system

Abstract

Provide a system. A means for acquiring business materials and communication data; means for analyzing the acquired data and extracting important information; means for generating a virtual worker based on the extracted critical information; means for transmitting communication data to the virtual worker and receiving a response; A system including:

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] In today's work environment, with the increase in maternity leave and work from home, work handover has become a major issue. In particular, when employees on leave or remote workers are absent for an extended period of time, it is necessary to efficiently hand over their work and important information to a new employee. However, current methods require a great deal of time and effort to manually organize information and explain materials, resulting in reduced productivity. Furthermore, there is a risk of work confusion and errors occurring due to insufficient handover. A new system that can effectively solve these issues is needed. [Means for solving the problem]

[0005] To solve the above problems, we provide a system that automatically acquires and analyzes business documents and communication data to extract important information. This system generates a virtual worker (persona) based on the extracted information and automatically responds to questions from users. In addition, the communication data and responses are saved and used as learning data for generating the next response, improving the accuracy and efficiency of the system. This configuration ensures smooth work handover even in cases of long-term leave of absence or teleworking, improving work productivity.

[0006] Specifically, the system includes (1) a means for automatically acquiring business documents and communication data, (2) a means for analyzing the acquired data using natural language processing technology and extracting important business information, (3) a means for generating virtual workers based on the extracted information, (4) a means for the virtual workers to automatically respond to questions from users, and (5) a means for saving response history and using it as learning data. This will automate and streamline the process of business handover, thereby increasing productivity and peace of mind in the workplace.

[0007] "Business materials" includes documents, spreadsheets, presentation files, and other electronic file formats that contain information or data related to business.

[0008] "Communication data" refers to business-related communication history and content, such as emails, chat logs, text messages and media files sent and received via messaging apps.

[0009] The term "means for acquiring" includes a device or program for uploading or transmitting business materials and communication data from users and importing and storing them in the system.

[0010] "Natural language processing" is a technology that allows computers to analyze, understand, and generate human language, and is used to analyze text data, understand its meaning, and extract information.

[0011] "Important information" refers to specific and useful information such as data and instructions essential to business operations, contact information for relevant parties, project progress, deadlines, etc.

[0012] A "virtual worker (persona)" is a computer-generated agent that acts as a substitute for a real worker and answers user questions based on specific job descriptions and information.

[0013] A "predefined answer pattern" is a predefined set of appropriate responses to specific questions, including rules and templates that allow virtual workers to automatically generate answers.

[0014] The "means for transmitting communication data and obtaining a response" includes functions and programs for having a virtual worker input a question from a user, generating a response to that question, and displaying the response to the user.

[0015] "Means for saving response history and using it as learning data" includes technologies and devices that record past interactions with users and analyze them to improve the accuracy of the system and the quality of responses. [Brief explanation of the drawings]

[0016] [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

[0017] 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.

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

[0019] 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).

[0020] 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.

[0021] 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.

[0022] 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.

[0023] 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."

[0024] [First embodiment]

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

[0026] 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.

[0027] 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).

[0028] 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.

[0029] 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.

[0030] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form 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.

[0031] 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.

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

[0033] 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.

[0034] 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.

[0035] 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.

[0036] 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."

[0037] The present invention relates to a system for efficiently and effectively communicating information during a task handover. This system acquires and analyzes business documents and communication data to generate a virtual worker (persona) that can respond appropriately to questions from users.

[0038] The whole system program consists of three main components: server, terminal and user.

[0039] Data collection phase:

[0040] 1. The user specifies business materials (documents, spreadsheets, presentation files, etc.) and communication data (emails, chat logs, etc.) on the system's upload screen.

[0041] 2. The device temporarily stores these files and obtains the file metadata (file name, file size, upload date and time, etc.).

[0042] 3. The server receives the temporarily stored files and metadata from the device and stores them in a database.

[0043] Data analysis phase:

[0044] 1. The server sequentially retrieves the saved files and sends them to a natural language processing (NLP) engine for analysis.

[0045] 2. The NLP engine analyzes the contents of each file and extracts important business information (project name, person in charge, deadline, etc.).

[0046] 3. The server receives the analyzed data obtained from the NLP engine and stores it in a database as structured data.

[0047] Persona generation phase:

[0048] 1. The server uses structured data from the database to generate virtual workers (personas) specialized for specific tasks.

[0049] 2. The server sets answer patterns and response styles for the generated persona based on predefined rules.

[0050] User interaction phase:

[0051] 1. A user types a question into the system in chat format: "Where are the materials for next week's meeting?"

[0052] 2. The device sends the user's question to the server.

[0053] 3. The server receives the user's question and uses the persona data to generate an appropriate answer, such as "The meeting materials are in the meeting folder on the shared drive."

[0054] 4. The terminal displays the answer returned by the server to the user.

[0055] Learning phase:

[0056] 1. The server stores the user's interaction history with the persona in a database.

[0057] 2. The stored interaction history is periodically analyzed and used as learning data for the next response generation, allowing the system to continuously improve and provide more accurate responses.

[0058] As a concrete example, before User A goes on a one-month maternity leave, she uploads her work documents and email archives to the system. The server analyzes these and generates a persona for User A. When User B, who is in charge of taking over during her leave, enters a question into the system, the server provides a quick and accurate answer based on User A's persona. The next time User B asks a new question, a more accurate answer is generated by referring to the previous interaction history.

[0059] In this way, the present invention can improve the efficiency of business handover and business productivity.

[0060] The processing flow will be explained below.

[0061] Step 1:

[0062] Users upload their business materials (e.g., documents, spreadsheets, presentation files) and communication data (e.g., emails, chat logs) to the Persona Chat system.

[0063] Step 2:

[0064] The device temporarily stores the file uploaded by the user, and acquires the file's metadata (file name, file size, upload date and time, etc.).

[0065] Step 3:

[0066] The server receives the files and metadata acquired from the device, stores them in a database, and records the storage location (file path) of the saved files.

[0067] Step 4:

[0068] The server sequentially retrieves files stored in the database and sends them to a natural language processing (NLP) engine for analysis, which includes text extraction, semantic analysis, and keyword extraction.

[0069] Step 5:

[0070] The NLP engine analyzes the file contents and extracts business-critical information (e.g., project name, responsible person, deadline, etc.).

[0071] Step 6:

[0072] The server receives the parsed data returned by the NLP engine and stores it in a database as structured data, which contains the extracted key information.

[0073] Step 7:

[0074] The server generates virtual workers (personas) specialized for specific tasks based on structured data in the database, with response patterns and styles set for these persona models.

[0075] Step 8:

[0076] A user enters a question (for example, "Where are the materials for next week's meeting?") into the system in chat format.

[0077] Step 9:

[0078] The terminal processes the user's questions in real time and sends them to the server.

[0079] Step 10:

[0080] The server receives questions from users and generates appropriate answers based on persona data, querying relevant information from the database and formulating specific responses.

[0081] Step 11:

[0082] The terminal displays the answer returned from the server to the user. For example, it may present an answer such as "The meeting materials are in the meeting folder on the shared drive."

[0083] Step 12:

[0084] The server stores the interaction history between the user and the persona in a database, which can then be used as learning data when generating a response next time.

[0085] These steps allow the system to efficiently take over and respond quickly and accurately to user questions.

[0086] Example 1

[0087] 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."

[0088] This system solves the problem of difficulty in efficiently and effectively communicating information during task handover. Conventional systems often have difficulty efficiently using business documents and communication data to generate appropriate responses, and they often lack continuous learning capabilities to improve the quality and accuracy of responses.

[0089] 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.

[0090] In this invention, the server includes means for acquiring business materials and communication data, means for temporarily storing the acquired data and acquiring metadata, means for storing the temporarily stored data and metadata in a database, means for transmitting the stored data to a natural language processing engine for analysis and extracting important information, means for generating a virtual worker based on the extracted important information, means for transmitting communication data to the virtual worker and generating a response, means for displaying the generated response to a user, and means for saving an interaction history of the response and communication data and using it as learning data. This makes it possible to efficiently and effectively convey information during business handover and continuously improve the quality and accuracy of responses.

[0091] "Business Materials" refers to information including documents, spreadsheets, presentation files, etc. related to business.

[0092] "Communication data" refers to data including communication history such as emails and chat logs.

[0093] "Server" refers to a computer system that acquires, analyzes, stores, and generates responses to business materials and communication data.

[0094] "Terminal" refers to a computer device used by a user to upload business materials and communication data to the system.

[0095] "User" refers to a human operator authorized to operate the system.

[0096] "Database" refers to data storage for storing business documents, communication data, analysis results, virtual worker profile information, etc.

[0097] A "natural language processing engine" refers to a software module for analyzing text data and extracting important information.

[0098] "Virtual Worker" refers to a simulated worker profile generated based on acquired and analyzed data.

[0099] "Metadata" refers to attribute information that is not the data itself, such as file name, file size, and upload date and time.

[0100] "Interaction history" refers to log data of questions and responses made between a user and a virtual worker.

[0101] "Training data" refers to data used to improve the system's response generation based on the analysis of interaction history.

[0102] "Response style" refers to the format or manner in which the virtual worker responds to the user.

[0103] "Predefined answer patterns" refer to predetermined response templates for specific questions.

[0104] This invention relates to a system for efficiently and effectively communicating information during task handover. This system acquires and analyzes business documents and communication data to generate virtual workers (personas) that respond appropriately to user questions. The entire system consists of three main components: a server, a terminal, and a user.

[0105] Data Collection Phase

[0106] The user specifies work-related documents, spreadsheets, presentation files, emails, chat logs, etc. on the system's upload screen. The device temporarily stores these specified files and acquires metadata such as file name, file size, and upload date and time. The server receives the files and metadata sent from the device and stores them in a database.

[0107] Data analysis phase

[0108] The server sequentially retrieves files stored in the database and sends them to a natural language processing (NLP) engine for analysis. The NLP engine uses, for example, the Python library spaCy or the BERT model. The NLP engine analyzes the contents of the files and extracts important business information such as project names, responsible persons, and deadlines. The server stores the analysis results obtained from the NLP engine in the database as structured data.

[0109] Persona generation phase

[0110] The server uses the structured data stored in the database to generate virtual workers (personas) specialized for specific work tasks. Specifically, it runs a Python script based on information retrieved from the database to generate a persona profile (e.g., name, responsibilities, experience level, etc.). The server then sets answer patterns and response styles for the generated personas based on predefined rules.

[0111] User Interaction Phase

[0112] A user enters a question in chat format into the system, such as "Where are the materials for next week's meeting?" The device sends this question to the server. The server receives the question and uses the persona data to generate an appropriate answer. For example, it generates an answer such as "The meeting materials are in the meeting folder on the shared drive." The device displays the answer returned from the server to the user.

[0113] Learning Phase

[0114] The server stores the interaction history of users and personas in a database. The server periodically analyzes the stored interaction history and uses it as learning data for generating the next response. This allows the system to continuously improve and enable more accurate responses.

[0115] Specific examples

[0116] Before User A goes on a one-month maternity leave, she uploads her work documents and email archives to the system. The server analyzes this data and generates a persona specific to User A's work. While User A is on leave, User B, who is in charge of taking over, enters the question "Where is this month's sales report?" into the system. The server provides a quick and accurate answer based on User A's persona. The next time User B asks a new question, the server references the previous interaction history to generate a more accurate answer.

[0117] Prompt Sentence Examples

[0118] "How can I parse work handover documents and generate virtual worker personas?"

[0119] "When a user types a question into the system, what are the steps to generate an appropriate answer to that question?"

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

[0121] Step 1:

[0122] Users specify work-related documents, spreadsheets, presentation files, emails, chat logs, etc. on the system's upload screen.

[0123] Input: Business documents and correspondence data

[0124] Specific operation: The user accesses the system's upload screen using a web browser and clicks the file selection button to select the file to upload.

[0125] Output: Paths and metadata of selected files

[0126] Step 2:

[0127] The terminal temporarily stores the file designated by the user and acquires metadata such as the file name, file size, and upload date and time.

[0128] Input: User specified file

[0129] Specific operation: Uploaded files are temporarily saved in a specific directory on the device, and file metadata is retrieved using Python's os module, etc.

[0130] Output: Path and metadata of the temporarily saved file (file name, size, upload date, etc.)

[0131] Step 3:

[0132] The server receives the files and metadata sent from the terminal and stores them in a database.

[0133] Input: Path and metadata of the temporarily saved file

[0134] Specific operation: The server receives the files and metadata sent via HTTP request, stores the files in object storage, and stores the metadata in a relational database.

[0135] Output: Files and metadata stored in a database

[0136] Step 4:

[0137] The server sequentially retrieves files stored in the database and sends them to a natural language processing (NLP) engine for analysis.

[0138] Input: Path of the file stored in the database

[0139] Specific operation: Obtain the path to the saved file and send a POST request to the NLP engine using the Python requests module.

[0140] Output: File data passed to the NLP engine

[0141] Step 5:

[0142] The NLP engine analyzes the contents of the file and extracts important business information (project name, person in charge, deadline, etc.).

[0143] Input: File sent from the server

[0144] How it works: The NLP engine uses models such as spaCy and BERT to tokenize text and extract entities.

[0145] Output: Extracted important information (project name, person in charge, deadline, etc.)

[0146] Step 6:

[0147] The server stores the analysis results obtained from the NLP engine in a database as structured data.

[0148] Input: Analysis results provided by the NLP engine

[0149] Specific operation: The analysis results are received in JSON format, converted into a data frame using Python's pandas library, and stored in a relational database.

[0150] Output: Structured data stored in a database

[0151] Step 7:

[0152] The server uses structured data stored in a database to generate virtual workers (personas) specialized for specific tasks.

[0153] Input: Structured data stored in a database

[0154] Specific operation: Runs a Python script based on information retrieved from the database to generate a persona profile (e.g., name, responsibilities, experience, etc.).

[0155] Output: Generated persona profile

[0156] Step 8:

[0157] The server sets answer patterns and response styles for the generated personas based on predefined rules.

[0158] Input: Generated persona profile

[0159] What it does: Prepare and apply response templates based on persona profiles for response styling.

[0160] Output: Personas with response styles and answer patterns

[0161] Step 9:

[0162] A user types a question into the system in chat format: "Where are the materials for next week's meeting?"

[0163] Input: User question

[0164] Specific behavior: A user types a question into a chat window on a web application and clicks the send button.

[0165] Output: User question data

[0166] Step 10:

[0167] The terminal sends this question to the server.

[0168] Input: User question data

[0169] What happens: Chat input is sent to the server via WebSocket or HTTP request.

[0170] Output: Question data sent to the server

[0171] Step 11:

[0172] The server receives the question and uses the persona data to generate an appropriate answer, such as "The meeting materials are in the meeting folder on the shared drive."

[0173] Input: User question data, persona data

[0174] Specific operation: The server sends the question data to a natural language processing model, analyzes the intent of the question, searches for relevant information from a database, and generates an answer based on a template.

[0175] Output: Generated response data

[0176] Step 12:

[0177] The terminal displays the answer returned by the server to the user.

[0178] Input: Response data from the server

[0179] Specific behavior: Receives a WebSocket or HTTP response and displays the reply in the chat window.

[0180] Output: The answer displayed to the user

[0181] Step 13:

[0182] The server stores the user and persona interaction history in a database.

[0183] Input: User's question data, Server's answer data

[0184] Specific behavior: Save user questions and persona answers as logs in a database.

[0185] Output: Interaction history stored in a database

[0186] Step 14:

[0187] The server periodically analyzes the stored interaction history and uses it as learning data for generating responses the next time, allowing the system to continuously improve and provide more accurate responses.

[0188] Input: Interaction history stored in a database

[0189] What it does: Set up a batch processing job to parse the stored logs to generate new training data and retrain the model.

[0190] Output: A new trained response generation model

[0191] (Application example 1)

[0192] 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."

[0193] In conventional business handovers, when a person takes temporary leave or resigns, the successor must quickly and accurately understand the work. However, manually reviewing the vast amount of business documents and communication data to extract the necessary information is extremely inefficient and carries a high risk of information leaks and misunderstandings. Work in factories is particularly highly specialized, and a quick and accurate handover has a decisive impact on productivity and quality, making it an urgent task to resolve this issue.

[0194] 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.

[0195] In this invention, the server includes means for acquiring business documents and communication data, means for analyzing the acquired data and extracting important information, means for generating virtual workers based on the extracted important information, means for operators to upload business documents and communication data within the factory, means for analyzing the saved data with a natural language processing engine, means for responding to questions from a new operator based on the virtual workers, and means for saving an interaction history with the operator and using it to generate responses next time. This enables efficient and accurate handover of work within the factory, improving productivity and quality of work.

[0196] "Business materials" refers to all documents and files related to work within a factory, and specifically includes documents such as work procedures, inspection reports, and manuals.

[0197] "Communication data" refers to digital data used to exchange business-related information, and specifically includes records such as e-mails and chat logs.

[0198] A "natural language processing engine" refers to technology that analyzes natural language used by humans and extracts information, and is used to extract important information from text.

[0199] "Virtual workers" refer to virtual assistants created based on the data of real workers, who provide the information needed for task handover.

[0200] "Interaction history" refers to a record of questions and responses between the user and the system (virtual worker), which is used for future learning and response improvement.

[0201] "Server" refers to the primary hardware or software platform for running and managing the various means described above, including storing data, analyzing data, and generating virtual workers.

[0202] A "database" refers to a system for systematically storing collected and analyzed business documents and communication data, and holds information such as structured data and interaction history.

[0203] An "operator" refers to a person who uses the system to take over work, uploading business documents and communication data, entering new questions, and so on.

[0204] This invention relates to a system for efficiently and effectively carrying out task handover within a factory. This system acquires and analyzes work documents and communication data, and generates a virtual worker (persona) that can respond appropriately to questions from a new operator.

[0205] System configuration

[0206] The system is broadly composed of the following components:

[0207] Server: Stores data, analyzes it, and generates virtual workers.

[0208] Terminal: Provides an interface for operators to upload business documents and communication data and input questions into the system.

[0209] User: Uses the system as an operator to upload business documents and ask questions.

[0210] Program processing explanation

[0211] The program of this system executes the following processes in sequence.

[0212] 1. Data collection phase:

[0213] The user specifies business documents (e.g., work procedures, inspection reports, manuals, etc.) and communication data (e.g., emails, chat logs) on the upload screen of the terminal.

[0214] The device temporarily stores these files and obtains the file metadata (e.g., file name, file size, upload date and time, etc.).

[0215] The server receives the temporarily stored files and metadata from the terminal and stores them in a database.

[0216] 2. Data analysis phase:

[0217] The server sequentially retrieves the stored files and sends them to a natural language processing engine (e.g., an NLP engine) for analysis.

[0218] The NLP engine analyzes the contents of each file and extracts important business information (e.g., task content, responsible person, deadline, etc.).

[0219] The server receives the analyzed data obtained from the NLP engine and stores it in a database as structured data.

[0220] 3. Persona generation phase:

[0221] The server uses structured data from the database to generate virtual workers (personas) specialized for specific tasks.

[0222] The server sets answer patterns and response styles for the generated personas based on predefined rules.

[0223] 4. User interaction phase:

[0224] The user types the question into the system in chat format: "Where are last night's test results saved?"

[0225] The terminal sends the user's question to the server.

[0226] The server receives the user's question and uses the persona data to generate an appropriate answer (e.g., a specific answer such as "Your test results can be found in the 'September 2023 Test Results' folder on the shared drive").

[0227] The terminal displays the answer returned by the server to the user.

[0228] 5. Learning Phase:

[0229] The server stores the interaction history between the user and the persona in a database.

[0230] The stored interaction history is periodically analyzed and used as learning data for the next response generation, allowing the system to continuously improve and provide more accurate responses.

[0231] Examples of specific examples and prompts

[0232] For example, if an operator asks the system, "Where are last night's test results stored?", the system will search the database for the appropriate information and respond, "The test results are located in the 'September 2023 Test Results' folder on the shared drive."

[0233] An example of a prompt sentence is as follows:

[0234] "Prompt for a robotic assistant helping with factory handover:

[0235] User: 'I'd like to know the results of last night's test.'

[0236] Robot: 'Test results are stored in the folder 'September 2023 Test Results' on the shared drive.'"

[0237] In this way, the handover of work within the factory can be carried out efficiently and accurately.

[0238] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0239] Step 1:

[0240] The user specifies business documents (e.g., work procedures, inspection reports, manuals, etc.) or communication data (e.g., emails, chat logs) on the upload screen of the terminal and uploads the specified files. The terminal temporarily stores the uploaded files and collects the file metadata (e.g., file name, file size, upload date and time, etc.). The input is the file specified by the user, and the output is the temporarily stored file and metadata.

[0241] Step 2:

[0242] The terminal sends the temporarily stored files and metadata to the server. The server receives this data and stores it in a database. The input is the files and metadata sent from the terminal, and the output is the data stored in the database.

[0243] Step 3:

[0244] The server sequentially retrieves the saved files and sends them to a natural language processing engine (e.g., NLP engine) for analysis. The input is the saved files, and the output is the data analyzed by the NLP engine.

[0245] Step 4:

[0246] The NLP engine analyzes the contents of each file and extracts important business information (e.g., task details, responsible parties, deadlines, etc.). The input is the text data of the file, and the output is the extracted business information.

[0247] Step 5:

[0248] The server receives the analyzed data obtained from the NLP engine and stores it in a database as structured data. The input is the analyzed data, and the output is business information stored as structured data.

[0249] Step 6:

[0250] The server uses the structured data stored in the database to generate virtual workers (personas) specialized for specific tasks. The input is the structured data, and the output is the generated virtual worker.

[0251] Step 7:

[0252] The server sets answer patterns and response styles for the generated virtual workers based on predefined rules. The input is the virtual worker's generation data and predefined rules, and the output is the virtual worker with the set response patterns and styles.

[0253] Step 8:

[0254] A user types a question into the system in chat format: "Where are last night's test results saved?" The input is the user's question, and the output is the data sent from the terminal to the server.

[0255] Step 9:

[0256] The terminal sends the user's question to the server. The input is the user's question, and the output is the data to be sent to the server.

[0257] Step 10:

[0258] The server receives the user's question and generates an appropriate answer using the virtual worker's data (e.g., a specific answer such as "The inspection results can be found in the 'September 2023 Inspection Results' folder on the shared drive"). The input is the user's question, and the output is the generated answer.

[0259] Step 11:

[0260] The terminal displays the answer returned from the server to the user. The input is the answer from the server, and the output is the display data to the user.

[0261] Step 12:

[0262] The server stores the interaction history between the user and the virtual worker in a database. The input is the interaction data, and the output is the data stored in the database.

[0263] Step 13:

[0264] The server periodically analyzes the stored interaction history and uses it as learning data for generating the next response. This allows the system to continuously improve and provide more accurate responses. The input is the interaction history, and the output is the updated learning model.

[0265] 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.

[0266] This invention relates to a system for efficiently and effectively communicating information during task handover. This system not only acquires and analyzes business documents and communication data to generate virtual workers (personas), but also incorporates an emotion engine that recognizes the user's emotions, further improving the accuracy and appropriateness of responses.

[0267] The whole system program consists of three main components: server, terminal and user.

[0268] Data collection phase:

[0269] 1. A user uploads their business materials (documents, spreadsheets, presentation files, etc.) and communication data (emails, chat logs, etc.) to the Persona Chat system.

[0270] 2. The device temporarily stores these files, and at this time, obtains the file metadata (file name, file size, upload date and time, etc.).

[0271] 3. The server receives the files and metadata from the device, stores them in a database, and records the storage location (file path) of the saved files.

[0272] Data analysis phase:

[0273] 1. The server sequentially retrieves files stored in the database and sends them to a natural language processing (NLP) engine for analysis, which includes text extraction, semantic analysis, and keyword extraction.

[0274] 2. The NLP engine analyzes the file contents and extracts business-critical information (e.g., project name, responsible person, deadline, etc.).

[0275] 3. The server receives the analyzed data obtained from the NLP engine and stores it in a database as structured data.

[0276] Persona generation phase:

[0277] 1. The server uses structured data in the database to generate virtual workers (personas) specialized for specific tasks. These persona models are assigned response patterns and styles.

[0278] Emotion Recognition Phase:

[0279] 1. The terminal transmits communication data between the user and the virtual worker to the emotion engine, which analyzes the user's emotions (e.g., joy, anger, anxiety, etc.).

[0280] 2. The emotion engine extracts emotions from the user's communication data and feeds those emotions back to the server.

[0281] User interaction phase:

[0282] 1. A user types a question into the system in chat format (e.g., "Where are the materials for next week's meeting?").

[0283] 2. The terminal processes the user's question in real time and sends it to the server.

[0284] 3. The server receives the user's question and the emotion engine's analysis results, and generates an appropriate answer based on the persona data. For example, if the user is feeling anxious, the persona will respond in a gentle tone, saying, "Don't worry, the meeting materials are in the meeting folder on the shared drive."

[0285] 4. The terminal displays the response returned from the server to the user.

[0286] Learning phase:

[0287] 1. The server stores the user's interaction history with the persona and the analyzed emotion data in a database.

[0288] 2. The system periodically analyzes the stored interaction history and emotion data and uses this data as learning data when generating the next response, allowing the system to continuously improve and provide more accurate responses.

[0289] As a concrete example, before User A goes on a one-month maternity leave, she uploads her work documents and email archives to the system. The server analyzes these and generates a persona for User A. During her leave, User B, who is in charge of taking over, enters questions into the system, and the server provides a quick and appropriate response based on User A's persona and the emotion engine. The system also recognizes User B's emotion in response to her question and adjusts the tone and content of the response accordingly. The next time User B asks a new question, the system will refer to the previous interaction history and emotion data to generate a more accurate answer.

[0290] In this way, the present invention can improve the efficiency of task handover and provide responses that take into consideration the user's feelings, thereby improving task productivity and user satisfaction.

[0291] The processing flow will be explained below.

[0292] Step 1:

[0293] Users upload their business materials (e.g., documents, spreadsheets, presentation files) and communication data (e.g., emails, chat logs) to the Persona Chat system.

[0294] Step 2:

[0295] The device temporarily stores the uploaded file and obtains the file's metadata (file name, file size, upload date and time, etc.).

[0296] Step 3:

[0297] The server receives the files and metadata acquired from the device, stores them in a database, and records the storage location (file path) of the saved files.

[0298] Step 4:

[0299] The server sequentially retrieves files stored in the database and sends them to a natural language processing (NLP) engine for analysis, which includes text extraction, semantic analysis, and keyword extraction.

[0300] Step 5:

[0301] The NLP engine analyzes the file contents and extracts business-critical information (e.g., project name, responsible person, deadline, etc.).

[0302] Step 6:

[0303] The server receives the analyzed data obtained from the NLP engine and stores it in a database as structured data.

[0304] Step 7:

[0305] The server generates a virtual worker (persona) specialized for a specific job based on structured data from the database, and this persona model is assigned answer patterns and response styles.

[0306] Step 8:

[0307] The terminal transmits communication data between the user and the virtual worker to the emotion engine, which analyzes the user's emotions (e.g., joy, anger, anxiety, etc.).

[0308] Step 9:

[0309] The emotion engine extracts emotions from the user's communication data and feeds the emotions back to the server.

[0310] Step 10:

[0311] A user enters a question (for example, "Where are the materials for next week's meeting?") into the system in chat format.

[0312] Step 11:

[0313] The terminal processes the user's questions in real time and sends them to the server.

[0314] Step 12:

[0315] The server receives the user's question and the analysis results of the emotion engine, and generates an appropriate answer based on the persona data. For example, if the user is feeling anxious, the persona will respond in a gentle tone, saying, "Don't worry, the meeting materials are in the meeting folder on the shared drive."

[0316] Step 13:

[0317] The terminal displays the answer returned from the server to the user.

[0318] Step 14:

[0319] The server stores the user's interaction history with the persona and the analyzed emotion data in a database.

[0320] Step 15:

[0321] The server periodically analyzes the stored interaction history and emotion data and uses this data as learning data when generating the next response, allowing the system to continuously improve and provide more accurate responses.

[0322] Example 2

[0323] 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."

[0324] In conventional business handovers, the information being handed over is often insufficient, and the details of the handed over business are often not fully understood, resulting in problems with the smooth execution of business. Also, there are cases where information is inconsistent or missing during the handover, making it difficult to improve business efficiency and ensure an effective handover. Furthermore, there is a lack of systems that respond appropriately to user emotions, which is a factor that reduces user satisfaction.

[0325] 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. In this invention, the server includes means for acquiring business materials and communication data, means for analyzing the acquired data and extracting important information, means for generating a virtual worker based on the extracted important information, means for analyzing the user's emotions, and means for adjusting the virtual worker's response based on the emotions. This enables efficient and effective handover of work and provides appropriate responses that take the user's emotions into consideration, thereby improving work productivity and user satisfaction.

[0326] "Business materials" refers to any form of digital data related to business, such as documents, spreadsheets, presentation files, emails, chat logs, etc.

[0327] "Communications Data" refers to data that records communications between users, such as emails, chat logs, and messages from messaging applications.

[0328] "Means of acquisition" refers to a method of executing a process in which business materials and communication data are uploaded by users to the system, and the data is temporarily stored and sent to a server.

[0329] "Means of analysis" refers to a method of analyzing the acquired data based on natural language processing technology and performing text extraction, semantic analysis, and keyword extraction.

[0330] "Important information" refers to information that is considered particularly important in business (for example, project name, person in charge, deadline, etc.).

[0331] A "virtual worker" refers to a virtual person constructed based on a user's work documents and communication data, with response patterns and styles specialized for specific work content.

[0332] "Means for transmitting communication data" refers to a process for transmitting questions or requests from users to virtual workers and receiving their responses.

[0333] "Means for obtaining a response" refers to a method by which the system receives a response sent by a virtual worker and provides it to the user.

[0334] "Means for analyzing emotions" refers to technology for extracting emotions (e.g., joy, anger, anxiety, etc.) from a user's communication data.

[0335] "Means for adjusting the virtual worker's response based on emotion" refers to a method for appropriately adjusting the tone and content of the virtual worker's response by referring to the user's emotional data.

[0336] MODE FOR CARRYING OUT THE INVENTION

[0337] This invention relates to a system for efficiently and effectively communicating information during task handover. This system not only acquires and analyzes business documents and communication data to generate virtual workers (personas), but also incorporates an emotion engine that recognizes the user's emotions, further improving the accuracy and appropriateness of responses.

[0338] The whole system program consists of three main components: server, terminal and user.

[0339] Hardware and software configuration:

[0340] Server: A server with large storage capacity and high-performance computing power is required. A database management system (e.g., MySQL®, PostgreSQL) and a natural language processing (NLP) engine (e.g., Google® NLP, IBM Watson® NLP) are installed and used.

[0341] Device: A user-accessible computer or smartphone that has a web browser or dedicated client application installed.

[0342] User: Uploads and queries business data.

[0343] Data Collection Phase

[0344] A user uploads business documents or communication data to the system. For example, the user uploads files by dragging and dropping them using the interface on a web browser.

[0345] The device temporarily stores these files and obtains metadata such as file name, size, and upload date and time.

[0346] The server receives the files and metadata acquired from the device and stores them in a dedicated database.

[0347] Data analysis phase

[0348] The server sequentially retrieves files stored in the database and sends them to the NLP engine for analysis, which includes text extraction, semantic analysis, and keyword extraction.

[0349] The NLP engine analyzes the file contents and extracts business-critical information (e.g., project name, responsible person, deadline, etc.).

[0350] The server stores the analyzed data as structured data in a database.

[0351] Persona generation phase

[0352] The server uses the structured data in the database to generate virtual workers (personas) tailored to specific job tasks, with response patterns and styles set for each persona model.

[0353] Emotion Recognition Phase

[0354] The terminal transmits communication data between the user and the virtual worker to the emotion engine, which analyzes the user's emotions (e.g., joy, anger, anxiety, etc.).

[0355] The emotion engine extracts emotions from the user's communication data and feeds the emotions back to the server.

[0356] User Interaction Phase

[0357] A user enters a question (for example, "Where are the materials for next week's meeting?") into the system in chat format.

[0358] The terminal processes the user's questions in real time and sends them to the server.

[0359] The server receives the user's question and the analysis results of the emotion engine, and generates an appropriate answer based on the persona data. For example, if the user is feeling anxious, the persona will respond in a gentle tone, saying, "Don't worry, the meeting materials are in the meeting folder on the shared drive."

[0360] The terminal displays the answer content returned from the server to the user.

[0361] Learning Phase

[0362] The server stores the user's interaction history with the persona and the analyzed emotion data in a database.

[0363] The system periodically analyzes the interaction history and emotional data it has received and uses this data as learning data when generating the next response, allowing the system to continuously improve and generate more accurate responses.

[0364] Specific examples

[0365] Before User A goes on a one-month maternity leave, she uploads her work documents and email archives to the system. The system analyzes them and generates a persona for User A. During her leave, User B, who is in charge of handover, enters a question into the system: "Where are the documents for next week's meeting?" The server provides an appropriate response based on User A's persona and emotion engine. The system also recognizes User B's emotion toward the question and adjusts the tone and content of the response accordingly. The next time User B asks a new question, it refers to the previous interaction history and emotion data to generate a more accurate answer.

[0366] Example prompt sentence:

[0367] "Before User A goes on a one-month maternity leave, she uploads her work documents and email archives to the system. Generate a virtual worker (persona) for User A based on these documents. Then, imagine that User B, who is in charge of the handover, enters the question, "Where are the documents for next week's meeting?" into the system. Use the emotion engine to generate an appropriate response."

[0368] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0369] Step 1:

[0370] Users upload their business materials (documents, spreadsheets, presentation files, etc.) and communication data (emails, chat logs, etc.) to the system. The input business materials and communication data are imported into the system, and the system is now ready for data collection.

[0371] How it works: A user uploads a file using a web browser interface.

[0372] Step 2:

[0373] The device temporarily stores the uploaded file in local storage. The input is the user's business documents and communication data, and the output is the temporarily stored file and metadata (file name, size, upload date and time, etc.). This prepares the server to receive the file.

[0374] Behavior: Saves the file to a temporary file directory and collects metadata.

[0375] Step 3:

[0376] The server receives the files and metadata acquired from the terminal and stores them in a dedicated database. The input is the files and metadata sent from the terminal, and the output is the files and metadata stored in the database. This identifies the storage location of the files and enables database management.

[0377] How it works: The server receives the file via an API request, stores it in the file system, and records the metadata in a database.

[0378] Step 4:

[0379] The server sequentially retrieves files stored in the database and sends them to a natural language processing (NLP) engine for analysis. The input is the file retrieved from the database, and the output is the analyzed text data. This yields the text data of the business documents.

[0380] What it does: Sends the file via an API request to an NLP engine that performs text extraction, semantic analysis, and keyword extraction.

[0381] Step 5:

[0382] The NLP engine analyzes the file contents and extracts business-critical information. The input is the file sent from the server, and the output is structured analyzed data, which allows for further detailed data analysis.

[0383] How it works: Parses the text, extracts keywords such as project name, person in charge, deadline date, etc., and returns the parsed results in JSON format to the server.

[0384] Step 6:

[0385] The server receives the analyzed data obtained from the NLP engine and stores it in the database as structured data. The input is the analyzed data from the NLP engine, and the output is the structured data stored in the database. This updates the database and adds new data.

[0386] What it does: Receives parsed data in JSON format and inserts it into a database.

[0387] Step 7:

[0388] The server uses the structured data in the database to generate virtual workers (personas) specialized for specific tasks. The input is the structured data in the database, and the output is the generated persona model. This creates a persona that responds to the user's question.

[0389] How it works: Analyzes structured data and generates persona models including response patterns and response styles, which are stored in a model database.

[0390] Step 8:

[0391] The terminal sends communication data between the user and the virtual worker to the emotion engine, which analyzes the user's emotions. The input is the communication data, and the output is the emotion analysis results. This allows the user's emotional state to be understood.

[0392] How it works: Chat logs are captured in real time and sent to the emotion engine via API requests.

[0393] Step 9:

[0394] The emotion engine extracts emotions from the user's communication data and feeds them back to the server. The input is the user's communication data, and the output is emotion data. This makes it possible to respond based on the user's emotions.

[0395] Operation: Analyzes emotions from communication data and returns the emotion data in JSON format to the server.

[0396] Step 10:

[0397] A user inputs a question in chat format on the system. The input is the user's question, and the output is the question data sent to the terminal. This allows the system to recognize the question.

[0398] Action: Type and send text through the chat interface.

[0399] Step 11:

[0400] The terminal processes the user's question in real time and sends it to the server. The input is the user's question data, and the output is the question data sent to the server, allowing the server to analyze the question content.

[0401] What it does: Sends a text message to the server using WebSocket or HTTP request.

[0402] Step 12:

[0403] The server receives the user's question and the analysis results of the emotion engine, and generates an appropriate answer based on the persona data. The input is the user's question data and emotion data, and the output is the generated answer. This allows for an answer that takes emotion into consideration.

[0404] How it works: The question content and sentiment data are combined to generate an answer based on the persona's response patterns, which is then returned in JSON format.

[0405] Step 13:

[0406] The terminal displays the answer content returned from the server to the user. The input is the generated answer data, and the output is the answer text displayed to the user, allowing the user to confirm the answer.

[0407] Behavior: Displays the reply text in the chat interface.

[0408] Step 14:

[0409] The server stores the user's interaction history and the analyzed emotional data in a database. The input is the interaction history and emotional data, and the output is the history and data stored in the database. This enables continuous system learning.

[0410] Operation: Interaction logs and emotion data are stored in a database and updated as training data.

[0411] Step 15:

[0412] The system periodically analyzes the stored interaction history and emotion data and uses them as learning data for the next response generation. The input is the interaction history and emotion data, and the output is an updated response generation model. This allows the system to continuously improve.

[0413] How it works: Use periodic batch processing or online learning algorithms to update the training dataset and retrain the model.

[0414] (Application example 2)

[0415] 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."

[0416] When handing over work tasks, information is often not communicated efficiently, which can lead to reduced work efficiency, especially in places where there is a lot of information and a lot of movement. Furthermore, there is a lack of responses that take user feelings into consideration, which can have a negative impact on work productivity and user satisfaction. There is a need to solve these issues and make work smoother and more efficient.

[0417] Means to solve the problem

[0418] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for acquiring business materials and communication data, means for analyzing the acquired data and extracting important information, means for generating virtual workers based on the extracted important information, and means installed on smartphones for analyzing user emotions and providing feedback. This enables efficient information transmission and responses that take emotions into consideration in logistics centers.

[0419] Definitions of important terms included in the claims

[0420] "Business materials" refers to documents and data files required for the operation and business execution of a company or organization.

[0421] "Communications Data" refers to information exchanged electronically, such as emails, chat logs, and messages.

[0422] "Means" refer to the methods, tools, and techniques used to achieve a particular function or purpose.

[0423] "Analysis" refers to the process of examining data or information in detail to understand its structure and meaning.

[0424] "Important information" refers to information that is essential for carrying out business and that requires particular attention.

[0425] A "virtual worker" refers to a computer-generated model of a worker that is designed to perform a specific task.

[0426] A "smartphone" refers to a portable information terminal that combines the functions of a mobile phone and a personal computer.

[0427] "User emotion" refers to the emotional state of the user, including emotions such as joy, anger, and anxiety.

[0428] "Feedback" refers to providing a reaction or evaluation to an action or response.

[0429] A "system" refers to a collection of multiple components or means combined to achieve a specific purpose.

[0430] MODE FOR CARRYING OUT THE INVENTION

[0431] The present invention relates to a system for efficiently transferring work and transmitting information in a logistics center. Specific embodiments for carrying out the present invention will be described below.

[0432] System Overview

[0433] This system consists of three main components: a server, a terminal (smartphone), and a user. The server acquires business documents and communication data, analyzes them, and extracts important information. It also generates virtual workers (personas) and analyzes user emotions. The terminal acts as an interface with the user and is responsible for collecting and displaying data.

[0434] Hardware and software used

[0435] Server: Amazon Web Services (AWS (registered trademark)) is used to store and analyze data. In particular, AWS S3 is used to store files, and AWS RDS is used to store analysis results in a database.

[0436] NLP engine: Uses AWS Comprehend to analyze business documents and communication data.

[0437] Emotion engine: Analyzes user emotions using Microsoft® Azure® Cognitive Services.

[0438] Terminal: The application on the smartphone was developed using React Native and provides the data upload and user interface.

[0439] System program and processing flow

[0440] Data collection

[0441] Users use their smartphones to upload business documents (e.g., PDF, Excel, Word) and communication data (e.g., emails, chat logs) to the application. The device stores these files in AWS S3 and temporarily collects metadata. The server receives the files and metadata from the device and stores them in a database.

[0442] Data analysis

[0443] The server sequentially retrieves files stored in the database and sends them to the NLP engine for analysis. Specifically, it performs text extraction, semantic analysis, and keyword extraction. The NLP engine analyzes the file contents and extracts important business information (e.g., project name, person in charge, deadline, etc.). The extracted data is returned to the server and stored as structured data in the database.

[0444] Persona Generation

[0445] The server uses the structured data in the database to generate virtual workers (personas) tailored to specific tasks, with answer patterns and response styles set for each persona model.

[0446] emotion recognition

[0447] The terminal sends communication data between the user and the virtual worker to the emotion engine, which analyzes the user's emotions (e.g., joy, anger, anxiety, etc.). The emotion engine extracts emotions from the user's communication data and feeds the results back to the server.

[0448] User Interaction

[0449] The user enters a chat-style question into the system (e.g., "Please tell me the current stock level."). The device processes the user's question in real time and sends it to the server. The server receives the user's question and the analysis results of the emotion engine, and generates an appropriate answer based on the persona data. For example, if the user is feeling anxious, the persona will respond in a gentle tone, saying, "The current stock level is 500 units, so please don't worry." The device then displays the answer returned from the server to the user.

[0450] Learning Phase

[0451] The server stores the user's interaction history and analyzed emotional data in a database. The stored interaction history and emotional data are used as learning data when generating the next response. This allows the system to continuously improve, enabling more accurate responses.

[0452] Examples of specific examples and prompts

[0453] When an operator asks "What's the current inventory level?" on a smartphone app, the following process takes place:

[0454] 1. The operator enters a question, and the terminal sends it to the server.

[0455] 2. The server analyzes the question and the user's emotions (anxiety), and the persona generates an answer: "There are currently 500 items in stock, so don't worry."

[0456] 3. The device will display the answer.

[0457] Example prompt sentence:

[0458] Here is an example of a conversation to check inventory at a distribution center:

[0459] User: What is your current inventory?

[0460] Persona: We currently have 500 in stock, so don't worry.

[0461] User Emotion: Anxiety

[0462] Persona Response: Respond in a helpful and kind tone.

[0463] By implementing the present invention in the above manner, it is possible to improve the operational efficiency of a logistics center and the satisfaction of users.

[0464] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0465] Step 1:

[0466] The user uploads business documents and communication data to a smartphone app. The input is business documents such as PDF, Excel, and Word documents, and communication data such as emails and chat logs. The device saves these files to AWS S3 and temporarily stores file metadata (file name, file size, upload date and time, etc.). The output is the temporarily saved metadata and the file storage location.

[0467] Step 2:

[0468] The server receives the metadata and file storage location sent from the device. The server records this data in a database and retrieves the file from AWS S3. The input is the metadata and the AWS S3 file path, and the output is the file and metadata stored in the database.

[0469] Step 3:

[0470] The server sequentially retrieves files stored in the database and sends them to an NLP engine (AWS Comprehend) for analysis. The input is the file data, and the NLP engine performs text extraction, semantic analysis, and keyword extraction. The output is important information (e.g., project name, person in charge, deadline, etc.) as the analysis result.

[0471] Step 4:

[0472] The server receives the analysis results from the NLP engine and stores them as structured data in a database. The input is important information (project name, person in charge, deadline, etc.), and the output is structured data stored in the database.

[0473] Step 5:

[0474] The server uses the structured data in the database to generate a virtual worker (persona) specialized for a specific job. The input is the structured data, and the output is a model of the virtual worker with a set answer pattern and response style.

[0475] Step 6:

[0476] The user inputs a question through a smartphone app. For example, "Please tell me the current inventory levels." The input is the user's question text, which the device sends to the server in real time. The output is the user's question data.

[0477] Step 7:

[0478] The server receives the user's question and redirects it to a virtual worker (persona). At the same time, it sends the question text to an emotion engine (Microsoft Azure Cognitive Services) to analyze the user's emotions. The input is the user's question text, the persona model, and the emotion engine's analysis data, and the output is the analysis results.

[0479] Step 8:

[0480] The emotion engine analyzes the user's emotions and feeds the results back to the server. The input is the user's question text, and the output is the analyzed emotion data (joy, anger, anxiety, etc.).

[0481] Step 9:

[0482] The server uses the user's question and emotional data to generate an appropriate response from the persona. For example, if the user is feeling anxious, the persona will respond in a gentle tone, saying, "There are currently 500 items in stock. Don't worry." The input is the user's question, emotional data, and the persona's response style, and the output is the final answer text.

[0483] Step 10:

[0484] The terminal displays the answer content returned from the server to the user. The input is the final answer text, and the output is the answer text displayed on the smartphone screen.

[0485] Through the above processing steps, the handover of work and the transmission of information at the logistics center are carried out efficiently, and responses that take into consideration the feelings of the user are provided.

[0486] 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.

[0487] 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.

[0488] 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.

[0489] [Second embodiment]

[0490] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0491] 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.

[0492] 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).

[0493] 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.

[0494] 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.

[0495] 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).

[0496] 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.

[0497] 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.

[0498] 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.

[0499] 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.

[0500] In the smart glasses 214, 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.

[0501] 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."

[0502] The present invention relates to a system for efficiently and effectively communicating information during a task handover. This system acquires and analyzes business documents and communication data to generate a virtual worker (persona) that can respond appropriately to questions from users.

[0503] The whole system program consists of three main components: server, terminal and user.

[0504] Data collection phase:

[0505] 1. The user specifies business materials (documents, spreadsheets, presentation files, etc.) and communication data (emails, chat logs, etc.) on the system's upload screen.

[0506] 2. The device temporarily stores these files and obtains the file metadata (file name, file size, upload date and time, etc.).

[0507] 3. The server receives the temporarily stored files and metadata from the device and stores them in a database.

[0508] Data analysis phase:

[0509] 1. The server sequentially retrieves the saved files and sends them to a natural language processing (NLP) engine for analysis.

[0510] 2. The NLP engine analyzes the contents of each file and extracts important business information (project name, person in charge, deadline, etc.).

[0511] 3. The server receives the analyzed data obtained from the NLP engine and stores it in a database as structured data.

[0512] Persona generation phase:

[0513] 1. The server uses structured data from the database to generate virtual workers (personas) specialized for specific tasks.

[0514] 2. The server sets answer patterns and response styles for the generated persona based on predefined rules.

[0515] User interaction phase:

[0516] 1. A user types a question into the system in chat format: "Where are the materials for next week's meeting?"

[0517] 2. The device sends the user's question to the server.

[0518] 3. The server receives the user's question and uses the persona data to generate an appropriate answer, such as "The meeting materials are in the meeting folder on the shared drive."

[0519] 4. The terminal displays the answer returned by the server to the user.

[0520] Learning phase:

[0521] 1. The server stores the user's interaction history with the persona in a database.

[0522] 2. The stored interaction history is periodically analyzed and used as learning data for the next response generation, allowing the system to continuously improve and provide more accurate responses.

[0523] As a concrete example, before User A goes on a one-month maternity leave, she uploads her work documents and email archives to the system. The server analyzes these and generates a persona for User A. When User B, who is in charge of taking over during her leave, enters a question into the system, the server provides a quick and accurate answer based on User A's persona. The next time User B asks a new question, a more accurate answer is generated by referring to the previous interaction history.

[0524] In this way, the present invention can improve the efficiency of business handover and business productivity.

[0525] The processing flow will be explained below.

[0526] Step 1:

[0527] Users upload their business materials (e.g., documents, spreadsheets, presentation files) and communication data (e.g., emails, chat logs) to the Persona Chat system.

[0528] Step 2:

[0529] The device temporarily stores the file uploaded by the user, and acquires the file's metadata (file name, file size, upload date and time, etc.).

[0530] Step 3:

[0531] The server receives the files and metadata acquired from the device, stores them in a database, and records the storage location (file path) of the saved files.

[0532] Step 4:

[0533] The server sequentially retrieves files stored in the database and sends them to a natural language processing (NLP) engine for analysis, which includes text extraction, semantic analysis, and keyword extraction.

[0534] Step 5:

[0535] The NLP engine analyzes the file contents and extracts business-critical information (e.g., project name, responsible person, deadline, etc.).

[0536] Step 6:

[0537] The server receives the parsed data returned by the NLP engine and stores it in a database as structured data, which contains the extracted key information.

[0538] Step 7:

[0539] The server generates virtual workers (personas) specialized for specific tasks based on structured data in the database, with response patterns and styles set for these persona models.

[0540] Step 8:

[0541] A user enters a question (for example, "Where are the materials for next week's meeting?") into the system in chat format.

[0542] Step 9:

[0543] The terminal processes the user's questions in real time and sends them to the server.

[0544] Step 10:

[0545] The server receives questions from users and generates appropriate answers based on persona data, querying relevant information from the database and formulating specific responses.

[0546] Step 11:

[0547] The terminal displays the answer returned from the server to the user. For example, it may present an answer such as "The meeting materials are in the meeting folder on the shared drive."

[0548] Step 12:

[0549] The server stores the interaction history between the user and the persona in a database, which can then be used as learning data when generating a response next time.

[0550] These steps allow the system to efficiently take over and respond quickly and accurately to user questions.

[0551] Example 1

[0552] 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."

[0553] This system solves the problem of difficulty in efficiently and effectively communicating information during task handover. Conventional systems often have difficulty efficiently using business documents and communication data to generate appropriate responses, and they often lack continuous learning capabilities to improve the quality and accuracy of responses.

[0554] 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.

[0555] In this invention, the server includes means for acquiring business materials and communication data, means for temporarily storing the acquired data and acquiring metadata, means for storing the temporarily stored data and metadata in a database, means for transmitting the stored data to a natural language processing engine for analysis and extracting important information, means for generating a virtual worker based on the extracted important information, means for transmitting communication data to the virtual worker and generating a response, means for displaying the generated response to a user, and means for saving an interaction history of the response and communication data and using it as learning data. This makes it possible to efficiently and effectively convey information during business handover and continuously improve the quality and accuracy of responses.

[0556] "Business Materials" refers to information including documents, spreadsheets, presentation files, etc. related to business.

[0557] "Communication data" refers to data including communication history such as emails and chat logs.

[0558] "Server" refers to a computer system that acquires, analyzes, stores, and generates responses to business materials and communication data.

[0559] "Terminal" refers to a computer device used by a user to upload business materials and communication data to the system.

[0560] "User" refers to a human operator authorized to operate the system.

[0561] "Database" refers to data storage for storing business documents, communication data, analysis results, virtual worker profile information, etc.

[0562] A "natural language processing engine" refers to a software module for analyzing text data and extracting important information.

[0563] "Virtual Worker" refers to a simulated worker profile generated based on acquired and analyzed data.

[0564] "Metadata" refers to attribute information that is not the data itself, such as file name, file size, and upload date and time.

[0565] "Interaction history" refers to log data of questions and responses made between a user and a virtual worker.

[0566] "Training data" refers to data used to improve the system's response generation based on the analysis of interaction history.

[0567] "Response style" refers to the format or manner in which the virtual worker responds to the user.

[0568] "Predefined answer patterns" refer to predetermined response templates for specific questions.

[0569] This invention relates to a system for efficiently and effectively communicating information during task handover. This system acquires and analyzes business documents and communication data to generate virtual workers (personas) that respond appropriately to user questions. The entire system consists of three main components: a server, a terminal, and a user.

[0570] Data Collection Phase

[0571] The user specifies work-related documents, spreadsheets, presentation files, emails, chat logs, etc. on the system's upload screen. The device temporarily stores these specified files and acquires metadata such as file name, file size, and upload date and time. The server receives the files and metadata sent from the device and stores them in a database.

[0572] Data analysis phase

[0573] The server sequentially retrieves files stored in the database and sends them to a natural language processing (NLP) engine for analysis. The NLP engine uses, for example, the Python library spaCy or the BERT model. The NLP engine analyzes the contents of the files and extracts important business information such as project names, responsible persons, and deadlines. The server stores the analysis results obtained from the NLP engine in the database as structured data.

[0574] Persona generation phase

[0575] The server uses the structured data stored in the database to generate virtual workers (personas) specialized for specific work tasks. Specifically, it runs a Python script based on information retrieved from the database to generate a persona profile (e.g., name, responsibilities, experience level, etc.). The server then sets answer patterns and response styles for the generated personas based on predefined rules.

[0576] User Interaction Phase

[0577] A user enters a question in chat format into the system, such as "Where are the materials for next week's meeting?" The device sends this question to the server. The server receives the question and uses the persona data to generate an appropriate answer. For example, it generates an answer such as "The meeting materials are in the meeting folder on the shared drive." The device displays the answer returned from the server to the user.

[0578] Learning Phase

[0579] The server stores the interaction history of users and personas in a database. The server periodically analyzes the stored interaction history and uses it as learning data for generating the next response. This allows the system to continuously improve and enable more accurate responses.

[0580] Specific examples

[0581] Before User A goes on a one-month maternity leave, she uploads her work documents and email archives to the system. The server analyzes this data and generates a persona specific to User A's work. While User A is on leave, User B, who is in charge of taking over, enters the question "Where is this month's sales report?" into the system. The server provides a quick and accurate answer based on User A's persona. The next time User B asks a new question, the server references the previous interaction history to generate a more accurate answer.

[0582] Prompt Sentence Examples

[0583] "How can I parse work handover documents and generate virtual worker personas?"

[0584] "When a user types a question into the system, what are the steps to generate an appropriate answer to that question?"

[0585] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0586] Step 1:

[0587] Users specify work-related documents, spreadsheets, presentation files, emails, chat logs, etc. on the system's upload screen.

[0588] Input: Business documents and correspondence data

[0589] Specific operation: The user accesses the system's upload screen using a web browser and clicks the file selection button to select the file to upload.

[0590] Output: Paths and metadata of selected files

[0591] Step 2:

[0592] The terminal temporarily stores the file designated by the user and acquires metadata such as the file name, file size, and upload date and time.

[0593] Input: User specified file

[0594] Specific operation: Uploaded files are temporarily saved in a specific directory on the device, and file metadata is retrieved using Python's os module, etc.

[0595] Output: Path and metadata of the temporarily saved file (file name, size, upload date, etc.)

[0596] Step 3:

[0597] The server receives the files and metadata sent from the terminal and stores them in a database.

[0598] Input: Path and metadata of the temporarily saved file

[0599] Specific operation: The server receives the files and metadata sent via HTTP request, stores the files in object storage, and stores the metadata in a relational database.

[0600] Output: Files and metadata stored in a database

[0601] Step 4:

[0602] The server sequentially retrieves files stored in the database and sends them to a natural language processing (NLP) engine for analysis.

[0603] Input: Path of the file stored in the database

[0604] Specific operation: Obtain the path to the saved file and send a POST request to the NLP engine using the Python requests module.

[0605] Output: File data passed to the NLP engine

[0606] Step 5:

[0607] The NLP engine analyzes the contents of the file and extracts important business information (project name, person in charge, deadline, etc.).

[0608] Input: File sent from the server

[0609] How it works: The NLP engine uses models such as spaCy and BERT to tokenize text and extract entities.

[0610] Output: Extracted important information (project name, person in charge, deadline, etc.)

[0611] Step 6:

[0612] The server stores the analysis results obtained from the NLP engine in a database as structured data.

[0613] Input: Analysis results provided by the NLP engine

[0614] Specific operation: The analysis results are received in JSON format, converted into a data frame using Python's pandas library, and stored in a relational database.

[0615] Output: Structured data stored in a database

[0616] Step 7:

[0617] The server uses structured data stored in a database to generate virtual workers (personas) specialized for specific tasks.

[0618] Input: Structured data stored in a database

[0619] Specific operation: Runs a Python script based on information retrieved from the database to generate a persona profile (e.g., name, responsibilities, experience, etc.).

[0620] Output: Generated persona profile

[0621] Step 8:

[0622] The server sets answer patterns and response styles for the generated personas based on predefined rules.

[0623] Input: Generated persona profile

[0624] What it does: Prepare and apply response templates based on persona profiles for response styling.

[0625] Output: Personas with response styles and answer patterns

[0626] Step 9:

[0627] A user types a question into the system in chat format: "Where are the materials for next week's meeting?"

[0628] Input: User question

[0629] Specific behavior: A user types a question into a chat window on a web application and clicks the send button.

[0630] Output: User question data

[0631] Step 10:

[0632] The terminal sends this question to the server.

[0633] Input: User question data

[0634] What happens: Chat input is sent to the server via WebSocket or HTTP request.

[0635] Output: Question data sent to the server

[0636] Step 11:

[0637] The server receives the question and uses the persona data to generate an appropriate answer, such as "The meeting materials are in the meeting folder on the shared drive."

[0638] Input: User question data, persona data

[0639] Specific operation: The server sends the question data to a natural language processing model, analyzes the intent of the question, searches for relevant information from a database, and generates an answer based on a template.

[0640] Output: Generated response data

[0641] Step 12:

[0642] The terminal displays the answer returned by the server to the user.

[0643] Input: Response data from the server

[0644] Specific behavior: Receives a WebSocket or HTTP response and displays the reply in the chat window.

[0645] Output: The answer displayed to the user

[0646] Step 13:

[0647] The server stores the user and persona interaction history in a database.

[0648] Input: User's question data, Server's answer data

[0649] Specific behavior: Save user questions and persona answers as logs in a database.

[0650] Output: Interaction history stored in a database

[0651] Step 14:

[0652] The server periodically analyzes the stored interaction history and uses it as learning data for generating responses the next time, allowing the system to continuously improve and provide more accurate responses.

[0653] Input: Interaction history stored in a database

[0654] What it does: Set up a batch processing job to parse the stored logs to generate new training data and retrain the model.

[0655] Output: A new trained response generation model

[0656] (Application example 1)

[0657] 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."

[0658] In conventional business handovers, when a person takes temporary leave or resigns, the successor must quickly and accurately understand the work. However, manually reviewing the vast amount of business documents and communication data to extract the necessary information is extremely inefficient and carries a high risk of information leaks and misunderstandings. Work in factories is particularly highly specialized, and a quick and accurate handover has a decisive impact on productivity and quality, making it an urgent task to resolve this issue.

[0659] 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.

[0660] In this invention, the server includes means for acquiring business documents and communication data, means for analyzing the acquired data and extracting important information, means for generating virtual workers based on the extracted important information, means for operators to upload business documents and communication data within the factory, means for analyzing the saved data with a natural language processing engine, means for responding to questions from a new operator based on the virtual workers, and means for saving an interaction history with the operator and using it to generate responses next time. This enables efficient and accurate handover of work within the factory, improving productivity and quality of work.

[0661] "Business materials" refers to all documents and files related to work within a factory, and specifically includes documents such as work procedures, inspection reports, and manuals.

[0662] "Communication data" refers to digital data used to exchange business-related information, and specifically includes records such as e-mails and chat logs.

[0663] A "natural language processing engine" refers to technology that analyzes natural language used by humans and extracts information, and is used to extract important information from text.

[0664] "Virtual workers" refer to virtual assistants created based on the data of real workers, who provide the information needed for task handover.

[0665] "Interaction history" refers to a record of questions and responses between the user and the system (virtual worker), which is used for future learning and response improvement.

[0666] "Server" refers to the primary hardware or software platform for running and managing the various means described above, including storing data, analyzing data, and generating virtual workers.

[0667] A "database" refers to a system for systematically storing collected and analyzed business documents and communication data, and holds information such as structured data and interaction history.

[0668] An "operator" refers to a person who uses the system to take over work, uploading business documents and communication data, entering new questions, and so on.

[0669] This invention relates to a system for efficiently and effectively carrying out task handover within a factory. This system acquires and analyzes work documents and communication data, and generates a virtual worker (persona) that can respond appropriately to questions from a new operator.

[0670] System configuration

[0671] The system is broadly composed of the following components:

[0672] Server: Stores data, analyzes it, and generates virtual workers.

[0673] Terminal: Provides an interface for operators to upload business documents and communication data and input questions into the system.

[0674] User: Uses the system as an operator to upload business documents and ask questions.

[0675] Program processing explanation

[0676] The program of this system executes the following processes in sequence.

[0677] 1. Data collection phase:

[0678] The user specifies business documents (e.g., work procedures, inspection reports, manuals, etc.) and communication data (e.g., emails, chat logs) on the upload screen of the terminal.

[0679] The device temporarily stores these files and obtains the file metadata (e.g., file name, file size, upload date and time, etc.).

[0680] The server receives the temporarily stored files and metadata from the terminal and stores them in a database.

[0681] 2. Data analysis phase:

[0682] The server sequentially retrieves the stored files and sends them to a natural language processing engine (e.g., an NLP engine) for analysis.

[0683] The NLP engine analyzes the contents of each file and extracts important business information (e.g., task content, responsible person, deadline, etc.).

[0684] The server receives the analyzed data obtained from the NLP engine and stores it in a database as structured data.

[0685] 3. Persona generation phase:

[0686] The server uses structured data from the database to generate virtual workers (personas) specialized for specific tasks.

[0687] The server sets answer patterns and response styles for the generated personas based on predefined rules.

[0688] 4. User interaction phase:

[0689] The user types the question into the system in chat format: "Where are last night's test results saved?"

[0690] The terminal sends the user's question to the server.

[0691] The server receives the user's question and uses the persona data to generate an appropriate answer (e.g., a specific answer such as "Your test results can be found in the 'September 2023 Test Results' folder on the shared drive").

[0692] The terminal displays the answer returned by the server to the user.

[0693] 5. Learning Phase:

[0694] The server stores the interaction history between the user and the persona in a database.

[0695] The stored interaction history is periodically analyzed and used as learning data for the next response generation, allowing the system to continuously improve and provide more accurate responses.

[0696] Examples of specific examples and prompts

[0697] For example, if an operator asks the system, "Where are last night's test results stored?", the system will search the database for the appropriate information and respond, "The test results are located in the 'September 2023 Test Results' folder on the shared drive."

[0698] An example of a prompt sentence is as follows:

[0699] "Prompt for a robotic assistant helping with factory handover:

[0700] User: 'I'd like to know the results of last night's test.'

[0701] Robot: 'Test results are stored in the folder 'September 2023 Test Results' on the shared drive.'"

[0702] In this way, the handover of work within the factory can be carried out efficiently and accurately.

[0703] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0704] Step 1:

[0705] The user specifies business documents (e.g., work procedures, inspection reports, manuals, etc.) or communication data (e.g., emails, chat logs) on the upload screen of the terminal and uploads the specified files. The terminal temporarily stores the uploaded files and collects the file metadata (e.g., file name, file size, upload date and time, etc.). The input is the file specified by the user, and the output is the temporarily stored file and metadata.

[0706] Step 2:

[0707] The terminal sends the temporarily stored files and metadata to the server. The server receives this data and stores it in a database. The input is the files and metadata sent from the terminal, and the output is the data stored in the database.

[0708] Step 3:

[0709] The server sequentially retrieves the saved files and sends them to a natural language processing engine (e.g., NLP engine) for analysis. The input is the saved files, and the output is the data analyzed by the NLP engine.

[0710] Step 4:

[0711] The NLP engine analyzes the contents of each file and extracts important business information (e.g., task details, responsible parties, deadlines, etc.). The input is the text data of the file, and the output is the extracted business information.

[0712] Step 5:

[0713] The server receives the analyzed data obtained from the NLP engine and stores it in a database as structured data. The input is the analyzed data, and the output is business information stored as structured data.

[0714] Step 6:

[0715] The server uses the structured data stored in the database to generate virtual workers (personas) specialized for specific tasks. The input is the structured data, and the output is the generated virtual worker.

[0716] Step 7:

[0717] The server sets answer patterns and response styles for the generated virtual workers based on predefined rules. The input is the virtual worker's generation data and predefined rules, and the output is the virtual worker with the set response patterns and styles.

[0718] Step 8:

[0719] A user types a question into the system in chat format: "Where are last night's test results saved?" The input is the user's question, and the output is the data sent from the terminal to the server.

[0720] Step 9:

[0721] The terminal sends the user's question to the server. The input is the user's question, and the output is the data to be sent to the server.

[0722] Step 10:

[0723] The server receives the user's question and generates an appropriate answer using the virtual worker's data (e.g., a specific answer such as "The inspection results can be found in the 'September 2023 Inspection Results' folder on the shared drive"). The input is the user's question, and the output is the generated answer.

[0724] Step 11:

[0725] The terminal displays the answer returned from the server to the user. The input is the answer from the server, and the output is the display data to the user.

[0726] Step 12:

[0727] The server stores the interaction history between the user and the virtual worker in a database. The input is the interaction data, and the output is the data stored in the database.

[0728] Step 13:

[0729] The server periodically analyzes the stored interaction history and uses it as learning data for generating the next response. This allows the system to continuously improve and provide more accurate responses. The input is the interaction history, and the output is the updated learning model.

[0730] 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.

[0731] This invention relates to a system for efficiently and effectively communicating information during task handover. This system not only acquires and analyzes business documents and communication data to generate virtual workers (personas), but also incorporates an emotion engine that recognizes the user's emotions, further improving the accuracy and appropriateness of responses.

[0732] The whole system program consists of three main components: server, terminal and user.

[0733] Data collection phase:

[0734] 1. A user uploads their business materials (documents, spreadsheets, presentation files, etc.) and communication data (emails, chat logs, etc.) to the Persona Chat system.

[0735] 2. The device temporarily stores these files, and at this time, obtains the file metadata (file name, file size, upload date and time, etc.).

[0736] 3. The server receives the files and metadata from the device, stores them in a database, and records the storage location (file path) of the saved files.

[0737] Data analysis phase:

[0738] 1. The server sequentially retrieves files stored in the database and sends them to a natural language processing (NLP) engine for analysis, which includes text extraction, semantic analysis, and keyword extraction.

[0739] 2. The NLP engine analyzes the file contents and extracts business-critical information (e.g., project name, responsible person, deadline, etc.).

[0740] 3. The server receives the analyzed data obtained from the NLP engine and stores it in a database as structured data.

[0741] Persona generation phase:

[0742] 1. The server uses structured data in the database to generate virtual workers (personas) specialized for specific tasks. These persona models are assigned response patterns and styles.

[0743] Emotion Recognition Phase:

[0744] 1. The terminal transmits communication data between the user and the virtual worker to the emotion engine, which analyzes the user's emotions (e.g., joy, anger, anxiety, etc.).

[0745] 2. The emotion engine extracts emotions from the user's communication data and feeds those emotions back to the server.

[0746] User interaction phase:

[0747] 1. A user types a question into the system in chat format (e.g., "Where are the materials for next week's meeting?").

[0748] 2. The terminal processes the user's question in real time and sends it to the server.

[0749] 3. The server receives the user's question and the emotion engine's analysis results, and generates an appropriate answer based on the persona data. For example, if the user is feeling anxious, the persona will respond in a gentle tone, saying, "Don't worry, the meeting materials are in the meeting folder on the shared drive."

[0750] 4. The terminal displays the response returned from the server to the user.

[0751] Learning phase:

[0752] 1. The server stores the user's interaction history with the persona and the analyzed emotion data in a database.

[0753] 2. The system periodically analyzes the stored interaction history and emotion data and uses this data as learning data when generating the next response, allowing the system to continuously improve and provide more accurate responses.

[0754] As a concrete example, before User A goes on a one-month maternity leave, she uploads her work documents and email archives to the system. The server analyzes these and generates a persona for User A. During her leave, User B, who is in charge of taking over, enters questions into the system, and the server provides a quick and appropriate response based on User A's persona and the emotion engine. The system also recognizes User B's emotion in response to her question and adjusts the tone and content of the response accordingly. The next time User B asks a new question, the system will refer to the previous interaction history and emotion data to generate a more accurate answer.

[0755] In this way, the present invention can improve the efficiency of task handover and provide responses that take into consideration the user's feelings, thereby improving task productivity and user satisfaction.

[0756] The processing flow will be explained below.

[0757] Step 1:

[0758] Users upload their business materials (e.g., documents, spreadsheets, presentation files) and communication data (e.g., emails, chat logs) to the Persona Chat system.

[0759] Step 2:

[0760] The device temporarily stores the uploaded file and obtains the file's metadata (file name, file size, upload date and time, etc.).

[0761] Step 3:

[0762] The server receives the files and metadata acquired from the device, stores them in a database, and records the storage location (file path) of the saved files.

[0763] Step 4:

[0764] The server sequentially retrieves files stored in the database and sends them to a natural language processing (NLP) engine for analysis, which includes text extraction, semantic analysis, and keyword extraction.

[0765] Step 5:

[0766] The NLP engine analyzes the file contents and extracts business-critical information (e.g., project name, responsible person, deadline, etc.).

[0767] Step 6:

[0768] The server receives the analyzed data obtained from the NLP engine and stores it in a database as structured data.

[0769] Step 7:

[0770] The server generates a virtual worker (persona) specialized for a specific job based on structured data from the database, and this persona model is assigned answer patterns and response styles.

[0771] Step 8:

[0772] The terminal transmits communication data between the user and the virtual worker to the emotion engine, which analyzes the user's emotions (e.g., joy, anger, anxiety, etc.).

[0773] Step 9:

[0774] The emotion engine extracts emotions from the user's communication data and feeds the emotions back to the server.

[0775] Step 10:

[0776] A user enters a question (for example, "Where are the materials for next week's meeting?") into the system in chat format.

[0777] Step 11:

[0778] The terminal processes the user's questions in real time and sends them to the server.

[0779] Step 12:

[0780] The server receives the user's question and the analysis results of the emotion engine, and generates an appropriate answer based on the persona data. For example, if the user is feeling anxious, the persona will respond in a gentle tone, saying, "Don't worry, the meeting materials are in the meeting folder on the shared drive."

[0781] Step 13:

[0782] The terminal displays the answer returned from the server to the user.

[0783] Step 14:

[0784] The server stores the user's interaction history with the persona and the analyzed emotion data in a database.

[0785] Step 15:

[0786] The server periodically analyzes the stored interaction history and emotion data and uses this data as learning data when generating the next response, allowing the system to continuously improve and provide more accurate responses.

[0787] Example 2

[0788] 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."

[0789] In conventional business handovers, the information being handed over is often insufficient, and the details of the handed over business are often not fully understood, resulting in problems with the smooth execution of business. Also, there are cases where information is inconsistent or missing during the handover, making it difficult to improve business efficiency and ensure an effective handover. Furthermore, there is a lack of systems that respond appropriately to user emotions, which is a factor that reduces user satisfaction.

[0790] 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. In this invention, the server includes means for acquiring business materials and communication data, means for analyzing the acquired data and extracting important information, means for generating a virtual worker based on the extracted important information, means for analyzing the user's emotions, and means for adjusting the virtual worker's response based on the emotions. This enables efficient and effective handover of work and provides appropriate responses that take the user's emotions into consideration, thereby improving work productivity and user satisfaction.

[0791] "Business materials" refers to any form of digital data related to business, such as documents, spreadsheets, presentation files, emails, chat logs, etc.

[0792] "Communications Data" refers to data that records communications between users, such as emails, chat logs, and messages from messaging applications.

[0793] "Means of acquisition" refers to a method of executing a process in which business materials and communication data are uploaded by users to the system, and the data is temporarily stored and sent to a server.

[0794] "Means of analysis" refers to a method of analyzing the acquired data based on natural language processing technology and performing text extraction, semantic analysis, and keyword extraction.

[0795] "Important information" refers to information that is considered particularly important in business (for example, project name, person in charge, deadline, etc.).

[0796] A "virtual worker" refers to a virtual person who has response patterns and styles specialized for a specific job content, constructed based on the user's work documents and communication data.

[0797] "Means for transmitting communication data" refers to a process for transmitting questions or requests from users to virtual workers and receiving their responses.

[0798] "Means for obtaining a response" refers to a method by which the system receives a response sent by a virtual worker and provides it to the user.

[0799] "Means for analyzing emotions" refers to technology for extracting emotions (e.g., joy, anger, anxiety, etc.) from a user's communication data.

[0800] "Means for adjusting the virtual worker's response based on emotion" refers to a method for appropriately adjusting the tone and content of the virtual worker's response by referring to the user's emotional data.

[0801] MODE FOR CARRYING OUT THE INVENTION

[0802] This invention relates to a system for efficiently and effectively communicating information during task handover. This system not only acquires and analyzes business documents and communication data to generate virtual workers (personas), but also incorporates an emotion engine that recognizes the user's emotions, further improving the accuracy and appropriateness of responses.

[0803] The whole system program consists of three main components: server, terminal and user.

[0804] Hardware and software configuration:

[0805] Server: A server with large storage capacity and high-performance computing power is required. A database management system (e.g., MySQL, PostgreSQL) and a natural language processing (NLP) engine (e.g., Google NLP, IBM Watson NLP) are installed and used.

[0806] Device: A user-accessible computer or smartphone that has a web browser or dedicated client application installed.

[0807] User: Uploads and queries business data.

[0808] Data Collection Phase

[0809] A user uploads business documents or communication data to the system. For example, the user uploads files by dragging and dropping them using the interface on a web browser.

[0810] The device temporarily stores these files and obtains metadata such as file name, size, and upload date and time.

[0811] The server receives the files and metadata acquired from the device and stores them in a dedicated database.

[0812] Data analysis phase

[0813] The server sequentially retrieves files stored in the database and sends them to the NLP engine for analysis, which includes text extraction, semantic analysis, and keyword extraction.

[0814] The NLP engine analyzes the file contents and extracts business-critical information (e.g., project name, responsible person, deadline, etc.).

[0815] The server stores the analyzed data as structured data in a database.

[0816] Persona generation phase

[0817] The server uses the structured data in the database to generate virtual workers (personas) tailored to specific job tasks, with response patterns and styles set for each persona model.

[0818] Emotion Recognition Phase

[0819] The terminal transmits communication data between the user and the virtual worker to the emotion engine, which analyzes the user's emotions (e.g., joy, anger, anxiety, etc.).

[0820] The emotion engine extracts emotions from the user's communication data and feeds the emotions back to the server.

[0821] User Interaction Phase

[0822] A user enters a question (for example, "Where are the materials for next week's meeting?") into the system in chat format.

[0823] The terminal processes the user's questions in real time and sends them to the server.

[0824] The server receives the user's question and the analysis results of the emotion engine, and generates an appropriate answer based on the persona data. For example, if the user is feeling anxious, the persona will respond in a gentle tone, saying, "Don't worry, the meeting materials are in the meeting folder on the shared drive."

[0825] The terminal displays the answer content returned from the server to the user.

[0826] Learning Phase

[0827] The server stores the user's interaction history with the persona and the analyzed emotion data in a database.

[0828] The system periodically analyzes the interaction history and emotional data it has received and uses this data as learning data when generating the next response, allowing the system to continuously improve and generate more accurate responses.

[0829] Specific examples

[0830] Before User A goes on a one-month maternity leave, she uploads her work documents and email archives to the system. The system analyzes them and generates a persona for User A. During her leave, User B, who is in charge of handover, enters a question into the system: "Where are the documents for next week's meeting?" The server provides an appropriate response based on User A's persona and emotion engine. The system also recognizes User B's emotion toward the question and adjusts the tone and content of the response accordingly. The next time User B asks a new question, it refers to the previous interaction history and emotion data to generate a more accurate answer.

[0831] Example prompt sentence:

[0832] "Before User A goes on a one-month maternity leave, she uploads her work documents and email archives to the system. Generate a virtual worker (persona) for User A based on these documents. Then, imagine that User B, who is in charge of the handover, enters the question, "Where are the documents for next week's meeting?" into the system. Use the emotion engine to generate an appropriate response."

[0833] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0834] Step 1:

[0835] Users upload their business materials (documents, spreadsheets, presentation files, etc.) and communication data (emails, chat logs, etc.) to the system. The input business materials and communication data are imported into the system, and the system is now ready for data collection.

[0836] How it works: A user uploads a file using a web browser interface.

[0837] Step 2:

[0838] The device temporarily stores the uploaded file in local storage. The input is the user's business documents and communication data, and the output is the temporarily stored file and metadata (file name, size, upload date and time, etc.). This prepares the server to receive the file.

[0839] Behavior: Saves the file to a temporary file directory and collects metadata.

[0840] Step 3:

[0841] The server receives the files and metadata acquired from the terminal and stores them in a dedicated database. The input is the files and metadata sent from the terminal, and the output is the files and metadata stored in the database. This identifies the storage location of the files and enables database management.

[0842] How it works: The server receives the file via an API request, stores it in the file system, and records the metadata in a database.

[0843] Step 4:

[0844] The server sequentially retrieves files stored in the database and sends them to a natural language processing (NLP) engine for analysis. The input is the file retrieved from the database, and the output is the analyzed text data. This yields the text data of the business documents.

[0845] What it does: Sends the file via an API request to an NLP engine that performs text extraction, semantic analysis, and keyword extraction.

[0846] Step 5:

[0847] The NLP engine analyzes the file contents and extracts business-critical information. The input is the file sent from the server, and the output is structured analyzed data, which allows for further detailed data analysis.

[0848] How it works: Parses the text, extracts keywords such as project name, person in charge, deadline date, etc., and returns the parsed results in JSON format to the server.

[0849] Step 6:

[0850] The server receives the analyzed data obtained from the NLP engine and stores it in the database as structured data. The input is the analyzed data from the NLP engine, and the output is the structured data stored in the database. This updates the database and adds new data.

[0851] What it does: Receives parsed data in JSON format and inserts it into a database.

[0852] Step 7:

[0853] The server uses the structured data in the database to generate virtual workers (personas) specialized for specific tasks. The input is the structured data in the database, and the output is the generated persona model. This creates a persona that responds to the user's question.

[0854] How it works: Analyzes structured data and generates persona models including response patterns and response styles, which are stored in a model database.

[0855] Step 8:

[0856] The terminal sends communication data between the user and the virtual worker to the emotion engine, which analyzes the user's emotions. The input is the communication data, and the output is the emotion analysis results. This allows the user's emotional state to be understood.

[0857] How it works: Chat logs are captured in real time and sent to the emotion engine via API requests.

[0858] Step 9:

[0859] The emotion engine extracts emotions from the user's communication data and feeds them back to the server. The input is the user's communication data, and the output is emotion data. This makes it possible to respond based on the user's emotions.

[0860] Operation: Analyzes emotions from communication data and returns the emotion data in JSON format to the server.

[0861] Step 10:

[0862] A user inputs a question in chat format on the system. The input is the user's question, and the output is the question data sent to the terminal. This allows the system to recognize the question.

[0863] Action: Type and send text through the chat interface.

[0864] Step 11:

[0865] The terminal processes the user's question in real time and sends it to the server. The input is the user's question data, and the output is the question data sent to the server, allowing the server to analyze the question content.

[0866] What it does: Sends a text message to the server using WebSocket or HTTP request.

[0867] Step 12:

[0868] The server receives the user's question and the analysis results of the emotion engine, and generates an appropriate answer based on the persona data. The input is the user's question data and emotion data, and the output is the generated answer. This allows for an answer that takes emotion into consideration.

[0869] How it works: The question content and sentiment data are combined to generate an answer based on the persona's response patterns, which is then returned in JSON format.

[0870] Step 13:

[0871] The terminal displays the answer content returned from the server to the user. The input is the generated answer data, and the output is the answer text displayed to the user, allowing the user to confirm the answer.

[0872] Behavior: Displays the reply text in the chat interface.

[0873] Step 14:

[0874] The server stores the user's interaction history and the analyzed emotional data in a database. The input is the interaction history and emotional data, and the output is the history and data stored in the database. This enables continuous system learning.

[0875] Operation: Interaction logs and emotion data are stored in a database and updated as training data.

[0876] Step 15:

[0877] The system periodically analyzes the stored interaction history and emotion data and uses them as learning data for the next response generation. The input is the interaction history and emotion data, and the output is an updated response generation model. This allows the system to continuously improve.

[0878] How it works: Use periodic batch processing or online learning algorithms to update the training dataset and retrain the model.

[0879] (Application example 2)

[0880] 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."

[0881] When handing over work tasks, information is often not communicated efficiently, which can lead to reduced work efficiency, especially in places where there is a lot of information and a lot of movement. Furthermore, there is a lack of responses that take user feelings into consideration, which can have a negative impact on work productivity and user satisfaction. There is a need to solve these issues and make work smoother and more efficient.

[0882] Means to solve the problem

[0883] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for acquiring business materials and communication data, means for analyzing the acquired data and extracting important information, means for generating virtual workers based on the extracted important information, and means installed on smartphones for analyzing user emotions and providing feedback. This enables efficient information transmission and responses that take emotions into consideration in logistics centers.

[0884] Definitions of important terms included in the claims

[0885] "Business materials" refers to documents and data files required for the operation and business execution of a company or organization.

[0886] "Communications Data" refers to information exchanged electronically, such as emails, chat logs, and messages.

[0887] "Means" refer to the methods, tools, and techniques used to achieve a particular function or purpose.

[0888] "Analysis" refers to the process of examining data or information in detail to understand its structure and meaning.

[0889] "Important information" refers to information that is essential for carrying out business and that requires particular attention.

[0890] A "virtual worker" refers to a computer-generated model of a worker that is designed to perform a specific task.

[0891] A "smartphone" refers to a portable information terminal that combines the functions of a mobile phone and a personal computer.

[0892] "User emotion" refers to the emotional state of the user, including emotions such as joy, anger, and anxiety.

[0893] "Feedback" refers to providing a reaction or evaluation to an action or response.

[0894] A "system" refers to a collection of multiple components or means combined to achieve a specific purpose.

[0895] MODE FOR CARRYING OUT THE INVENTION

[0896] The present invention relates to a system for efficiently transferring work and transmitting information in a logistics center. Specific embodiments for carrying out the present invention will be described below.

[0897] System Overview

[0898] This system consists of three main components: a server, a terminal (smartphone), and a user. The server acquires business documents and communication data, analyzes them, and extracts important information. It also generates virtual workers (personas) and analyzes user emotions. The terminal acts as an interface with the user and is responsible for collecting and displaying data.

[0899] Hardware and software used

[0900] Server: Amazon Web Services (AWS) is used to store and analyze data. In particular, AWS S3 is used to store files, and AWS RDS is used to store analysis results in a database.

[0901] NLP engine: Uses AWS Comprehend to analyze business documents and communication data.

[0902] Emotion engine: Uses Microsoft Azure Cognitive Services to analyze user emotions.

[0903] Terminal: The application on the smartphone was developed using React Native and provides the data upload and user interface.

[0904] System program and processing flow

[0905] Data collection

[0906] Users use their smartphones to upload business documents (e.g., PDF, Excel, Word) and communication data (e.g., emails, chat logs) to the application. The device stores these files in AWS S3 and temporarily collects metadata. The server receives the files and metadata from the device and stores them in a database.

[0907] Data analysis

[0908] The server sequentially retrieves files stored in the database and sends them to the NLP engine for analysis. Specifically, it performs text extraction, semantic analysis, and keyword extraction. The NLP engine analyzes the file contents and extracts important business information (e.g., project name, person in charge, deadline, etc.). The extracted data is returned to the server and stored as structured data in the database.

[0909] Persona Generation

[0910] The server uses the structured data in the database to generate virtual workers (personas) tailored to specific tasks, with answer patterns and response styles set for each persona model.

[0911] emotion recognition

[0912] The terminal sends communication data between the user and the virtual worker to the emotion engine, which analyzes the user's emotions (e.g., joy, anger, anxiety, etc.). The emotion engine extracts emotions from the user's communication data and feeds the results back to the server.

[0913] User Interaction

[0914] The user enters a chat-style question into the system (e.g., "Please tell me the current stock level."). The device processes the user's question in real time and sends it to the server. The server receives the user's question and the analysis results of the emotion engine, and generates an appropriate answer based on the persona data. For example, if the user is feeling anxious, the persona will respond in a gentle tone, saying, "The current stock level is 500 units, so please don't worry." The device then displays the answer returned from the server to the user.

[0915] Learning Phase

[0916] The server stores the user's interaction history and analyzed emotional data in a database. The stored interaction history and emotional data are used as learning data when generating the next response. This allows the system to continuously improve, enabling more accurate responses.

[0917] Examples of specific examples and prompts

[0918] When an operator asks "What's the current inventory level?" on a smartphone app, the following process takes place:

[0919] 1. The operator enters a question, and the terminal sends it to the server.

[0920] 2. The server analyzes the question and the user's emotions (anxiety), and the persona generates an answer: "There are currently 500 items in stock, so don't worry."

[0921] 3. The device will display the answer.

[0922] Example prompt sentence:

[0923] Here is an example of a conversation to check inventory at a distribution center:

[0924] User: What is your current inventory?

[0925] Persona: We currently have 500 in stock, so don't worry.

[0926] User Emotion: Anxiety

[0927] Persona Response: Respond in a helpful and kind tone.

[0928] By implementing the present invention in the above manner, it is possible to improve the operational efficiency of a logistics center and the satisfaction of users.

[0929] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0930] Step 1:

[0931] The user uploads business documents and communication data to a smartphone app. The input is business documents such as PDF, Excel, and Word documents, and communication data such as emails and chat logs. The device saves these files to AWS S3 and temporarily stores file metadata (file name, file size, upload date and time, etc.). The output is the temporarily saved metadata and the file storage location.

[0932] Step 2:

[0933] The server receives the metadata and file storage location sent from the device. The server records this data in a database and retrieves the file from AWS S3. The input is the metadata and the AWS S3 file path, and the output is the file and metadata stored in the database.

[0934] Step 3:

[0935] The server sequentially retrieves files stored in the database and sends them to an NLP engine (AWS Comprehend) for analysis. The input is the file data, and the NLP engine performs text extraction, semantic analysis, and keyword extraction. The output is important information (e.g., project name, person in charge, deadline, etc.) as the analysis result.

[0936] Step 4:

[0937] The server receives the analysis results from the NLP engine and stores them as structured data in a database. The input is important information (project name, person in charge, deadline, etc.), and the output is structured data stored in the database.

[0938] Step 5:

[0939] The server uses the structured data in the database to generate a virtual worker (persona) specialized for a specific job. The input is the structured data, and the output is a model of the virtual worker with a set answer pattern and response style.

[0940] Step 6:

[0941] The user inputs a question through a smartphone app. For example, "Please tell me the current inventory levels." The input is the user's question text, which the device sends to the server in real time. The output is the user's question data.

[0942] Step 7:

[0943] The server receives the user's question and redirects it to a virtual worker (persona). At the same time, it sends the question text to an emotion engine (Microsoft Azure Cognitive Services) to analyze the user's emotions. The input is the user's question text, the persona model, and the emotion engine's analysis data, and the output is the analysis results.

[0944] Step 8:

[0945] The emotion engine analyzes the user's emotions and feeds the results back to the server. The input is the user's question text, and the output is the analyzed emotion data (joy, anger, anxiety, etc.).

[0946] Step 9:

[0947] The server uses the user's question and emotional data to generate an appropriate response from the persona. For example, if the user is feeling anxious, the persona will respond in a gentle tone, saying, "There are currently 500 items in stock. Don't worry." The input is the user's question, emotional data, and the persona's response style, and the output is the final answer text.

[0948] Step 10:

[0949] The terminal displays the answer content returned from the server to the user. The input is the final answer text, and the output is the answer text displayed on the smartphone screen.

[0950] Through the above processing steps, the handover of work and the transmission of information at the logistics center are carried out efficiently, and responses that take into consideration the feelings of the user are provided.

[0951] 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.

[0952] 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.

[0953] 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.

[0954] [Third embodiment]

[0955] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0956] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0957] 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).

[0958] 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.

[0959] 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.

[0960] 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).

[0961] 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.

[0962] 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.

[0963] 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.

[0964] 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.

[0965] 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.

[0966] 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."

[0967] The present invention relates to a system for efficiently and effectively communicating information during a task handover. This system acquires and analyzes business documents and communication data to generate a virtual worker (persona) that can respond appropriately to questions from users.

[0968] The whole system program consists of three main components: server, terminal and user.

[0969] Data collection phase:

[0970] 1. The user specifies business materials (documents, spreadsheets, presentation files, etc.) and communication data (emails, chat logs, etc.) on the system's upload screen.

[0971] 2. The device temporarily stores these files and obtains the file metadata (file name, file size, upload date and time, etc.).

[0972] 3. The server receives the temporarily stored files and metadata from the device and stores them in a database.

[0973] Data analysis phase:

[0974] 1. The server sequentially retrieves the saved files and sends them to a natural language processing (NLP) engine for analysis.

[0975] 2. The NLP engine analyzes the contents of each file and extracts important business information (project name, person in charge, deadline, etc.).

[0976] 3. The server receives the analyzed data obtained from the NLP engine and stores it in a database as structured data.

[0977] Persona generation phase:

[0978] 1. The server uses structured data from the database to generate virtual workers (personas) specialized for specific tasks.

[0979] 2. The server sets answer patterns and response styles for the generated persona based on predefined rules.

[0980] User interaction phase:

[0981] 1. A user types a question into the system in chat format: "Where are the materials for next week's meeting?"

[0982] 2. The device sends the user's question to the server.

[0983] 3. The server receives the user's question and uses the persona data to generate an appropriate answer, such as "The meeting materials are in the meeting folder on the shared drive."

[0984] 4. The terminal displays the answer returned by the server to the user.

[0985] Learning phase:

[0986] 1. The server stores the user's interaction history with the persona in a database.

[0987] 2. The stored interaction history is periodically analyzed and used as learning data for the next response generation, allowing the system to continuously improve and provide more accurate responses.

[0988] As a concrete example, before User A goes on a one-month maternity leave, she uploads her work documents and email archives to the system. The server analyzes these and generates a persona for User A. When User B, who is in charge of taking over during her leave, enters a question into the system, the server provides a quick and accurate answer based on User A's persona. The next time User B asks a new question, a more accurate answer is generated by referring to the previous interaction history.

[0989] In this way, the present invention can improve the efficiency of business handover and business productivity.

[0990] The processing flow will be explained below.

[0991] Step 1:

[0992] Users upload their business materials (e.g., documents, spreadsheets, presentation files) and communication data (e.g., emails, chat logs) to the Persona Chat system.

[0993] Step 2:

[0994] The device temporarily stores the file uploaded by the user, and acquires the file's metadata (file name, file size, upload date and time, etc.).

[0995] Step 3:

[0996] The server receives the files and metadata acquired from the device, stores them in a database, and records the storage location (file path) of the saved files.

[0997] Step 4:

[0998] The server sequentially retrieves files stored in the database and sends them to a natural language processing (NLP) engine for analysis, which includes text extraction, semantic analysis, and keyword extraction.

[0999] Step 5:

[1000] The NLP engine analyzes the file contents and extracts business-critical information (e.g., project name, responsible person, deadline, etc.).

[1001] Step 6:

[1002] The server receives the parsed data returned by the NLP engine and stores it in a database as structured data, which contains the extracted key information.

[1003] Step 7:

[1004] The server generates virtual workers (personas) specialized for specific tasks based on structured data in the database, with response patterns and styles set for these persona models.

[1005] Step 8:

[1006] A user enters a question (for example, "Where are the materials for next week's meeting?") into the system in chat format.

[1007] Step 9:

[1008] The terminal processes the user's questions in real time and sends them to the server.

[1009] Step 10:

[1010] The server receives questions from users and generates appropriate answers based on persona data, querying relevant information from the database and formulating specific responses.

[1011] Step 11:

[1012] The terminal displays the answer returned from the server to the user. For example, it may present an answer such as "The meeting materials are in the meeting folder on the shared drive."

[1013] Step 12:

[1014] The server stores the interaction history between the user and the persona in a database, which can then be used as learning data when generating a response next time.

[1015] These steps allow the system to efficiently take over and respond quickly and accurately to user questions.

[1016] Example 1

[1017] 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."

[1018] This system solves the problem of difficulty in efficiently and effectively communicating information during task handover. Conventional systems often have difficulty efficiently using business documents and communication data to generate appropriate responses, and they often lack continuous learning capabilities to improve the quality and accuracy of responses.

[1019] 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.

[1020] In this invention, the server includes means for acquiring business materials and communication data, means for temporarily storing the acquired data and acquiring metadata, means for storing the temporarily stored data and metadata in a database, means for transmitting the stored data to a natural language processing engine for analysis and extracting important information, means for generating a virtual worker based on the extracted important information, means for transmitting communication data to the virtual worker and generating a response, means for displaying the generated response to a user, and means for saving an interaction history of the response and communication data and using it as learning data. This makes it possible to efficiently and effectively convey information during business handover and continuously improve the quality and accuracy of responses.

[1021] "Business Materials" refers to information including documents, spreadsheets, presentation files, etc. related to business.

[1022] "Communication data" refers to data including communication history such as emails and chat logs.

[1023] "Server" refers to a computer system that acquires, analyzes, stores, and generates responses to business materials and communication data.

[1024] "Terminal" refers to a computer device used by a user to upload business materials and communication data to the system.

[1025] "User" refers to a human operator authorized to operate the system.

[1026] "Database" refers to data storage for storing business documents, communication data, analysis results, virtual worker profile information, etc.

[1027] A "natural language processing engine" refers to a software module for analyzing text data and extracting important information.

[1028] "Virtual Worker" refers to a simulated worker profile generated based on acquired and analyzed data.

[1029] "Metadata" refers to attribute information that is not the data itself, such as file name, file size, and upload date and time.

[1030] "Interaction history" refers to log data of questions and responses made between a user and a virtual worker.

[1031] "Training data" refers to data used to improve the system's response generation based on the analysis of interaction history.

[1032] "Response style" refers to the format or manner in which the virtual worker responds to the user.

[1033] "Predefined answer patterns" refer to predetermined response templates for specific questions.

[1034] This invention relates to a system for efficiently and effectively communicating information during task handover. This system acquires and analyzes business documents and communication data to generate virtual workers (personas) that respond appropriately to user questions. The entire system consists of three main components: a server, a terminal, and a user.

[1035] Data Collection Phase

[1036] The user specifies work-related documents, spreadsheets, presentation files, emails, chat logs, etc. on the system's upload screen. The device temporarily stores these specified files and acquires metadata such as file name, file size, and upload date and time. The server receives the files and metadata sent from the device and stores them in a database.

[1037] Data analysis phase

[1038] The server sequentially retrieves files stored in the database and sends them to a natural language processing (NLP) engine for analysis. The NLP engine uses, for example, the Python library spaCy or the BERT model. The NLP engine analyzes the contents of the files and extracts important business information such as project names, responsible persons, and deadlines. The server stores the analysis results obtained from the NLP engine in the database as structured data.

[1039] Persona generation phase

[1040] The server uses the structured data stored in the database to generate virtual workers (personas) specialized for specific work tasks. Specifically, it runs a Python script based on information retrieved from the database to generate a persona profile (e.g., name, responsibilities, experience level, etc.). The server then sets answer patterns and response styles for the generated personas based on predefined rules.

[1041] User Interaction Phase

[1042] A user enters a question in chat format into the system, such as "Where are the materials for next week's meeting?" The device sends this question to the server. The server receives the question and uses the persona data to generate an appropriate answer. For example, it generates an answer such as "The meeting materials are in the meeting folder on the shared drive." The device displays the answer returned from the server to the user.

[1043] Learning Phase

[1044] The server stores the interaction history of users and personas in a database. The server periodically analyzes the stored interaction history and uses it as learning data for generating the next response. This allows the system to continuously improve and enable more accurate responses.

[1045] Specific examples

[1046] Before User A goes on a one-month maternity leave, she uploads her work documents and email archives to the system. The server analyzes this data and generates a persona specific to User A's work. While User A is on leave, User B, who is in charge of taking over, enters the question "Where is this month's sales report?" into the system. The server provides a quick and accurate answer based on User A's persona. The next time User B asks a new question, the server references the previous interaction history to generate a more accurate answer.

[1047] Prompt Sentence Examples

[1048] "How can I parse work handover documents and generate virtual worker personas?"

[1049] "When a user types a question into the system, what are the steps to generate an appropriate answer to that question?"

[1050] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1051] Step 1:

[1052] Users specify work-related documents, spreadsheets, presentation files, emails, chat logs, etc. on the system's upload screen.

[1053] Input: Business documents and correspondence data

[1054] Specific operation: The user accesses the system's upload screen using a web browser and clicks the file selection button to select the file to upload.

[1055] Output: Paths and metadata of selected files

[1056] Step 2:

[1057] The terminal temporarily stores the file designated by the user and acquires metadata such as the file name, file size, and upload date and time.

[1058] Input: User specified file

[1059] Specific operation: Uploaded files are temporarily saved in a specific directory on the device, and file metadata is retrieved using Python's os module, etc.

[1060] Output: Path and metadata of the temporarily saved file (file name, size, upload date, etc.)

[1061] Step 3:

[1062] The server receives the files and metadata sent from the terminal and stores them in a database.

[1063] Input: Path and metadata of the temporarily saved file

[1064] Specific operation: The server receives the files and metadata sent via HTTP request, stores the files in object storage, and stores the metadata in a relational database.

[1065] Output: Files and metadata stored in a database

[1066] Step 4:

[1067] The server sequentially retrieves files stored in the database and sends them to a natural language processing (NLP) engine for analysis.

[1068] Input: Path of the file stored in the database

[1069] Specific operation: Obtain the path to the saved file and send a POST request to the NLP engine using the Python requests module.

[1070] Output: File data passed to the NLP engine

[1071] Step 5:

[1072] The NLP engine analyzes the contents of the file and extracts important business information (project name, person in charge, deadline, etc.).

[1073] Input: File sent from the server

[1074] How it works: The NLP engine uses models such as spaCy and BERT to tokenize text and extract entities.

[1075] Output: Extracted important information (project name, person in charge, deadline, etc.)

[1076] Step 6:

[1077] The server stores the analysis results obtained from the NLP engine in a database as structured data.

[1078] Input: Analysis results provided by the NLP engine

[1079] Specific operation: The analysis results are received in JSON format, converted into a data frame using Python's pandas library, and stored in a relational database.

[1080] Output: Structured data stored in a database

[1081] Step 7:

[1082] The server uses structured data stored in a database to generate virtual workers (personas) specialized for specific tasks.

[1083] Input: Structured data stored in a database

[1084] Specific operation: Runs a Python script based on information retrieved from the database to generate a persona profile (e.g., name, responsibilities, experience, etc.).

[1085] Output: Generated persona profile

[1086] Step 8:

[1087] The server sets answer patterns and response styles for the generated personas based on predefined rules.

[1088] Input: Generated persona profile

[1089] What it does: Prepare and apply response templates based on persona profiles for response styling.

[1090] Output: Personas with response styles and answer patterns

[1091] Step 9:

[1092] A user types a question into the system in chat format: "Where are the materials for next week's meeting?"

[1093] Input: User question

[1094] Specific behavior: A user types a question into a chat window on a web application and clicks the send button.

[1095] Output: User question data

[1096] Step 10:

[1097] The terminal sends this question to the server.

[1098] Input: User question data

[1099] What happens: Chat input is sent to the server via WebSocket or HTTP request.

[1100] Output: Question data sent to the server

[1101] Step 11:

[1102] The server receives the question and uses the persona data to generate an appropriate answer, such as "The meeting materials are in the meeting folder on the shared drive."

[1103] Input: User question data, persona data

[1104] Specific operation: The server sends the question data to a natural language processing model, analyzes the intent of the question, searches for relevant information from a database, and generates an answer based on a template.

[1105] Output: Generated response data

[1106] Step 12:

[1107] The terminal displays the answer returned by the server to the user.

[1108] Input: Response data from the server

[1109] Specific behavior: Receives a WebSocket or HTTP response and displays the reply in the chat window.

[1110] Output: The answer displayed to the user

[1111] Step 13:

[1112] The server stores the user and persona interaction history in a database.

[1113] Input: User's question data, Server's answer data

[1114] Specific behavior: Save user questions and persona answers as logs in a database.

[1115] Output: Interaction history stored in a database

[1116] Step 14:

[1117] The server periodically analyzes the stored interaction history and uses it as learning data for generating responses the next time, allowing the system to continuously improve and provide more accurate responses.

[1118] Input: Interaction history stored in a database

[1119] What it does: Set up a batch processing job to parse the stored logs to generate new training data and retrain the model.

[1120] Output: A new trained response generation model

[1121] (Application example 1)

[1122] 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."

[1123] In conventional business handovers, when a person takes temporary leave or resigns, the successor must quickly and accurately understand the work. However, manually reviewing the vast amount of business documents and communication data to extract the necessary information is extremely inefficient and carries a high risk of information leaks and misunderstandings. Work in factories is particularly highly specialized, and a quick and accurate handover has a decisive impact on productivity and quality, making it an urgent task to resolve this issue.

[1124] 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.

[1125] In this invention, the server includes means for acquiring business documents and communication data, means for analyzing the acquired data and extracting important information, means for generating virtual workers based on the extracted important information, means for operators to upload business documents and communication data within the factory, means for analyzing the saved data with a natural language processing engine, means for responding to questions from a new operator based on the virtual workers, and means for saving an interaction history with the operator and using it to generate responses next time. This enables efficient and accurate handover of work within the factory, improving productivity and quality of work.

[1126] "Business materials" refers to all documents and files related to work within a factory, and specifically includes documents such as work procedures, inspection reports, and manuals.

[1127] "Communication data" refers to digital data used to exchange business-related information, and specifically includes records such as e-mails and chat logs.

[1128] A "natural language processing engine" refers to technology that analyzes natural language used by humans and extracts information, and is used to extract important information from text.

[1129] "Virtual workers" refer to virtual assistants created based on the data of real workers, who provide the information needed for task handover.

[1130] "Interaction history" refers to a record of questions and responses between the user and the system (virtual worker), which is used for future learning and response improvement.

[1131] "Server" refers to the primary hardware or software platform for running and managing the various means described above, including storing data, analyzing data, and generating virtual workers.

[1132] A "database" refers to a system for systematically storing collected and analyzed business documents and communication data, and holds information such as structured data and interaction history.

[1133] An "operator" refers to a person who uses the system to take over work, uploading business documents and communication data, entering new questions, and so on.

[1134] This invention relates to a system for efficiently and effectively carrying out task handover within a factory. This system acquires and analyzes work documents and communication data, and generates a virtual worker (persona) that can respond appropriately to questions from a new operator.

[1135] System configuration

[1136] The system is broadly composed of the following components:

[1137] Server: Stores data, analyzes it, and generates virtual workers.

[1138] Terminal: Provides an interface for operators to upload business documents and communication data and input questions into the system.

[1139] User: Uses the system as an operator to upload business documents and ask questions.

[1140] Program processing explanation

[1141] The program of this system executes the following processes in sequence.

[1142] 1. Data collection phase:

[1143] The user specifies business documents (e.g., work procedures, inspection reports, manuals, etc.) and communication data (e.g., emails, chat logs) on the upload screen of the terminal.

[1144] The device temporarily stores these files and obtains the file metadata (e.g., file name, file size, upload date and time, etc.).

[1145] The server receives the temporarily stored files and metadata from the terminal and stores them in a database.

[1146] 2. Data analysis phase:

[1147] The server sequentially retrieves the stored files and sends them to a natural language processing engine (e.g., an NLP engine) for analysis.

[1148] The NLP engine analyzes the contents of each file and extracts important business information (e.g., task content, responsible person, deadline, etc.).

[1149] The server receives the analyzed data obtained from the NLP engine and stores it in a database as structured data.

[1150] 3. Persona generation phase:

[1151] The server uses structured data from the database to generate virtual workers (personas) specialized for specific tasks.

[1152] The server sets answer patterns and response styles for the generated personas based on predefined rules.

[1153] 4. User interaction phase:

[1154] The user types the question into the system in chat format: "Where are last night's test results saved?"

[1155] The terminal sends the user's question to the server.

[1156] The server receives the user's question and uses the persona data to generate an appropriate answer (e.g., a specific answer such as "Your test results can be found in the 'September 2023 Test Results' folder on the shared drive").

[1157] The terminal displays the answer returned by the server to the user.

[1158] 5. Learning Phase:

[1159] The server stores the interaction history between the user and the persona in a database.

[1160] The stored interaction history is periodically analyzed and used as learning data for the next response generation, allowing the system to continuously improve and provide more accurate responses.

[1161] Examples of specific examples and prompts

[1162] For example, if an operator asks the system, "Where are last night's test results stored?", the system will search the database for the appropriate information and respond, "The test results are located in the 'September 2023 Test Results' folder on the shared drive."

[1163] An example of a prompt sentence is as follows:

[1164] "Prompt for a robotic assistant helping with factory handover:

[1165] User: 'I'd like to know the results of last night's test.'

[1166] Robot: 'Test results are stored in the folder 'September 2023 Test Results' on the shared drive.'"

[1167] In this way, the handover of work within the factory can be carried out efficiently and accurately.

[1168] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1169] Step 1:

[1170] The user specifies business documents (e.g., work procedures, inspection reports, manuals, etc.) or communication data (e.g., emails, chat logs) on the upload screen of the terminal and uploads the specified files. The terminal temporarily stores the uploaded files and collects the file metadata (e.g., file name, file size, upload date and time, etc.). The input is the file specified by the user, and the output is the temporarily stored file and metadata.

[1171] Step 2:

[1172] The terminal sends the temporarily stored files and metadata to the server. The server receives this data and stores it in a database. The input is the files and metadata sent from the terminal, and the output is the data stored in the database.

[1173] Step 3:

[1174] The server sequentially retrieves the saved files and sends them to a natural language processing engine (e.g., NLP engine) for analysis. The input is the saved files, and the output is the data analyzed by the NLP engine.

[1175] Step 4:

[1176] The NLP engine analyzes the contents of each file and extracts important business information (e.g., task details, responsible parties, deadlines, etc.). The input is the text data of the file, and the output is the extracted business information.

[1177] Step 5:

[1178] The server receives the analyzed data obtained from the NLP engine and stores it in a database as structured data. The input is the analyzed data, and the output is business information stored as structured data.

[1179] Step 6:

[1180] The server uses the structured data stored in the database to generate virtual workers (personas) specialized for specific tasks. The input is the structured data, and the output is the generated virtual worker.

[1181] Step 7:

[1182] The server sets answer patterns and response styles for the generated virtual workers based on predefined rules. The input is the virtual worker's generation data and predefined rules, and the output is the virtual worker with the set response patterns and styles.

[1183] Step 8:

[1184] A user types a question into the system in chat format: "Where are last night's test results saved?" The input is the user's question, and the output is the data sent from the terminal to the server.

[1185] Step 9:

[1186] The terminal sends the user's question to the server. The input is the user's question, and the output is the data to be sent to the server.

[1187] Step 10:

[1188] The server receives the user's question and generates an appropriate answer using the virtual worker's data (e.g., a specific answer such as "The inspection results can be found in the 'September 2023 Inspection Results' folder on the shared drive"). The input is the user's question, and the output is the generated answer.

[1189] Step 11:

[1190] The terminal displays the answer returned from the server to the user. The input is the answer from the server, and the output is the display data to the user.

[1191] Step 12:

[1192] The server stores the interaction history between the user and the virtual worker in a database. The input is the interaction data, and the output is the data stored in the database.

[1193] Step 13:

[1194] The server periodically analyzes the stored interaction history and uses it as learning data for generating the next response. This allows the system to continuously improve and provide more accurate responses. The input is the interaction history, and the output is the updated learning model.

[1195] 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.

[1196] This invention relates to a system for efficiently and effectively communicating information during task handover. This system not only acquires and analyzes business documents and communication data to generate virtual workers (personas), but also incorporates an emotion engine that recognizes the user's emotions, further improving the accuracy and appropriateness of responses.

[1197] The whole system program consists of three main components: server, terminal and user.

[1198] Data collection phase:

[1199] 1. A user uploads their business materials (documents, spreadsheets, presentation files, etc.) and communication data (emails, chat logs, etc.) to the Persona Chat system.

[1200] 2. The device temporarily stores these files, and at this time, obtains the file metadata (file name, file size, upload date and time, etc.).

[1201] 3. The server receives the files and metadata from the device, stores them in a database, and records the storage location (file path) of the saved files.

[1202] Data analysis phase:

[1203] 1. The server sequentially retrieves files stored in the database and sends them to a natural language processing (NLP) engine for analysis, which includes text extraction, semantic analysis, and keyword extraction.

[1204] 2. The NLP engine analyzes the file contents and extracts business-critical information (e.g., project name, responsible person, deadline, etc.).

[1205] 3. The server receives the analyzed data obtained from the NLP engine and stores it in a database as structured data.

[1206] Persona generation phase:

[1207] 1. The server uses structured data in the database to generate virtual workers (personas) specialized for specific tasks. These persona models are assigned response patterns and styles.

[1208] Emotion Recognition Phase:

[1209] 1. The terminal transmits communication data between the user and the virtual worker to the emotion engine, which analyzes the user's emotions (e.g., joy, anger, anxiety, etc.).

[1210] 2. The emotion engine extracts emotions from the user's communication data and feeds those emotions back to the server.

[1211] User interaction phase:

[1212] 1. A user types a question into the system in chat format (e.g., "Where are the materials for next week's meeting?").

[1213] 2. The terminal processes the user's question in real time and sends it to the server.

[1214] 3. The server receives the user's question and the emotion engine's analysis results, and generates an appropriate answer based on the persona data. For example, if the user is feeling anxious, the persona will respond in a gentle tone, saying, "Don't worry, the meeting materials are in the meeting folder on the shared drive."

[1215] 4. The terminal displays the response returned from the server to the user.

[1216] Learning phase:

[1217] 1. The server stores the user's interaction history with the persona and the analyzed emotion data in a database.

[1218] 2. The system periodically analyzes the stored interaction history and emotion data and uses this data as learning data when generating the next response, allowing the system to continuously improve and provide more accurate responses.

[1219] As a concrete example, before User A goes on a one-month maternity leave, she uploads her work documents and email archives to the system. The server analyzes these and generates a persona for User A. During her leave, User B, who is in charge of taking over, enters questions into the system, and the server provides a quick and appropriate response based on User A's persona and the emotion engine. The system also recognizes User B's emotion in response to her question and adjusts the tone and content of the response accordingly. The next time User B asks a new question, the system will refer to the previous interaction history and emotion data to generate a more accurate answer.

[1220] In this way, the present invention can improve the efficiency of task handover and provide responses that take into consideration the user's feelings, thereby improving task productivity and user satisfaction.

[1221] The processing flow will be explained below.

[1222] Step 1:

[1223] Users upload their business materials (e.g., documents, spreadsheets, presentation files) and communication data (e.g., emails, chat logs) to the Persona Chat system.

[1224] Step 2:

[1225] The device temporarily stores the uploaded file and obtains the file's metadata (file name, file size, upload date and time, etc.).

[1226] Step 3:

[1227] The server receives the files and metadata acquired from the device, stores them in a database, and records the storage location (file path) of the saved files.

[1228] Step 4:

[1229] The server sequentially retrieves files stored in the database and sends them to a natural language processing (NLP) engine for analysis, which includes text extraction, semantic analysis, and keyword extraction.

[1230] Step 5:

[1231] The NLP engine analyzes the file contents and extracts business-critical information (e.g., project name, responsible person, deadline, etc.).

[1232] Step 6:

[1233] The server receives the analyzed data obtained from the NLP engine and stores it in a database as structured data.

[1234] Step 7:

[1235] The server generates a virtual worker (persona) specialized for a specific job based on structured data from the database, and this persona model is assigned answer patterns and response styles.

[1236] Step 8:

[1237] The terminal transmits communication data between the user and the virtual worker to the emotion engine, which analyzes the user's emotions (e.g., joy, anger, anxiety, etc.).

[1238] Step 9:

[1239] The emotion engine extracts emotions from the user's communication data and feeds the emotions back to the server.

[1240] Step 10:

[1241] A user enters a question (for example, "Where are the materials for next week's meeting?") into the system in chat format.

[1242] Step 11:

[1243] The terminal processes the user's questions in real time and sends them to the server.

[1244] Step 12:

[1245] The server receives the user's question and the analysis results of the emotion engine, and generates an appropriate answer based on the persona data. For example, if the user is feeling anxious, the persona will respond in a gentle tone, saying, "Don't worry, the meeting materials are in the meeting folder on the shared drive."

[1246] Step 13:

[1247] The terminal displays the answer returned from the server to the user.

[1248] Step 14:

[1249] The server stores the user's interaction history with the persona and the analyzed emotion data in a database.

[1250] Step 15:

[1251] The server periodically analyzes the stored interaction history and emotion data and uses this data as learning data when generating the next response, allowing the system to continuously improve and provide more accurate responses.

[1252] Example 2

[1253] 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."

[1254] In conventional business handovers, the information being handed over is often insufficient, and the details of the handed over business are often not fully understood, resulting in problems with the smooth execution of business. Also, there are cases where information is inconsistent or missing during the handover, making it difficult to improve business efficiency and ensure an effective handover. Furthermore, there is a lack of systems that respond appropriately to user emotions, which is a factor that reduces user satisfaction.

[1255] 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. In this invention, the server includes means for acquiring business materials and communication data, means for analyzing the acquired data and extracting important information, means for generating a virtual worker based on the extracted important information, means for analyzing the user's emotions, and means for adjusting the virtual worker's response based on the emotions. This enables efficient and effective handover of work and provides appropriate responses that take the user's emotions into consideration, thereby improving work productivity and user satisfaction.

[1256] "Business materials" refers to any form of digital data related to business, such as documents, spreadsheets, presentation files, emails, chat logs, etc.

[1257] "Communications Data" refers to data that records communications between users, such as emails, chat logs, and messages from messaging applications.

[1258] "Means of acquisition" refers to a method of executing a process in which business materials and communication data are uploaded by users to the system, and the data is temporarily stored and sent to a server.

[1259] "Means of analysis" refers to a method of analyzing the acquired data based on natural language processing technology and performing text extraction, semantic analysis, and keyword extraction.

[1260] "Important information" refers to information that is considered particularly important in business (for example, project name, person in charge, deadline, etc.).

[1261] A "virtual worker" refers to a virtual person who has response patterns and styles specialized for a specific job content, constructed based on the user's work documents and communication data.

[1262] "Means for transmitting communication data" refers to a process for transmitting questions or requests from users to virtual workers and receiving their responses.

[1263] "Means for obtaining a response" refers to a method by which the system receives a response sent by a virtual worker and provides it to the user.

[1264] "Means for analyzing emotions" refers to technology for extracting emotions (e.g., joy, anger, anxiety, etc.) from a user's communication data.

[1265] "Means for adjusting the virtual worker's response based on emotion" refers to a method for appropriately adjusting the tone and content of the virtual worker's response by referring to the user's emotional data.

[1266] MODE FOR CARRYING OUT THE INVENTION

[1267] This invention relates to a system for efficiently and effectively communicating information during task handover. This system not only acquires and analyzes business documents and communication data to generate virtual workers (personas), but also incorporates an emotion engine that recognizes the user's emotions, further improving the accuracy and appropriateness of responses.

[1268] The whole system program consists of three main components: server, terminal and user.

[1269] Hardware and software configuration:

[1270] Server: A server with large storage capacity and high-performance computing power is required. A database management system (e.g., MySQL, PostgreSQL) and a natural language processing (NLP) engine (e.g., Google NLP, IBM Watson NLP) are installed and used.

[1271] Device: A user-accessible computer or smartphone that has a web browser or dedicated client application installed.

[1272] User: Uploads and queries business data.

[1273] Data Collection Phase

[1274] A user uploads business documents or communication data to the system. For example, the user uploads files by dragging and dropping them using the interface on a web browser.

[1275] The device temporarily stores these files and obtains metadata such as file name, size, and upload date and time.

[1276] The server receives the files and metadata acquired from the device and stores them in a dedicated database.

[1277] Data analysis phase

[1278] The server sequentially retrieves files stored in the database and sends them to the NLP engine for analysis, which includes text extraction, semantic analysis, and keyword extraction.

[1279] The NLP engine analyzes the file contents and extracts business-critical information (e.g., project name, responsible person, deadline, etc.).

[1280] The server stores the analyzed data as structured data in a database.

[1281] Persona generation phase

[1282] The server uses the structured data in the database to generate virtual workers (personas) tailored to specific job tasks, with response patterns and styles set for each persona model.

[1283] Emotion Recognition Phase

[1284] The terminal transmits communication data between the user and the virtual worker to the emotion engine, which analyzes the user's emotions (e.g., joy, anger, anxiety, etc.).

[1285] The emotion engine extracts emotions from the user's communication data and feeds the emotions back to the server.

[1286] User Interaction Phase

[1287] A user enters a question (for example, "Where are the materials for next week's meeting?") into the system in chat format.

[1288] The terminal processes the user's questions in real time and sends them to the server.

[1289] The server receives the user's question and the analysis results of the emotion engine, and generates an appropriate answer based on the persona data. For example, if the user is feeling anxious, the persona will respond in a gentle tone, saying, "Don't worry, the meeting materials are in the meeting folder on the shared drive."

[1290] The terminal displays the answer content returned from the server to the user.

[1291] Learning Phase

[1292] The server stores the user's interaction history with the persona and the analyzed emotion data in a database.

[1293] The system periodically analyzes the interaction history and emotional data it has received and uses this data as learning data when generating the next response, allowing the system to continuously improve and generate more accurate responses.

[1294] Specific examples

[1295] Before User A goes on a one-month maternity leave, she uploads her work documents and email archives to the system. The system analyzes them and generates a persona for User A. During her leave, User B, who is in charge of handover, enters a question into the system: "Where are the documents for next week's meeting?" The server provides an appropriate response based on User A's persona and emotion engine. The system also recognizes User B's emotion toward the question and adjusts the tone and content of the response accordingly. The next time User B asks a new question, it refers to the previous interaction history and emotion data to generate a more accurate answer.

[1296] Example prompt sentence:

[1297] "Before User A goes on a one-month maternity leave, she uploads her work documents and email archives to the system. Generate a virtual worker (persona) for User A based on these documents. Then, imagine that User B, who is in charge of the handover, enters the question, "Where are the documents for next week's meeting?" into the system. Use the emotion engine to generate an appropriate response."

[1298] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1299] Step 1:

[1300] Users upload their business materials (documents, spreadsheets, presentation files, etc.) and communication data (emails, chat logs, etc.) to the system. The input business materials and communication data are imported into the system, and the system is now ready for data collection.

[1301] How it works: A user uploads a file using a web browser interface.

[1302] Step 2:

[1303] The device temporarily stores the uploaded file in local storage. The input is the user's business documents and communication data, and the output is the temporarily stored file and metadata (file name, size, upload date and time, etc.). This prepares the server to receive the file.

[1304] Behavior: Saves the file to a temporary file directory and collects metadata.

[1305] Step 3:

[1306] The server receives the files and metadata acquired from the terminal and stores them in a dedicated database. The input is the files and metadata sent from the terminal, and the output is the files and metadata stored in the database. This identifies the storage location of the files and enables database management.

[1307] How it works: The server receives the file via an API request, stores it in the file system, and records the metadata in a database.

[1308] Step 4:

[1309] The server sequentially retrieves files stored in the database and sends them to a natural language processing (NLP) engine for analysis. The input is the file retrieved from the database, and the output is the analyzed text data. This yields the text data of the business documents.

[1310] What it does: Sends the file via an API request to an NLP engine that performs text extraction, semantic analysis, and keyword extraction.

[1311] Step 5:

[1312] The NLP engine analyzes the file contents and extracts business-critical information. The input is the file sent from the server, and the output is structured analyzed data, which allows for further detailed data analysis.

[1313] How it works: Parses the text, extracts keywords such as project name, person in charge, deadline date, etc., and returns the parsed results in JSON format to the server.

[1314] Step 6:

[1315] The server receives the analyzed data obtained from the NLP engine and stores it in the database as structured data. The input is the analyzed data from the NLP engine, and the output is the structured data stored in the database. This updates the database and adds new data.

[1316] What it does: Receives parsed data in JSON format and inserts it into a database.

[1317] Step 7:

[1318] The server uses the structured data in the database to generate virtual workers (personas) specialized for specific tasks. The input is the structured data in the database, and the output is the generated persona model. This creates a persona that responds to the user's question.

[1319] How it works: Analyzes structured data and generates persona models including response patterns and response styles, which are stored in a model database.

[1320] Step 8:

[1321] The terminal sends communication data between the user and the virtual worker to the emotion engine, which analyzes the user's emotions. The input is the communication data, and the output is the emotion analysis results. This allows the user's emotional state to be understood.

[1322] How it works: Chat logs are captured in real time and sent to the emotion engine via API requests.

[1323] Step 9:

[1324] The emotion engine extracts emotions from the user's communication data and feeds them back to the server. The input is the user's communication data, and the output is emotion data. This makes it possible to respond based on the user's emotions.

[1325] Operation: Analyzes emotions from communication data and returns the emotion data in JSON format to the server.

[1326] Step 10:

[1327] A user inputs a question in chat format on the system. The input is the user's question, and the output is the question data sent to the terminal. This allows the system to recognize the question.

[1328] Action: Type and send text through the chat interface.

[1329] Step 11:

[1330] The terminal processes the user's question in real time and sends it to the server. The input is the user's question data, and the output is the question data sent to the server, allowing the server to analyze the question content.

[1331] What it does: Sends a text message to the server using WebSocket or HTTP request.

[1332] Step 12:

[1333] The server receives the user's question and the analysis results of the emotion engine, and generates an appropriate answer based on the persona data. The input is the user's question data and emotion data, and the output is the generated answer. This allows for an answer that takes emotion into consideration.

[1334] How it works: The question content and sentiment data are combined to generate an answer based on the persona's response patterns, which is then returned in JSON format.

[1335] Step 13:

[1336] The terminal displays the answer content returned from the server to the user. The input is the generated answer data, and the output is the answer text displayed to the user, allowing the user to confirm the answer.

[1337] Behavior: Displays the reply text in the chat interface.

[1338] Step 14:

[1339] The server stores the user's interaction history and the analyzed emotional data in a database. The input is the interaction history and emotional data, and the output is the history and data stored in the database. This enables continuous system learning.

[1340] Operation: Interaction logs and emotion data are stored in a database and updated as training data.

[1341] Step 15:

[1342] The system periodically analyzes the stored interaction history and emotion data and uses them as learning data for the next response generation. The input is the interaction history and emotion data, and the output is an updated response generation model. This allows the system to continuously improve.

[1343] How it works: Use periodic batch processing or online learning algorithms to update the training dataset and retrain the model.

[1344] (Application example 2)

[1345] 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."

[1346] When handing over work tasks, information is often not communicated efficiently, which can lead to reduced work efficiency, especially in places where there is a lot of information and a lot of movement. Furthermore, there is a lack of responses that take user feelings into consideration, which can have a negative impact on work productivity and user satisfaction. There is a need to solve these issues and make work smoother and more efficient.

[1347] Means to solve the problem

[1348] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for acquiring business materials and communication data, means for analyzing the acquired data and extracting important information, means for generating virtual workers based on the extracted important information, and means installed on smartphones for analyzing user emotions and providing feedback. This enables efficient information transmission and responses that take emotions into consideration in logistics centers.

[1349] Definitions of important terms included in the claims

[1350] "Business materials" refers to documents and data files required for the operation and business execution of a company or organization.

[1351] "Communications Data" refers to information exchanged electronically, such as emails, chat logs, and messages.

[1352] "Means" refer to the methods, tools, and techniques used to achieve a particular function or purpose.

[1353] "Analysis" refers to the process of examining data or information in detail to understand its structure and meaning.

[1354] "Important information" refers to information that is essential for carrying out business and that requires particular attention.

[1355] A "virtual worker" refers to a computer-generated model of a worker that is designed to perform a specific task.

[1356] A "smartphone" refers to a portable information terminal that combines the functions of a mobile phone and a personal computer.

[1357] "User emotion" refers to the emotional state of the user, including emotions such as joy, anger, and anxiety.

[1358] "Feedback" refers to providing a reaction or evaluation to an action or response.

[1359] A "system" refers to a collection of multiple components or means combined to achieve a specific purpose.

[1360] MODE FOR CARRYING OUT THE INVENTION

[1361] The present invention relates to a system for efficiently transferring work and transmitting information in a logistics center. Specific embodiments for carrying out the present invention will be described below.

[1362] System Overview

[1363] This system consists of three main components: a server, a terminal (smartphone), and a user. The server acquires business documents and communication data, analyzes them, and extracts important information. It also generates virtual workers (personas) and analyzes user emotions. The terminal acts as an interface with the user and is responsible for collecting and displaying data.

[1364] Hardware and software used

[1365] Server: Amazon Web Services (AWS) is used to store and analyze data. In particular, AWS S3 is used to store files, and AWS RDS is used to store analysis results in a database.

[1366] NLP engine: Uses AWS Comprehend to analyze business documents and communication data.

[1367] Emotion engine: Uses Microsoft Azure Cognitive Services to analyze user emotions.

[1368] Terminal: The application on the smartphone was developed using React Native and provides the data upload and user interface.

[1369] System program and processing flow

[1370] Data collection

[1371] Users use their smartphones to upload business documents (e.g., PDF, Excel, Word) and communication data (e.g., emails, chat logs) to the application. The device stores these files in AWS S3 and temporarily collects metadata. The server receives the files and metadata from the device and stores them in a database.

[1372] Data analysis

[1373] The server sequentially retrieves files stored in the database and sends them to the NLP engine for analysis. Specifically, it performs text extraction, semantic analysis, and keyword extraction. The NLP engine analyzes the file contents and extracts important business information (e.g., project name, person in charge, deadline, etc.). The extracted data is returned to the server and stored as structured data in the database.

[1374] Persona Generation

[1375] The server uses the structured data in the database to generate virtual workers (personas) tailored to specific tasks, with answer patterns and response styles set for each persona model.

[1376] emotion recognition

[1377] The terminal sends communication data between the user and the virtual worker to the emotion engine, which analyzes the user's emotions (e.g., joy, anger, anxiety, etc.). The emotion engine extracts emotions from the user's communication data and feeds the results back to the server.

[1378] User Interaction

[1379] The user enters a chat-style question into the system (e.g., "Please tell me the current stock level."). The device processes the user's question in real time and sends it to the server. The server receives the user's question and the analysis results of the emotion engine, and generates an appropriate answer based on the persona data. For example, if the user is feeling anxious, the persona will respond in a gentle tone, saying, "The current stock level is 500 units, so please don't worry." The device then displays the answer returned from the server to the user.

[1380] Learning Phase

[1381] The server stores the user's interaction history and analyzed emotional data in a database. The stored interaction history and emotional data are used as learning data when generating the next response. This allows the system to continuously improve, enabling more accurate responses.

[1382] Examples of specific examples and prompts

[1383] When an operator asks "What's the current inventory level?" on a smartphone app, the following process takes place:

[1384] 1. The operator enters a question, and the terminal sends it to the server.

[1385] 2. The server analyzes the question and the user's emotions (anxiety), and the persona generates an answer: "There are currently 500 items in stock, so don't worry."

[1386] 3. The device will display the answer.

[1387] Example prompt sentence:

[1388] Here is an example of a conversation to check inventory at a distribution center:

[1389] User: What is your current inventory?

[1390] Persona: We currently have 500 in stock, so don't worry.

[1391] User Emotion: Anxiety

[1392] Persona Response: Respond in a helpful and kind tone.

[1393] By implementing the present invention in the above manner, it is possible to improve the operational efficiency of a logistics center and the satisfaction of users.

[1394] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1395] Step 1:

[1396] The user uploads business documents and communication data to a smartphone app. The input is business documents such as PDF, Excel, and Word documents, and communication data such as emails and chat logs. The device saves these files to AWS S3 and temporarily stores file metadata (file name, file size, upload date and time, etc.). The output is the temporarily saved metadata and the file storage location.

[1397] Step 2:

[1398] The server receives the metadata and file storage location sent from the device. The server records this data in a database and retrieves the file from AWS S3. The input is the metadata and the AWS S3 file path, and the output is the file and metadata stored in the database.

[1399] Step 3:

[1400] The server sequentially retrieves files stored in the database and sends them to an NLP engine (AWS Comprehend) for analysis. The input is the file data, and the NLP engine performs text extraction, semantic analysis, and keyword extraction. The output is important information (e.g., project name, person in charge, deadline, etc.) as the analysis result.

[1401] Step 4:

[1402] The server receives the analysis results from the NLP engine and stores them as structured data in a database. The input is important information (project name, person in charge, deadline, etc.), and the output is structured data stored in the database.

[1403] Step 5:

[1404] The server uses the structured data in the database to generate a virtual worker (persona) specialized for a specific job. The input is the structured data, and the output is a model of the virtual worker with a set answer pattern and response style.

[1405] Step 6:

[1406] The user inputs a question through a smartphone app. For example, "Please tell me the current inventory levels." The input is the user's question text, which the device sends to the server in real time. The output is the user's question data.

[1407] Step 7:

[1408] The server receives the user's question and redirects it to a virtual worker (persona). At the same time, it sends the question text to an emotion engine (Microsoft Azure Cognitive Services) to analyze the user's emotions. The input is the user's question text, the persona model, and the emotion engine's analysis data, and the output is the analysis results.

[1409] Step 8:

[1410] The emotion engine analyzes the user's emotions and feeds the results back to the server. The input is the user's question text, and the output is the analyzed emotion data (joy, anger, anxiety, etc.).

[1411] Step 9:

[1412] The server uses the user's question and emotional data to generate an appropriate response from the persona. For example, if the user is feeling anxious, the persona will respond in a gentle tone, saying, "There are currently 500 items in stock. Don't worry." The input is the user's question, emotional data, and the persona's response style, and the output is the final answer text.

[1413] Step 10:

[1414] The terminal displays the answer content returned from the server to the user. The input is the final answer text, and the output is the answer text displayed on the smartphone screen.

[1415] Through the above processing steps, the handover of work and the transmission of information at the logistics center are carried out efficiently, and responses that take into consideration the feelings of the user are provided.

[1416] 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.

[1417] 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.

[1418] 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.

[1419] [Fourth embodiment]

[1420] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1421] 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.

[1422] 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).

[1423] 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.

[1424] 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.

[1425] 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).

[1426] 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.

[1427] 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.

[1428] 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.

[1429] 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.

[1430] 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.

[1431] 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.

[1432] 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."

[1433] The present invention relates to a system for efficiently and effectively communicating information during a task handover. This system acquires and analyzes business documents and communication data to generate a virtual worker (persona) that can respond appropriately to questions from users.

[1434] The whole system program consists of three main components: server, terminal and user.

[1435] Data collection phase:

[1436] 1. The user specifies business materials (documents, spreadsheets, presentation files, etc.) and communication data (emails, chat logs, etc.) on the system's upload screen.

[1437] 2. The device temporarily stores these files and obtains the file metadata (file name, file size, upload date and time, etc.).

[1438] 3. The server receives the temporarily stored files and metadata from the device and stores them in a database.

[1439] Data analysis phase:

[1440] 1. The server sequentially retrieves the saved files and sends them to a natural language processing (NLP) engine for analysis.

[1441] 2. The NLP engine analyzes the contents of each file and extracts important business information (project name, person in charge, deadline, etc.).

[1442] 3. The server receives the analyzed data obtained from the NLP engine and stores it in a database as structured data.

[1443] Persona generation phase:

[1444] 1. The server uses structured data from the database to generate virtual workers (personas) specialized for specific tasks.

[1445] 2. The server sets answer patterns and response styles for the generated persona based on predefined rules.

[1446] User interaction phase:

[1447] 1. A user types a question into the system in chat format: "Where are the materials for next week's meeting?"

[1448] 2. The device sends the user's question to the server.

[1449] 3. The server receives the user's question and uses the persona data to generate an appropriate answer, such as "The meeting materials are in the meeting folder on the shared drive."

[1450] 4. The terminal displays the answer returned by the server to the user.

[1451] Learning phase:

[1452] 1. The server stores the user's interaction history with the persona in a database.

[1453] 2. The stored interaction history is periodically analyzed and used as learning data for the next response generation, allowing the system to continuously improve and provide more accurate responses.

[1454] As a concrete example, before User A goes on a one-month maternity leave, she uploads her work documents and email archives to the system. The server analyzes these and generates a persona for User A. When User B, who is in charge of taking over during her leave, enters a question into the system, the server provides a quick and accurate answer based on User A's persona. The next time User B asks a new question, a more accurate answer is generated by referring to the previous interaction history.

[1455] In this way, the present invention can improve the efficiency of business handover and business productivity.

[1456] The processing flow will be explained below.

[1457] Step 1:

[1458] Users upload their business materials (e.g., documents, spreadsheets, presentation files) and communication data (e.g., emails, chat logs) to the Persona Chat system.

[1459] Step 2:

[1460] The device temporarily stores the file uploaded by the user, and acquires the file's metadata (file name, file size, upload date and time, etc.).

[1461] Step 3:

[1462] The server receives the files and metadata acquired from the device, stores them in a database, and records the storage location (file path) of the saved files.

[1463] Step 4:

[1464] The server sequentially retrieves files stored in the database and sends them to a natural language processing (NLP) engine for analysis, which includes text extraction, semantic analysis, and keyword extraction.

[1465] Step 5:

[1466] The NLP engine analyzes the file contents and extracts business-critical information (e.g., project name, responsible person, deadline, etc.).

[1467] Step 6:

[1468] The server receives the parsed data returned by the NLP engine and stores it in a database as structured data, which contains the extracted key information.

[1469] Step 7:

[1470] The server generates virtual workers (personas) specialized for specific tasks based on structured data in the database, with response patterns and styles set for these persona models.

[1471] Step 8:

[1472] A user enters a question (for example, "Where are the materials for next week's meeting?") into the system in chat format.

[1473] Step 9:

[1474] The terminal processes the user's questions in real time and sends them to the server.

[1475] Step 10:

[1476] The server receives questions from users and generates appropriate answers based on persona data, querying relevant information from the database and formulating specific responses.

[1477] Step 11:

[1478] The terminal displays the answer returned from the server to the user. For example, it may present an answer such as "The meeting materials are in the meeting folder on the shared drive."

[1479] Step 12:

[1480] The server stores the interaction history between the user and the persona in a database, which can then be used as learning data when generating a response next time.

[1481] These steps allow the system to efficiently take over and respond quickly and accurately to user questions.

[1482] Example 1

[1483] 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."

[1484] This system solves the problem of difficulty in efficiently and effectively communicating information during task handover. Conventional systems often have difficulty efficiently using business documents and communication data to generate appropriate responses, and they often lack continuous learning capabilities to improve the quality and accuracy of responses.

[1485] 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.

[1486] In this invention, the server includes means for acquiring business materials and communication data, means for temporarily storing the acquired data and acquiring metadata, means for storing the temporarily stored data and metadata in a database, means for transmitting the stored data to a natural language processing engine for analysis and extracting important information, means for generating a virtual worker based on the extracted important information, means for transmitting communication data to the virtual worker and generating a response, means for displaying the generated response to a user, and means for saving an interaction history of the response and communication data and using it as learning data. This makes it possible to efficiently and effectively convey information during business handover and continuously improve the quality and accuracy of responses.

[1487] "Business Materials" refers to information including documents, spreadsheets, presentation files, etc. related to business.

[1488] "Communication data" refers to data including communication history such as emails and chat logs.

[1489] "Server" refers to a computer system that acquires, analyzes, stores, and generates responses to business materials and communication data.

[1490] "Terminal" refers to a computer device used by a user to upload business materials and communication data to the system.

[1491] "User" refers to a human operator authorized to operate the system.

[1492] "Database" refers to data storage for storing business documents, communication data, analysis results, virtual worker profile information, etc.

[1493] A "natural language processing engine" refers to a software module for analyzing text data and extracting important information.

[1494] "Virtual Worker" refers to a simulated worker profile generated based on acquired and analyzed data.

[1495] "Metadata" refers to attribute information that is not the data itself, such as file name, file size, and upload date and time.

[1496] "Interaction history" refers to log data of questions and responses made between a user and a virtual worker.

[1497] "Training data" refers to data used to improve the system's response generation based on the analysis of interaction history.

[1498] "Response style" refers to the format or manner in which the virtual worker responds to the user.

[1499] "Predefined answer patterns" refer to predetermined response templates for specific questions.

[1500] This invention relates to a system for efficiently and effectively communicating information during task handover. This system acquires and analyzes business documents and communication data to generate virtual workers (personas) that respond appropriately to user questions. The entire system consists of three main components: a server, a terminal, and a user.

[1501] Data Collection Phase

[1502] The user specifies work-related documents, spreadsheets, presentation files, emails, chat logs, etc. on the system's upload screen. The device temporarily stores these specified files and acquires metadata such as file name, file size, and upload date and time. The server receives the files and metadata sent from the device and stores them in a database.

[1503] Data analysis phase

[1504] The server sequentially retrieves files stored in the database and sends them to a natural language processing (NLP) engine for analysis. The NLP engine uses, for example, the Python library spaCy or the BERT model. The NLP engine analyzes the contents of the files and extracts important business information such as project names, responsible persons, and deadlines. The server stores the analysis results obtained from the NLP engine in the database as structured data.

[1505] Persona generation phase

[1506] The server uses the structured data stored in the database to generate virtual workers (personas) specialized for specific work tasks. Specifically, it runs a Python script based on information retrieved from the database to generate a persona profile (e.g., name, responsibilities, experience level, etc.). The server then sets answer patterns and response styles for the generated personas based on predefined rules.

[1507] User Interaction Phase

[1508] A user enters a question in chat format into the system, such as "Where are the materials for next week's meeting?" The device sends this question to the server. The server receives the question and uses the persona data to generate an appropriate answer. For example, it generates an answer such as "The meeting materials are in the meeting folder on the shared drive." The device displays the answer returned from the server to the user.

[1509] Learning Phase

[1510] The server stores the interaction history of users and personas in a database. The server periodically analyzes the stored interaction history and uses it as learning data for generating the next response. This allows the system to continuously improve and enable more accurate responses.

[1511] Specific examples

[1512] Before User A goes on a one-month maternity leave, she uploads her work documents and email archives to the system. The server analyzes this data and generates a persona specific to User A's work. While User A is on leave, User B, who is in charge of taking over, enters the question "Where is this month's sales report?" into the system. The server provides a quick and accurate answer based on User A's persona. The next time User B asks a new question, the server references the previous interaction history to generate a more accurate answer.

[1513] Prompt Sentence Examples

[1514] "How can I parse work handover documents and generate virtual worker personas?"

[1515] "When a user types a question into the system, what are the steps to generate an appropriate answer to that question?"

[1516] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1517] Step 1:

[1518] Users specify work-related documents, spreadsheets, presentation files, emails, chat logs, etc. on the system's upload screen.

[1519] Input: Business documents and correspondence data

[1520] Specific operation: The user accesses the system's upload screen using a web browser and clicks the file selection button to select the file to upload.

[1521] Output: Paths and metadata of selected files

[1522] Step 2:

[1523] The terminal temporarily stores the file designated by the user and acquires metadata such as the file name, file size, and upload date and time.

[1524] Input: User specified file

[1525] Specific operation: Uploaded files are temporarily saved in a specific directory on the device, and file metadata is retrieved using Python's os module, etc.

[1526] Output: Path and metadata of the temporarily saved file (file name, size, upload date, etc.)

[1527] Step 3:

[1528] The server receives the files and metadata sent from the terminal and stores them in a database.

[1529] Input: Path and metadata of the temporarily saved file

[1530] Specific operation: The server receives the files and metadata sent via HTTP request, stores the files in object storage, and stores the metadata in a relational database.

[1531] Output: Files and metadata stored in a database

[1532] Step 4:

[1533] The server sequentially retrieves files stored in the database and sends them to a natural language processing (NLP) engine for analysis.

[1534] Input: Path of the file stored in the database

[1535] Specific operation: Obtain the path to the saved file and send a POST request to the NLP engine using the Python requests module.

[1536] Output: File data passed to the NLP engine

[1537] Step 5:

[1538] The NLP engine analyzes the contents of the file and extracts important business information (project name, person in charge, deadline, etc.).

[1539] Input: File sent from the server

[1540] How it works: The NLP engine uses models such as spaCy and BERT to tokenize text and extract entities.

[1541] Output: Extracted important information (project name, person in charge, deadline, etc.)

[1542] Step 6:

[1543] The server stores the analysis results obtained from the NLP engine in a database as structured data.

[1544] Input: Analysis results provided by the NLP engine

[1545] Specific operation: The analysis results are received in JSON format, converted into a data frame using Python's pandas library, and stored in a relational database.

[1546] Output: Structured data stored in a database

[1547] Step 7:

[1548] The server uses structured data stored in a database to generate virtual workers (personas) specialized for specific tasks.

[1549] Input: Structured data stored in a database

[1550] Specific operation: Runs a Python script based on information retrieved from the database to generate a persona profile (e.g., name, responsibilities, experience, etc.).

[1551] Output: Generated persona profile

[1552] Step 8:

[1553] The server sets answer patterns and response styles for the generated personas based on predefined rules.

[1554] Input: Generated persona profile

[1555] What it does: Prepare and apply response templates based on persona profiles for response styling.

[1556] Output: Personas with response styles and answer patterns

[1557] Step 9:

[1558] A user types a question into the system in chat format: "Where are the materials for next week's meeting?"

[1559] Input: User question

[1560] Specific behavior: A user types a question into a chat window on a web application and clicks the send button.

[1561] Output: User question data

[1562] Step 10:

[1563] The terminal sends this question to the server.

[1564] Input: User question data

[1565] What happens: Chat input is sent to the server via WebSocket or HTTP request.

[1566] Output: Question data sent to the server

[1567] Step 11:

[1568] The server receives the question and uses the persona data to generate an appropriate answer, such as "The meeting materials are in the meeting folder on the shared drive."

[1569] Input: User question data, persona data

[1570] Specific operation: The server sends the question data to a natural language processing model, analyzes the intent of the question, searches for relevant information from a database, and generates an answer based on a template.

[1571] Output: Generated response data

[1572] Step 12:

[1573] The terminal displays the answer returned by the server to the user.

[1574] Input: Response data from the server

[1575] Specific behavior: Receives a WebSocket or HTTP response and displays the reply in the chat window.

[1576] Output: The answer displayed to the user

[1577] Step 13:

[1578] The server stores the user and persona interaction history in a database.

[1579] Input: User's question data, Server's answer data

[1580] Specific behavior: Save user questions and persona answers as logs in a database.

[1581] Output: Interaction history stored in a database

[1582] Step 14:

[1583] The server periodically analyzes the stored interaction history and uses it as learning data for generating responses the next time, allowing the system to continuously improve and provide more accurate responses.

[1584] Input: Interaction history stored in a database

[1585] What it does: Set up a batch processing job to parse the stored logs to generate new training data and retrain the model.

[1586] Output: A new trained response generation model

[1587] (Application example 1)

[1588] 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."

[1589] In conventional business handovers, when a person takes temporary leave or resigns, the successor must quickly and accurately understand the work. However, manually reviewing the vast amount of business documents and communication data to extract the necessary information is extremely inefficient and carries a high risk of information leaks and misunderstandings. Work in factories is particularly highly specialized, and a quick and accurate handover has a decisive impact on productivity and quality, making it an urgent task to resolve this issue.

[1590] 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.

[1591] In this invention, the server includes means for acquiring business documents and communication data, means for analyzing the acquired data and extracting important information, means for generating virtual workers based on the extracted important information, means for operators to upload business documents and communication data within the factory, means for analyzing the saved data with a natural language processing engine, means for responding to questions from a new operator based on the virtual workers, and means for saving an interaction history with the operator and using it to generate responses next time. This enables efficient and accurate handover of work within the factory, improving productivity and quality of work.

[1592] "Business materials" refers to all documents and files related to work within a factory, and specifically includes documents such as work procedures, inspection reports, and manuals.

[1593] "Communication data" refers to digital data used to exchange business-related information, and specifically includes records such as e-mails and chat logs.

[1594] A "natural language processing engine" refers to technology that analyzes natural language used by humans and extracts information, and is used to extract important information from text.

[1595] "Virtual workers" refer to virtual assistants created based on the data of real workers, who provide the information needed for task handover.

[1596] "Interaction history" refers to a record of questions and responses between the user and the system (virtual worker), which is used for future learning and response improvement.

[1597] "Server" refers to the primary hardware or software platform for running and managing the various means described above, including storing data, analyzing data, and generating virtual workers.

[1598] A "database" refers to a system for systematically storing collected and analyzed business documents and communication data, and holds information such as structured data and interaction history.

[1599] An "operator" refers to a person who uses the system to take over work, uploading business documents and communication data, entering new questions, and so on.

[1600] This invention relates to a system for efficiently and effectively carrying out task handover within a factory. This system acquires and analyzes work documents and communication data, and generates a virtual worker (persona) that can respond appropriately to questions from a new operator.

[1601] System configuration

[1602] The system is broadly composed of the following components:

[1603] Server: Stores data, analyzes it, and generates virtual workers.

[1604] Terminal: Provides an interface for operators to upload business documents and communication data and input questions into the system.

[1605] User: Uses the system as an operator to upload business documents and ask questions.

[1606] Program processing explanation

[1607] The program of this system executes the following processes in sequence.

[1608] 1. Data collection phase:

[1609] The user specifies business documents (e.g., work procedures, inspection reports, manuals, etc.) and communication data (e.g., emails, chat logs) on the upload screen of the terminal.

[1610] The device temporarily stores these files and obtains the file metadata (e.g., file name, file size, upload date and time, etc.).

[1611] The server receives the temporarily stored files and metadata from the terminal and stores them in a database.

[1612] 2. Data analysis phase:

[1613] The server sequentially retrieves the stored files and sends them to a natural language processing engine (e.g., an NLP engine) for analysis.

[1614] The NLP engine analyzes the contents of each file and extracts important business information (e.g., task content, responsible person, deadline, etc.).

[1615] The server receives the analyzed data obtained from the NLP engine and stores it in a database as structured data.

[1616] 3. Persona generation phase:

[1617] The server uses structured data from the database to generate virtual workers (personas) specialized for specific tasks.

[1618] The server sets answer patterns and response styles for the generated personas based on predefined rules.

[1619] 4. User interaction phase:

[1620] The user types the question into the system in chat format: "Where are last night's test results saved?"

[1621] The terminal sends the user's question to the server.

[1622] The server receives the user's question and uses the persona data to generate an appropriate answer (e.g., a specific answer such as "Your test results can be found in the 'September 2023 Test Results' folder on the shared drive").

[1623] The terminal displays the answer returned by the server to the user.

[1624] 5. Learning Phase:

[1625] The server stores the interaction history between the user and the persona in a database.

[1626] The stored interaction history is periodically analyzed and used as learning data for the next response generation, allowing the system to continuously improve and provide more accurate responses.

[1627] Examples of specific examples and prompts

[1628] For example, if an operator asks the system, "Where are last night's test results stored?", the system will search the database for the appropriate information and respond, "The test results are located in the 'September 2023 Test Results' folder on the shared drive."

[1629] An example of a prompt sentence is as follows:

[1630] "Prompt for a robotic assistant helping with factory handover:

[1631] User: 'I'd like to know the results of last night's test.'

[1632] Robot: 'Test results are stored in the folder 'September 2023 Test Results' on the shared drive.'"

[1633] In this way, the handover of work within the factory can be carried out efficiently and accurately.

[1634] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1635] Step 1:

[1636] The user specifies business documents (e.g., work procedures, inspection reports, manuals, etc.) or communication data (e.g., emails, chat logs) on the upload screen of the terminal and uploads the specified files. The terminal temporarily stores the uploaded files and collects the file metadata (e.g., file name, file size, upload date and time, etc.). The input is the file specified by the user, and the output is the temporarily stored file and metadata.

[1637] Step 2:

[1638] The terminal sends the temporarily stored files and metadata to the server. The server receives this data and stores it in a database. The input is the files and metadata sent from the terminal, and the output is the data stored in the database.

[1639] Step 3:

[1640] The server sequentially retrieves the saved files and sends them to a natural language processing engine (e.g., NLP engine) for analysis. The input is the saved files, and the output is the data analyzed by the NLP engine.

[1641] Step 4:

[1642] The NLP engine analyzes the contents of each file and extracts important business information (e.g., task details, responsible parties, deadlines, etc.). The input is the text data of the file, and the output is the extracted business information.

[1643] Step 5:

[1644] The server receives the analyzed data obtained from the NLP engine and stores it in a database as structured data. The input is the analyzed data, and the output is business information stored as structured data.

[1645] Step 6:

[1646] The server uses the structured data stored in the database to generate virtual workers (personas) specialized for specific tasks. The input is the structured data, and the output is the generated virtual worker.

[1647] Step 7:

[1648] The server sets answer patterns and response styles for the generated virtual workers based on predefined rules. The input is the virtual worker's generation data and predefined rules, and the output is the virtual worker with the set response patterns and styles.

[1649] Step 8:

[1650] A user types a question into the system in chat format: "Where are last night's test results saved?" The input is the user's question, and the output is the data sent from the terminal to the server.

[1651] Step 9:

[1652] The terminal sends the user's question to the server. The input is the user's question, and the output is the data to be sent to the server.

[1653] Step 10:

[1654] The server receives the user's question and generates an appropriate answer using the virtual worker's data (e.g., a specific answer such as "The inspection results can be found in the 'September 2023 Inspection Results' folder on the shared drive"). The input is the user's question, and the output is the generated answer.

[1655] Step 11:

[1656] The terminal displays the answer returned from the server to the user. The input is the answer from the server, and the output is the display data to the user.

[1657] Step 12:

[1658] The server stores the interaction history between the user and the virtual worker in a database. The input is the interaction data, and the output is the data stored in the database.

[1659] Step 13:

[1660] The server periodically analyzes the stored interaction history and uses it as learning data for generating the next response. This allows the system to continuously improve and provide more accurate responses. The input is the interaction history, and the output is the updated learning model.

[1661] 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.

[1662] This invention relates to a system for efficiently and effectively communicating information during task handover. This system not only acquires and analyzes business documents and communication data to generate virtual workers (personas), but also incorporates an emotion engine that recognizes the user's emotions, further improving the accuracy and appropriateness of responses.

[1663] The whole system program consists of three main components: server, terminal and user.

[1664] Data collection phase:

[1665] 1. A user uploads their business materials (documents, spreadsheets, presentation files, etc.) and communication data (emails, chat logs, etc.) to the Persona Chat system.

[1666] 2. The device temporarily stores these files, and at this time, obtains the file metadata (file name, file size, upload date and time, etc.).

[1667] 3. The server receives the files and metadata from the device, stores them in a database, and records the storage location (file path) of the saved files.

[1668] Data analysis phase:

[1669] 1. The server sequentially retrieves files stored in the database and sends them to a natural language processing (NLP) engine for analysis, which includes text extraction, semantic analysis, and keyword extraction.

[1670] 2. The NLP engine analyzes the file contents and extracts business-critical information (e.g., project name, responsible person, deadline, etc.).

[1671] 3. The server receives the analyzed data obtained from the NLP engine and stores it in a database as structured data.

[1672] Persona generation phase:

[1673] 1. The server uses structured data in the database to generate virtual workers (personas) specialized for specific tasks. These persona models are assigned response patterns and styles.

[1674] Emotion Recognition Phase:

[1675] 1. The terminal transmits communication data between the user and the virtual worker to the emotion engine, which analyzes the user's emotions (e.g., joy, anger, anxiety, etc.).

[1676] 2. The emotion engine extracts emotions from the user's communication data and feeds those emotions back to the server.

[1677] User interaction phase:

[1678] 1. A user types a question into the system in chat format (e.g., "Where are the materials for next week's meeting?").

[1679] 2. The terminal processes the user's question in real time and sends it to the server.

[1680] 3. The server receives the user's question and the emotion engine's analysis results, and generates an appropriate answer based on the persona data. For example, if the user is feeling anxious, the persona will respond in a gentle tone, saying, "Don't worry, the meeting materials are in the meeting folder on the shared drive."

[1681] 4. The terminal displays the response returned from the server to the user.

[1682] Learning phase:

[1683] 1. The server stores the user's interaction history with the persona and the analyzed emotion data in a database.

[1684] 2. The system periodically analyzes the stored interaction history and emotion data and uses this data as learning data when generating the next response, allowing the system to continuously improve and provide more accurate responses.

[1685] As a concrete example, before User A goes on a one-month maternity leave, she uploads her work documents and email archives to the system. The server analyzes these and generates a persona for User A. During her leave, User B, who is in charge of taking over, enters questions into the system, and the server provides a quick and appropriate response based on User A's persona and the emotion engine. The system also recognizes User B's emotion in response to her question and adjusts the tone and content of the response accordingly. The next time User B asks a new question, the system will refer to the previous interaction history and emotion data to generate a more accurate answer.

[1686] In this way, the present invention can improve the efficiency of task handover and provide responses that take into consideration the user's feelings, thereby improving task productivity and user satisfaction.

[1687] The processing flow will be explained below.

[1688] Step 1:

[1689] Users upload their business materials (e.g., documents, spreadsheets, presentation files) and communication data (e.g., emails, chat logs) to the Persona Chat system.

[1690] Step 2:

[1691] The device temporarily stores the uploaded file and obtains the file's metadata (file name, file size, upload date and time, etc.).

[1692] Step 3:

[1693] The server receives the files and metadata acquired from the device, stores them in a database, and records the storage location (file path) of the saved files.

[1694] Step 4:

[1695] The server sequentially retrieves files stored in the database and sends them to a natural language processing (NLP) engine for analysis, which includes text extraction, semantic analysis, and keyword extraction.

[1696] Step 5:

[1697] The NLP engine analyzes the file contents and extracts business-critical information (e.g., project name, responsible person, deadline, etc.).

[1698] Step 6:

[1699] The server receives the analyzed data obtained from the NLP engine and stores it in a database as structured data.

[1700] Step 7:

[1701] The server generates a virtual worker (persona) specialized for a specific job based on structured data from the database, and this persona model is assigned answer patterns and response styles.

[1702] Step 8:

[1703] The terminal transmits communication data between the user and the virtual worker to the emotion engine, which analyzes the user's emotions (e.g., joy, anger, anxiety, etc.).

[1704] Step 9:

[1705] The emotion engine extracts emotions from the user's communication data and feeds the emotions back to the server.

[1706] Step 10:

[1707] A user enters a question (for example, "Where are the materials for next week's meeting?") into the system in chat format.

[1708] Step 11:

[1709] The terminal processes the user's questions in real time and sends them to the server.

[1710] Step 12:

[1711] The server receives the user's question and the analysis results of the emotion engine, and generates an appropriate answer based on the persona data. For example, if the user is feeling anxious, the persona will respond in a gentle tone, saying, "Don't worry, the meeting materials are in the meeting folder on the shared drive."

[1712] Step 13:

[1713] The terminal displays the answer returned from the server to the user.

[1714] Step 14:

[1715] The server stores the user's interaction history with the persona and the analyzed emotion data in a database.

[1716] Step 15:

[1717] The server periodically analyzes the stored interaction history and emotion data and uses this data as learning data when generating the next response, allowing the system to continuously improve and provide more accurate responses.

[1718] Example 2

[1719] 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."

[1720] In conventional business handovers, the information being handed over is often insufficient, and the details of the handed over business are often not fully understood, resulting in problems with the smooth execution of business. Also, there are cases where information is inconsistent or missing during the handover, making it difficult to improve business efficiency and ensure an effective handover. Furthermore, there is a lack of systems that respond appropriately to user emotions, which is a factor that reduces user satisfaction.

[1721] 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. In this invention, the server includes means for acquiring business materials and communication data, means for analyzing the acquired data and extracting important information, means for generating a virtual worker based on the extracted important information, means for analyzing the user's emotions, and means for adjusting the virtual worker's response based on the emotions. This enables efficient and effective handover of work and provides appropriate responses that take the user's emotions into consideration, thereby improving work productivity and user satisfaction.

[1722] "Business materials" refers to any form of digital data related to business, such as documents, spreadsheets, presentation files, emails, chat logs, etc.

[1723] "Communications Data" refers to data that records communications between users, such as emails, chat logs, and messages from messaging applications.

[1724] "Means of acquisition" refers to a method of executing a process in which business materials and communication data are uploaded by users to the system, and the data is temporarily stored and sent to a server.

[1725] "Means of analysis" refers to a method of analyzing the acquired data based on natural language processing technology and performing text extraction, semantic analysis, and keyword extraction.

[1726] "Important information" refers to information that is considered particularly important in business (for example, project name, person in charge, deadline, etc.).

[1727] A "virtual worker" refers to a virtual person who has response patterns and styles specialized for a specific job content, constructed based on the user's work documents and communication data.

[1728] "Means for transmitting communication data" refers to a process for transmitting questions or requests from users to virtual workers and receiving their responses.

[1729] "Means for obtaining a response" refers to a method by which the system receives a response sent by a virtual worker and provides it to the user.

[1730] "Means for analyzing emotions" refers to technology for extracting emotions (e.g., joy, anger, anxiety, etc.) from a user's communication data.

[1731] "Means for adjusting the virtual worker's response based on emotion" refers to a method for appropriately adjusting the tone and content of the virtual worker's response by referring to the user's emotional data.

[1732] MODE FOR CARRYING OUT THE INVENTION

[1733] This invention relates to a system for efficiently and effectively communicating information during task handover. This system not only acquires and analyzes business documents and communication data to generate virtual workers (personas), but also incorporates an emotion engine that recognizes the user's emotions, further improving the accuracy and appropriateness of responses.

[1734] The whole system program consists of three main components: server, terminal and user.

[1735] Hardware and software configuration:

[1736] Server: A server with large storage capacity and high-performance computing power is required. A database management system (e.g., MySQL, PostgreSQL) and a natural language processing (NLP) engine (e.g., Google NLP, IBM Watson NLP) are installed and used.

[1737] Device: A user-accessible computer or smartphone that has a web browser or dedicated client application installed.

[1738] User: Uploads and queries business data.

[1739] Data Collection Phase

[1740] A user uploads business documents or communication data to the system. For example, the user uploads files by dragging and dropping them using the interface on a web browser.

[1741] The device temporarily stores these files and obtains metadata such as file name, size, and upload date and time.

[1742] The server receives the files and metadata acquired from the device and stores them in a dedicated database.

[1743] Data analysis phase

[1744] The server sequentially retrieves files stored in the database and sends them to the NLP engine for analysis, which includes text extraction, semantic analysis, and keyword extraction.

[1745] The NLP engine analyzes the file contents and extracts business-critical information (e.g., project name, responsible person, deadline, etc.).

[1746] The server stores the analyzed data as structured data in a database.

[1747] Persona generation phase

[1748] The server uses the structured data in the database to generate virtual workers (personas) tailored to specific job tasks, with response patterns and styles set for each persona model.

[1749] Emotion Recognition Phase

[1750] The terminal transmits communication data between the user and the virtual worker to the emotion engine, which analyzes the user's emotions (e.g., joy, anger, anxiety, etc.).

[1751] The emotion engine extracts emotions from the user's communication data and feeds the emotions back to the server.

[1752] User Interaction Phase

[1753] A user enters a question (for example, "Where are the materials for next week's meeting?") into the system in chat format.

[1754] The terminal processes the user's questions in real time and sends them to the server.

[1755] The server receives the user's question and the analysis results of the emotion engine, and generates an appropriate answer based on the persona data. For example, if the user is feeling anxious, the persona will respond in a gentle tone, saying, "Don't worry, the meeting materials are in the meeting folder on the shared drive."

[1756] The terminal displays the answer content returned from the server to the user.

[1757] Learning Phase

[1758] The server stores the user's interaction history with the persona and the analyzed emotion data in a database.

[1759] The system periodically analyzes the interaction history and emotional data it has received and uses this data as learning data when generating the next response, allowing the system to continuously improve and generate more accurate responses.

[1760] Specific examples

[1761] Before User A goes on a one-month maternity leave, she uploads her work documents and email archives to the system. The system analyzes them and generates a persona for User A. During her leave, User B, who is in charge of handover, enters a question into the system: "Where are the documents for next week's meeting?" The server provides an appropriate response based on User A's persona and emotion engine. The system also recognizes User B's emotion toward the question and adjusts the tone and content of the response accordingly. The next time User B asks a new question, it refers to the previous interaction history and emotion data to generate a more accurate answer.

[1762] Example prompt sentence:

[1763] "Before User A goes on a one-month maternity leave, she uploads her work documents and email archives to the system. Generate a virtual worker (persona) for User A based on these documents. Then, imagine that User B, who is in charge of the handover, enters the question, "Where are the documents for next week's meeting?" into the system. Use the emotion engine to generate an appropriate response."

[1764] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1765] Step 1:

[1766] Users upload their business materials (documents, spreadsheets, presentation files, etc.) and communication data (emails, chat logs, etc.) to the system. The input business materials and communication data are imported into the system, and the system is now ready for data collection.

[1767] How it works: A user uploads a file using a web browser interface.

[1768] Step 2:

[1769] The device temporarily stores the uploaded file in local storage. The input is the user's business documents and communication data, and the output is the temporarily stored file and metadata (file name, size, upload date and time, etc.). This prepares the server to receive the file.

[1770] Behavior: Saves the file to a temporary file directory and collects metadata.

[1771] Step 3:

[1772] The server receives the files and metadata acquired from the terminal and stores them in a dedicated database. The input is the files and metadata sent from the terminal, and the output is the files and metadata stored in the database. This identifies the storage location of the files and enables database management.

[1773] How it works: The server receives the file via an API request, stores it in the file system, and records the metadata in a database.

[1774] Step 4:

[1775] The server sequentially retrieves files stored in the database and sends them to a natural language processing (NLP) engine for analysis. The input is the file retrieved from the database, and the output is the analyzed text data. This yields the text data of the business documents.

[1776] What it does: Sends the file via an API request to an NLP engine that performs text extraction, semantic analysis, and keyword extraction.

[1777] Step 5:

[1778] The NLP engine analyzes the file contents and extracts business-critical information. The input is the file sent from the server, and the output is structured analyzed data, which allows for further detailed data analysis.

[1779] How it works: Parses the text, extracts keywords such as project name, person in charge, deadline date, etc., and returns the parsed results in JSON format to the server.

[1780] Step 6:

[1781] The server receives the analyzed data obtained from the NLP engine and stores it in the database as structured data. The input is the analyzed data from the NLP engine, and the output is the structured data stored in the database. This updates the database and adds new data.

[1782] What it does: Receives parsed data in JSON format and inserts it into a database.

[1783] Step 7:

[1784] The server uses the structured data in the database to generate virtual workers (personas) specialized for specific tasks. The input is the structured data in the database, and the output is the generated persona model. This creates a persona that responds to the user's question.

[1785] How it works: Analyzes structured data and generates persona models including response patterns and response styles, which are stored in a model database.

[1786] Step 8:

[1787] The terminal sends communication data between the user and the virtual worker to the emotion engine, which analyzes the user's emotions. The input is the communication data, and the output is the emotion analysis results. This allows the user's emotional state to be understood.

[1788] How it works: Chat logs are captured in real time and sent to the emotion engine via API requests.

[1789] Step 9:

[1790] The emotion engine extracts emotions from the user's communication data and feeds them back to the server. The input is the user's communication data, and the output is emotion data. This makes it possible to respond based on the user's emotions.

[1791] Operation: Analyzes emotions from communication data and returns the emotion data in JSON format to the server.

[1792] Step 10:

[1793] A user inputs a question in chat format on the system. The input is the user's question, and the output is the question data sent to the terminal. This allows the system to recognize the question.

[1794] Action: Type and send text through the chat interface.

[1795] Step 11:

[1796] The terminal processes the user's question in real time and sends it to the server. The input is the user's question data, and the output is the question data sent to the server, allowing the server to analyze the question content.

[1797] What it does: Sends a text message to the server using WebSocket or HTTP request.

[1798] Step 12:

[1799] The server receives the user's question and the analysis results of the emotion engine, and generates an appropriate answer based on the persona data. The input is the user's question data and emotion data, and the output is the generated answer. This allows for an answer that takes emotion into consideration.

[1800] How it works: The question content and sentiment data are combined to generate an answer based on the persona's response patterns, which is then returned in JSON format.

[1801] Step 13:

[1802] The terminal displays the answer content returned from the server to the user. The input is the generated answer data, and the output is the answer text displayed to the user, allowing the user to confirm the answer.

[1803] Behavior: Displays the reply text in the chat interface.

[1804] Step 14:

[1805] The server stores the user's interaction history and the analyzed emotional data in a database. The input is the interaction history and emotional data, and the output is the history and data stored in the database. This enables continuous system learning.

[1806] Operation: Interaction logs and emotion data are stored in a database and updated as training data.

[1807] Step 15:

[1808] The system periodically analyzes the stored interaction history and emotion data and uses them as learning data for the next response generation. The input is the interaction history and emotion data, and the output is an updated response generation model. This allows the system to continuously improve.

[1809] How it works: Use periodic batch processing or online learning algorithms to update the training dataset and retrain the model.

[1810] (Application example 2)

[1811] 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."

[1812] When handing over work tasks, information is often not communicated efficiently, which can lead to reduced work efficiency, especially in places where there is a lot of information and a lot of movement. Furthermore, there is a lack of responses that take user feelings into consideration, which can have a negative impact on work productivity and user satisfaction. There is a need to solve these issues and make work smoother and more efficient.

[1813] Means to solve the problem

[1814] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for acquiring business materials and communication data, means for analyzing the acquired data and extracting important information, means for generating virtual workers based on the extracted important information, and means installed on smartphones for analyzing user emotions and providing feedback. This enables efficient information transmission and responses that take emotions into consideration in logistics centers.

[1815] Definitions of important terms included in the claims

[1816] "Business materials" refers to documents and data files required for the operation and business execution of a company or organization.

[1817] "Communications Data" refers to information exchanged electronically, such as emails, chat logs, and messages.

[1818] "Means" refer to the methods, tools, and techniques used to achieve a particular function or purpose.

[1819] "Analysis" refers to the process of examining data or information in detail to understand its structure and meaning.

[1820] "Important information" refers to information that is essential for carrying out business and that requires particular attention.

[1821] A "virtual worker" refers to a computer-generated model of a worker that is designed to perform a specific task.

[1822] A "smartphone" refers to a portable information terminal that combines the functions of a mobile phone and a personal computer.

[1823] "User emotion" refers to the emotional state of the user, including emotions such as joy, anger, and anxiety.

[1824] "Feedback" refers to providing a reaction or evaluation to an action or response.

[1825] A "system" refers to a collection of multiple components or means combined to achieve a specific purpose.

[1826] MODE FOR CARRYING OUT THE INVENTION

[1827] The present invention relates to a system for efficiently transferring work and transmitting information in a logistics center. Specific embodiments for carrying out the present invention will be described below.

[1828] System Overview

[1829] This system consists of three main components: a server, a terminal (smartphone), and a user. The server acquires business documents and communication data, analyzes them, and extracts important information. It also generates virtual workers (personas) and analyzes user emotions. The terminal acts as an interface with the user and is responsible for collecting and displaying data.

[1830] Hardware and software used

[1831] Server: Amazon Web Services (AWS) is used to store and analyze data. In particular, AWS S3 is used to store files, and AWS RDS is used to store analysis results in a database.

[1832] NLP engine: Uses AWS Comprehend to analyze business documents and communication data.

[1833] Emotion engine: Uses Microsoft Azure Cognitive Services to analyze user emotions.

[1834] Terminal: The application on the smartphone was developed using React Native and provides the data upload and user interface.

[1835] System program and processing flow

[1836] Data collection

[1837] Users use their smartphones to upload business documents (e.g., PDF, Excel, Word) and communication data (e.g., emails, chat logs) to the application. The device stores these files in AWS S3 and temporarily collects metadata. The server receives the files and metadata from the device and stores them in a database.

[1838] Data analysis

[1839] The server sequentially retrieves files stored in the database and sends them to the NLP engine for analysis. Specifically, it performs text extraction, semantic analysis, and keyword extraction. The NLP engine analyzes the file contents and extracts important business information (e.g., project name, person in charge, deadline, etc.). The extracted data is returned to the server and stored as structured data in the database.

[1840] Persona Generation

[1841] The server uses the structured data in the database to generate virtual workers (personas) tailored to specific tasks, with answer patterns and response styles set for each persona model.

[1842] emotion recognition

[1843] The terminal sends communication data between the user and the virtual worker to the emotion engine, which analyzes the user's emotions (e.g., joy, anger, anxiety, etc.). The emotion engine extracts emotions from the user's communication data and feeds the results back to the server.

[1844] User Interaction

[1845] The user enters a chat-style question into the system (e.g., "Please tell me the current stock level."). The device processes the user's question in real time and sends it to the server. The server receives the user's question and the analysis results of the emotion engine, and generates an appropriate answer based on the persona data. For example, if the user is feeling anxious, the persona will respond in a gentle tone, saying, "The current stock level is 500 units, so please don't worry." The device then displays the answer returned from the server to the user.

[1846] Learning Phase

[1847] The server stores the user's interaction history and analyzed emotional data in a database. The stored interaction history and emotional data are used as learning data when generating the next response. This allows the system to continuously improve, enabling more accurate responses.

[1848] Examples of specific examples and prompts

[1849] When an operator asks "What's the current inventory level?" on a smartphone app, the following process takes place:

[1850] 1. The operator enters a question, and the terminal sends it to the server.

[1851] 2. The server analyzes the question and the user's emotions (anxiety), and the persona generates an answer: "There are currently 500 items in stock, so don't worry."

[1852] 3. The device will display the answer.

[1853] Example prompt sentence:

[1854] Here is an example of a conversation to check inventory at a distribution center:

[1855] User: What is your current inventory?

[1856] Persona: We currently have 500 in stock, so don't worry.

[1857] User Emotion: Anxiety

[1858] Persona Response: Respond in a helpful and kind tone.

[1859] By implementing the present invention in the above manner, it is possible to improve the operational efficiency of a logistics center and the satisfaction of users.

[1860] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1861] Step 1:

[1862] The user uploads business documents and communication data to a smartphone app. The input is business documents such as PDF, Excel, and Word documents, and communication data such as emails and chat logs. The device saves these files to AWS S3 and temporarily stores file metadata (file name, file size, upload date and time, etc.). The output is the temporarily saved metadata and the file storage location.

[1863] Step 2:

[1864] The server receives the metadata and file storage location sent from the device. The server records this data in a database and retrieves the file from AWS S3. The input is the metadata and the AWS S3 file path, and the output is the file and metadata stored in the database.

[1865] Step 3:

[1866] The server sequentially retrieves files stored in the database and sends them to an NLP engine (AWS Comprehend) for analysis. The input is the file data, and the NLP engine performs text extraction, semantic analysis, and keyword extraction. The output is important information (e.g., project name, person in charge, deadline, etc.) as the analysis result.

[1867] Step 4:

[1868] The server receives the analysis results from the NLP engine and stores them as structured data in a database. The input is important information (project name, person in charge, deadline, etc.), and the output is structured data stored in the database.

[1869] Step 5:

[1870] The server uses the structured data in the database to generate a virtual worker (persona) specialized for a specific job. The input is the structured data, and the output is a model of the virtual worker with a set answer pattern and response style.

[1871] Step 6:

[1872] The user inputs a question through a smartphone app. For example, "Please tell me the current inventory levels." The input is the user's question text, which the device sends to the server in real time. The output is the user's question data.

[1873] Step 7:

[1874] The server receives the user's question and redirects it to a virtual worker (persona). At the same time, it sends the question text to an emotion engine (Microsoft Azure Cognitive Services) to analyze the user's emotions. The input is the user's question text, the persona model, and the emotion engine's analysis data, and the output is the analysis results.

[1875] Step 8:

[1876] The emotion engine analyzes the user's emotions and feeds the results back to the server. The input is the user's question text, and the output is the analyzed emotion data (joy, anger, anxiety, etc.).

[1877] Step 9:

[1878] The server uses the user's question and emotional data to generate an appropriate response from the persona. For example, if the user is feeling anxious, the persona will respond in a gentle tone, saying, "There are currently 500 items in stock. Don't worry." The input is the user's question, emotional data, and the persona's response style, and the output is the final answer text.

[1879] Step 10:

[1880] The terminal displays the answer content returned from the server to the user. The input is the final answer text, and the output is the answer text displayed on the smartphone screen.

[1881] Through the above processing steps, the handover of work and the transmission of information at the logistics center are carried out efficiently, and responses that take into consideration the feelings of the user are provided.

[1882] 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.

[1883] 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.

[1884] 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 robot 414.

[1885] 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.

[1886] FIG. 9 illustrates 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 behaviors 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.

[1887] 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.

[1888] 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).

[1889] 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.

[1890] 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."

[1891] 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.

[1892] 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).

[1893] 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.

[1894] 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.

[1895] 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.

[1896] 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.

[1897] 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.

[1898] 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.

[1899] 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.

[1900] 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.

[1901] 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.

[1902] 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.

[1903] The following is further disclosed regarding the above embodiment.

[1904] (Claim 1)

[1905] a means for obtaining business materials and communication data;

[1906] means for analyzing the acquired data and extracting important information;

[1907] means for generating a virtual worker based on the extracted critical information;

[1908] means for transmitting communication data to the virtual worker and receiving a response;

[1909] A system including:

[1910] (Claim 2)

[1911] 10. The system of claim 1, wherein the virtual worker includes predefined response patterns for providing automated responses.

[1912] (Claim 3)

[1913] 2. The system according to claim 1, further comprising means for storing the acquired communication data and responses and using them as learning data for use in later response generation.

[1914] "Example 1"

[1915] (Claim 1)

[1916] a means for obtaining business materials and communication data;

[1917] means for temporarily storing the acquired data and acquiring metadata;

[1918] means for storing the temporarily stored data and metadata in a database;

[1919] means for transmitting the stored data to a natural language processing engine for analysis and extracting important information;

[1920] means for generating a virtual worker based on the extracted critical information;

[1921] means for transmitting communication data to said virtual worker and generating a response;

[1922] means for displaying the generated response to a user;

[1923] means for storing an interaction history of the responses and communication data and using the history as learning data;

[1924] A system including:

[1925] (Claim 2)

[1926] 10. The system of claim 1, wherein the virtual worker includes predefined answer patterns and response styles for providing automated responses.

[1927] (Claim 3)

[1928] 2. The system according to claim 1, further comprising means for periodically analyzing the acquired communication data and responses and using the data as learning data for use in generating the next response.

[1929] "Application Example 1"

[1930] (Claim 1)

[1931] a means for obtaining business materials and communication data;

[1932] means for analyzing the acquired data and extracting important information;

[1933] means for generating a virtual worker based on the extracted critical information;

[1934] means for transmitting communication data to the virtual worker and receiving a response;

[1935] A means for operators to upload work documents and communication data within the factory,

[1936] A means of analyzing the stored data using a natural language processing engine;

[1937] means for responding to inquiries from new operators based on said virtual workers;

[1938] a means for storing an interaction history with the operator and using the history to generate a next response;

[1939] A system including:

[1940] (Claim 2)

[1941] 10. The system of claim 1, wherein the virtual worker includes predefined response patterns for providing automated responses.

[1942] (Claim 3)

[1943] 2. The system according to claim 1, further comprising means for storing the acquired communication data and responses and using them as learning data for use in later response generation.

[1944] "Example 2: Combining Emotion Engines"

[1945] (Claim 1)

[1946] a means for obtaining business materials and communication data;

[1947] means for analyzing the acquired data and extracting important information;

[1948] means for generating a virtual worker based on the extracted critical information;

[1949] means for transmitting communication data to the virtual worker and r...

Claims

1. a means for obtaining business materials and communication data; means for analyzing the acquired data and extracting important information; means for generating a virtual worker based on the extracted critical information; means for transmitting communication data to the virtual worker and receiving a response; A system including:

2. The system of claim 1 , wherein the virtual worker includes predefined response patterns for providing automated responses.

3. 2. The system according to claim 1, further comprising means for storing the acquired communication data and responses and using the stored data as learning data for use in later response generation.

Images