System

Abstract

A system is provided.SOLUTION: A system comprising: means for receiving and analyzing a natural language request; means for retrieving and identifying a plurality of suitable generative artificial intelligences based on the content of the analyzed request; means for invoking the identified plurality of generative artificial intelligences to perform a service according to the request, and providing the result to the user; means for confirming the user's satisfaction, and automating an event-service routine; and means for developing the automated routine as a Web application, thereby automatically processing future requests.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Corporate workflows and business processes involve many labor-intensive tasks, and the use of generative artificial intelligence (AI) is seen as a promising way to streamline them. However, there are a huge number of generative AIs, and searching, selecting, and combining the appropriate ones is time-consuming, contributing to reduced productivity. Furthermore, it is not easy to utilize the optimal AI to reduce errors and achieve operational efficiency and automation. Therefore, there is a need for a system that seamlessly integrates multiple AIs and supports workforce optimization. [Means for solving the problem]

[0005] The present invention provides the following means: a system including means for receiving and analyzing natural language requests, means for searching for and identifying multiple appropriate generative artificial intelligences based on the content of the analyzed request, means for calling the identified multiple generative artificial intelligences to execute a service according to the request and providing the result to the user, means for checking user satisfaction and automating an event-service routine, and means for deploying the automated routine as a web application to automatically process future requests. This allows companies to efficiently process labor-intensive tasks, reduce errors, and optimize operations.

[0006] A "natural language request" is a command or request entered by a user in spoken or written form.

[0007] "Parsing" refers to the process of understanding the content of a natural language request and extracting tasks and requirements.

[0008] "Generative artificial intelligence" refers to machine learning models that are trained to perform specific tasks.

[0009] "Search" refers to the process of finding a model that meets specific criteria within a database or store of generative artificial intelligence.

[0010] "Identification" means selecting the most suitable generative artificial intelligence from the search results.

[0011] "Performing a service" means using a specified generative artificial intelligence to accomplish a task in response to a request.

[0012] "User" refers to the person who interacts with the system, enters natural language requests, and reviews the results.

[0013] "User satisfaction" refers to a user's evaluation of the results of the provided service.

[0014] An "event-service routine" refers to a set of procedures that automatically executes a service in response to a particular event.

[0015] "Automation" means configuring a system to perform repetitive tasks without human intervention.

[0016] "Web Application" means a software application available over the Internet.

[0017] "Voice, text, sensor, date and time" refers to some of the event input formats.

[0018] "Text, images, videos, spreadsheets, code, and documents" refer to some of the output formats of the services offered. [Brief explanation of the drawings]

[0019] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0020] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

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

[0022] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0023] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0024] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0025] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0026] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0027] [First embodiment]

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

[0029] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0030] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0031] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0032] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0033] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0034] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

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

[0036] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0037] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0038] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0039] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0040] The present invention is a system in which a user inputs a request in natural language, a server analyzes the request, calls an appropriate generative artificial intelligence, and ultimately provides a service to the user. Specific embodiments of the present invention will be described below.

[0041] Basic system configuration

[0042] 1. User Input

[0043] The user inputs a request in natural language using a terminal, such as "Please create a schedule for the next project meeting and notify the relevant parties."

[0044] 2. Receiving and parsing the request

[0045] The server receives a natural language request from the user, which is then parsed by a natural language processing (NLP) module to extract the gist of the request, which includes "schedule" and "notify relevant parties."

[0046] 3. Searching and Identifying Generative AI

[0047] Based on the analysis of the request, the server searches a database of generative AI models for suitable models, for example, identifying a "generative AI for scheduling" and a "generative AI for notifications."

[0048] 4. Calling generative AI and providing services

[0049] The server calls the identified generative AI and executes the service according to the request. The generative AI for schedule creation generates a schedule for the project meeting, and the generative AI for notification notifies the relevant parties of the schedule.

[0050] 5. User satisfaction check

[0051] The server provides the results of the service performed by the generative AI to the user and confirms their satisfaction. For example, the user checks the provided schedule and notification content and provides feedback on whether they are satisfied.

[0052] 6. Routine Automation and Web Application Deployment

[0053] If the user is satisfied, the server automates the event-service routine, which is then deployed as a web application to automatically process similar requests in the future.

[0054] Specific examples

[0055] For example, if a user requests, "Please create a schedule for a project meeting next Monday and notify the relevant parties," the following processing is performed.

[0056] 1. Input Reception

[0057] The user inputs a request through the terminal.

[0058] 2. Request Analysis

[0059] The server receives the request, analyzes it, and extracts the information that needs to be "scheduled" and "notified."

[0060] 3. Identifying the right generative AI

[0061] The server searches for and identifies "generative artificial intelligence for schedule creation" and "generative artificial intelligence for notifications."

[0062] 4. Calling generative AI and providing services

[0063] The server calls each generative AI, creates a schedule, and notifies relevant parties.

[0064] 5. Check the results

[0065] The server provides the generated schedule and notification contents to the user, who then confirms them.

[0066] 6. Deployment of automated routines

[0067] If the user is satisfied, the server automates the process and deploys it as a web application.

[0068] In this way, the present invention efficiently operates generative artificial intelligence based on the user's natural language requests, thereby optimizing workflow.

[0069] The processing flow will be explained below.

[0070] Step 1:

[0071] A user inputs a natural language request using a terminal, for example, a request such as "Please create a schedule for the next project meeting and notify the relevant parties."

[0072] Step 2:

[0073] The terminal sends this natural language request to the server, which transmits the request over the Internet and receives it at the server's receiving module.

[0074] Step 3:

[0075] The server passes the received natural language request to an NLP (Natural Language Processing) module for analysis. This analysis module understands the content of the request and extracts related tasks and requirements. For example, "schedule creation" and "notify relevant parties" are extracted as tasks.

[0076] Step 4:

[0077] The server searches a database of generative artificial intelligences based on the extracted tasks. This search module accesses a database such as the OpenAI (registered trademark) store and identifies "generative artificial intelligences for scheduling" and "generative artificial intelligences for notifications."

[0078] Step 5:

[0079] The server then calls the identified generative AI and performs specific processing in response to the user's request. For example, a generative AI for schedule creation generates a schedule for a project meeting and returns the schedule data. Similarly, a generative AI for notification notifies relevant parties of the generated schedule.

[0080] Step 6:

[0081] The server compiles the generated schedule and notification results and provides feedback to the user. The user can then check the results of the services provided by the server through their terminal and evaluate their level of satisfaction.

[0082] Step 7:

[0083] If the user is satisfied with the results of the provided service, the server automates the event-service routine. The automation module does this and implements the routine as a web application.

[0084] Step 8:

[0085] The server provides the built web application to the user's device and automatically processes similar requests in the future. The web application receives events in the form of voice, text, sensors, date and time, and provides services in the form of text, images, videos, spreadsheets, code, and documents.

[0086] Thus, through each step centered on the server, the terminal, and the user, the present invention provides an efficient and flexible service for natural language requests from users.

[0087] Example 1

[0088] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0089] In conventional systems, the process of users entering requests in natural language, searching for and identifying appropriate generative AI models based on those requests, and providing services was not automated, making it difficult to achieve an efficient workflow. It was also difficult to build a system that could confirm satisfaction with the generated results and automatically respond to similar requests in the future. This resulted in a major issue of the time and effort required for the process.

[0090] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0091] In this invention, the server includes means for receiving and analyzing a natural language request, means for searching for and identifying multiple appropriate generative AI models based on the content of the analyzed request, means for extracting a gist of the request using a natural language processing module, means for searching for and selecting a generative AI model from a database based on the extracted gist, means for calling the selected generative AI and providing a specific service in response to the request, means for providing the generated results to a user and confirming the user's satisfaction, means for automating the post-confirmation routine and deploying it as a Web application, and means for publishing it as a Web application for automatic processing of subsequent requests. This makes it possible to efficiently and automatically operate a generative AI model based on a user's natural language request, significantly reducing the effort and time required for the work.

[0092] A "natural language request" is a request or instruction entered by a user using human language.

[0093] "Parsing" is the act of receiving a natural language request and converting its content into an understandable format.

[0094] "Generative AI" is a type of AI model that is trained to perform specific tasks and has the ability to generate new information and results.

[0095] "Searching" is the act of finding items in a database or repository that match specific criteria.

[0096] "Identifying" is the act of selecting the most appropriate item from those found in a search.

[0097] A "natural language processing module" is a program or tool that converts natural language into a format that is easy for a computer to understand.

[0098] The "Summary" is a brief summary of the main content and purpose of the request.

[0099] A "database" is a system that stores large amounts of data in an organized manner and makes it easy to search and manage.

[0100] "Invocation" refers to sending a command or operation to execute a program or service.

[0101] "Satisfaction confirmation" is the act of checking whether the provided services and results meet the user's expectations.

[0102] A "routine" is a series of processing steps in a program that defines certain operations and processing procedures.

[0103] "Automation" refers to making a system operate automatically without human intervention.

[0104] A "web application" is application software that is used over the Internet and operates via a web browser.

[0105] The present invention is a system that allows a user to input a natural language request, and then calls an appropriate generative AI based on the request to provide a service. This system is primarily composed of a server, a terminal, and a generative AI model. A specific implementation method is described below.

[0106] A user inputs a request in natural language using a device (e.g., a personal computer, smartphone, tablet, etc.). This request is sent to the server via a web browser. A specific example is a request such as, "Please create a schedule for the next project meeting and notify the relevant parties."

[0107] The server uses natural language processing modules such as Python's NLTK or spaCy to parse the incoming request, extracting the gist of the request, such as "schedule" and "notify relevant parties."

[0108] Next, the server searches and selects an appropriate model from a database of generative artificial intelligence models based on the extracted summary. It is possible to use OpenAI's GPT-3 (registered trademark) for schedule creation and the Twilio API for notifications.

[0109] The selected generative AI model is then called by the server to provide a specific service: first, GPT-3 is used to generate a schedule for a project meeting, and then the generated schedule is notified to relevant parties using the Twilio API.

[0110] The generated results are provided to the user by the server, who checks them through their terminal. If the user is satisfied with the results, they send their feedback to the server, which uses this feedback to automate the routine.

[0111] Finally, the automated routine can be deployed as a web application, so that similar requests can be processed automatically in the future. For example, the web application can be built using the Python Flask framework.

[0112] Examples of prompt statements

[0113] Prompt for scheduling a project meeting:

[0114] "Please schedule a project meeting for next Monday. Please include an announcement to stakeholders."

[0115] Prompt for generating notification content:

[0116] "Create a meeting notification based on the following schedule: [generated schedule content]"

[0117] In this way, the present invention efficiently operates generative artificial intelligence based on the user's natural language request, significantly reducing the effort and time required for the work.

[0118] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0119] Step 1:

[0120] User Input

[0121] The user uses a terminal to input a natural language request, for example, "Please create a schedule for the next project meeting and notify the relevant parties," into an input form in a web browser and presses the send button.

[0122] Input: A natural language request typed by a user using a terminal.

[0123] Output: When the submit button is pressed, the request is sent to the server.

[0124] Step 2:

[0125] Receiving and parsing the request

[0126] The server receives the request and then uses a natural language processing module such as Python's NLTK or spaCy to tokenize the request and extract keywords, extracting the gist of the request: "schedule" and "notify."

[0127] Input: A natural language request sent by the user.

[0128] Output: A brief summary of the parsed request (e.g., "Schedule", "Notify").

[0129] Step 3:

[0130] Search and Identify Generative Artificial Intelligence

[0131] Based on the analysis results, the server searches for and selects an appropriate model from a database of generative AI models. For example, for "schedule creation," it identifies high-performance natural language generation software as a generative AI model, and for "notification," it identifies a high-performance communication platform.

[0132] Input: A summary of the parsed request.

[0133] Output: A selected generative artificial intelligence model (e.g., high-performance natural language generation software, high-performance communications platform).

[0134] Step 4:

[0135] Calling generative AI and providing services

[0136] The server invokes selected generative artificial intelligence models to provide specific services, first generating a project meeting schedule using high-performance natural language generation software, and then communicating the generated schedule to stakeholders using a high-performance communication platform.

[0137] Input: Selected generative artificial intelligence model and a summary of the request.

[0138] Output: Generated schedule contents and notification results.

[0139] Step 5:

[0140] Check user satisfaction

[0141] The server provides the generated schedule and notification content to the user, who then checks the results through their terminal. If the user is satisfied with the results, they send their feedback to the server.

[0142] Input: Generated schedule contents and notification results.

[0143] Output: User feedback (confirmation of satisfaction).

[0144] Step 6:

[0145] Routine automation and web application deployment

[0146] The server automates this process based on user feedback by building and deploying a web application using the Python Flask framework, which will allow it to process similar requests automatically in the future.

[0147] Input: User feedback and verified routines.

[0148] Output: A web application as an automated routine.

[0149] (Application example 1)

[0150] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0151] As automation in factories continues to increase, many tasks are now performed by robots and automated equipment. However, efficiently controlling and managing these devices remains a complex process, and it is particularly difficult to achieve fast and efficient control using natural language. Furthermore, because users must clearly instruct each task, training is required and operation can be cumbersome. This can result in reduced factory productivity.

[0152] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0153] In this invention, the server includes means for receiving and analyzing natural language requests, means for searching for and identifying multiple appropriate generative AIs based on the content of the analyzed request, means for calling the identified multiple generative AIs to generate operating procedures for controlling equipment in the factory and providing the procedures to the equipment in the factory, means for confirming user satisfaction and automating event-service routines, and means for deploying the automated routines as web applications to automatically process future requests. This enables efficient control of robots and automated equipment in the factory and improves productivity throughout the factory.

[0154] A "natural language request" refers to a request entered by a user in a language that they normally use.

[0155] "Parsing" is the process of interpreting an input natural language request and understanding its content.

[0156] "Generative AI" refers to an AI model that generates outputs appropriate for a specific task.

[0157] "Search" is the process of finding the model that best suits a request from a generative artificial intelligence database.

[0158] "Identification" refers to determining the generative artificial intelligence that best suits the request from the search results.

[0159] "Invoking" refers to putting a specified generative artificial intelligence into operation and starting processing in response to a request.

[0160] "Operation procedure" refers to the specific procedure for operating equipment in a factory, generated by generative artificial intelligence.

[0161] "Provision" is the process of sending the generated operating procedure to the equipment in the factory and making it executable.

[0162] "Satisfaction" refers to the evaluation and impressions that users have of the results of the service provided.

[0163] A "routine" is a sequence of automatically and repeatedly performed steps or processes.

[0164] "Automation" refers to the state in which a process proceeds without the need for manual intervention.

[0165] "Web application" refers to a software application that is available over the Internet.

[0166] "Factory equipment" refers to robots and automated equipment used in factories in general.

[0167] This invention is a system that allows users to input requests in natural language and efficiently operate and manage automated equipment in a factory based on those requests. Specifically, a server analyzes the user's request, generates operating procedures using appropriate generative artificial intelligence, and provides them to various equipment in the factory.

[0168] Basic system configuration

[0169] 1. Receiving and parsing natural language requests

[0170] Users input requests in natural language through devices such as smartphones or tablets.

[0171] The server uses a Natural Language Processing (NLP) module to parse the request and extract the gist of the request, e.g., "Please assemble the following parts."

[0172] 2. Searching and Identifying Generative AI

[0173] Based on the analysis results, the server searches and identifies appropriate models from a database of generative AI, including "generative AI for scheduling" and "generative AI for notifications."

[0174] 3. Calling the generative AI and generating operational procedures

[0175] The server then calls the identified generative artificial intelligence to generate operational procedures based on the analyzed request, such as assembly procedures or device operation procedures.

[0176] 4. Providing operating procedures

[0177] The generated operation procedures are provided to robot arms and other in-factory equipment so that they can be executed.

[0178] 5. User satisfaction checks and routine automation

[0179] The user reviews the results provided and provides feedback on their satisfaction to the server, which, if satisfied, automates this routine and serves future requests as a web application.

[0180] Hardware and software used

[0181] Hardware

[0182] Smartphones, tablets, and PCs: for inputting users' natural language requests

[0183] Robotic arms and other factory equipment: Equipment that performs the actual movements

[0184] software

[0185] Natural Language Processing (NLP) modules: e.g., NLTK, spaCy, etc.

[0186] Generative AI modules: e.g., GPT-3 or other custom AI models

[0187] Robot control module: e.g., ROS (Robot Operating System)

[0188] Specific examples

[0189] The user inputs a request into their smartphone, such as "Please assemble the following parts." This request is analyzed by the NLP module, and the task "assembly of parts" is extracted. The generative AI generates an assembly procedure suitable for this task and sends it to the robot control module. Finally, the robot arm assembles the specified parts.

[0190] Example prompts to input to the generative AI model

[0191] "Please generate the steps to assemble parts A and B and attach part C."

[0192] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0193] Step 1:

[0194] A user inputs a natural language request into a terminal. The terminal sends the request to a server. The input is a specific natural language request such as "Please assemble the following parts," and the output is that the request is sent to the server. Here, text data entered on the terminal's operating screen is formed.

[0195] Step 2:

[0196] The server analyzes the received natural language request. Specifically, a natural language processing (NLP) module analyzes the request and extracts its gist. In this step, the input natural language sentence is analyzed and the action instructions for "assembly" are extracted. The input is the request data sent from the terminal, and the output is structured data containing the task instruction "assembly."

[0197] Step 3:

[0198] The server searches for and identifies an appropriate generative AI based on the analysis results. The server references a database of pre-registered generative AI models to search for a model suitable for the request. The input to this step is a structured task instruction, and the output is an appropriate generative AI model. Here, a generative AI model that matches the recognized task is searched for.

[0199] Step 4:

[0200] The server calls the identified generative AI to generate the required operating procedure. The called generative AI generates the operating procedure based on the given task instructions. In this step, specific instructions such as "Please assemble the following parts" are input as a prompt, and the output is a specific assembly procedure. The operating procedure is expressed as a series of control commands generated by the generative AI.

[0201] Step 5:

[0202] The server provides the generated operating procedures to the equipment in the factory. Here, a robot control module is used to send the generated instructions to a robot arm or other automated equipment. The input of this step is the generated operating procedure, and the output is the specific control signal received by the equipment. Here, the generated procedure is sent to the actual equipment and execution begins.

[0203] Step 6:

[0204] The user checks the provided results and provides feedback on their satisfaction to the server. The item the user checks is whether the operation was actually performed correctly. The input of this step is the operation result, and the output is the user's satisfaction feedback. The confirmation process is completed when the user provides feedback based on the actual operation result.

[0205] Step 7:

[0206] The server automates the event-service routine based on the user's satisfaction feedback and deploys it as a web application. In this step, past feedback is analyzed and an automated routine is designed. The input of this step is satisfaction feedback data, and the output is a web application containing the automated routine. The automated routine is configured to respond immediately to similar requests from the next time onwards.

[0207] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0208] The present invention is a system that combines natural language request processing with an emotion engine to optimize the user experience. Specific embodiments of the present invention are described below.

[0209] Basic system configuration

[0210] 1. User Input

[0211] The user uses a terminal to input a natural language request, for example, "Please schedule the next project meeting and notify the relevant parties."

[0212] 2. Receiving and parsing the request

[0213] The server receives a natural language request from the user, which is then parsed by a natural language processing (NLP) module to extract the gist of the request, which includes "schedule" and "notify relevant parties."

[0214] 3. Searching and Identifying Generative AI

[0215] Based on the analysis results of the request, the server searches for appropriate models from a database of generative artificial intelligence and identifies a "generative artificial intelligence for schedule creation" and a "generative artificial intelligence for notifications."

[0216] 4. Calling generative AI and providing services

[0217] The server calls the identified generative AI and performs specific processing according to the request. The generative AI for schedule creation generates a schedule for the project meeting and returns the schedule data. Similarly, the generative AI for notification notifies the relevant parties of the generated schedule.

[0218] 5. User Emotion Recognition by Emotion Engine

[0219] The server recognizes the user's emotions through the request and the result of the provided service, for example, the emotion engine analyzes whether the user is satisfied or dissatisfied.

[0220] 6. User Satisfaction Check

[0221] The server checks the user's satisfaction level based on the analysis results of the emotion engine, and if the user is satisfied, collects their feedback and uses it to optimize the automation routines.

[0222] 7. Routine Automation and Web Application Deployment

[0223] The server automates the event-service routines, which are deployed as web applications and automatically process future requests from users.

[0224] Specific examples

[0225] For example, if a user requests, "Please create a schedule for a project meeting next Monday and notify the relevant parties," the following processing is performed.

[0226] 1. Input Reception

[0227] The user inputs a request through the terminal.

[0228] 2. Request Analysis

[0229] The server receives the request, analyzes it, and extracts the information that needs to be "scheduled" and "notified."

[0230] 3. Identifying the right generative AI

[0231] The server searches for and identifies "generative artificial intelligence for schedule creation" and "generative artificial intelligence for notifications."

[0232] 4. Calling generative AI and providing services

[0233] The server calls each generative AI, creates a schedule, and notifies relevant parties.

[0234] 5. Emotion recognition

[0235] After the server provides the service, it uses an emotion engine to analyze the user's emotions and measure their satisfaction.

[0236] 6. Check the results and optimize

[0237] The user checks the provided schedule and notification content, and the server collects feedback based on the analysis results of the emotion engine.

[0238] 7. Deployment of automated routines

[0239] If the user is satisfied, the server automates the process and deploys it as a web application.

[0240] In this way, the present invention is a system that combines generative artificial intelligence and an emotion engine to effectively provide services based on a user's natural language request, thereby optimizing workflow.

[0241] The processing flow will be explained below.

[0242] Step 1:

[0243] A user inputs a natural language request using a terminal, for example, a request such as "Please create a schedule for the next project meeting and notify the relevant parties."

[0244] Step 2:

[0245] The terminal sends this natural language request to the server, which transmits the request over the Internet and receives it at the server's receiving module.

[0246] Step 3:

[0247] The server passes the received natural language request to an NLP (Natural Language Processing) module for analysis. This analysis module understands the content of the request and extracts related tasks and requirements. For example, "schedule creation" and "notify relevant parties" are extracted as tasks.

[0248] Step 4:

[0249] The server searches a database of generative AIs based on the extracted tasks. This search module accesses databases such as the OpenAI Store to find suitable generative AIs, and identifies "generative AIs for scheduling" and "generative AIs for notifications."

[0250] Step 5:

[0251] The server then calls the identified generative AI and performs specific processing in response to the user's request. The generative AI for schedule creation generates a schedule for the project meeting and returns the schedule data. Similarly, the generative AI for notification notifies the relevant parties of the generated schedule.

[0252] Step 6:

[0253] The server compiles the generated schedule and notification results and provides feedback to the user. The user can then check the results of the services provided by the server through their terminal and evaluate their level of satisfaction.

[0254] Step 7:

[0255] The server uses an emotion engine to analyze the user's emotions based on the feedback, for example, through natural language requests and the results of the provided service, the emotion engine recognizes whether the user is satisfied or dissatisfied.

[0256] Step 8:

[0257] Based on the analysis results of the emotion engine, the server optimizes user satisfaction. For example, if the user is dissatisfied, it will select a different generative AI or improve the existing process.

[0258] Step 9:

[0259] If the user is satisfied with the results of the provided service, the server automates the event-service routine. The automation module does this and implements the routine as a web application.

[0260] Step 10:

[0261] The server provides the built web application to the user's device and automatically processes similar requests in the future. The web application receives events in the form of voice, text, sensors, date and time, and provides services in the form of text, images, videos, spreadsheets, code, and documents.

[0262] In this way, through each step centered on the server, terminal, and user, the present invention is a system that provides efficient and flexible services in response to natural language requests from users and further increases user satisfaction using an emotion engine.

[0263] Example 2

[0264] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0265] Currently, many systems only provide simple information or limited services in response to user requests, and do not optimize the user experience or continuously improve processes based on feedback. As a result, the problems and frustrations faced by users are not fully resolved. In particular, processing complex requests, collecting feedback using emotion recognition, and automating processes are difficult challenges.

[0266] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0267] In this invention, the server includes means for receiving and analyzing natural language requests, means for searching for and identifying multiple appropriate generative AI models based on the content of the analyzed request, means for calling the identified generative AI model to execute a service according to the request and providing the results to the target, means for analyzing the target's emotions and confirming satisfaction, and means for automating an event-service routine, deploying the automated routine as a web application, and automatically processing future requests. This enables the provision of appropriate services based on user requests, optimization based on emotional feedback, and efficient processing of future requests.

[0268] A "natural language request" is a request that a user enters into a system in human language.

[0269] "Parsing" is the process of understanding a natural language request based on its meaning and structure.

[0270] A "generative artificial intelligence model" is an artificial intelligence model that has the ability to generate new data and information from given input.

[0271] "Search" is the process of finding models that meet specific criteria from a database of generative artificial intelligence models.

[0272] "Identification" is the process of selecting the most suitable generative artificial intelligence model from the searched results.

[0273] "Executing a service" means performing processing in response to a user request using a specified generative artificial intelligence model.

[0274] "Providing" means delivering the results of the performed service to the user.

[0275] "Analyzing emotions" is the process of analyzing a user's feedback and behavior to infer their emotional state.

[0276] "Confirming satisfaction level" refers to confirming the degree of satisfaction of the user with the provided service based on the results of sentiment analysis.

[0277] An "event-service routine" is a systematized flow of a series of service processes that correspond to a specific event.

[0278] An "automated routine" refers to a series of processes that a system executes automatically without human intervention.

[0279] A "web application" is a software application available over the Internet.

[0280] "Future requests" refers to additional requests that the user may make in the future.

[0281] The present invention is a system that efficiently processes natural language requests from users, provides services, and optimizes the process based on emotional feedback. The system consists of three main components: a server, a terminal, and a user.

[0282] System configuration

[0283] 1. User Input

[0284] A user inputs a natural language request using a device, such as a PC, smartphone, or tablet, using keyboard or voice input. For example, a request might be, "Please schedule the next project meeting and notify the relevant parties."

[0285] 2. Receiving and parsing the request

[0286] The server receives requests from users. This server may be a physical server equipped with an Intel Xeon processor or a cloud service instance (e.g., AWS (registered trademark) EC2). The received request is analyzed using a natural language processing (NLP) module such as Google (registered trademark) Cloud Natural Language API or SpaCy, and the gist of the request is extracted. For example, "schedule creation" and "notify relevant parties" are extracted as the gist.

[0287] 3. Searching and identifying generative AI models

[0288] The server searches and identifies appropriate generative AI models from the database based on the analysis results of the NLP module. Examples of generative AI models that can be used include OpenAI's GPT-4 (registered trademark) and similar generative systems. A "generative AI model for schedule creation" and a "generative AI model for notifications" are identified.

[0289] 4. Calling the generative AI model and providing it as a service

[0290] The server then calls the identified generative AI model and performs specific processing according to the user's request. For example, GPT-4 generates a schedule for a project meeting and returns the schedule data. Similarly, the notification AI model notifies relevant parties of the generated schedule.

[0291] 5. User Emotion Recognition by Emotion Engine

[0292] The server recognizes the user's emotions based on the execution results. The emotion engine used is the IBM Watson (registered trademark) Tone Analyzer, which analyzes whether the user is satisfied or dissatisfied. For example, emotions can be estimated by analyzing user feedback and behavioral data.

[0293] 6. User Satisfaction Check

[0294] The server checks the user's satisfaction with the service provided based on the output of the emotion engine, and if so, collects their feedback to help improve the process.

[0295] 7. Routine Automation and Web Application Deployment

[0296] The server then templates the process of successful requests and deploys them as automated routines in frameworks such as Django or Flask as web applications that automatically process similar requests in the future.

[0297] Specific examples

[0298] If a user requests, "Please schedule a project meeting for next Monday and notify the relevant parties," the following processing occurs:

[0299] 1. Enter your natural language request

[0300] The user uses a PC and inputs data using the keyboard.

[0301] 2. Receiving and parsing the request

[0302] The server receives this request and analyzes it using the NLP module (Google Cloud Natural Language API). It extracts the necessary information: "schedule creation" and "notification."

[0303] 3. Searching and identifying generative AI models

[0304] The server identifies GPT-4 from its database as the generative AI model for scheduling and T5 as the generative AI model for notifications.

[0305] 4. Calling the generative AI model and providing it as a service

[0306] The server calls GPT-4 to generate the schedule and calls T5 to notify relevant parties of the schedule.

[0307] 5. Emotion Recognition by Emotion Engine

[0308] The server uses IBM Watson Tone Analyzer to analyze the user's emotions and measure their satisfaction.

[0309] 6. User satisfaction confirmation and optimization

[0310] The user checks the provided schedule and notification content, and the server collects feedback based on the analysis results of the emotion engine.

[0311] 7. Automate routines

[0312] If the user is satisfied, the server will template this process and deploy it as a web application using the Django framework.

[0313] This allows us to provide appropriate services based on users' natural language requests, continuously improve our processes based on feedback, and efficiently process requests.

[0314] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0315] Step 1:

[0316] A user uses a terminal to input a natural language request.

[0317] Specific operation: A user uses a PC or smartphone to input a request using a keyboard or voice input, such as "Please schedule the next project meeting and notify relevant parties."

[0318] Input: A request expressed in natural language.

[0319] Output: Natural language text data.

[0320] Step 2:

[0321] The server receives a natural language request from a user.

[0322] Specific operation: The server receives the text data sent from the terminal and stores it in a database.

[0323] Input: Natural language text data sent by the user from their device.

[0324] Output: Request data stored in a database on the server.

[0325] Step 3:

[0326] The server analyzes the received request.

[0327] What happens: The server uses the Google Cloud Natural Language API to parse the request text and extract the main points: "Create a schedule" and "Notify relevant parties."

[0328] Input: Request data stored in the database.

[0329] Output: A brief summary of the extracted request.

[0330] Step 4:

[0331] The server searches for and identifies an appropriate generative AI model based on the analysis results.

[0332] Specific operation: The server identifies GPT-4 as the scheduling model and T5 as the notification model from the database.

[0333] Input: Summary of analysis results ("Schedule Creation" and "Notify Stakeholders").

[0334] Output: The identified generative AI model.

[0335] Step 5:

[0336] The server calls the identified generative AI model and performs specific processing in response to the user's request.

[0337] How it works: The server uses GPT-4 to generate a project meeting schedule and retrieves the data. Similarly, the notification AI model notifies relevant parties of the generated schedule.

[0338] Input: The identified generative AI model and a brief summary of the request.

[0339] Output: Generated schedule data and notification execution results.

[0340] Step 6:

[0341] The server uses an emotion engine to recognize the user's emotion.

[0342] Specific operation: The server uses IBM Watson Tone Analyzer to analyze the user's feedback and behavioral data after the service is provided and estimate the user's emotional state.

[0343] Input: User feedback and behavioral data after the service is provided.

[0344] Output: Estimated emotional state.

[0345] Step 7:

[0346] The server checks the user's satisfaction level based on the analysis results of the emotion engine.

[0347] What happens: The server measures satisfaction and, if the user is satisfied, collects that feedback and uses it to improve the process.

[0348] Input: Estimated emotional state.

[0349] Output: Satisfaction ratings and feedback data.

[0350] Step 8:

[0351] The server templatizes the process of successful requests and deploys the web application as an automated routine.

[0352] What it does: The server uses Django or Flask to automate the generated schedule and notification process and deploy it as a web application.

[0353] Input: Process and feedback data for successful requests.

[0354] Output: An automated web application.

[0355] This allows for the provision of appropriate services based on the user's natural language requests, and for process optimization and efficient request handling based on emotional feedback.

[0356] (Application example 2)

[0357] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0358] In food delivery services, when users communicate their orders in natural language, it is difficult to analyze their requests in detail and provide optimal suggestions. Furthermore, to increase user satisfaction, it is necessary to recognize the extent to which the service provided affected the user's emotions and improve the quality of the service based on that information. Conventional systems are unable to effectively integrate these elements, limiting their ability to optimize services and improve user satisfaction.

[0359] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0360] In this invention, the server includes means for receiving and analyzing natural language requests, means for searching for and identifying multiple appropriate generative artificial intelligences based on the content of the analyzed request, means for calling the identified multiple generative artificial intelligences to execute a service according to the request and providing the result to the user, means for checking user satisfaction and automating an event-service routine, means for deploying the automated routine as a web application and thereby automatically processing future requests, means for proposing food delivery menus and presenting estimated delivery times based on the user's natural language request, and means for analyzing user emotions from user feedback after providing the service to improve satisfaction. This enables a quick and accurate response to the user's natural language request and, at the same time, improves the quality of the service by recognizing how the provided service affected the user's emotions.

[0361] A "natural language request" is a request or command that a user enters into a system using everyday language.

[0362] "Parsing" is the process of breaking down the content of an incoming natural language request to determine its gist and purpose.

[0363] "Generative AI" refers to AI that has the ability to generate appropriate content and data based on input information.

[0364] "Searching and identifying" means finding a generative artificial intelligence from a database that matches the content of the analyzed request and preparing it for use.

[0365] "Request-based services" refer to specific processing and results provided by generative artificial intelligence based on a user's natural language request.

[0366] "User satisfaction" is an evaluation index that indicates how well a provided service meets the user's expectations and requirements.

[0367] An "event-service routine" is a set of service delivery procedures or processes that respond to a particular event or request.

[0368] An "automated routine" is a series of actions that the system takes to service an event or request automatically, without manual intervention.

[0369] A "web application" is software that runs on the Internet and is accessed by users through a browser.

[0370] "Food delivery menu suggestions" is a service that provides appropriate food and dish options based on user requests.

[0371] "Expected Delivery Time" means calculating and providing an estimate of the time it will take for an item to be delivered after an order is completed.

[0372] "Feedback" refers to the evaluations and impressions that users make about the services provided.

[0373] "Analyzing emotions" is the process of identifying and analyzing emotions (satisfaction, dissatisfaction, joy, etc.) from user feedback.

[0374] The present invention is a food delivery service system that receives natural language requests from users, analyzes them, and uses appropriate generative artificial intelligence to make optimal suggestions and provide delivery services. This system is operated through a smartphone application.

[0375] Basic system configuration

[0376] 1. User Input

[0377] Users input requests in natural language through a smartphone application, for example, "I'd like to order sushi. Please tell me what's recommended today. I'd also like to know how long it will take to deliver."

[0378] 2. Receiving and parsing the request

[0379] The server receives a natural language request from the user. The request is analyzed by a natural language processing (NLP) module (e.g., SpaCy or BERT) to extract the gist of the request. In this example, the gist is "Recommend sushi menu" and "Confirm delivery time."

[0380] 3. Searching and Identifying Generative AI

[0381] Based on the analysis of the request, the server searches and identifies appropriate models from its database of generative artificial intelligence. Specifically, it identifies a "generative artificial intelligence for menu suggestions" (e.g., GPT-4) and a "generative artificial intelligence for delivery time prediction."

[0382] 4. Calling generative AI and providing services

[0383] The server calls the identified generative AI and performs specific processing in response to the request. The generative AI for menu suggestions generates the optimal sushi menu and returns the data. The generative AI for delivery time prediction calculates the predicted delivery time and returns the result. This information is provided to the user via a smartphone application.

[0384] 5. User Emotion Recognition by Emotion Engine

[0385] The server receives the user's feedback after the service is provided and analyzes the user's emotions using an emotion engine (e.g., Google Cloud Natural Language API). The emotion engine analyzes whether the user is satisfied or dissatisfied.

[0386] 6. User Satisfaction Check

[0387] The server checks the user's satisfaction level based on the analysis results of the emotion engine, and if the user is satisfied, collects the feedback and stores it in a database.

[0388] 7. Routine Automation and Web Application Deployment

[0389] The server automates the event-service routines, which are deployed as web applications and perform automated processing when future requests are received from the user.

[0390] Specific examples

[0391] A user types a natural language request into a smartphone application, such as:

[0392] "I'm planning a family party next Sunday. Could you recommend a sushi set? Also, let me know the delivery time if I order now."

[0393] The server analyzes this request and uses appropriate generative artificial intelligence to suggest a sushi set and provide a predicted delivery time. The user then reviews the suggested menu and delivery time and makes their order decision based on that. After providing the service, the emotion engine analyzes the user's feedback and uses it to improve the service in the future.

[0394] This process allows for a fast and accurate response to the user's natural language requests and improves the quality of the service by recognizing how the provided service affected the user's emotions.

[0395] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0396] Step 1:

[0397] A user inputs a request in natural language through a smartphone application. For example, "I'd like to order sushi. Please tell me what's recommended today. I'd also like to know how long it will take to deliver." This request is then sent to the server by the application.

[0398] Input: User's natural language request

[0399] Output: The natural language request sent to the server

[0400] Step 2:

[0401] The server receives a natural language request from the user. The request is analyzed by a natural language processing (NLP) module (e.g., SpaCy or BERT) to extract the gist of the request. Specifically, the gist is extracted as "recommended sushi menu" and "confirm delivery time."

[0402] Input: A natural language request sent to the server

[0403] Output: A summary of the parsed request

[0404] Step 3:

[0405] Based on the analysis results of the request, the server searches and identifies an appropriate model from its database of generative AI. Specifically, it identifies a "generative AI for menu suggestions" (e.g., GPT-4) and a "generative AI for delivery time prediction." At this point, it selects the AI ​​model that best suits the content of the request.

[0406] Input: Parsed request summary

[0407] Output: Identified generative artificial intelligence model

[0408] Step 4:

[0409] The server calls the identified generative AI and performs specific processing in response to the request. The generative AI for menu suggestions generates the optimal sushi menu and returns the data. The generative AI for delivery time prediction calculates the predicted delivery time and returns the result. This information is provided to the user via a smartphone application.

[0410] Input: Identified generative artificial intelligence model, parsed request abstract

[0411] Output: Generated sushi menu, predicted delivery time

[0412] Step 5:

[0413] The server receives the user's feedback after the service is provided. It analyzes the user's emotions using an emotion engine (e.g., Google Cloud Natural Language API). The emotion engine analyzes whether the user is satisfied or dissatisfied.

[0414] Input: User feedback

[0415] Output: Analyzed user sentiment results

[0416] Step 6:

[0417] The server checks the user's satisfaction level based on the analysis results of the emotion engine. If the user is satisfied, the server collects the feedback and stores it in a database. This information can be used to improve the service in the future.

[0418] Input: Analyzed user sentiment results

[0419] Output: Saved feedback

[0420] Step 7:

[0421] The server automates the event-service routine. The automated routine is deployed as a web application and automatically processes future requests from users. This allows users to always receive optimized service.

[0422] Input: Stored feedback, parsed request summary

[0423] Output: Automated routines, deployed web applications

[0424] This enables a fast and accurate response to the user's natural language requests, and improves the quality of the service by recognizing how the provided service affected the user's emotions.

[0425] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0426] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0427] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0428] [Second embodiment]

[0429] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0430] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0431] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0432] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0433] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0434] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0435] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0436] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0437] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0438] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0439] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0440] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0441] The present invention is a system in which a user inputs a request in natural language, a server analyzes the request, calls an appropriate generative artificial intelligence, and ultimately provides a service to the user. Specific embodiments of the present invention will be described below.

[0442] Basic system configuration

[0443] 1. User Input

[0444] The user inputs a request in natural language using a terminal, such as "Please create a schedule for the next project meeting and notify the relevant parties."

[0445] 2. Receiving and parsing the request

[0446] The server receives a natural language request from the user, which is then parsed by a natural language processing (NLP) module to extract the gist of the request, which includes "schedule" and "notify relevant parties."

[0447] 3. Searching and Identifying Generative AI

[0448] Based on the analysis of the request, the server searches a database of generative AI models for suitable models, for example, identifying a "generative AI for scheduling" and a "generative AI for notifications."

[0449] 4. Calling generative AI and providing services

[0450] The server calls the identified generative AI and executes the service according to the request. The generative AI for schedule creation generates a schedule for the project meeting, and the generative AI for notification notifies the relevant parties of the schedule.

[0451] 5. User satisfaction check

[0452] The server provides the results of the service performed by the generative AI to the user and confirms their satisfaction. For example, the user checks the provided schedule and notification content and provides feedback on whether they are satisfied.

[0453] 6. Routine Automation and Web Application Deployment

[0454] If the user is satisfied, the server automates the event-service routine, which is then deployed as a web application to automatically process similar requests in the future.

[0455] Specific examples

[0456] For example, if a user requests, "Please create a schedule for a project meeting next Monday and notify the relevant parties," the following processing is performed.

[0457] 1. Input Reception

[0458] The user inputs a request through the terminal.

[0459] 2. Request Analysis

[0460] The server receives the request, analyzes it, and extracts the information that needs to be "scheduled" and "notified."

[0461] 3. Identifying the right generative AI

[0462] The server searches for and identifies "generative artificial intelligence for schedule creation" and "generative artificial intelligence for notifications."

[0463] 4. Calling generative AI and providing services

[0464] The server calls each generative AI, creates a schedule, and notifies relevant parties.

[0465] 5. Check the results

[0466] The server provides the generated schedule and notification contents to the user, who then confirms them.

[0467] 6. Deployment of automated routines

[0468] If the user is satisfied, the server automates the process and deploys it as a web application.

[0469] In this way, the present invention efficiently operates generative artificial intelligence based on the user's natural language requests, thereby optimizing workflow.

[0470] The processing flow will be explained below.

[0471] Step 1:

[0472] A user inputs a natural language request using a terminal, for example, a request such as "Please create a schedule for the next project meeting and notify the relevant parties."

[0473] Step 2:

[0474] The terminal sends this natural language request to the server, which transmits the request over the Internet and receives it at the server's receiving module.

[0475] Step 3:

[0476] The server passes the received natural language request to an NLP (Natural Language Processing) module for analysis. This analysis module understands the content of the request and extracts related tasks and requirements. For example, "schedule creation" and "notify relevant parties" are extracted as tasks.

[0477] Step 4:

[0478] The server searches a database of generative artificial intelligence (GA) models based on the extracted tasks. This search module accesses databases such as the OpenAI Store to identify "GA for scheduling" and "GA for notifications."

[0479] Step 5:

[0480] The server then calls the identified generative AI and performs specific processing in response to the user's request. For example, a generative AI for schedule creation generates a schedule for a project meeting and returns the schedule data. Similarly, a generative AI for notification notifies relevant parties of the generated schedule.

[0481] Step 6:

[0482] The server compiles the generated schedule and notification results and provides feedback to the user. The user can then check the results of the services provided by the server through their terminal and evaluate their level of satisfaction.

[0483] Step 7:

[0484] If the user is satisfied with the results of the provided service, the server automates the event-service routine. The automation module does this and implements the routine as a web application.

[0485] Step 8:

[0486] The server provides the built web application to the user's device and automatically processes similar requests in the future. The web application receives events in the form of voice, text, sensors, date and time, and provides services in the form of text, images, videos, spreadsheets, code, and documents.

[0487] Thus, through each step centered on the server, the terminal, and the user, the present invention provides an efficient and flexible service for natural language requests from users.

[0488] Example 1

[0489] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0490] In conventional systems, the process of users entering requests in natural language, searching for and identifying appropriate generative AI models based on those requests, and providing services was not automated, making it difficult to achieve an efficient workflow. It was also difficult to build a system that could confirm satisfaction with the generated results and automatically respond to similar requests in the future. This resulted in a major issue of the time and effort required for the process.

[0491] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0492] In this invention, the server includes means for receiving and analyzing a natural language request, means for searching for and identifying multiple appropriate generative AI models based on the content of the analyzed request, means for extracting a gist of the request using a natural language processing module, means for searching for and selecting a generative AI model from a database based on the extracted gist, means for calling the selected generative AI and providing a specific service in response to the request, means for providing the generated results to a user and confirming the user's satisfaction, means for automating the post-confirmation routine and deploying it as a Web application, and means for publishing it as a Web application for automatic processing of subsequent requests. This makes it possible to efficiently and automatically operate a generative AI model based on a user's natural language request, significantly reducing the effort and time required for the work.

[0493] A "natural language request" is a request or instruction entered by a user using human language.

[0494] "Parsing" is the act of receiving a natural language request and converting its content into an understandable format.

[0495] "Generative AI" is a type of AI model that is trained to perform specific tasks and has the ability to generate new information and results.

[0496] "Searching" is the act of finding items in a database or repository that match specific criteria.

[0497] "Identifying" is the act of selecting the most appropriate item from those found in a search.

[0498] A "natural language processing module" is a program or tool that converts natural language into a format that is easy for a computer to understand.

[0499] The "Summary" is a brief summary of the main content and purpose of the request.

[0500] A "database" is a system that stores large amounts of data in an organized manner and makes it easy to search and manage.

[0501] "Invocation" refers to sending a command or operation to execute a program or service.

[0502] "Satisfaction confirmation" is the act of checking whether the provided services and results meet the user's expectations.

[0503] A "routine" is a series of processing steps in a program that defines certain operations and processing procedures.

[0504] "Automation" refers to making a system operate automatically without human intervention.

[0505] A "web application" is application software that is used over the Internet and operates via a web browser.

[0506] The present invention is a system that allows a user to input a natural language request, and then calls an appropriate generative AI based on the request to provide a service. This system is primarily composed of a server, a terminal, and a generative AI model. A specific implementation method is described below.

[0507] A user inputs a request in natural language using a device (e.g., a personal computer, smartphone, tablet, etc.). This request is sent to the server via a web browser. A specific example is a request such as, "Please create a schedule for the next project meeting and notify the relevant parties."

[0508] The server uses natural language processing modules such as Python's NLTK or spaCy to parse the incoming request, extracting the gist of the request, such as "schedule" and "notify relevant parties."

[0509] Next, the server searches and selects an appropriate model from a database of generative AI models based on the extracted summary. It is possible to use OpenAI's GPT-3 for scheduling and Twilio API for notifications.

[0510] The selected generative AI model is then called by the server to provide a specific service: first, GPT-3 is used to generate a schedule for a project meeting, and then the generated schedule is notified to relevant parties using the Twilio API.

[0511] The generated results are provided to the user by the server, who checks them through their terminal. If the user is satisfied with the results, they send their feedback to the server, which uses this feedback to automate the routine.

[0512] Finally, the automated routine can be deployed as a web application, so that similar requests can be processed automatically in the future. For example, the web application can be built using the Python Flask framework.

[0513] Examples of prompt statements

[0514] Prompt for scheduling a project meeting:

[0515] "Please schedule a project meeting for next Monday. Please include an announcement to stakeholders."

[0516] Prompt for generating notification content:

[0517] "Create a meeting notification based on the following schedule: [generated schedule content]"

[0518] In this way, the present invention efficiently operates generative artificial intelligence based on the user's natural language request, significantly reducing the effort and time required for the work.

[0519] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0520] Step 1:

[0521] User Input

[0522] The user uses a terminal to input a natural language request, for example, "Please create a schedule for the next project meeting and notify the relevant parties," into an input form in a web browser and presses the send button.

[0523] Input: A natural language request typed by a user using a terminal.

[0524] Output: When the submit button is pressed, the request is sent to the server.

[0525] Step 2:

[0526] Receiving and parsing the request

[0527] The server receives the request and then uses a natural language processing module such as Python's NLTK or spaCy to tokenize the request and extract keywords, extracting the gist of the request: "schedule" and "notify."

[0528] Input: A natural language request sent by the user.

[0529] Output: A brief summary of the parsed request (e.g., "Schedule", "Notify").

[0530] Step 3:

[0531] Search and Identify Generative Artificial Intelligence

[0532] Based on the analysis results, the server searches for and selects an appropriate model from a database of generative AI models. For example, for "schedule creation," it identifies high-performance natural language generation software as a generative AI model, and for "notification," it identifies a high-performance communication platform.

[0533] Input: A summary of the parsed request.

[0534] Output: A selected generative artificial intelligence model (e.g., high-performance natural language generation software, high-performance communications platform).

[0535] Step 4:

[0536] Calling generative AI and providing services

[0537] The server invokes selected generative artificial intelligence models to provide specific services, first generating a project meeting schedule using high-performance natural language generation software, and then communicating the generated schedule to stakeholders using a high-performance communication platform.

[0538] Input: Selected generative artificial intelligence model and a summary of the request.

[0539] Output: Generated schedule contents and notification results.

[0540] Step 5:

[0541] Check user satisfaction

[0542] The server provides the generated schedule and notification content to the user, who then checks the results through their terminal. If the user is satisfied with the results, they send their feedback to the server.

[0543] Input: Generated schedule contents and notification results.

[0544] Output: User feedback (confirmation of satisfaction).

[0545] Step 6:

[0546] Routine automation and web application deployment

[0547] The server automates this process based on user feedback by building and deploying a web application using the Python Flask framework, which will allow it to process similar requests automatically in the future.

[0548] Input: User feedback and verified routines.

[0549] Output: A web application as an automated routine.

[0550] (Application example 1)

[0551] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0552] As automation in factories continues to increase, many tasks are now performed by robots and automated equipment. However, efficiently controlling and managing these devices remains a complex process, and it is particularly difficult to achieve fast and efficient control using natural language. Furthermore, because users must clearly instruct each task, training is required and operation can be cumbersome. This can result in reduced factory productivity.

[0553] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0554] In this invention, the server includes means for receiving and analyzing natural language requests, means for searching for and identifying multiple appropriate generative AIs based on the content of the analyzed request, means for calling the identified multiple generative AIs to generate operating procedures for controlling equipment in the factory and providing the procedures to the equipment in the factory, means for confirming user satisfaction and automating event-service routines, and means for deploying the automated routines as web applications to automatically process future requests. This enables efficient control of robots and automated equipment in the factory and improves productivity throughout the factory.

[0555] A "natural language request" refers to a request entered by a user in a language that they normally use.

[0556] "Parsing" is the process of interpreting an input natural language request and understanding its content.

[0557] "Generative AI" refers to an AI model that generates outputs appropriate for a specific task.

[0558] "Search" is the process of finding the model that best suits a request from a generative artificial intelligence database.

[0559] "Identification" refers to determining the generative artificial intelligence that best suits the request from the search results.

[0560] "Invoking" refers to putting a specified generative artificial intelligence into operation and starting processing in response to a request.

[0561] "Operation procedure" refers to the specific procedure for operating equipment in a factory, generated by generative artificial intelligence.

[0562] "Provision" is the process of sending the generated operating procedure to the equipment in the factory and making it executable.

[0563] "Satisfaction" refers to the evaluation and impressions that users have of the results of the service provided.

[0564] A "routine" is a sequence of automatically and repeatedly performed steps or processes.

[0565] "Automation" refers to the state in which a process proceeds without the need for manual intervention.

[0566] "Web application" refers to a software application that is available over the Internet.

[0567] "Factory equipment" refers to robots and automated equipment used in factories in general.

[0568] This invention is a system that allows users to input requests in natural language and efficiently operate and manage automated equipment in a factory based on those requests. Specifically, a server analyzes the user's request, generates operating procedures using appropriate generative artificial intelligence, and provides them to various equipment in the factory.

[0569] Basic system configuration

[0570] 1. Receiving and parsing natural language requests

[0571] Users input requests in natural language through devices such as smartphones or tablets.

[0572] The server uses a Natural Language Processing (NLP) module to parse the request and extract the gist of the request, e.g., "Please assemble the following parts."

[0573] 2. Searching and Identifying Generative AI

[0574] Based on the analysis results, the server searches and identifies appropriate models from a database of generative AI, including "generative AI for scheduling" and "generative AI for notifications."

[0575] 3. Calling the generative AI and generating operational procedures

[0576] The server then calls the identified generative artificial intelligence to generate operational procedures based on the analyzed request, such as assembly procedures or device operation procedures.

[0577] 4. Providing operating procedures

[0578] The generated operation procedures are provided to robot arms and other in-factory equipment so that they can be executed.

[0579] 5. User satisfaction checks and routine automation

[0580] The user reviews the results provided and provides feedback on their satisfaction to the server, which, if satisfied, automates this routine and serves future requests as a web application.

[0581] Hardware and software used

[0582] Hardware

[0583] Smartphones, tablets, and PCs: for inputting users' natural language requests

[0584] Robotic arms and other factory equipment: Equipment that performs the actual movements

[0585] software

[0586] Natural Language Processing (NLP) modules: e.g., NLTK, spaCy, etc.

[0587] Generative AI modules: e.g., GPT-3 or other custom AI models

[0588] Robot control module: e.g., ROS (Robot Operating System)

[0589] Specific examples

[0590] The user inputs a request into their smartphone, such as "Please assemble the following parts." This request is analyzed by the NLP module, and the task "assembly of parts" is extracted. The generative AI generates an assembly procedure suitable for this task and sends it to the robot control module. Finally, the robot arm assembles the specified parts.

[0591] Example prompts to input to the generative AI model

[0592] "Please generate the steps to assemble parts A and B and attach part C."

[0593] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0594] Step 1:

[0595] A user inputs a natural language request into a terminal. The terminal sends the request to a server. The input is a specific natural language request such as "Please assemble the following parts," and the output is that the request is sent to the server. Here, text data entered on the terminal's operating screen is formed.

[0596] Step 2:

[0597] The server analyzes the received natural language request. Specifically, a natural language processing (NLP) module analyzes the request and extracts its gist. In this step, the input natural language sentence is analyzed and the action instructions for "assembly" are extracted. The input is the request data sent from the terminal, and the output is structured data containing the task instruction "assembly."

[0598] Step 3:

[0599] The server searches for and identifies an appropriate generative AI based on the analysis results. The server references a database of pre-registered generative AI models to search for a model suitable for the request. The input to this step is a structured task instruction, and the output is an appropriate generative AI model. Here, a generative AI model that matches the recognized task is searched for.

[0600] Step 4:

[0601] The server calls the identified generative AI to generate the required operating procedure. The called generative AI generates the operating procedure based on the given task instructions. In this step, specific instructions such as "Please assemble the following parts" are input as a prompt, and the output is a specific assembly procedure. The operating procedure is expressed as a series of control commands generated by the generative AI.

[0602] Step 5:

[0603] The server provides the generated operating procedures to the equipment in the factory. Here, a robot control module is used to send the generated instructions to a robot arm or other automated equipment. The input of this step is the generated operating procedure, and the output is the specific control signal received by the equipment. Here, the generated procedure is sent to the actual equipment and execution begins.

[0604] Step 6:

[0605] The user checks the provided results and provides feedback on their satisfaction to the server. The item the user checks is whether the operation was actually performed correctly. The input of this step is the operation result, and the output is the user's satisfaction feedback. The confirmation process is completed when the user provides feedback based on the actual operation result.

[0606] Step 7:

[0607] The server automates the event-service routine based on the user's satisfaction feedback and deploys it as a web application. In this step, past feedback is analyzed and an automated routine is designed. The input of this step is satisfaction feedback data, and the output is a web application containing the automated routine. The automated routine is configured to respond immediately to similar requests from the next time onwards.

[0608] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0609] The present invention is a system that combines natural language request processing with an emotion engine to optimize the user experience. Specific embodiments of the present invention are described below.

[0610] Basic system configuration

[0611] 1. User Input

[0612] The user uses a terminal to input a natural language request, for example, "Please schedule the next project meeting and notify the relevant parties."

[0613] 2. Receiving and parsing the request

[0614] The server receives a natural language request from the user, which is then parsed by a natural language processing (NLP) module to extract the gist of the request, which includes "schedule" and "notify relevant parties."

[0615] 3. Searching and Identifying Generative AI

[0616] Based on the analysis results of the request, the server searches for appropriate models from a database of generative artificial intelligence and identifies a "generative artificial intelligence for schedule creation" and a "generative artificial intelligence for notifications."

[0617] 4. Calling generative AI and providing services

[0618] The server calls the identified generative AI and performs specific processing according to the request. The generative AI for schedule creation generates a schedule for the project meeting and returns the schedule data. Similarly, the generative AI for notification notifies the relevant parties of the generated schedule.

[0619] 5. User Emotion Recognition by Emotion Engine

[0620] The server recognizes the user's emotions through the request and the result of the provided service, for example, the emotion engine analyzes whether the user is satisfied or dissatisfied.

[0621] 6. User Satisfaction Check

[0622] The server checks the user's satisfaction level based on the analysis results of the emotion engine, and if the user is satisfied, collects their feedback and uses it to optimize the automation routines.

[0623] 7. Routine Automation and Web Application Deployment

[0624] The server automates the event-service routines, which are deployed as web applications and automatically process future requests from users.

[0625] Specific examples

[0626] For example, if a user requests, "Please create a schedule for a project meeting next Monday and notify the relevant parties," the following processing is performed.

[0627] 1. Input Reception

[0628] The user inputs a request through the terminal.

[0629] 2. Request Analysis

[0630] The server receives the request, analyzes it, and extracts the information that needs to be "scheduled" and "notified."

[0631] 3. Identifying the right generative AI

[0632] The server searches for and identifies "generative artificial intelligence for schedule creation" and "generative artificial intelligence for notifications."

[0633] 4. Calling generative AI and providing services

[0634] The server calls each generative AI, creates a schedule, and notifies relevant parties.

[0635] 5. Emotion recognition

[0636] After the server provides the service, it uses an emotion engine to analyze the user's emotions and measure their satisfaction.

[0637] 6. Check the results and optimize

[0638] The user checks the provided schedule and notification content, and the server collects feedback based on the analysis results of the emotion engine.

[0639] 7. Deployment of automated routines

[0640] If the user is satisfied, the server automates the process and deploys it as a web application.

[0641] In this way, the present invention is a system that combines generative artificial intelligence and an emotion engine to effectively provide services based on a user's natural language request, thereby optimizing workflow.

[0642] The processing flow will be explained below.

[0643] Step 1:

[0644] A user inputs a natural language request using a terminal, for example, a request such as "Please create a schedule for the next project meeting and notify the relevant parties."

[0645] Step 2:

[0646] The terminal sends this natural language request to the server, which transmits the request over the Internet and receives it at the server's receiving module.

[0647] Step 3:

[0648] The server passes the received natural language request to an NLP (Natural Language Processing) module for analysis. This analysis module understands the content of the request and extracts related tasks and requirements. For example, "schedule creation" and "notify relevant parties" are extracted as tasks.

[0649] Step 4:

[0650] The server searches a database of generative AIs based on the extracted tasks. This search module accesses databases such as the OpenAI Store to find suitable generative AIs, and identifies "generative AIs for scheduling" and "generative AIs for notifications."

[0651] Step 5:

[0652] The server then calls the identified generative AI and performs specific processing in response to the user's request. The generative AI for schedule creation generates a schedule for the project meeting and returns the schedule data. Similarly, the generative AI for notification notifies the relevant parties of the generated schedule.

[0653] Step 6:

[0654] The server compiles the generated schedule and notification results and provides feedback to the user. The user can then check the results of the services provided by the server through their terminal and evaluate their level of satisfaction.

[0655] Step 7:

[0656] The server uses an emotion engine to analyze the user's emotions based on the feedback, for example, through natural language requests and the results of the provided service, the emotion engine recognizes whether the user is satisfied or dissatisfied.

[0657] Step 8:

[0658] Based on the analysis results of the emotion engine, the server optimizes user satisfaction. For example, if the user is dissatisfied, it will select a different generative AI or improve the existing process.

[0659] Step 9:

[0660] If the user is satisfied with the results of the provided service, the server automates the event-service routine. The automation module does this and implements the routine as a web application.

[0661] Step 10:

[0662] The server provides the built web application to the user's device and automatically processes similar requests in the future. The web application receives events in the form of voice, text, sensors, date and time, and provides services in the form of text, images, videos, spreadsheets, code, and documents.

[0663] In this way, through each step centered on the server, terminal, and user, the present invention is a system that provides efficient and flexible services in response to natural language requests from users and further increases user satisfaction using an emotion engine.

[0664] Example 2

[0665] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0666] Currently, many systems only provide simple information or limited services in response to user requests, and do not optimize the user experience or continuously improve processes based on feedback. As a result, the problems and frustrations faced by users are not fully resolved. In particular, processing complex requests, collecting feedback using emotion recognition, and automating processes are difficult challenges.

[0667] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0668] In this invention, the server includes means for receiving and analyzing natural language requests, means for searching for and identifying multiple appropriate generative AI models based on the content of the analyzed request, means for calling the identified generative AI model to execute a service according to the request and providing the results to the target, means for analyzing the target's emotions and confirming satisfaction, and means for automating an event-service routine, deploying the automated routine as a web application, and automatically processing future requests. This enables the provision of appropriate services based on user requests, optimization based on emotional feedback, and efficient processing of future requests.

[0669] A "natural language request" is a request that a user enters into a system in human language.

[0670] "Parsing" is the process of understanding a natural language request based on its meaning and structure.

[0671] A "generative artificial intelligence model" is an artificial intelligence model that has the ability to generate new data and information from given input.

[0672] "Search" is the process of finding models that meet specific criteria from a database of generative artificial intelligence models.

[0673] "Identification" is the process of selecting the most suitable generative artificial intelligence model from the searched results.

[0674] "Executing a service" means performing processing in response to a user request using a specified generative artificial intelligence model.

[0675] "Providing" means delivering the results of the performed service to the user.

[0676] "Analyzing emotions" is the process of analyzing a user's feedback and behavior to infer their emotional state.

[0677] "Confirming satisfaction level" refers to confirming the degree of satisfaction of the user with the provided service based on the results of sentiment analysis.

[0678] An "event-service routine" is a systematized flow of a series of service processes that correspond to a specific event.

[0679] An "automated routine" refers to a series of processes that a system executes automatically without human intervention.

[0680] A "web application" is a software application available over the Internet.

[0681] "Future requests" refers to additional requests that the user may make in the future.

[0682] The present invention is a system that efficiently processes natural language requests from users, provides services, and optimizes the process based on emotional feedback. The system consists of three main components: a server, a terminal, and a user.

[0683] System configuration

[0684] 1. User Input

[0685] A user inputs a natural language request using a device, such as a PC, smartphone, or tablet, using keyboard or voice input. For example, a request might be, "Please schedule the next project meeting and notify the relevant parties."

[0686] 2. Receiving and parsing the request

[0687] The server receives requests from users. This server can be a physical server equipped with an Intel Xeon processor or a cloud service instance (e.g., AWS EC2). The received request is analyzed using a natural language processing (NLP) module such as Google's Cloud Natural Language API or SpaCy, and the gist of the request is extracted. For example, "Create a schedule" and "Notify relevant parties" are extracted as the gist.

[0688] 3. Searching and identifying generative AI models

[0689] Based on the analysis results of the NLP module, the server searches and identifies appropriate generative AI models from the database. Examples of generative AI models that can be used include OpenAI's GPT-4 and similar generative systems. A "generative AI model for schedule creation" and a "generative AI model for notifications" are identified.

[0690] 4. Calling the generative AI model and providing it as a service

[0691] The server then calls the identified generative AI model and performs specific processing according to the user's request. For example, GPT-4 generates a schedule for a project meeting and returns the schedule data. Similarly, the notification AI model notifies relevant parties of the generated schedule.

[0692] 5. User Emotion Recognition by Emotion Engine

[0693] The server recognizes the user's emotions based on the execution results. The emotion engine uses IBM Watson Tone Analyzer, among others. This engine analyzes whether the user is satisfied or dissatisfied. For example, it can estimate emotions by analyzing user feedback and behavioral data.

[0694] 6. User Satisfaction Check

[0695] The server checks the user's satisfaction with the service provided based on the output of the emotion engine, and if so, collects their feedback to help improve the process.

[0696] 7. Routine Automation and Web Application Deployment

[0697] The server then templates the process of successful requests and deploys them as automated routines in frameworks such as Django or Flask as web applications that automatically process similar requests in the future.

[0698] Specific examples

[0699] If a user requests, "Please schedule a project meeting for next Monday and notify the relevant parties," the following processing occurs:

[0700] 1. Enter your natural language request

[0701] The user uses a PC and inputs data using the keyboard.

[0702] 2. Receiving and parsing the request

[0703] The server receives this request and analyzes it using the NLP module (Google Cloud Natural Language API). It extracts the necessary information: "schedule creation" and "notification."

[0704] 3. Searching and identifying generative AI models

[0705] The server identifies GPT-4 from its database as the generative AI model for scheduling and T5 as the generative AI model for notifications.

[0706] 4. Calling the generative AI model and providing it as a service

[0707] The server calls GPT-4 to generate the schedule and calls T5 to notify relevant parties of the schedule.

[0708] 5. Emotion Recognition by Emotion Engine

[0709] The server uses IBM Watson Tone Analyzer to analyze the user's emotions and measure their satisfaction.

[0710] 6. User satisfaction confirmation and optimization

[0711] The user checks the provided schedule and notification content, and the server collects feedback based on the analysis results of the emotion engine.

[0712] 7. Automate routines

[0713] If the user is satisfied, the server will template this process and deploy it as a web application using the Django framework.

[0714] This allows us to provide appropriate services based on users' natural language requests, continuously improve our processes based on feedback, and efficiently process requests.

[0715] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0716] Step 1:

[0717] A user uses a terminal to input a natural language request.

[0718] Specific operation: A user uses a PC or smartphone to input a request using a keyboard or voice input, such as "Please schedule the next project meeting and notify relevant parties."

[0719] Input: A request expressed in natural language.

[0720] Output: Natural language text data.

[0721] Step 2:

[0722] The server receives a natural language request from a user.

[0723] Specific operation: The server receives the text data sent from the terminal and stores it in a database.

[0724] Input: Natural language text data sent by the user from their device.

[0725] Output: Request data stored in a database on the server.

[0726] Step 3:

[0727] The server analyzes the received request.

[0728] What happens: The server uses the Google Cloud Natural Language API to parse the request text and extract the main points: "Create a schedule" and "Notify relevant parties."

[0729] Input: Request data stored in the database.

[0730] Output: A brief summary of the extracted request.

[0731] Step 4:

[0732] The server searches for and identifies an appropriate generative AI model based on the analysis results.

[0733] Specific operation: The server identifies GPT-4 as the scheduling model and T5 as the notification model from the database.

[0734] Input: Summary of analysis results ("Schedule Creation" and "Notify Stakeholders").

[0735] Output: The identified generative AI model.

[0736] Step 5:

[0737] The server calls the identified generative AI model and performs specific processing in response to the user's request.

[0738] How it works: The server uses GPT-4 to generate a project meeting schedule and retrieves the data. Similarly, the notification AI model notifies relevant parties of the generated schedule.

[0739] Input: The identified generative AI model and a brief summary of the request.

[0740] Output: Generated schedule data and notification execution results.

[0741] Step 6:

[0742] The server uses an emotion engine to recognize the user's emotion.

[0743] Specific operation: The server uses IBM Watson Tone Analyzer to analyze the user's feedback and behavioral data after the service is provided and estimate the user's emotional state.

[0744] Input: User feedback and behavioral data after the service is provided.

[0745] Output: Estimated emotional state.

[0746] Step 7:

[0747] The server checks the user's satisfaction level based on the analysis results of the emotion engine.

[0748] What happens: The server measures satisfaction and, if the user is satisfied, collects that feedback and uses it to improve the process.

[0749] Input: Estimated emotional state.

[0750] Output: Satisfaction ratings and feedback data.

[0751] Step 8:

[0752] The server templatizes the process of successful requests and deploys the web application as an automated routine.

[0753] What it does: The server uses Django or Flask to automate the generated schedule and notification process and deploy it as a web application.

[0754] Input: Process and feedback data for successful requests.

[0755] Output: An automated web application.

[0756] This allows for the provision of appropriate services based on the user's natural language requests, and for process optimization and efficient request handling based on emotional feedback.

[0757] (Application example 2)

[0758] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0759] In food delivery services, when users communicate their orders in natural language, it is difficult to analyze their requests in detail and provide optimal suggestions. Furthermore, to increase user satisfaction, it is necessary to recognize the extent to which the service provided affected the user's emotions and improve the quality of the service based on that information. Conventional systems are unable to effectively integrate these elements, limiting their ability to optimize services and improve user satisfaction.

[0760] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0761] In this invention, the server includes means for receiving and analyzing natural language requests, means for searching for and identifying multiple appropriate generative artificial intelligences based on the content of the analyzed request, means for calling the identified multiple generative artificial intelligences to execute a service according to the request and providing the result to the user, means for checking user satisfaction and automating an event-service routine, means for deploying the automated routine as a web application and thereby automatically processing future requests, means for proposing food delivery menus and presenting estimated delivery times based on the user's natural language request, and means for analyzing user emotions from user feedback after providing the service to improve satisfaction. This enables a quick and accurate response to the user's natural language request and, at the same time, improves the quality of the service by recognizing how the provided service affected the user's emotions.

[0762] A "natural language request" is a request or command that a user enters into a system using everyday language.

[0763] "Parsing" is the process of breaking down the content of an incoming natural language request to determine its gist and purpose.

[0764] "Generative AI" refers to AI that has the ability to generate appropriate content and data based on input information.

[0765] "Searching and identifying" means finding a generative artificial intelligence from a database that matches the content of the analyzed request and preparing it for use.

[0766] "Request-based services" refer to specific processing and results provided by generative artificial intelligence based on a user's natural language request.

[0767] "User satisfaction" is an evaluation index that indicates how well a provided service meets the user's expectations and requirements.

[0768] An "event-service routine" is a set of service delivery procedures or processes that respond to a particular event or request.

[0769] An "automated routine" is a series of actions that the system takes to service an event or request automatically, without manual intervention.

[0770] A "web application" is software that runs on the Internet and is accessed by users through a browser.

[0771] "Food delivery menu suggestions" is a service that provides appropriate food and dish options based on user requests.

[0772] "Expected Delivery Time" means calculating and providing an estimate of the time it will take for an item to be delivered after an order is completed.

[0773] "Feedback" refers to the evaluations and impressions that users make about the services provided.

[0774] "Analyzing emotions" is the process of identifying and analyzing emotions (satisfaction, dissatisfaction, joy, etc.) from user feedback.

[0775] The present invention is a food delivery service system that receives natural language requests from users, analyzes them, and uses appropriate generative artificial intelligence to make optimal suggestions and provide delivery services. This system is operated through a smartphone application.

[0776] Basic system configuration

[0777] 1. User Input

[0778] Users input requests in natural language through a smartphone application, for example, "I'd like to order sushi. Please tell me what's recommended today. I'd also like to know how long it will take to deliver."

[0779] 2. Receiving and parsing the request

[0780] The server receives a natural language request from the user. The request is analyzed by a natural language processing (NLP) module (e.g., SpaCy or BERT) to extract the gist of the request. In this example, the gist is "Recommend sushi menu" and "Confirm delivery time."

[0781] 3. Searching and Identifying Generative AI

[0782] Based on the analysis of the request, the server searches and identifies appropriate models from its database of generative artificial intelligence. Specifically, it identifies a "generative artificial intelligence for menu suggestions" (e.g., GPT-4) and a "generative artificial intelligence for delivery time prediction."

[0783] 4. Calling generative AI and providing services

[0784] The server calls the identified generative AI and performs specific processing in response to the request. The generative AI for menu suggestions generates the optimal sushi menu and returns the data. The generative AI for delivery time prediction calculates the predicted delivery time and returns the result. This information is provided to the user via a smartphone application.

[0785] 5. User Emotion Recognition by Emotion Engine

[0786] The server receives the user's feedback after the service is provided and analyzes the user's emotions using an emotion engine (e.g., Google Cloud Natural Language API). The emotion engine analyzes whether the user is satisfied or dissatisfied.

[0787] 6. User Satisfaction Check

[0788] The server checks the user's satisfaction level based on the analysis results of the emotion engine, and if the user is satisfied, collects the feedback and stores it in a database.

[0789] 7. Routine Automation and Web Application Deployment

[0790] The server automates the event-service routines, which are deployed as web applications and perform automated processing when future requests are received from the user.

[0791] Specific examples

[0792] A user types a natural language request into a smartphone application, such as:

[0793] "I'm planning a family party next Sunday. Could you recommend a sushi set? Also, let me know the delivery time if I order now."

[0794] The server analyzes this request and uses appropriate generative artificial intelligence to suggest a sushi set and provide a predicted delivery time. The user then reviews the suggested menu and delivery time and makes their order decision based on that. After providing the service, the emotion engine analyzes the user's feedback and uses it to improve the service in the future.

[0795] This process allows for a fast and accurate response to the user's natural language requests and improves the quality of the service by recognizing how the provided service affected the user's emotions.

[0796] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0797] Step 1:

[0798] A user inputs a request in natural language through a smartphone application. For example, "I'd like to order sushi. Please tell me what's recommended today. I'd also like to know how long it will take to deliver." This request is then sent to the server by the application.

[0799] Input: User's natural language request

[0800] Output: The natural language request sent to the server

[0801] Step 2:

[0802] The server receives a natural language request from the user. The request is analyzed by a natural language processing (NLP) module (e.g., SpaCy or BERT) to extract the gist of the request. Specifically, the gist is extracted as "recommended sushi menu" and "confirm delivery time."

[0803] Input: A natural language request sent to the server

[0804] Output: A summary of the parsed request

[0805] Step 3:

[0806] Based on the analysis results of the request, the server searches and identifies an appropriate model from its database of generative AI. Specifically, it identifies a "generative AI for menu suggestions" (e.g., GPT-4) and a "generative AI for delivery time prediction." At this point, it selects the AI ​​model that best suits the content of the request.

[0807] Input: Parsed request summary

[0808] Output: Identified generative artificial intelligence model

[0809] Step 4:

[0810] The server calls the identified generative AI and performs specific processing in response to the request. The generative AI for menu suggestions generates the optimal sushi menu and returns the data. The generative AI for delivery time prediction calculates the predicted delivery time and returns the result. This information is provided to the user via a smartphone application.

[0811] Input: Identified generative artificial intelligence model, parsed request abstract

[0812] Output: Generated sushi menu, predicted delivery time

[0813] Step 5:

[0814] The server receives the user's feedback after the service is provided. It analyzes the user's emotions using an emotion engine (e.g., Google Cloud Natural Language API). The emotion engine analyzes whether the user is satisfied or dissatisfied.

[0815] Input: User feedback

[0816] Output: Analyzed user sentiment results

[0817] Step 6:

[0818] The server checks the user's satisfaction level based on the analysis results of the emotion engine. If the user is satisfied, the server collects the feedback and stores it in a database. This information can be used to improve the service in the future.

[0819] Input: Analyzed user sentiment results

[0820] Output: Saved feedback

[0821] Step 7:

[0822] The server automates the event-service routine. The automated routine is deployed as a web application and automatically processes future requests from users. This allows users to always receive optimized service.

[0823] Input: Stored feedback, parsed request summary

[0824] Output: Automated routines, deployed web applications

[0825] This enables a fast and accurate response to the user's natural language requests, and improves the quality of the service by recognizing how the provided service affected the user's emotions.

[0826] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0827] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0828] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0829] [Third embodiment]

[0830] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0831] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0832] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0833] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0834] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0835] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0836] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0837] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0838] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0839] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0840] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0841] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0842] The present invention is a system in which a user inputs a request in natural language, a server analyzes the request, calls an appropriate generative artificial intelligence, and ultimately provides a service to the user. Specific embodiments of the present invention will be described below.

[0843] Basic system configuration

[0844] 1. User Input

[0845] The user inputs a request in natural language using a terminal, such as "Please create a schedule for the next project meeting and notify the relevant parties."

[0846] 2. Receiving and parsing the request

[0847] The server receives a natural language request from the user, which is then parsed by a natural language processing (NLP) module to extract the gist of the request, which includes "schedule" and "notify relevant parties."

[0848] 3. Searching and Identifying Generative AI

[0849] Based on the analysis of the request, the server searches a database of generative AI models for suitable models, for example, identifying a "generative AI for scheduling" and a "generative AI for notifications."

[0850] 4. Calling generative AI and providing services

[0851] The server calls the identified generative AI and executes the service according to the request. The generative AI for schedule creation generates a schedule for the project meeting, and the generative AI for notification notifies the relevant parties of the schedule.

[0852] 5. User satisfaction check

[0853] The server provides the results of the service performed by the generative AI to the user and confirms their satisfaction. For example, the user checks the provided schedule and notification content and provides feedback on whether they are satisfied.

[0854] 6. Routine Automation and Web Application Deployment

[0855] If the user is satisfied, the server automates the event-service routine, which is then deployed as a web application to automatically process similar requests in the future.

[0856] Specific examples

[0857] For example, if a user requests, "Please create a schedule for a project meeting next Monday and notify the relevant parties," the following processing is performed.

[0858] 1. Input Reception

[0859] The user inputs a request through the terminal.

[0860] 2. Request Analysis

[0861] The server receives the request, analyzes it, and extracts the information that needs to be "scheduled" and "notified."

[0862] 3. Identifying the right generative AI

[0863] The server searches for and identifies "generative artificial intelligence for schedule creation" and "generative artificial intelligence for notifications."

[0864] 4. Calling generative AI and providing services

[0865] The server calls each generative AI, creates a schedule, and notifies relevant parties.

[0866] 5. Check the results

[0867] The server provides the generated schedule and notification contents to the user, who then confirms them.

[0868] 6. Deployment of automated routines

[0869] If the user is satisfied, the server automates the process and deploys it as a web application.

[0870] In this way, the present invention efficiently operates generative artificial intelligence based on the user's natural language requests, thereby optimizing workflow.

[0871] The processing flow will be explained below.

[0872] Step 1:

[0873] A user inputs a natural language request using a terminal, for example, a request such as "Please create a schedule for the next project meeting and notify the relevant parties."

[0874] Step 2:

[0875] The terminal sends this natural language request to the server, which transmits the request over the Internet and receives it at the server's receiving module.

[0876] Step 3:

[0877] The server passes the received natural language request to an NLP (Natural Language Processing) module for analysis. This analysis module understands the content of the request and extracts related tasks and requirements. For example, "schedule creation" and "notify relevant parties" are extracted as tasks.

[0878] Step 4:

[0879] The server searches a database of generative artificial intelligence (GA) models based on the extracted tasks. This search module accesses databases such as the OpenAI Store to identify "GA for scheduling" and "GA for notifications."

[0880] Step 5:

[0881] The server then calls the identified generative AI and performs specific processing in response to the user's request. For example, a generative AI for schedule creation generates a schedule for a project meeting and returns the schedule data. Similarly, a generative AI for notification notifies relevant parties of the generated schedule.

[0882] Step 6:

[0883] The server compiles the generated schedule and notification results and provides feedback to the user. The user can then check the results of the services provided by the server through their terminal and evaluate their level of satisfaction.

[0884] Step 7:

[0885] If the user is satisfied with the results of the provided service, the server automates the event-service routine. The automation module does this and implements the routine as a web application.

[0886] Step 8:

[0887] The server provides the built web application to the user's device and automatically processes similar requests in the future. The web application receives events in the form of voice, text, sensors, date and time, and provides services in the form of text, images, videos, spreadsheets, code, and documents.

[0888] Thus, through each step centered on the server, the terminal, and the user, the present invention provides an efficient and flexible service for natural language requests from users.

[0889] Example 1

[0890] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0891] In conventional systems, the process of users entering requests in natural language, searching for and identifying appropriate generative AI models based on those requests, and providing services was not automated, making it difficult to achieve an efficient workflow. It was also difficult to build a system that could confirm satisfaction with the generated results and automatically respond to similar requests in the future. This resulted in a major issue of the time and effort required for the process.

[0892] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0893] In this invention, the server includes means for receiving and analyzing a natural language request, means for searching for and identifying multiple appropriate generative AI models based on the content of the analyzed request, means for extracting a gist of the request using a natural language processing module, means for searching for and selecting a generative AI model from a database based on the extracted gist, means for calling the selected generative AI and providing a specific service in response to the request, means for providing the generated results to a user and confirming the user's satisfaction, means for automating the post-confirmation routine and deploying it as a Web application, and means for publishing it as a Web application for automatic processing of subsequent requests. This makes it possible to efficiently and automatically operate a generative AI model based on a user's natural language request, significantly reducing the effort and time required for the work.

[0894] A "natural language request" is a request or instruction entered by a user using human language.

[0895] "Parsing" is the act of receiving a natural language request and converting its content into an understandable format.

[0896] "Generative AI" is a type of AI model that is trained to perform specific tasks and has the ability to generate new information and results.

[0897] "Searching" is the act of finding items in a database or repository that match specific criteria.

[0898] "Identifying" is the act of selecting the most appropriate item from those found in a search.

[0899] A "natural language processing module" is a program or tool that converts natural language into a format that is easy for a computer to understand.

[0900] The "Summary" is a brief summary of the main content and purpose of the request.

[0901] A "database" is a system that stores large amounts of data in an organized manner and makes it easy to search and manage.

[0902] "Invocation" refers to sending a command or operation to execute a program or service.

[0903] "Satisfaction confirmation" is the act of checking whether the provided services and results meet the user's expectations.

[0904] A "routine" is a series of processing steps in a program that defines certain operations and processing procedures.

[0905] "Automation" refers to making a system operate automatically without human intervention.

[0906] A "web application" is application software that is used over the Internet and operates via a web browser.

[0907] The present invention is a system that allows a user to input a natural language request, and then calls an appropriate generative AI based on the request to provide a service. This system is primarily composed of a server, a terminal, and a generative AI model. A specific implementation method is described below.

[0908] A user inputs a request in natural language using a device (e.g., a personal computer, smartphone, tablet, etc.). This request is sent to the server via a web browser. A specific example is a request such as, "Please create a schedule for the next project meeting and notify the relevant parties."

[0909] The server uses natural language processing modules such as Python's NLTK or spaCy to parse the incoming request, extracting the gist of the request, such as "schedule" and "notify relevant parties."

[0910] Next, the server searches and selects an appropriate model from a database of generative AI models based on the extracted summary. It is possible to use OpenAI's GPT-3 for scheduling and Twilio API for notifications.

[0911] The selected generative AI model is then called by the server to provide a specific service: first, GPT-3 is used to generate a schedule for a project meeting, and then the generated schedule is notified to relevant parties using the Twilio API.

[0912] The generated results are provided to the user by the server, who checks them through their terminal. If the user is satisfied with the results, they send their feedback to the server, which uses this feedback to automate the routine.

[0913] Finally, the automated routine can be deployed as a web application, so that similar requests can be processed automatically in the future. For example, the web application can be built using the Python Flask framework.

[0914] Examples of prompt statements

[0915] Prompt for scheduling a project meeting:

[0916] "Please schedule a project meeting for next Monday. Please include an announcement to stakeholders."

[0917] Prompt for generating notification content:

[0918] "Create a meeting notification based on the following schedule: [generated schedule content]"

[0919] In this way, the present invention efficiently operates generative artificial intelligence based on the user's natural language request, significantly reducing the effort and time required for the work.

[0920] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0921] Step 1:

[0922] User Input

[0923] The user uses a terminal to input a natural language request, for example, "Please create a schedule for the next project meeting and notify the relevant parties," into an input form in a web browser and presses the send button.

[0924] Input: A natural language request typed by a user using a terminal.

[0925] Output: When the submit button is pressed, the request is sent to the server.

[0926] Step 2:

[0927] Receiving and parsing the request

[0928] The server receives the request and then uses a natural language processing module such as Python's NLTK or spaCy to tokenize the request and extract keywords, extracting the gist of the request: "schedule" and "notify."

[0929] Input: A natural language request sent by the user.

[0930] Output: A brief summary of the parsed request (e.g., "Schedule", "Notify").

[0931] Step 3:

[0932] Search and Identify Generative Artificial Intelligence

[0933] Based on the analysis results, the server searches for and selects an appropriate model from a database of generative AI models. For example, for "schedule creation," it identifies high-performance natural language generation software as a generative AI model, and for "notification," it identifies a high-performance communication platform.

[0934] Input: A summary of the parsed request.

[0935] Output: A selected generative artificial intelligence model (e.g., high-performance natural language generation software, high-performance communications platform).

[0936] Step 4:

[0937] Calling generative AI and providing services

[0938] The server invokes selected generative artificial intelligence models to provide specific services, first generating a project meeting schedule using high-performance natural language generation software, and then communicating the generated schedule to stakeholders using a high-performance communication platform.

[0939] Input: Selected generative artificial intelligence model and a summary of the request.

[0940] Output: Generated schedule contents and notification results.

[0941] Step 5:

[0942] Check user satisfaction

[0943] The server provides the generated schedule and notification content to the user, who then checks the results through their terminal. If the user is satisfied with the results, they send their feedback to the server.

[0944] Input: Generated schedule contents and notification results.

[0945] Output: User feedback (confirmation of satisfaction).

[0946] Step 6:

[0947] Routine automation and web application deployment

[0948] The server automates this process based on user feedback by building and deploying a web application using the Python Flask framework, which will allow it to process similar requests automatically in the future.

[0949] Input: User feedback and verified routines.

[0950] Output: A web application as an automated routine.

[0951] (Application example 1)

[0952] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0953] As automation in factories continues to increase, many tasks are now performed by robots and automated equipment. However, efficiently controlling and managing these devices remains a complex process, and it is particularly difficult to achieve fast and efficient control using natural language. Furthermore, because users must clearly instruct each task, training is required and operation can be cumbersome. This can result in reduced factory productivity.

[0954] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0955] In this invention, the server includes means for receiving and analyzing natural language requests, means for searching for and identifying multiple appropriate generative AIs based on the content of the analyzed request, means for calling the identified multiple generative AIs to generate operating procedures for controlling equipment in the factory and providing the procedures to the equipment in the factory, means for confirming user satisfaction and automating event-service routines, and means for deploying the automated routines as web applications to automatically process future requests. This enables efficient control of robots and automated equipment in the factory and improves productivity throughout the factory.

[0956] A "natural language request" refers to a request entered by a user in a language that they normally use.

[0957] "Parsing" is the process of interpreting an input natural language request and understanding its content.

[0958] "Generative AI" refers to an AI model that generates outputs appropriate for a specific task.

[0959] "Search" is the process of finding the model that best suits a request from a generative artificial intelligence database.

[0960] "Identification" refers to determining the generative artificial intelligence that best suits the request from the search results.

[0961] "Invoking" refers to putting a specified generative artificial intelligence into operation and starting processing in response to a request.

[0962] "Operation procedure" refers to the specific procedure for operating equipment in a factory, generated by generative artificial intelligence.

[0963] "Provision" is the process of sending the generated operating procedure to the equipment in the factory and making it executable.

[0964] "Satisfaction" refers to the evaluation and impressions that users have of the results of the service provided.

[0965] A "routine" is a sequence of automatically and repeatedly performed steps or processes.

[0966] "Automation" refers to the state in which a process proceeds without the need for manual intervention.

[0967] "Web application" refers to a software application that is available over the Internet.

[0968] "Factory equipment" refers to robots and automated equipment used in factories in general.

[0969] This invention is a system that allows users to input requests in natural language and efficiently operate and manage automated equipment in a factory based on those requests. Specifically, a server analyzes the user's request, generates operating procedures using appropriate generative artificial intelligence, and provides them to various equipment in the factory.

[0970] Basic system configuration

[0971] 1. Receiving and parsing natural language requests

[0972] Users input requests in natural language through devices such as smartphones or tablets.

[0973] The server uses a Natural Language Processing (NLP) module to parse the request and extract the gist of the request, e.g., "Please assemble the following parts."

[0974] 2. Searching and Identifying Generative AI

[0975] Based on the analysis results, the server searches and identifies appropriate models from a database of generative AI, including "generative AI for scheduling" and "generative AI for notifications."

[0976] 3. Calling the generative AI and generating operational procedures

[0977] The server then calls the identified generative artificial intelligence to generate operational procedures based on the analyzed request, such as assembly procedures or device operation procedures.

[0978] 4. Providing operating procedures

[0979] The generated operation procedures are provided to robot arms and other in-factory equipment so that they can be executed.

[0980] 5. User satisfaction checks and routine automation

[0981] The user reviews the results provided and provides feedback on their satisfaction to the server, which, if satisfied, automates this routine and serves future requests as a web application.

[0982] Hardware and software used

[0983] Hardware

[0984] Smartphones, tablets, and PCs: for inputting users' natural language requests

[0985] Robotic arms and other factory equipment: Equipment that performs the actual movements

[0986] software

[0987] Natural Language Processing (NLP) modules: e.g., NLTK, spaCy, etc.

[0988] Generative AI modules: e.g., GPT-3 or other custom AI models

[0989] Robot control module: e.g., ROS (Robot Operating System)

[0990] Specific examples

[0991] The user inputs a request into their smartphone, such as "Please assemble the following parts." This request is analyzed by the NLP module, and the task "assembly of parts" is extracted. The generative AI generates an assembly procedure suitable for this task and sends it to the robot control module. Finally, the robot arm assembles the specified parts.

[0992] Example prompts to input to the generative AI model

[0993] "Please generate the steps to assemble parts A and B and attach part C."

[0994] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0995] Step 1:

[0996] A user inputs a natural language request into a terminal. The terminal sends the request to a server. The input is a specific natural language request such as "Please assemble the following parts," and the output is that the request is sent to the server. Here, text data entered on the terminal's operating screen is formed.

[0997] Step 2:

[0998] The server analyzes the received natural language request. Specifically, a natural language processing (NLP) module analyzes the request and extracts its gist. In this step, the input natural language sentence is analyzed and the action instructions for "assembly" are extracted. The input is the request data sent from the terminal, and the output is structured data containing the task instruction "assembly."

[0999] Step 3:

[1000] The server searches for and identifies an appropriate generative AI based on the analysis results. The server references a database of pre-registered generative AI models to search for a model suitable for the request. The input to this step is a structured task instruction, and the output is an appropriate generative AI model. Here, a generative AI model that matches the recognized task is searched for.

[1001] Step 4:

[1002] The server calls the identified generative AI to generate the required operating procedure. The called generative AI generates the operating procedure based on the given task instructions. In this step, specific instructions such as "Please assemble the following parts" are input as a prompt, and the output is a specific assembly procedure. The operating procedure is expressed as a series of control commands generated by the generative AI.

[1003] Step 5:

[1004] The server provides the generated operating procedures to the equipment in the factory. Here, a robot control module is used to send the generated instructions to a robot arm or other automated equipment. The input of this step is the generated operating procedure, and the output is the specific control signal received by the equipment. Here, the generated procedure is sent to the actual equipment and execution begins.

[1005] Step 6:

[1006] The user checks the provided results and provides feedback on their satisfaction to the server. The item the user checks is whether the operation was actually performed correctly. The input of this step is the operation result, and the output is the user's satisfaction feedback. The confirmation process is completed when the user provides feedback based on the actual operation result.

[1007] Step 7:

[1008] The server automates the event-service routine based on the user's satisfaction feedback and deploys it as a web application. In this step, past feedback is analyzed and an automated routine is designed. The input of this step is satisfaction feedback data, and the output is a web application containing the automated routine. The automated routine is configured to respond immediately to similar requests from the next time onwards.

[1009] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1010] The present invention is a system that combines natural language request processing with an emotion engine to optimize the user experience. Specific embodiments of the present invention are described below.

[1011] Basic system configuration

[1012] 1. User Input

[1013] The user uses a terminal to input a natural language request, for example, "Please schedule the next project meeting and notify the relevant parties."

[1014] 2. Receiving and parsing the request

[1015] The server receives a natural language request from the user, which is then parsed by a natural language processing (NLP) module to extract the gist of the request, which includes "schedule" and "notify relevant parties."

[1016] 3. Searching and Identifying Generative AI

[1017] Based on the analysis results of the request, the server searches for appropriate models from a database of generative artificial intelligence and identifies a "generative artificial intelligence for schedule creation" and a "generative artificial intelligence for notifications."

[1018] 4. Calling generative AI and providing services

[1019] The server calls the identified generative AI and performs specific processing according to the request. The generative AI for schedule creation generates a schedule for the project meeting and returns the schedule data. Similarly, the generative AI for notification notifies the relevant parties of the generated schedule.

[1020] 5. User Emotion Recognition by Emotion Engine

[1021] The server recognizes the user's emotions through the request and the result of the provided service, for example, the emotion engine analyzes whether the user is satisfied or dissatisfied.

[1022] 6. User Satisfaction Check

[1023] The server checks the user's satisfaction level based on the analysis results of the emotion engine, and if the user is satisfied, collects their feedback and uses it to optimize the automation routines.

[1024] 7. Routine Automation and Web Application Deployment

[1025] The server automates the event-service routines, which are deployed as web applications and automatically process future requests from users.

[1026] Specific examples

[1027] For example, if a user requests, "Please create a schedule for a project meeting next Monday and notify the relevant parties," the following processing is performed.

[1028] 1. Input Reception

[1029] The user inputs a request through the terminal.

[1030] 2. Request Analysis

[1031] The server receives the request, analyzes it, and extracts the information that needs to be "scheduled" and "notified."

[1032] 3. Identifying the right generative AI

[1033] The server searches for and identifies "generative artificial intelligence for schedule creation" and "generative artificial intelligence for notifications."

[1034] 4. Calling generative AI and providing services

[1035] The server calls each generative AI, creates a schedule, and notifies relevant parties.

[1036] 5. Emotion recognition

[1037] After the server provides the service, it uses an emotion engine to analyze the user's emotions and measure their satisfaction.

[1038] 6. Check the results and optimize

[1039] The user checks the provided schedule and notification content, and the server collects feedback based on the analysis results of the emotion engine.

[1040] 7. Deployment of automated routines

[1041] If the user is satisfied, the server automates the process and deploys it as a web application.

[1042] In this way, the present invention is a system that combines generative artificial intelligence and an emotion engine to effectively provide services based on a user's natural language request, thereby optimizing workflow.

[1043] The processing flow will be explained below.

[1044] Step 1:

[1045] A user inputs a natural language request using a terminal, for example, a request such as "Please create a schedule for the next project meeting and notify the relevant parties."

[1046] Step 2:

[1047] The terminal sends this natural language request to the server, which transmits the request over the Internet and receives it at the server's receiving module.

[1048] Step 3:

[1049] The server passes the received natural language request to an NLP (Natural Language Processing) module for analysis. This analysis module understands the content of the request and extracts related tasks and requirements. For example, "schedule creation" and "notify relevant parties" are extracted as tasks.

[1050] Step 4:

[1051] The server searches a database of generative AIs based on the extracted tasks. This search module accesses databases such as the OpenAI Store to find suitable generative AIs, and identifies "generative AIs for scheduling" and "generative AIs for notifications."

[1052] Step 5:

[1053] The server then calls the identified generative AI and performs specific processing in response to the user's request. The generative AI for schedule creation generates a schedule for the project meeting and returns the schedule data. Similarly, the generative AI for notification notifies the relevant parties of the generated schedule.

[1054] Step 6:

[1055] The server compiles the generated schedule and notification results and provides feedback to the user. The user can then check the results of the services provided by the server through their terminal and evaluate their level of satisfaction.

[1056] Step 7:

[1057] The server uses an emotion engine to analyze the user's emotions based on the feedback, for example, through natural language requests and the results of the provided service, the emotion engine recognizes whether the user is satisfied or dissatisfied.

[1058] Step 8:

[1059] Based on the analysis results of the emotion engine, the server optimizes user satisfaction. For example, if the user is dissatisfied, it will select a different generative AI or improve the existing process.

[1060] Step 9:

[1061] If the user is satisfied with the results of the provided service, the server automates the event-service routine. The automation module does this and implements the routine as a web application.

[1062] Step 10:

[1063] The server provides the built web application to the user's device and automatically processes similar requests in the future. The web application receives events in the form of voice, text, sensors, date and time, and provides services in the form of text, images, videos, spreadsheets, code, and documents.

[1064] In this way, through each step centered on the server, terminal, and user, the present invention is a system that provides efficient and flexible services in response to natural language requests from users and further increases user satisfaction using an emotion engine.

[1065] Example 2

[1066] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1067] Currently, many systems only provide simple information or limited services in response to user requests, and do not optimize the user experience or continuously improve processes based on feedback. As a result, the problems and frustrations faced by users are not fully resolved. In particular, processing complex requests, collecting feedback using emotion recognition, and automating processes are difficult challenges.

[1068] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1069] In this invention, the server includes means for receiving and analyzing natural language requests, means for searching for and identifying multiple appropriate generative AI models based on the content of the analyzed request, means for calling the identified generative AI model to execute a service according to the request and providing the results to the target, means for analyzing the target's emotions and confirming satisfaction, and means for automating an event-service routine, deploying the automated routine as a web application, and automatically processing future requests. This enables the provision of appropriate services based on user requests, optimization based on emotional feedback, and efficient processing of future requests.

[1070] A "natural language request" is a request that a user enters into a system in human language.

[1071] "Parsing" is the process of understanding a natural language request based on its meaning and structure.

[1072] A "generative artificial intelligence model" is an artificial intelligence model that has the ability to generate new data and information from given input.

[1073] "Search" is the process of finding models that meet specific criteria from a database of generative artificial intelligence models.

[1074] "Identification" is the process of selecting the most suitable generative artificial intelligence model from the searched results.

[1075] "Executing a service" means performing processing in response to a user request using a specified generative artificial intelligence model.

[1076] "Providing" means delivering the results of the performed service to the user.

[1077] "Analyzing emotions" is the process of analyzing a user's feedback and behavior to infer their emotional state.

[1078] "Confirming satisfaction level" refers to confirming the degree of satisfaction of the user with the provided service based on the results of sentiment analysis.

[1079] An "event-service routine" is a systematized flow of a series of service processes that correspond to a specific event.

[1080] An "automated routine" refers to a series of processes that a system executes automatically without human intervention.

[1081] A "web application" is a software application available over the Internet.

[1082] "Future requests" refers to additional requests that the user may make in the future.

[1083] The present invention is a system that efficiently processes natural language requests from users, provides services, and optimizes the process based on emotional feedback. The system consists of three main components: a server, a terminal, and a user.

[1084] System configuration

[1085] 1. User Input

[1086] A user inputs a natural language request using a device, such as a PC, smartphone, or tablet, using keyboard or voice input. For example, a request might be, "Please schedule the next project meeting and notify the relevant parties."

[1087] 2. Receiving and parsing the request

[1088] The server receives requests from users. This server can be a physical server equipped with an Intel Xeon processor or a cloud service instance (e.g., AWS EC2). The received request is analyzed using a natural language processing (NLP) module such as Google's Cloud Natural Language API or SpaCy, and the gist of the request is extracted. For example, "Create a schedule" and "Notify relevant parties" are extracted as the gist.

[1089] 3. Searching and identifying generative AI models

[1090] Based on the analysis results of the NLP module, the server searches and identifies appropriate generative AI models from the database. Examples of generative AI models that can be used include OpenAI's GPT-4 and similar generative systems. A "generative AI model for schedule creation" and a "generative AI model for notifications" are identified.

[1091] 4. Calling the generative AI model and providing it as a service

[1092] The server then calls the identified generative AI model and performs specific processing according to the user's request. For example, GPT-4 generates a schedule for a project meeting and returns the schedule data. Similarly, the notification AI model notifies relevant parties of the generated schedule.

[1093] 5. User Emotion Recognition by Emotion Engine

[1094] The server recognizes the user's emotions based on the execution results. The emotion engine uses IBM Watson Tone Analyzer, among others. This engine analyzes whether the user is satisfied or dissatisfied. For example, it can estimate emotions by analyzing user feedback and behavioral data.

[1095] 6. User Satisfaction Check

[1096] The server checks the user's satisfaction with the service provided based on the output of the emotion engine, and if so, collects their feedback to help improve the process.

[1097] 7. Routine Automation and Web Application Deployment

[1098] The server then templates the process of successful requests and deploys them as automated routines in frameworks such as Django or Flask as web applications that automatically process similar requests in the future.

[1099] Specific examples

[1100] If a user requests, "Please schedule a project meeting for next Monday and notify the relevant parties," the following processing occurs:

[1101] 1. Enter your natural language request

[1102] The user uses a PC and inputs data using the keyboard.

[1103] 2. Receiving and parsing the request

[1104] The server receives this request and analyzes it using the NLP module (Google Cloud Natural Language API). It extracts the necessary information: "schedule creation" and "notification."

[1105] 3. Searching and identifying generative AI models

[1106] The server identifies GPT-4 from its database as the generative AI model for scheduling and T5 as the generative AI model for notifications.

[1107] 4. Calling the generative AI model and providing it as a service

[1108] The server calls GPT-4 to generate the schedule and calls T5 to notify relevant parties of the schedule.

[1109] 5. Emotion Recognition by Emotion Engine

[1110] The server uses IBM Watson Tone Analyzer to analyze the user's emotions and measure their satisfaction.

[1111] 6. User satisfaction confirmation and optimization

[1112] The user checks the provided schedule and notification content, and the server collects feedback based on the analysis results of the emotion engine.

[1113] 7. Automate routines

[1114] If the user is satisfied, the server will template this process and deploy it as a web application using the Django framework.

[1115] This allows us to provide appropriate services based on users' natural language requests, continuously improve our processes based on feedback, and efficiently process requests.

[1116] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1117] Step 1:

[1118] A user uses a terminal to input a natural language request.

[1119] Specific operation: A user uses a PC or smartphone to input a request using a keyboard or voice input, such as "Please schedule the next project meeting and notify relevant parties."

[1120] Input: A request expressed in natural language.

[1121] Output: Natural language text data.

[1122] Step 2:

[1123] The server receives a natural language request from a user.

[1124] Specific operation: The server receives the text data sent from the terminal and stores it in a database.

[1125] Input: Natural language text data sent by the user from their device.

[1126] Output: Request data stored in a database on the server.

[1127] Step 3:

[1128] The server analyzes the received request.

[1129] What happens: The server uses the Google Cloud Natural Language API to parse the request text and extract the main points: "Create a schedule" and "Notify relevant parties."

[1130] Input: Request data stored in the database.

[1131] Output: A brief summary of the extracted request.

[1132] Step 4:

[1133] The server searches for and identifies an appropriate generative AI model based on the analysis results.

[1134] Specific operation: The server identifies GPT-4 as the scheduling model and T5 as the notification model from the database.

[1135] Input: Summary of analysis results ("Schedule Creation" and "Notify Stakeholders").

[1136] Output: The identified generative AI model.

[1137] Step 5:

[1138] The server calls the identified generative AI model and performs specific processing in response to the user's request.

[1139] How it works: The server uses GPT-4 to generate a project meeting schedule and retrieves the data. Similarly, the notification AI model notifies relevant parties of the generated schedule.

[1140] Input: The identified generative AI model and a brief summary of the request.

[1141] Output: Generated schedule data and notification execution results.

[1142] Step 6:

[1143] The server uses an emotion engine to recognize the user's emotion.

[1144] Specific operation: The server uses IBM Watson Tone Analyzer to analyze the user's feedback and behavioral data after the service is provided and estimate the user's emotional state.

[1145] Input: User feedback and behavioral data after the service is provided.

[1146] Output: Estimated emotional state.

[1147] Step 7:

[1148] The server checks the user's satisfaction level based on the analysis results of the emotion engine.

[1149] What happens: The server measures satisfaction and, if the user is satisfied, collects that feedback and uses it to improve the process.

[1150] Input: Estimated emotional state.

[1151] Output: Satisfaction ratings and feedback data.

[1152] Step 8:

[1153] The server templatizes the process of successful requests and deploys the web application as an automated routine.

[1154] What it does: The server uses Django or Flask to automate the generated schedule and notification process and deploy it as a web application.

[1155] Input: Process and feedback data for successful requests.

[1156] Output: An automated web application.

[1157] This allows for the provision of appropriate services based on the user's natural language requests, and for process optimization and efficient request handling based on emotional feedback.

[1158] (Application example 2)

[1159] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1160] In food delivery services, when users communicate their orders in natural language, it is difficult to analyze their requests in detail and provide optimal suggestions. Furthermore, to increase user satisfaction, it is necessary to recognize the extent to which the service provided affected the user's emotions and improve the quality of the service based on that information. Conventional systems are unable to effectively integrate these elements, limiting their ability to optimize services and improve user satisfaction.

[1161] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1162] In this invention, the server includes means for receiving and analyzing natural language requests, means for searching for and identifying multiple appropriate generative artificial intelligences based on the content of the analyzed request, means for calling the identified multiple generative artificial intelligences to execute a service according to the request and providing the result to the user, means for checking user satisfaction and automating an event-service routine, means for deploying the automated routine as a web application and thereby automatically processing future requests, means for proposing food delivery menus and presenting estimated delivery times based on the user's natural language request, and means for analyzing user emotions from user feedback after providing the service to improve satisfaction. This enables a quick and accurate response to the user's natural language request and, at the same time, improves the quality of the service by recognizing how the provided service affected the user's emotions.

[1163] A "natural language request" is a request or command that a user enters into a system using everyday language.

[1164] "Parsing" is the process of breaking down the content of an incoming natural language request to determine its gist and purpose.

[1165] "Generative AI" refers to AI that has the ability to generate appropriate content and data based on input information.

[1166] "Searching and identifying" means finding a generative artificial intelligence from a database that matches the content of the analyzed request and preparing it for use.

[1167] "Request-based services" refer to specific processing and results provided by generative artificial intelligence based on a user's natural language request.

[1168] "User satisfaction" is an evaluation index that indicates how well a provided service meets the user's expectations and requirements.

[1169] An "event-service routine" is a set of service delivery procedures or processes that respond to a particular event or request.

[1170] An "automated routine" is a series of actions that the system takes to service an event or request automatically, without manual intervention.

[1171] A "web application" is software that runs on the Internet and is accessed by users through a browser.

[1172] "Food delivery menu suggestions" is a service that provides appropriate food and dish options based on user requests.

[1173] "Expected Delivery Time" means calculating and providing an estimate of the time it will take for an item to be delivered after an order is completed.

[1174] "Feedback" refers to the evaluations and impressions that users make about the services provided.

[1175] "Analyzing emotions" is the process of identifying and analyzing emotions (satisfaction, dissatisfaction, joy, etc.) from user feedback.

[1176] The present invention is a food delivery service system that receives natural language requests from users, analyzes them, and uses appropriate generative artificial intelligence to make optimal suggestions and provide delivery services. This system is operated through a smartphone application.

[1177] Basic system configuration

[1178] 1. User Input

[1179] Users input requests in natural language through a smartphone application, for example, "I'd like to order sushi. Please tell me what's recommended today. I'd also like to know how long it will take to deliver."

[1180] 2. Receiving and parsing the request

[1181] The server receives a natural language request from the user. The request is analyzed by a natural language processing (NLP) module (e.g., SpaCy or BERT) to extract the gist of the request. In this example, the gist is "Recommend sushi menu" and "Confirm delivery time."

[1182] 3. Searching and Identifying Generative AI

[1183] Based on the analysis of the request, the server searches and identifies appropriate models from its database of generative artificial intelligence. Specifically, it identifies a "generative artificial intelligence for menu suggestions" (e.g., GPT-4) and a "generative artificial intelligence for delivery time prediction."

[1184] 4. Calling generative AI and providing services

[1185] The server calls the identified generative AI and performs specific processing in response to the request. The generative AI for menu suggestions generates the optimal sushi menu and returns the data. The generative AI for delivery time prediction calculates the predicted delivery time and returns the result. This information is provided to the user via a smartphone application.

[1186] 5. User Emotion Recognition by Emotion Engine

[1187] The server receives the user's feedback after the service is provided and analyzes the user's emotions using an emotion engine (e.g., Google Cloud Natural Language API). The emotion engine analyzes whether the user is satisfied or dissatisfied.

[1188] 6. User Satisfaction Check

[1189] The server checks the user's satisfaction level based on the analysis results of the emotion engine, and if the user is satisfied, collects the feedback and stores it in a database.

[1190] 7. Routine Automation and Web Application Deployment

[1191] The server automates the event-service routines, which are deployed as web applications and perform automated processing when future requests are received from the user.

[1192] Specific examples

[1193] A user types a natural language request into a smartphone application, such as:

[1194] "I'm planning a family party next Sunday. Could you recommend a sushi set? Also, let me know the delivery time if I order now."

[1195] The server analyzes this request and uses appropriate generative artificial intelligence to suggest a sushi set and provide a predicted delivery time. The user then reviews the suggested menu and delivery time and makes their order decision based on that. After providing the service, the emotion engine analyzes the user's feedback and uses it to improve the service in the future.

[1196] This process allows for a fast and accurate response to the user's natural language requests and improves the quality of the service by recognizing how the provided service affected the user's emotions.

[1197] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1198] Step 1:

[1199] A user inputs a request in natural language through a smartphone application. For example, "I'd like to order sushi. Please tell me what's recommended today. I'd also like to know how long it will take to deliver." This request is then sent to the server by the application.

[1200] Input: User's natural language request

[1201] Output: The natural language request sent to the server

[1202] Step 2:

[1203] The server receives a natural language request from the user. The request is analyzed by a natural language processing (NLP) module (e.g., SpaCy or BERT) to extract the gist of the request. Specifically, the gist is extracted as "recommended sushi menu" and "confirm delivery time."

[1204] Input: A natural language request sent to the server

[1205] Output: A summary of the parsed request

[1206] Step 3:

[1207] Based on the analysis results of the request, the server searches and identifies an appropriate model from its database of generative AI. Specifically, it identifies a "generative AI for menu suggestions" (e.g., GPT-4) and a "generative AI for delivery time prediction." At this point, it selects the AI ​​model that best suits the content of the request.

[1208] Input: Parsed request summary

[1209] Output: Identified generative artificial intelligence model

[1210] Step 4:

[1211] The server calls the identified generative AI and performs specific processing in response to the request. The generative AI for menu suggestions generates the optimal sushi menu and returns the data. The generative AI for delivery time prediction calculates the predicted delivery time and returns the result. This information is provided to the user via a smartphone application.

[1212] Input: Identified generative artificial intelligence model, parsed request abstract

[1213] Output: Generated sushi menu, predicted delivery time

[1214] Step 5:

[1215] The server receives the user's feedback after the service is provided. It analyzes the user's emotions using an emotion engine (e.g., Google Cloud Natural Language API). The emotion engine analyzes whether the user is satisfied or dissatisfied.

[1216] Input: User feedback

[1217] Output: Analyzed user sentiment results

[1218] Step 6:

[1219] The server checks the user's satisfaction level based on the analysis results of the emotion engine. If the user is satisfied, the server collects the feedback and stores it in a database. This information can be used to improve the service in the future.

[1220] Input: Analyzed user sentiment results

[1221] Output: Saved feedback

[1222] Step 7:

[1223] The server automates the event-service routine. The automated routine is deployed as a web application and automatically processes future requests from users. This allows users to always receive optimized service.

[1224] Input: Stored feedback, parsed request summary

[1225] Output: Automated routines, deployed web applications

[1226] This enables a fast and accurate response to the user's natural language requests, and improves the quality of the service by recognizing how the provided service affected the user's emotions.

[1227] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1228] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1229] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1230] [Fourth embodiment]

[1231] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1232] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1233] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1234] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1235] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1236] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1237] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1238] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1239] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1240] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1241] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1242] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1243] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1244] The present invention is a system in which a user inputs a request in natural language, a server analyzes the request, calls an appropriate generative artificial intelligence, and ultimately provides a service to the user. Specific embodiments of the present invention will be described below.

[1245] Basic system configuration

[1246] 1. User Input

[1247] The user inputs a request in natural language using a terminal, such as "Please create a schedule for the next project meeting and notify the relevant parties."

[1248] 2. Receiving and parsing the request

[1249] The server receives a natural language request from the user, which is then parsed by a natural language processing (NLP) module to extract the gist of the request, which includes "schedule" and "notify relevant parties."

[1250] 3. Searching and Identifying Generative AI

[1251] Based on the analysis of the request, the server searches a database of generative AI models for suitable models, for example, identifying a "generative AI for scheduling" and a "generative AI for notifications."

[1252] 4. Calling generative AI and providing services

[1253] The server calls the identified generative AI and executes the service according to the request. The generative AI for schedule creation generates a schedule for the project meeting, and the generative AI for notification notifies the relevant parties of the schedule.

[1254] 5. User satisfaction check

[1255] The server provides the results of the service performed by the generative AI to the user and confirms their satisfaction. For example, the user checks the provided schedule and notification content and provides feedback on whether they are satisfied.

[1256] 6. Routine Automation and Web Application Deployment

[1257] If the user is satisfied, the server automates the event-service routine, which is then deployed as a web application to automatically process similar requests in the future.

[1258] Specific examples

[1259] For example, if a user requests, "Please create a schedule for a project meeting next Monday and notify the relevant parties," the following processing is performed.

[1260] 1. Input Reception

[1261] The user inputs a request through the terminal.

[1262] 2. Request Analysis

[1263] The server receives the request, analyzes it, and extracts the information that needs to be "scheduled" and "notified."

[1264] 3. Identifying the right generative AI

[1265] The server searches for and identifies "generative artificial intelligence for schedule creation" and "generative artificial intelligence for notifications."

[1266] 4. Calling generative AI and providing services

[1267] The server calls each generative AI, creates a schedule, and notifies relevant parties.

[1268] 5. Check the results

[1269] The server provides the generated schedule and notification contents to the user, who then confirms them.

[1270] 6. Deployment of automated routines

[1271] If the user is satisfied, the server automates the process and deploys it as a web application.

[1272] In this way, the present invention efficiently operates generative artificial intelligence based on the user's natural language requests, thereby optimizing workflow.

[1273] The processing flow will be explained below.

[1274] Step 1:

[1275] A user inputs a natural language request using a terminal, for example, a request such as "Please create a schedule for the next project meeting and notify the relevant parties."

[1276] Step 2:

[1277] The terminal sends this natural language request to the server, which transmits the request over the Internet and receives it at the server's receiving module.

[1278] Step 3:

[1279] The server passes the received natural language request to an NLP (Natural Language Processing) module for analysis. This analysis module understands the content of the request and extracts related tasks and requirements. For example, "schedule creation" and "notify relevant parties" are extracted as tasks.

[1280] Step 4:

[1281] The server searches a database of generative artificial intelligence (GA) models based on the extracted tasks. This search module accesses databases such as the OpenAI Store to identify "GA for scheduling" and "GA for notifications."

[1282] Step 5:

[1283] The server then calls the identified generative AI and performs specific processing in response to the user's request. For example, a generative AI for schedule creation generates a schedule for a project meeting and returns the schedule data. Similarly, a generative AI for notification notifies relevant parties of the generated schedule.

[1284] Step 6:

[1285] The server compiles the generated schedule and notification results and provides feedback to the user. The user can then check the results of the services provided by the server through their terminal and evaluate their level of satisfaction.

[1286] Step 7:

[1287] If the user is satisfied with the results of the provided service, the server automates the event-service routine. The automation module does this and implements the routine as a web application.

[1288] Step 8:

[1289] The server provides the built web application to the user's device and automatically processes similar requests in the future. The web application receives events in the form of voice, text, sensors, date and time, and provides services in the form of text, images, videos, spreadsheets, code, and documents.

[1290] Thus, through each step centered on the server, the terminal, and the user, the present invention provides an efficient and flexible service for natural language requests from users.

[1291] Example 1

[1292] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1293] In conventional systems, the process of users entering requests in natural language, searching for and identifying appropriate generative AI models based on those requests, and providing services was not automated, making it difficult to achieve an efficient workflow. It was also difficult to build a system that could confirm satisfaction with the generated results and automatically respond to similar requests in the future. This resulted in a major issue of the time and effort required for the process.

[1294] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1295] In this invention, the server includes means for receiving and analyzing a natural language request, means for searching for and identifying multiple appropriate generative AI models based on the content of the analyzed request, means for extracting a gist of the request using a natural language processing module, means for searching for and selecting a generative AI model from a database based on the extracted gist, means for calling the selected generative AI and providing a specific service in response to the request, means for providing the generated results to a user and confirming the user's satisfaction, means for automating the post-confirmation routine and deploying it as a Web application, and means for publishing it as a Web application for automatic processing of subsequent requests. This makes it possible to efficiently and automatically operate a generative AI model based on a user's natural language request, significantly reducing the effort and time required for the work.

[1296] A "natural language request" is a request or instruction entered by a user using human language.

[1297] "Parsing" is the act of receiving a natural language request and converting its content into an understandable format.

[1298] "Generative AI" is a type of AI model that is trained to perform specific tasks and has the ability to generate new information and results.

[1299] "Searching" is the act of finding items in a database or repository that match specific criteria.

[1300] "Identifying" is the act of selecting the most appropriate item from those found in a search.

[1301] A "natural language processing module" is a program or tool that converts natural language into a format that is easy for a computer to understand.

[1302] The "Summary" is a brief summary of the main content and purpose of the request.

[1303] A "database" is a system that stores large amounts of data in an organized manner and makes it easy to search and manage.

[1304] "Invocation" refers to sending a command or operation to execute a program or service.

[1305] "Satisfaction confirmation" is the act of checking whether the provided services and results meet the user's expectations.

[1306] A "routine" is a series of processing steps in a program that defines certain operations and processing procedures.

[1307] "Automation" refers to making a system operate automatically without human intervention.

[1308] A "web application" is application software that is used over the Internet and operates via a web browser.

[1309] The present invention is a system that allows a user to input a natural language request, and then calls an appropriate generative AI based on the request to provide a service. This system is primarily composed of a server, a terminal, and a generative AI model. A specific implementation method is described below.

[1310] A user inputs a request in natural language using a device (e.g., a personal computer, smartphone, tablet, etc.). This request is sent to the server via a web browser. A specific example is a request such as, "Please create a schedule for the next project meeting and notify the relevant parties."

[1311] The server uses natural language processing modules such as Python's NLTK or spaCy to parse the incoming request, extracting the gist of the request, such as "schedule" and "notify relevant parties."

[1312] Next, the server searches and selects an appropriate model from a database of generative AI models based on the extracted summary. It is possible to use OpenAI's GPT-3 for scheduling and Twilio API for notifications.

[1313] The selected generative AI model is then called by the server to provide a specific service: first, GPT-3 is used to generate a schedule for a project meeting, and then the generated schedule is notified to relevant parties using the Twilio API.

[1314] The generated results are provided to the user by the server, who checks them through their terminal. If the user is satisfied with the results, they send their feedback to the server, which uses this feedback to automate the routine.

[1315] Finally, the automated routine can be deployed as a web application, so that similar requests can be processed automatically in the future. For example, the web application can be built using the Python Flask framework.

[1316] Examples of prompt statements

[1317] Prompt for scheduling a project meeting:

[1318] "Please schedule a project meeting for next Monday. Please include an announcement to stakeholders."

[1319] Prompt for generating notification content:

[1320] "Create a meeting notification based on the following schedule: [generated schedule content]"

[1321] In this way, the present invention efficiently operates generative artificial intelligence based on the user's natural language request, significantly reducing the effort and time required for the work.

[1322] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1323] Step 1:

[1324] User Input

[1325] The user uses a terminal to input a natural language request, for example, "Please create a schedule for the next project meeting and notify the relevant parties," into an input form in a web browser and presses the send button.

[1326] Input: A natural language request typed by a user using a terminal.

[1327] Output: When the submit button is pressed, the request is sent to the server.

[1328] Step 2:

[1329] Receiving and parsing the request

[1330] The server receives the request and then uses a natural language processing module such as Python's NLTK or spaCy to tokenize the request and extract keywords, extracting the gist of the request: "schedule" and "notify."

[1331] Input: A natural language request sent by the user.

[1332] Output: A brief summary of the parsed request (e.g., "Schedule", "Notify").

[1333] Step 3:

[1334] Search and Identify Generative Artificial Intelligence

[1335] Based on the analysis results, the server searches for and selects an appropriate model from a database of generative AI models. For example, for "schedule creation," it identifies high-performance natural language generation software as a generative AI model, and for "notification," it identifies a high-performance communication platform.

[1336] Input: A summary of the parsed request.

[1337] Output: A selected generative artificial intelligence model (e.g., high-performance natural language generation software, high-performance communications platform).

[1338] Step 4:

[1339] Calling generative AI and providing services

[1340] The server invokes selected generative artificial intelligence models to provide specific services, first generating a project meeting schedule using high-performance natural language generation software, and then communicating the generated schedule to stakeholders using a high-performance communication platform.

[1341] Input: Selected generative artificial intelligence model and a summary of the request.

[1342] Output: Generated schedule contents and notification results.

[1343] Step 5:

[1344] Check user satisfaction

[1345] The server provides the generated schedule and notification content to the user, who then checks the results through their terminal. If the user is satisfied with the results, they send their feedback to the server.

[1346] Input: Generated schedule contents and notification results.

[1347] Output: User feedback (confirmation of satisfaction).

[1348] Step 6:

[1349] Routine automation and web application deployment

[1350] The server automates this process based on user feedback by building and deploying a web application using the Python Flask framework, which will allow it to process similar requests automatically in the future.

[1351] Input: User feedback and verified routines.

[1352] Output: A web application as an automated routine.

[1353] (Application example 1)

[1354] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1355] As automation in factories continues to increase, many tasks are now performed by robots and automated equipment. However, efficiently controlling and managing these devices remains a complex process, and it is particularly difficult to achieve fast and efficient control using natural language. Furthermore, because users must clearly instruct each task, training is required and operation can be cumbersome. This can result in reduced factory productivity.

[1356] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1357] In this invention, the server includes means for receiving and analyzing natural language requests, means for searching for and identifying multiple appropriate generative AIs based on the content of the analyzed request, means for calling the identified multiple generative AIs to generate operating procedures for controlling equipment in the factory and providing the procedures to the equipment in the factory, means for confirming user satisfaction and automating event-service routines, and means for deploying the automated routines as web applications to automatically process future requests. This enables efficient control of robots and automated equipment in the factory and improves productivity throughout the factory.

[1358] A "natural language request" refers to a request entered by a user in a language that they normally use.

[1359] "Parsing" is the process of interpreting an input natural language request and understanding its content.

[1360] "Generative AI" refers to an AI model that generates outputs appropriate for a specific task.

[1361] "Search" is the process of finding the model that best suits a request from a generative artificial intelligence database.

[1362] "Identification" refers to determining the generative artificial intelligence that best suits the request from the search results.

[1363] "Invoking" refers to putting a specified generative artificial intelligence into operation and starting processing in response to a request.

[1364] "Operation procedure" refers to the specific procedure for operating equipment in a factory, generated by generative artificial intelligence.

[1365] "Provision" is the process of sending the generated operating procedure to the equipment in the factory and making it executable.

[1366] "Satisfaction" refers to the evaluation and impressions that users have of the results of the service provided.

[1367] A "routine" is a sequence of automatically and repeatedly performed steps or processes.

[1368] "Automation" refers to the state in which a process proceeds without the need for manual intervention.

[1369] "Web application" refers to a software application that is available over the Internet.

[1370] "Factory equipment" refers to robots and automated equipment used in factories in general.

[1371] This invention is a system that allows users to input requests in natural language and efficiently operate and manage automated equipment in a factory based on those requests. Specifically, a server analyzes the user's request, generates operating procedures using appropriate generative artificial intelligence, and provides them to various equipment in the factory.

[1372] Basic system configuration

[1373] 1. Receiving and parsing natural language requests

[1374] Users input requests in natural language through devices such as smartphones or tablets.

[1375] The server uses a Natural Language Processing (NLP) module to parse the request and extract the gist of the request, e.g., "Please assemble the following parts."

[1376] 2. Searching and Identifying Generative AI

[1377] Based on the analysis results, the server searches and identifies appropriate models from a database of generative AI, including "generative AI for scheduling" and "generative AI for notifications."

[1378] 3. Calling the generative AI and generating operational procedures

[1379] The server then calls the identified generative artificial intelligence to generate operational procedures based on the analyzed request, such as assembly procedures or device operation procedures.

[1380] 4. Providing operating procedures

[1381] The generated operation procedures are provided to robot arms and other in-factory equipment so that they can be executed.

[1382] 5. User satisfaction checks and routine automation

[1383] The user reviews the results provided and provides feedback on their satisfaction to the server, which, if satisfied, automates this routine and serves future requests as a web application.

[1384] Hardware and software used

[1385] Hardware

[1386] Smartphones, tablets, and PCs: for inputting users' natural language requests

[1387] Robotic arms and other factory equipment: Equipment that performs the actual movements

[1388] software

[1389] Natural Language Processing (NLP) modules: e.g., NLTK, spaCy, etc.

[1390] Generative AI modules: e.g., GPT-3 or other custom AI models

[1391] Robot control module: e.g., ROS (Robot Operating System)

[1392] Specific examples

[1393] The user inputs a request into their smartphone, such as "Please assemble the following parts." This request is analyzed by the NLP module, and the task "assembly of parts" is extracted. The generative AI generates an assembly procedure suitable for this task and sends it to the robot control module. Finally, the robot arm assembles the specified parts.

[1394] Example prompts to input to the generative AI model

[1395] "Please generate the steps to assemble parts A and B and attach part C."

[1396] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1397] Step 1:

[1398] A user inputs a natural language request into a terminal. The terminal sends the request to a server. The input is a specific natural language request such as "Please assemble the following parts," and the output is that the request is sent to the server. Here, text data entered on the terminal's operating screen is formed.

[1399] Step 2:

[1400] The server analyzes the received natural language request. Specifically, a natural language processing (NLP) module analyzes the request and extracts its gist. In this step, the input natural language sentence is analyzed and the action instructions for "assembly" are extracted. The input is the request data sent from the terminal, and the output is structured data containing the task instruction "assembly."

[1401] Step 3:

[1402] The server searches for and identifies an appropriate generative AI based on the analysis results. The server references a database of pre-registered generative AI models to search for a model suitable for the request. The input to this step is a structured task instruction, and the output is an appropriate generative AI model. Here, a generative AI model that matches the recognized task is searched for.

[1403] Step 4:

[1404] The server calls the identified generative AI to generate the required operating procedure. The called generative AI generates the operating procedure based on the given task instructions. In this step, specific instructions such as "Please assemble the following parts" are input as a prompt, and the output is a specific assembly procedure. The operating procedure is expressed as a series of control commands generated by the generative AI.

[1405] Step 5:

[1406] The server provides the generated operating procedures to the equipment in the factory. Here, a robot control module is used to send the generated instructions to a robot arm or other automated equipment. The input of this step is the generated operating procedure, and the output is the specific control signal received by the equipment. Here, the generated procedure is sent to the actual equipment and execution begins.

[1407] Step 6:

[1408] The user checks the provided results and provides feedback on their satisfaction to the server. The item the user checks is whether the operation was actually performed correctly. The input of this step is the operation result, and the output is the user's satisfaction feedback. The confirmation process is completed when the user provides feedback based on the actual operation result.

[1409] Step 7:

[1410] The server automates the event-service routine based on the user's satisfaction feedback and deploys it as a web application. In this step, past feedback is analyzed and an automated routine is designed. The input of this step is satisfaction feedback data, and the output is a web application containing the automated routine. The automated routine is configured to respond immediately to similar requests from the next time onwards.

[1411] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1412] The present invention is a system that combines natural language request processing with an emotion engine to optimize the user experience. Specific embodiments of the present invention are described below.

[1413] Basic system configuration

[1414] 1. User Input

[1415] The user uses a terminal to input a natural language request, for example, "Please schedule the next project meeting and notify the relevant parties."

[1416] 2. Receiving and parsing the request

[1417] The server receives a natural language request from the user, which is then parsed by a natural language processing (NLP) module to extract the gist of the request, which includes "schedule" and "notify relevant parties."

[1418] 3. Searching and Identifying Generative AI

[1419] Based on the analysis results of the request, the server searches for appropriate models from a database of generative artificial intelligence and identifies a "generative artificial intelligence for schedule creation" and a "generative artificial intelligence for notifications."

[1420] 4. Calling generative AI and providing services

[1421] The server calls the identified generative AI and performs specific processing according to the request. The generative AI for schedule creation generates a schedule for the project meeting and returns the schedule data. Similarly, the generative AI for notification notifies the relevant parties of the generated schedule.

[1422] 5. User Emotion Recognition by Emotion Engine

[1423] The server recognizes the user's emotions through the request and the result of the provided service, for example, the emotion engine analyzes whether the user is satisfied or dissatisfied.

[1424] 6. User Satisfaction Check

[1425] The server checks the user's satisfaction level based on the analysis results of the emotion engine, and if the user is satisfied, collects their feedback and uses it to optimize the automation routines.

[1426] 7. Routine Automation and Web Application Deployment

[1427] The server automates the event-service routines, which are deployed as web applications and automatically process future requests from users.

[1428] Specific examples

[1429] For example, if a user requests, "Please create a schedule for a project meeting next Monday and notify the relevant parties," the following processing is performed.

[1430] 1. Input Reception

[1431] The user inputs a request through the terminal.

[1432] 2. Request Analysis

[1433] The server receives the request, analyzes it, and extracts the information that needs to be "scheduled" and "notified."

[1434] 3. Identifying the right generative AI

[1435] The server searches for and identifies "generative artificial intelligence for schedule creation" and "generative artificial intelligence for notifications."

[1436] 4. Calling generative AI and providing services

[1437] The server calls each generative AI, creates a schedule, and notifies relevant parties.

[1438] 5. Emotion recognition

[1439] After the server provides the service, it uses an emotion engine to analyze the user's emotions and measure their satisfaction.

[1440] 6. Check the results and optimize

[1441] The user checks the provided schedule and notification content, and the server collects feedback based on the analysis results of the emotion engine.

[1442] 7. Deployment of automated routines

[1443] If the user is satisfied, the server automates the process and deploys it as a web application.

[1444] In this way, the present invention is a system that combines generative artificial intelligence and an emotion engine to effectively provide services based on a user's natural language request, thereby optimizing workflow.

[1445] The processing flow will be explained below.

[1446] Step 1:

[1447] A user inputs a natural language request using a terminal, for example, a request such as "Please create a schedule for the next project meeting and notify the relevant parties."

[1448] Step 2:

[1449] The terminal sends this natural language request to the server, which transmits the request over the Internet and receives it at the server's receiving module.

[1450] Step 3:

[1451] The server passes the received natural language request to an NLP (Natural Language Processing) module for analysis. This analysis module understands the content of the request and extracts related tasks and requirements. For example, "schedule creation" and "notify relevant parties" are extracted as tasks.

[1452] Step 4:

[1453] The server searches a database of generative AIs based on the extracted tasks. This search module accesses databases such as the OpenAI Store to find suitable generative AIs, and identifies "generative AIs for scheduling" and "generative AIs for notifications."

[1454] Step 5:

[1455] The server then calls the identified generative AI and performs specific processing in response to the user's request. The generative AI for schedule creation generates a schedule for the project meeting and returns the schedule data. Similarly, the generative AI for notification notifies the relevant parties of the generated schedule.

[1456] Step 6:

[1457] The server compiles the generated schedule and notification results and provides feedback to the user. The user can then check the results of the services provided by the server through their terminal and evaluate their level of satisfaction.

[1458] Step 7:

[1459] The server uses an emotion engine to analyze the user's emotions based on the feedback, for example, through natural language requests and the results of the provided service, the emotion engine recognizes whether the user is satisfied or dissatisfied.

[1460] Step 8:

[1461] Based on the analysis results of the emotion engine, the server optimizes user satisfaction. For example, if the user is dissatisfied, it will select a different generative AI or improve the existing process.

[1462] Step 9:

[1463] If the user is satisfied with the results of the provided service, the server automates the event-service routine. The automation module does this and implements the routine as a web application.

[1464] Step 10:

[1465] The server provides the built web application to the user's device and automatically processes similar requests in the future. The web application receives events in the form of voice, text, sensors, date and time, and provides services in the form of text, images, videos, spreadsheets, code, and documents.

[1466] In this way, through each step centered on the server, terminal, and user, the present invention is a system that provides efficient and flexible services in response to natural language requests from users and further increases user satisfaction using an emotion engine.

[1467] Example 2

[1468] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1469] Currently, many systems only provide simple information or limited services in response to user requests, and do not optimize the user experience or continuously improve processes based on feedback. As a result, the problems and frustrations faced by users are not fully resolved. In particular, processing complex requests, collecting feedback using emotion recognition, and automating processes are difficult challenges.

[1470] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1471] In this invention, the server includes means for receiving and analyzing natural language requests, means for searching for and identifying multiple appropriate generative AI models based on the content of the analyzed request, means for calling the identified generative AI model to execute a service according to the request and providing the results to the target, means for analyzing the target's emotions and confirming satisfaction, and means for automating an event-service routine, deploying the automated routine as a web application, and automatically processing future requests. This enables the provision of appropriate services based on user requests, optimization based on emotional feedback, and efficient processing of future requests.

[1472] A "natural language request" is a request that a user enters into a system in human language.

[1473] "Parsing" is the process of understanding a natural language request based on its meaning and structure.

[1474] A "generative artificial intelligence model" is an artificial intelligence model that has the ability to generate new data and information from given input.

[1475] "Search" is the process of finding models that meet specific criteria from a database of generative artificial intelligence models.

[1476] "Identification" is the process of selecting the most suitable generative artificial intelligence model from the searched results.

[1477] "Executing a service" means performing processing in response to a user request using a specified generative artificial intelligence model.

[1478] "Providing" means delivering the results of the performed service to the user.

[1479] "Analyzing emotions" is the process of analyzing a user's feedback and behavior to infer their emotional state.

[1480] "Confirming satisfaction level" refers to confirming the degree of satisfaction of the user with the provided service based on the results of sentiment analysis.

[1481] An "event-service routine" is a systematized flow of a series of service processes that correspond to a specific event.

[1482] An "automated routine" refers to a series of processes that a system executes automatically without human intervention.

[1483] A "web application" is a software application available over the Internet.

[1484] "Future requests" refers to additional requests that the user may make in the future.

[1485] The present invention is a system that efficiently processes natural language requests from users, provides services, and optimizes the process based on emotional feedback. The system consists of three main components: a server, a terminal, and a user.

[1486] System configuration

[1487] 1. User Input

[1488] A user inputs a natural language request using a device, such as a PC, smartphone, or tablet, using keyboard or voice input. For example, a request might be, "Please schedule the next project meeting and notify the relevant parties."

[1489] 2. Receiving and parsing the request

[1490] The server receives requests from users. This server can be a physical server equipped with an Intel Xeon processor or a cloud service instance (e.g., AWS EC2). The received request is analyzed using a natural language processing (NLP) module such as Google's Cloud Natural Language API or SpaCy, and the gist of the request is extracted. For example, "Create a schedule" and "Notify relevant parties" are extracted as the gist.

[1491] 3. Searching and identifying generative AI models

[1492] Based on the analysis results of the NLP module, the server searches and identifies appropriate generative AI models from the database. Examples of generative AI models that can be used include OpenAI's GPT-4 and similar generative systems. A "generative AI model for schedule creation" and a "generative AI model for notifications" are identified.

[1493] 4. Calling the generative AI model and providing it as a service

[1494] The server then calls the identified generative AI model and performs specific processing according to the user's request. For example, GPT-4 generates a schedule for a project meeting and returns the schedule data. Similarly, the notification AI model notifies relevant parties of the generated schedule.

[1495] 5. User Emotion Recognition by Emotion Engine

[1496] The server recognizes the user's emotions based on the execution results. The emotion engine uses IBM Watson Tone Analyzer, among others. This engine analyzes whether the user is satisfied or dissatisfied. For example, it can estimate emotions by analyzing user feedback and behavioral data.

[1497] 6. User Satisfaction Check

[1498] The server checks the user's satisfaction with the service provided based on the output of the emotion engine, and if so, collects their feedback to help improve the process.

[1499] 7. Routine Automation and Web Application Deployment

[1500] The server then templates the process of successful requests and deploys them as automated routines in frameworks such as Django or Flask as web applications that automatically process similar requests in the future.

[1501] Specific examples

[1502] If a user requests, "Please schedule a project meeting for next Monday and notify the relevant parties," the following processing occurs:

[1503] 1. Enter your natural language request

[1504] The user uses a PC and inputs data using the keyboard.

[1505] 2. Receiving and parsing the request

[1506] The server receives this request and analyzes it using the NLP module (Google Cloud Natural Language API). It extracts the necessary information: "schedule creation" and "notification."

[1507] 3. Searching and identifying generative AI models

[1508] The server identifies GPT-4 from its database as the generative AI model for scheduling and T5 as the generative AI model for notifications.

[1509] 4. Calling the generative AI model and providing it as a service

[1510] The server calls GPT-4 to generate the schedule and calls T5 to notify relevant parties of the schedule.

[1511] 5. Emotion Recognition by Emotion Engine

[1512] The server uses IBM Watson Tone Analyzer to analyze the user's emotions and measure their satisfaction.

[1513] 6. User satisfaction confirmation and optimization

[1514] The user checks the provided schedule and notification content, and the server collects feedback based on the analysis results of the emotion engine.

[1515] 7. Automate routines

[1516] If the user is satisfied, the server will template this process and deploy it as a web application using the Django framework.

[1517] This allows us to provide appropriate services based on users' natural language requests, continuously improve our processes based on feedback, and efficiently process requests.

[1518] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1519] Step 1:

[1520] A user uses a terminal to input a natural language request.

[1521] Specific operation: A user uses a PC or smartphone to input a request using a keyboard or voice input, such as "Please schedule the next project meeting and notify relevant parties."

[1522] Input: A request expressed in natural language.

[1523] Output: Natural language text data.

[1524] Step 2:

[1525] The server receives a natural language request from a user.

[1526] Specific operation: The server receives the text data sent from the terminal and stores it in a database.

[1527] Input: Natural language text data sent by the user from their device.

[1528] Output: Request data stored in a database on the server.

[1529] Step 3:

[1530] The server analyzes the received request.

[1531] What happens: The server uses the Google Cloud Natural Language API to parse the request text and extract the main points: "Create a schedule" and "Notify relevant parties."

[1532] Input: Request data stored in the database.

[1533] Output: A brief summary of the extracted request.

[1534] Step 4:

[1535] The server searches for and identifies an appropriate generative AI model based on the analysis results.

[1536] Specific operation: The server identifies GPT-4 as the scheduling model and T5 as the notification model from the database.

[1537] Input: Summary of analysis results ("Schedule Creation" and "Notify Stakeholders").

[1538] Output: The identified generative AI model.

[1539] Step 5:

[1540] The server calls the identified generative AI model and performs specific processing in response to the user's request.

[1541] How it works: The server uses GPT-4 to generate a project meeting schedule and retrieves the data. Similarly, the notification AI model notifies relevant parties of the generated schedule.

[1542] Input: The identified generative AI model and a brief summary of the request.

[1543] Output: Generated schedule data and notification execution results.

[1544] Step 6:

[1545] The server uses an emotion engine to recognize the user's emotion.

[1546] Specific operation: The server uses IBM Watson Tone Analyzer to analyze the user's feedback and behavioral data after the service is provided and estimate the user's emotional state.

[1547] Input: User feedback and behavioral data after the service is provided.

[1548] Output: Estimated emotional state.

[1549] Step 7:

[1550] The server checks the user's satisfaction level based on the analysis results of the emotion engine.

[1551] What happens: The server measures satisfaction and, if the user is satisfied, collects that feedback and uses it to improve the process.

[1552] Input: Estimated emotional state.

[1553] Output: Satisfaction ratings and feedback data.

[1554] Step 8:

[1555] The server templatizes the process of successful requests and deploys the web application as an automated routine.

[1556] What it does: The server uses Django or Flask to automate the generated schedule and notification process and deploy it as a web application.

[1557] Input: Process and feedback data for successful requests.

[1558] Output: An automated web application.

[1559] This allows for the provision of appropriate services based on the user's natural language requests, and for process optimization and efficient request handling based on emotional feedback.

[1560] (Application example 2)

[1561] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1562] In food delivery services, when users communicate their orders in natural language, it is difficult to analyze their requests in detail and provide optimal suggestions. Furthermore, to increase user satisfaction, it is necessary to recognize the extent to which the service provided affected the user's emotions and improve the quality of the service based on that information. Conventional systems are unable to effectively integrate these elements, limiting their ability to optimize services and improve user satisfaction.

[1563] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1564] In this invention, the server includes means for receiving and analyzing natural language requests, means for searching for and identifying multiple appropriate generative artificial intelligences based on the content of the analyzed request, means for calling the identified multiple generative artificial intelligences to execute a service according to the request and providing the result to the user, means for checking user satisfaction and automating an event-service routine, means for deploying the automated routine as a web application and thereby automatically processing future requests, means for proposing food delivery menus and presenting estimated delivery times based on the user's natural language request, and means for analyzing user emotions from user feedback after providing the service to improve satisfaction. This enables a quick and accurate response to the user's natural language request and, at the same time, improves the quality of the service by recognizing how the provided service affected the user's emotions.

[1565] A "natural language request" is a request or command that a user enters into a system using everyday language.

[1566] "Parsing" is the process of breaking down the content of an incoming natural language request to determine its gist and purpose.

[1567] "Generative AI" refers to AI that has the ability to generate appropriate content and data based on input information.

[1568] "Searching and identifying" means finding a generative artificial intelligence from a database that matches the content of the analyzed request and preparing it for use.

[1569] "Request-based services" refer to specific processing and results provided by generative artificial intelligence based on a user's natural language request.

[1570] "User satisfaction" is an evaluation index that indicates how well a provided service meets the user's expectations and requirements.

[1571] An "event-service routine" is a set of service delivery procedures or processes that respond to a particular event or request.

[1572] An "automated routine" is a series of actions that the system takes to service an event or request automatically, without manual intervention.

[1573] A "web application" is software that runs on the Internet and is accessed by users through a browser.

[1574] "Food delivery menu suggestions" is a service that provides appropriate food and dish options based on user requests.

[1575] "Expected Delivery Time" means calculating and providing an estimate of the time it will take for an item to be delivered after an order is completed.

[1576] "Feedback" refers to the evaluations and impressions that users make about the services provided.

[1577] "Analyzing emotions" is the process of identifying and analyzing emotions (satisfaction, dissatisfaction, joy, etc.) from user feedback.

[1578] The present invention is a food delivery service system that receives natural language requests from users, analyzes them, and uses appropriate generative artificial intelligence to make optimal suggestions and provide delivery services. This system is operated through a smartphone application.

[1579] Basic system configuration

[1580] 1. User Input

[1581] Users input requests in natural language through a smartphone application, for example, "I'd like to order sushi. Please tell me what's recommended today. I'd also like to know how long it will take to deliver."

[1582] 2. Receiving and parsing the request

[1583] The server receives a natural language request from the user. The request is analyzed by a natural language processing (NLP) module (e.g., SpaCy or BERT) to extract the gist of the request. In this example, the gist is "Recommend sushi menu" and "Confirm delivery time."

[1584] 3. Searching and Identifying Generative AI

[1585] Based on the analysis of the request, the server searches and identifies appropriate models from its database of generative artificial intelligence. Specifically, it identifies a "generative artificial intelligence for menu suggestions" (e.g., GPT-4) and a "generative artificial intelligence for delivery time prediction."

[1586] 4. Calling generative AI and providing services

[1587] The server calls the identified generative AI and performs specific processing in response to the request. The generative AI for menu suggestions generates the optimal sushi menu and returns the data. The generative AI for delivery time prediction calculates the predicted delivery time and returns the result. This information is provided to the user via a smartphone application.

[1588] 5. User Emotion Recognition by Emotion Engine

[1589] The server receives the user's feedback after the service is provided and analyzes the user's emotions using an emotion engine (e.g., Google Cloud Natural Language API). The emotion engine analyzes whether the user is satisfied or dissatisfied.

[1590] 6. User Satisfaction Check

[1591] The server checks the user's satisfaction level based on the analysis results of the emotion engine, and if the user is satisfied, collects the feedback and stores it in a database.

[1592] 7. Routine Automation and Web Application Deployment

[1593] The server automates the event-service routines, which are deployed as web applications and perform automated processing when future requests are received from the user.

[1594] Specific examples

[1595] A user types a natural language request into a smartphone application, such as:

[1596] "I'm planning a family party next Sunday. Could you recommend a sushi set? Also, let me know the delivery time if I order now."

[1597] The server analyzes this request and uses appropriate generative artificial intelligence to suggest a sushi set and provide a predicted delivery time. The user then reviews the suggested menu and delivery time and makes their order decision based on that. After providing the service, the emotion engine analyzes the user's feedback and uses it to improve the service in the future.

[1598] This process allows for a fast and accurate response to the user's natural language requests and improves the quality of the service by recognizing how the provided service affected the user's emotions.

[1599] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1600] Step 1:

[1601] A user inputs a request in natural language through a smartphone application. For example, "I'd like to order sushi. Please tell me what's recommended today. I'd also like to know how long it will take to deliver." This request is then sent to the server by the application.

[1602] Input: User's natural language request

[1603] Output: The natural language request sent to the server

[1604] Step 2:

[1605] The server receives a natural language request from the user. The request is analyzed by a natural language processing (NLP) module (e.g., SpaCy or BERT) to extract the gist of the request. Specifically, the gist is extracted as "recommended sushi menu" and "confirm delivery time."

[1606] Input: A natural language request sent to the server

[1607] Output: A summary of the parsed request

[1608] Step 3:

[1609] Based on the analysis results of the request, the server searches and identifies an appropriate model from its database of generative AI. Specifically, it identifies a "generative AI for menu suggestions" (e.g., GPT-4) and a "generative AI for delivery time prediction." At this point, it selects the AI ​​model that best suits the content of the request.

[1610] Input: Parsed request summary

[1611] Output: Identified generative artificial intelligence model

[1612] Step 4:

[1613] The server calls the identified generative AI and performs specific processing in response to the request. The generative AI for menu suggestions generates the optimal sushi menu and returns the data. The generative AI for delivery time prediction calculates the predicted delivery time and returns the result. This information is provided to the user via a smartphone application.

[1614] Input: Identified generative artificial intelligence model, parsed request abstract

[1615] Output: Generated sushi menu, predicted delivery time

[1616] Step 5:

[1617] The server receives the user's feedback after the service is provided. It analyzes the user's emotions using an emotion engine (e.g., Google Cloud Natural Language API). The emotion engine analyzes whether the user is satisfied or dissatisfied.

[1618] Input: User feedback

[1619] Output: Analyzed user sentiment results

[1620] Step 6:

[1621] The server checks the user's satisfaction level based on the analysis results of the emotion engine. If the user is satisfied, the server collects the feedback and stores it in a database. This information can be used to improve the service in the future.

[1622] Input: Analyzed user sentiment results

[1623] Output: Saved feedback

[1624] Step 7:

[1625] The server automates the event-service routine. The automated routine is deployed as a web application and automatically processes future requests from users. This allows users to always receive optimized service.

[1626] Input: Stored feedback, parsed request summary

[1627] Output: Automated routines, deployed web applications

[1628] This enables a fast and accurate response to the user's natural language requests, and improves the quality of the service by recognizing how the provided service affected the user's emotions.

[1629] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1630] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1631] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1632] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1633] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1634] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1635] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1636] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1637] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1638] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1639] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1640] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1641] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1642] 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.

[1643] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1644] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1645] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1646] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1647] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1648] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1649] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1650] The following is further disclosed regarding the above embodiment.

[1651] (Claim 1)

[1652] means for receiving and parsing natural language requests;

[1653] A means for searching for and identifying a plurality of appropriate generative artificial intelligences based on the content of the analyzed request;

[1654] A means for calling the identified plurality of generative artificial intelligences to execute a service according to the request and providing the result to the user;

[1655] A means to check user satisfaction and automate event-service routines;

[1656] A means to deploy the automated routine as a web application, thereby automatically handling future requests;

[1657] A system including:

[1658] (Claim 2)

[1659] 10. The system of claim 1, further comprising means for combining and using appropriate generative artificial intelligences based on the analyzed content of the request.

[1660] (Claim 3)

[1661] 10. The system of claim 1, further comprising means for the web application to receive events in the form of audio, text, sensors, date and time, and provide services in the form of text, images, videos, spreadsheets, code, and documents.

[1662] "Example 1"

[1663] (Claim 1)

[1664] means for receiving and parsing natural language requests;

[1665] A means for searching for and identifying a plurality of appropriate generative artificial intelligences based on the content of the analyzed request;

[1666] means for extracting a gist of the request using a natural language processing module;

[1667] A means for searching and selecting a generative artificial intelligence model from a database based on the extracted summary;

[1668] A means for invoking a selected generative artificial intelligence to provide a specific service in response to a request;

[1669] a means for providing the generated results to the user and confirming their satisfaction;

[1670] A means to automate the post-check routine and deploy it as a web application;

[1671] A method to publish it as a web application to automatically process subsequent requests, and

[1672] A system including:

[1673] (Claim 2)

[1674] 10. The system of claim 1, further comprising means for combining and using appropriate generative artificial intelligences based on the analyzed content of the request.

[1675] (Claim 3)

[1676] 10. The system of claim 1, further comprising means for the web application to receive events in the form of audio, text, sensors, date and time, and provide services in the form of text, images, videos, spreadsheets, code, and documents.

[1677] "Application Example 1"

[1678] (Claim 1)

[1679] means for receiving and parsing natural language requests;

[1680] A means for searching for and identifying a plurality of appropriate generative artificial intelligences based on the content of the analyzed request;

[1681] a means for calling the identified plurality of generative artificial intelligences, generating an operating procedure for controlling the equipment in the factory, and providing the operating procedure to the equipment in the factory;

[1682] A means to check user satisfaction and automate event-service routines;

[1683] A means to deploy the automated routine as a web application, thereby automatically handling future requests;

[1684] A system including:

[1685] (Claim 2)

[1686] The system of claim 1, further comprising means for combining and using appropriate generative artificial intelligences based on the content of the analyzed request to generate an operating procedure for controlling equipment in a factory.

[1687] (Claim 3)

[1688] 10. The system of claim 1, further comprising means for the web application to receive events in the form of voice, text, sensor, date and time, generate equipment control procedures, and provide the operating procedures to equipment in the factory.

[1689] "Example 2: Combining Emotion Engines"

[1690] (Claim 1)

[1691] means for receiving and parsing natural language requests;

[1692] A means for searching for and identifying a plurality of appropriate generative artificial intelligence models based on the analyzed content of the request;

[1693] a means for invoking the identified generative artificial intelligence model to execute the requested service and provide the result to a target;

[1694] A means of analyzing the subject's emotions and confirming their satisfaction;

[1695] Events - A means of automating routines in the service and deploying the automated routines as web applications, thereby automatically handling future requests;

[1696] A system including:

[1697] (Claim 2)

[1698] 10. The system of claim 1, further comprising means for combining and using appropriate generative artificial intelligence models based on the analyzed request content.

[1699] (Claim 3)

[1700] 10. The system of claim 1, further comprising means for the web application to receive events in the form of audio, text, sensors, date and time, and provide services in the form of text, images, videos, spreadsheets, code, and documents.

[1701] "Application example 2 when combining emotion engines"

[1702] (Claim 1)

[1703] means for receiving and parsing natural language requests;

[1704] A means for searching for and identifying a plurality of appropriate generative artificial intelligences based on the content of the analyzed request;

[1705] A means for calling the identified plurality of generative artificial intelligences to execute a service according to the request and providing the result to the user;

[1706] A means to check user satisfaction and automate event-service routines;

[1707] A means to deploy the automated routine as a web application, thereby automatically handling future requests;

[1708] means for suggesting food delivery menus and providing estimated delivery times based on a user's natural language request;

[1709] A method for analyzing emotions from user feedback after providing services and improving satisfaction;

[1710] A system including:

[1711] (Claim 2)

[1712] 10. The system of claim 1, further comprising means for combining and using appropriate generative artificial intelligences based on the analyzed content of the request.

[1713] (Claim 3)

[1714] 10. The system of claim 1, further comprising means for the web application to receive events in the form of audio, text, sensors, date and time, and provide services in the form of text, images, videos, spreadsheets, code, and documents. [Explanation of symbols]

[1715] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for receiving and parsing natural language requests; A means for searching for and identifying a plurality of appropriate generative artificial intelligences based on the content of the analyzed request; A means for calling the identified plurality of generative artificial intelligences to execute a service according to the request and providing the result to the user; A means to check user satisfaction and automate event-service routines; A means to deploy the automated routine as a web application, thereby automatically handling future requests; A system including:

2. The system of claim 1 , further comprising means for combining and using a plurality of appropriate generative artificial intelligences based on the analyzed content of the request.

3. The system of claim 1 , further comprising means for the web application to receive events in the form of audio, text, sensors, date and time, and provide services in the form of text, images, videos, spreadsheets, code, and documents.

Images