

[0832] Step 2:

[0833] The server automatically generates a first draft of the report using a natural language generation AI model based on the received prompt text. In this process, it analyzes the prompt text provided as input and outputs a logically structured document according to its content.

[0834] Step 3:

[0835] The server uses an emotion engine to analyze the user's emotional data. It receives emotional data sent from the terminal as input and uses an emotion analysis API to evaluate the user's current emotional state. This evaluation result is used to adjust the tone and style of the report.

[0836] Step 4:

[0837] The server adjusts the tone and style of the report based on the sentiment analysis results. This adjustment generates an appropriate report that matches the user's emotions, and the adjusted content is sent to the terminal as the final output.

[0838] Step 5:

[0839] The user reviews the initial draft of the report sent from the server on their device. The user then inputs revision requests or sentiment feedback for specific parts of the report, as needed. This feedback is then sent back to the server for further adjustments and improvements.

[0840] Step 6:

[0841] The server readjusts the report based on the received feedback and revision requests. Specifically, it analyzes the feedback, applies it again to the natural language generation AI model, and generates a revised report. The improved report is then sent to the terminal as the final deliverable.

[0842] Step 7:

[0843] The server uses accumulated feedback and sentiment data to train its generative AI model. This improves the accuracy and adaptability of subsequent report generation. By using feedback and sentiment analysis results as input and executing the learning process, the quality of suggestions is continuously improved.

[0844] (Application Example 2)

[0845] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0846] In modern society, generating personalized content that responds to the diverse feelings and needs of users is a crucial challenge. In particular, content distribution services require more natural and effective communication by adjusting content to reflect the user's emotional state. However, current systems struggle to generate content that appropriately reflects user emotions, making it difficult to improve user satisfaction. Therefore, it is necessary to enable the adjustment of tone and style according to the user's emotional state and to efficiently generate personalized content.

[0847] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0848] In this invention, the server includes a generation means for generating an initial draft of a document using natural language generation technology, an storage means for accumulating and analyzing past revision history and user feedback, and an adjustment means for recognizing the user's emotional state and adjusting the tone and style of the document based on that. This makes it possible to generate personalized content that reflects the user's emotional state.

[0849] "Natural language generation technology" is a technology that allows computers to automatically generate text using human language.

[0850] The term "first draft of a document" refers to the initial, basic form of the document, before any further revisions or adjustments are made.

[0851] "Generation means" refers to a device or software that has the function of creating a first draft of a document using natural language generation technology.

[0852] A "storage method" refers to a device or software that has the function of storing past revision history and user feedback in a database, etc., and analyzing this data.

[0853] "Opinions" refer to feedback and suggestions for corrections that users provide to the system.

[0854] A "dialogue means" is a device or software that interactively exchanges information with the user and presents suggestions for correction.

[0855] A "learning tool" is a device or software that has the function of adjusting the system to improve the accuracy of natural language generation based on collected opinions.

[0856] "Emotional state" refers to the user's current feelings and psychological state, and recognizing this is an element that allows for adjustment of the tone and style of the content.

[0857] "Adjustment means" refers to a device or software that has the function of appropriately changing the tone and style of a generated document based on the user's emotional state.

[0858] To realize this invention, a system is constructed by appropriately combining three elements: a server, a terminal, and a user.

[0859] The server has the functionality to generate a first draft of a document using natural language generation technology. Specifically, it can use a natural language generation API (e.g., OpenAI GPT) to automatically generate a document in response to a user's request, using prompt text as input data. The server also analyzes the user's emotional state using emotion recognition technology and uses the emotion data to adjust the tone and style of the document. The server uses TensorFlow or similar tools to build an emotion recognition model. Furthermore, the server accumulates feedback and emotion data, and uses this information to improve the generation method through learning mechanisms.

[0860] The terminal provides an interface with the user and enables communication with the server. Using smart glasses or a dedicated device, users can input and send emotional data to the server in real time. The terminal easily collects user feedback, sends it to the server, and uses it to improve future content creation.

[0861] Users use the system to send document creation requests to the server via their terminal. For example, while relaxing in a park, they can request the creation of a document in a relaxed tone that matches their mood. The feedback and mood data provided by the user greatly contribute to the subsequent content generation on the server.

[0862] For example, if a user requests a "relaxing article," one could send a prompt message to the AI ​​model saying, "This user is relaxed. Please generate an article with relaxing content that matches this state."

[0863] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0864] Step 1:

[0865] The user enters a prompt message via smart glasses or a device, requesting the server to generate specific content. The entered prompt message is sent to the server as initial data to reflect the content and tone of the document desired by the user.

[0866] Step 2:

[0867] The device uses built-in sensors and cameras to analyze the user's real-time emotional state and transmits that data to a server. This emotional data is then used to adjust the tone and style of documents. Input data includes the user's facial expressions and voice tone, while output is numerical or categorical data corresponding to that emotion.

[0868] Step 3:

[0869] The server uses a generative AI model to generate a first draft of a document based on the received prompt text and sentiment data. This step uses natural language generation technology to optimize the tone to match the sentiment. The input is the prompt text and sentiment data, and the output is the first draft of the generated document.

[0870] Step 4:

[0871] The server sends the generated initial draft to the terminal and requests the user to review its contents. The user reviews the initial draft on the terminal and provides feedback. This feedback includes requests for document revisions as needed. The input consists of the user's feedback comments and revision instructions.

[0872] Step 5:

[0873] The server analyzes user feedback and modifies the document as needed. This process allows the server to learn and improve the accuracy of its suggestions. The input at this stage is feedback data, and the output is the modified document.

[0874] Step 6:

[0875] Finally, the server sends the revised document to the terminal and requests the user's final confirmation. The user reviews the final version of the document and approves or requests further revisions. The output at this step is the completed document.

[0876] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0877] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0878] 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 this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0879] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0880] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0881] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0882] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0883] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0884] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0885] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0886] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0887] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0888] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0889] 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.

[0890] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0891] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0892] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0893] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0894] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0895] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0896] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0897] The following is further disclosed regarding the embodiments described above.

[0898] (Claim 1)

[0899] A generation method for generating the first draft of a report using natural language generation technology,

[0900] A means of accumulating and analyzing past revision history and feedback,

[0901] An interaction mechanism that presents correction suggestions to users and collects feedback from them,

[0902] A learning method that improves the accuracy of proposed generation methods using feedback,

[0903] A system that includes this.

[0904] (Claim 2)

[0905] The system according to claim 1, further comprising a means for modifying the generated initial draft based on the user's request.

[0906] (Claim 3)

[0907] The system according to claim 1, further comprising optimization means for retraining or adjusting the generation technique based on user feedback.

[0908] "Example 1"

[0909] (Claim 1)

[0910] A generation method for generating the initial draft of a document using natural language,

[0911] A means of storing and analyzing past revision history and user ratings,

[0912] An interaction mechanism that presents revised proposals to users and collects feedback from users,

[0913] A learning method that improves the accuracy of proposed generation methods using evaluation,

[0914] A display means for displaying the initial draft generated on the user's terminal,

[0915] An input method for receiving user correction requests,

[0916] A system that includes this.

[0917] (Claim 2)

[0918] The system according to claim 1, further comprising a means for modifying the generated initial draft based on the user's request.

[0919] (Claim 3)

[0920] The system according to claim 1, further comprising an optimization means for readjusting the generation technology based on user evaluation.

[0921] "Application Example 1"

[0922] (Claim 1)

[0923] A generation method for generating the first draft of a report using natural language generation technology,

[0924] A means of accumulating and analyzing past revision history and feedback,

[0925] An interaction mechanism that presents correction suggestions to users and collects feedback from them,

[0926] A business information processing system that inputs sales information and inventory information and generates reports,

[0927] A learning method that improves the accuracy of proposed generation methods using feedback,

[0928] A system that includes this.

[0929] (Claim 2)

[0930] The system according to claim 1, further comprising a means for modifying the generated initial draft based on the user's request.

[0931] (Claim 3)

[0932] The system according to claim 1, further comprising optimization means for retraining or adjusting the generation technique based on user feedback.

[0933] "Example 2 of combining an emotion engine"

[0934] (Claim 1)

[0935] A generation method for generating the first draft of a report using natural language generation technology,

[0936] A means of accumulating and analyzing past revision history and feedback,

[0937] An interaction mechanism that presents correction suggestions to users and collects feedback from them,

[0938] A means of adjusting tone and style using feedback and emotional data,

[0939] An analytical method for analyzing the user's emotional state,

[0940] A learning method that improves the accuracy of proposed generation methods using feedback and sentiment data,

[0941] A system that includes this.

[0942] (Claim 2)

[0943] The system according to claim 1, further comprising a means for modifying the generated initial draft based on the user's requests and feelings.

[0944] (Claim 3)

[0945] The system according to claim 1, further comprising optimization means for retraining or adjusting the generation technique based on user feedback and sentiment data.

[0946] "Application example 2 when combining with an emotional engine"

[0947] (Claim 1)

[0948] A generation method for generating the first draft of a document using natural language generation technology,

[0949] A means of accumulating and analyzing past revision history and user feedback,

[0950] A dialogue method for presenting revision suggestions to users and collecting feedback from them,

[0951] A learning method that improves the accuracy of proposed generation methods using opinions,

[0952] An adjustment mechanism that recognizes the user's emotional state and adjusts the tone and style of the document based on that,

[0953] ...

[0954] A system that includes this.

[0955] (Claim 2)

[0956] The system according to claim 1, further comprising a means for modifying the generated initial draft based on the user's request.

[0957] (Claim 3)

[0958] The system according to claim 1, further comprising optimization means for retraining or adjusting the generation technique based on user feedback. [Explanation of Symbols]

[0959] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A generation method for generating the first draft of a report using natural language generation technology, A means of accumulating and analyzing past revision history and feedback, An interaction mechanism that presents correction suggestions to users and collects feedback from them, A learning method that improves the accuracy of proposed generation methods using feedback, A system that includes this.

2. The system according to claim 1, further comprising a means for modifying the generated initial draft based on the user's request.

3. The system according to claim 1, further comprising optimization means for retraining or adjusting the generation technique based on user feedback.

Images