Program, information processing device, method, and system

Abstract

[Problem] To assist a user in creating a document while minimizing a burden associated with system construction in cases where there are multiple tasks. [Solution] A program to be executed by a computer including a processor and a memory. This program causes the processor to execute: a step for analyzing a plurality of documents created in the past for a predetermined task, and extracting and storing a plurality of tags for each task; a step for accepting selection of a task linked to document creation; a step for presenting a plurality of tags associated with the selected task; a step for accepting designation for the presented tags; a step for inputting a word as the designated tag and an instruction sentence that instructs to create a document using the word into a large language model, and outputting the document using the word from the large language model; and a step for presenting the output document using the word.

Description

Program, information processing device, method and system

[0001] The present disclosure relates to a program, an information processing device, a method, and a system.

[0002] Technologies for assisting users in creating sentences have been researched and developed. For example, Patent Literature 1 discloses a technology for generating sentences containing words specified by a user by using a language model that has been trained on the order of words that appear in a sentence of a contract for learning.

[0003] Japanese Patent Application Laid-Open No. 2020-035019

[0004] However, in the technology disclosed in Patent Document 1, the server learns each sentence of the structured contract for each category and generates a separate language model, which imposes a heavy burden on the system construction including the language model.

[0005] An object of the present disclosure is to support users in creating documents when there are multiple tasks, while reducing the burden of system construction.

[0006] To solve the above-described problems, one aspect of the present disclosure provides a program for execution by a computer including a processor and a memory. The program causes the processor to execute the following steps: analyzing multiple documents created in the past for a specific task, extracting and storing multiple tags for each task, accepting a selection of a task associated with the creation of the document, presenting multiple tags associated with the selected task, accepting a specification for the presented tags, inputting words as the specified tags and an instruction statement instructing the creation of a document using the words into a large-scale language model, causing the large-scale language model to output a document using the words, and presenting the document using the output words.

[0007] In order to solve the above-described problems, a program according to one aspect of the present disclosure is a program to be executed by a computer including a processor and a memory. The program causes the processor to execute the following steps: analyzing multiple documents created in the past for predetermined tasks, extracting and storing multiple tags for each task, accepting a selection of a task associated with the creation of the document, presenting multiple tags associated with the selected task, accepting a designation for the presented tags and a predetermined image, or a designation for the image, inputting the designated tags and image, or the designated image, and an instruction for creating a document, into a multimodal generative AI model, and causing the multimodal generative AI model to output a document including at least content related to the image, and presenting the output document.

[0008] According to the present disclosure, when there are multiple tasks, it is possible to support a user in creating documents while reducing the burden of system construction.

[0009] 1 is a block diagram showing an example of the overall configuration of the system 1. FIG. 2 is a block diagram showing an example of the configuration of the terminal device 10 shown in FIG. 1. FIG. 3 is a block diagram showing an example of the functional configuration of the server 20 shown in FIG. 1. FIG. 4 is a diagram showing the data structure of a user information table 2021 shown in FIG. 3. FIG. 5 is a diagram showing the data structure of a document table 2022 shown in FIG. 3. FIG. 6 is a diagram showing the data structure of a tag table 2023 shown in FIG. 3. FIG. 7 is a diagram showing the data structure of a prompt table 2024 shown in FIG. 3. FIG. 8 is a flowchart showing an example of the operation of the server 20 when an automatically created document is presented to a user. FIG. 9 is a schematic diagram showing an example of the display screen of the display 141 when tags are presented to a user. FIG. 10 is a schematic diagram showing an example of the display screen of the display 141 when an automatically created document is presented to a user. FIG. 11 is a schematic diagram showing an example of the display screen of the display 141 when an automatically created document is created. FIG. 12 is a schematic diagram showing an example of the display screen of the display 141 when related tags are presented to a user. FIG. 13 is a block diagram showing the basic hardware configuration of a computer 90.

[0010] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. In all drawings describing the embodiments, common components are designated by the same reference numerals, and repeated description will be omitted. Note that the following embodiments do not unduly limit the content of the present disclosure described in the claims. Furthermore, not all components shown in the embodiments are necessarily essential components of the present disclosure. Furthermore, each drawing is a schematic diagram and is not necessarily a precise illustration.

[0011] [1. Overview] A system according to this embodiment analyzes documents previously created for a specific task and extracts tags. When the user creates a document related to the task, the system presents the extracted tags to the user. Tags selected by the user are input into large language models (LLMs), and documents using words as the selected tags are output from the LLMs.

[0012] There are no particular limitations on the types and contents of documents that can be applied to the system according to this embodiment. Examples of document types include resumes, notifications, contracts, minutes, applications, invoices, certificates, receipts, and reports. In this embodiment, a report will be used as an example of a document type. The report may be in any field, such as a medical report, a business report, or a research report.

[0013] The report describes, for example, the preconditions, the current state, specific aspects, points to note, measures being taken, etc. If the report is a medical report, for example, a nursing record, the preconditions, the current state, specific aspects, points to note, and measures being taken may correspond to, for example, information about the disease being suffered from, the patient's current condition, the state of the patient's life, events that require attention, and the treatments currently being performed, respectively.

[0014] 1 is a block diagram showing an example of the overall configuration of a system 1. The system 1 shown in Fig. 1 includes, for example, a terminal device 10, a server 20, and an NLP system 30. The terminal device 10, the server 20, and the NLP system 30 are communicatively connected via, for example, a network 80.

[0015] 1 shows an example in which the system 1 includes two terminal devices 10, but the number of terminal devices 10 included in the system 1 may be less than three or may be three or more. In FIG. 1, an example in which the system 1 includes one NLP system 30 is shown, but the number of NLP systems 30 included in the system 1 may be two or more.

[0016] 1 shows an example in which the NLP system 30 is independent from the server 20, the server 20 may include the functions of the NLP system 30. In other words, the server 20 may store the LLM and the NLP model.

[0017] In this embodiment, a collection of multiple devices may be one server 20. The way in which multiple functions required to realize the server 20 according to this embodiment are allocated to one or more pieces of hardware can be determined appropriately depending on the processing capacity of each piece of hardware and / or the specifications required for the server 20.

[0018] The terminal device 10 shown in FIG. 1 is, for example, an information processing device operated by a user who uses a document creation service. The document creation service is provided by a server 20 and an NLP system 30. If the document is, for example, a medical report, the users include nurses, doctors, medical staff, etc. The terminal device 10 is realized by, for example, a mobile terminal such as a smartphone or a tablet. In this embodiment, the terminal device 10 is assumed to be a tablet. The terminal device 10 may also be realized by, for example, a desktop personal computer (PC), a laptop PC, etc.

[0019] The terminal device 10 includes a communication interface (IF) 12, an input device 13, an output device 14, a memory 15, a storage 16, and a processor 19. The input device 13 is a device for receiving input operations from a user (for example, a touch panel, a touch pad, a pointing device such as a mouse, a keyboard, etc.). The output device 14 is a device for presenting information to the user (a display, a speaker, etc.). In this embodiment, the terminal device 10 is assumed to include a touch panel in which the input device 13 and the output device 14 are integrated.

[0020] The server 20 is, for example, an information processing device that provides a document creation service using the NLP system 30. The server 20 is, for example, an information processing device realized by a computer connected to a network 80. As shown in Fig. 1 , the server 20 includes a communication IF 22, an input / output IF 23, a memory 25, a storage 26, and a processor 29. The input / output IF 23 functions as an input device for receiving input operations from a user and as an interface for an output device for outputting information to the user.

[0021] The NLP system 30 is, for example, a system in which a natural language model is built. The natural language model is an artificial intelligence model used in the field of natural language processing (NLP), which learns from text data (web pages, books, articles, etc.) to understand patterns of human language and effectively perform natural language generation (NLG) tasks.

[0022] The NLP system 30 includes, for example, an LLM (Large Scale Language Model) and an NLP model. The NLP system 30 may be divided into, for example, a system including an LLM and a system including an NLP model. An LLM is a natural language model constructed by learning large amounts of text data, and is used for many NLG tasks, such as generating responses to specific questions, automatically generating sentences, summarizing text, translating, and sentiment analysis. Furthermore, an LLM can be used for a variety of purposes, such as education, entertainment, customer service, and product development. Examples of LLMs include the following:

[0023] OpenAI ChatGPT (registered trademark) Google Bart (registered trademark) Stable Diffusion Meta Llama2 Midjourney (registered trademark)

[0024] An NLP model is a natural language model that is not an LLM, such as a feedforward neural network, a custom model designed for a particular task or domain, etc.

[0025] The NLP system 30 inputs the text data transmitted from the server 20 into the NLP model or LLM, and causes the NLP model or LLM to output a response based on the input text data. The NLP system 30 transmits the response output from the NLP model or LLM to the server 20.

[0026] Each information processing device is configured by a computer 90 (see FIG. 13) equipped with an arithmetic unit and a storage device. The basic hardware configuration of the computer 90 and the basic functional configuration of the computer 90 realized by the basic hardware configuration will be described later. For each of the terminal device 10, the server 20, and the NLP system 30, descriptions that overlap with the basic hardware configuration and basic functional configuration of the computer 90 will be omitted.

[0027] 2.1 Configuration of Terminal Device Fig. 2 is a block diagram showing an example configuration of the terminal device 10 shown in Fig. 1. As shown in Fig. 2, the terminal device 10 includes a communication unit 120, an input device 13, an output device 14, a storage unit 180, and a control unit 190. The blocks included in the terminal device 10 are electrically connected by, for example, a bus. The terminal device 10 may also include an audio processing unit, a microphone, a speaker, a camera, a position information sensor, or a combination of at least two of these.

[0028] The communication unit 120 performs processes such as modulation and demodulation for the terminal device 10 to communicate with other devices. The communication unit 120 performs transmission processing on signals generated by the control unit 190 and transmits the signals to an external device (for example, the server 20). The communication unit 120 performs reception processing on signals received from an external device and outputs the signals to the control unit 190.

[0029] The input device 13 is a device through which a user operating the terminal device 10 inputs instructions or information. The input device 13 is realized, for example, by a touch-sensitive device 131 through which instructions are input by touching the operation surface. If the terminal device 10 is a PC or the like, the input device 13 may be realized by a reader, keyboard, mouse, or the like. The input device 13 converts instructions input by the user into electrical signals and outputs them to the control unit 190. Note that the input device 13 may include, for example, a receiving port that receives electrical signals input from an external input device.

[0030] The output device 14 is a device for presenting information to a user operating the terminal device 10. The output device 14 is realized, for example, by a display 141. The display 141 displays various types of information under the control of the control unit 190. The display 141 is realized, for example, by an LCD (Liquid Crystal Display) or an organic EL (Electro-Luminescence) display.

[0031] The storage unit 180 is realized by, for example, the memory 15 and the storage 16, and stores data and programs used by the terminal device 10. The storage unit 180 stores, for example, user information 181. The user information 181 includes, for example, various information related to the user who uses the terminal device 10. The various information related to the user includes, for example, the user's name, age, address, date of birth, and contact information.

[0032] The control unit 190 is realized by the processor 19 reading a program stored in the storage unit 180 and executing instructions included in the program. The control unit 190 controls the operation of the terminal device 10. The control unit 190 performs the functions of an operation reception unit 191, a transmission / reception unit 192, and a presentation control unit 193 by operating in accordance with the program.

[0033] The operation reception unit 191 performs processing for receiving instructions or information input from the input device 13. Specifically, the operation reception unit 191 receives instructions or information input from the touch-sensitive device 131. The transmission / reception unit 192 performs processing for the terminal device 10 to transmit and receive data to and from an external device such as the server 20 in accordance with a communication protocol. Specifically, the transmission / reception unit 192 transmits instructions or information input by the user to the server 20. The transmission / reception unit 192 receives information transmitted from the server 20. The presentation control unit 193 controls the output device 14 to present various pieces of information transmitted from the server 20 to the user. For example, the presentation control unit 193 displays multiple tags transmitted from the server 20 on the display 141. Details of the tags will be described later.

[0034] <2.2 Functional Configuration of Server> Fig. 3 is a block diagram showing an example of the functional configuration of the server 20 shown in Fig. 1. As shown in Fig. 3, the server 20 performs the functions of a communication unit 201, a storage unit 202, and a control unit 203.

[0035] The communication unit 201 performs processing for communication between the server 20 and external devices. The storage unit 202 includes, for example, a user information table 2021, a document table 2022, a tag table 2023, and a prompt table 2024.

[0036] The tables stored in the storage unit 202 are not limited to these. For example, the document table 2022 may not be stored in the storage unit 202. The services associated with documents may vary depending on their contents. Therefore, documents may be stored for each server that manages the service. Furthermore, documents may be stored in a server related to a service for sharing information among multiple users.

[0037] The user information table 2021 is a table that stores information about users. The document table 2022 is a table that stores information about multiple documents created in a specified task (hereinafter referred to as document information). The document information includes information about multiple documents created in the past in a specified task (hereinafter referred to as past document information). The tag table 2023 is a table that stores multiple tags. The prompt table 2024 is a table that stores information about instruction statements that serve as input statements to the NLP system 30. Details of these tables will be described later.

[0038] The control unit 203 is realized by the processor 29 reading a program stored in the storage unit 202 and executing instructions included in the program. The program includes an application such as a web browser application. The program includes a programming language such as JavaScript (registered trademark) that is executed on the web browser application stored in the terminal device 10. The control unit 203 operates in accordance with the program to fulfill the functions of a reception control module 2031, a transmission control module 2032, a service processing module 2033, and a presentation control module 2034.

[0039] The reception control module 2031 controls the process of the server 20 receiving signals from external devices in accordance with a communication protocol. The transmission control module 2032 controls the process of the server 20 transmitting signals to external devices in accordance with a communication protocol.

[0040] The service processing module 2033 analyzes a plurality of documents created in the past for a predetermined task, and controls the process of extracting and storing a plurality of tags for each task. Specific examples of tasks will be described later.

[0041] Specifically, the service processing module 2033 transmits the past document information read from the document table 2022 to the NLP system 30 via the transmission control module 2032. The document table 2022 stores, for example, past document information for each of a plurality of types of tasks.

[0042] At this time, the service processing module 2033 may transmit a predetermined instruction statement along with the past document information depending on the performance and characteristics of the NLP model, the content and amount of information of the past document information, etc. The predetermined instruction statement is a prompt that instructs the NLP model included in the NLP system 30 to extract multiple tags. In this case, for example, a second prompt table (not shown) stored in the storage unit 202 may store default predetermined instruction statements with different content for multiple types of tasks. The service processing module 2033 may, for example, read from the second prompt table a predetermined instruction statement associated with the type of task to which the past document information is linked, and transmit the predetermined instruction statement along with the past document information to the NLP system 30.

[0043] The NLP system 30 receives past document information transmitted from the transmission control module 2032, inputs the information into the NLP model, and outputs multiple tags from the NLP model. The NLP model is a natural language model that is trained to extract specific words from multiple documents as tags when the documents are input. Specific examples of tags will be described later.

[0044] The NLP system 30 transmits a plurality of tags output from the NLP model to the server 20. The service processing module 2033 receives a plurality of tags transmitted from the NLP system 30 via the reception control module 2031 and stores them in the tag table 2023. The service processing module 2033 executes this series of processes for all of the past document information linked to each of the plurality of types of tasks.

[0045] Note that the service processing module 2033 does not need to use an NLP model to extract multiple tags. For example, the service processing module 2033 may perform rule-based natural language analysis such as lexical semantic analysis on multiple documents created in the past for a specific task. The service processing module 2033 may then extract multiple tags using such well-known natural language analysis. Also, for example, the service processing module 2033 may extract tags using an LLM instead of an NLP model.

[0046] The service processing module 2033 inputs a user-specified tag word and instruction into the LLM, and causes the LLM to output a document using the word (hereinafter, an automatically generated document). The instruction is a prompt that instructs the LLM to generate an automatically generated document. Details of user tag specification will be described later. The LLM may be tuned so that when a specific word is input as a tag, the LLM generates a document using the word.

[0047] Specifically, the service processing module 2033 transmits tags specified by the user and instruction sentences read and edited from the prompt table 2024 to the NLP system 30 via the transmission control module 2032. The tag table 2023 stores, for example, multiple tags extracted by an NLP model for each of multiple types of tasks. The prompt table 2024 stores, for example, partially defaulted instruction sentences with different contents for each of multiple types of tasks. In other words, multiple tags and instruction sentences are associated with each of multiple types of tasks. Blank portions in the instruction sentences that are not defaulted are completed by, for example, specifying tags by the user. Details of this completion will be described later.

[0048] The NLP system 30 receives the tag and instruction sent from the transmission control module 2032, inputs them to the LLM, and causes the LLM to output an automatically created document. In other words, the service processing module 2033 causes the NLP system 30 to output an automatically created document.

[0049] It is not essential to input an instruction to the LLM using the prompt table 2024. For example, an instruction may be input to the input device 13, and the service processing module 2033 or the transmission control module 2032 may transmit the instruction to the NLP system 30.

[0050] The presentation control module 2034 controls the process of presenting information to the user. For example, the presentation control module 2034 reads out multiple tags associated with a task selected by the user from the tag table 2023 and controls the process of presenting the tags to the user. Details of the user's task selection will be described later. Also, for example, the presentation control module 2034 controls the process of presenting an automatically created document output from the LLM to the user.

[0051] 4 to 7 are diagrams showing the data structures of tables stored in the server 20. Note that Figures 4 to 7 are merely examples and do not exclude data that is not listed. Furthermore, even data listed in the same table may be stored in separate storage areas in the storage unit 202.

[0052] Fig. 4 is a diagram showing the data structure of the user information table 2021. The user information table 2021 shown in Fig. 4 is a table having columns of name, age, sex, date of birth, and contact information, with a user ID as a key.

[0053] The user ID is an item that stores an identifier for uniquely identifying a user. The name is an item that stores the name of the user. The age is an item that stores the age of the user. The gender is an item that stores the gender of the user. The date of birth is an item that stores the date of birth of the user. The contact information is an item that stores the contact information (e.g., telephone number, email address, etc.) of the terminal device 10 that the user has.

[0054] Fig. 5 is a diagram showing the data structure of the document table 2022. The document table 2022 shown in Fig. 5 is a table having columns of user ID, date and time, document data, task, and type ID, with a document ID as a key.

[0055] The document ID is an item that stores an identifier for uniquely identifying a document. The user ID is an item that stores an identifier for uniquely identifying the person who created the document. The "user ID" item in the document table 2022 corresponds to the "user ID" item in the user information table 2021, for example. Note that the "user ID" item may store the name of the document creator instead of the above-mentioned identifier. The date and time is an item that stores the date and time the document was created.

[0056] The document data is an item that stores data related to a document, i.e., document information. The document data is created by the creator by summarizing in text the underlying situation, current state, specific aspects, points to note, and measures taken. The document data includes data related to past documents in a specified task, i.e., past document information. The document data may include a specified numerical table, a photograph, a video, etc., or at least any combination of these, along with the text. The item "document data" may also store reference information (path) to a document data file located in another location.

[0057] The "task" is an item that stores the type of task associated with the creation of a document. For example, if the document is a medical report, the "task" item stores patient reception / registration, patient condition assessment / diagnosis, prescription, drug management, treatment / procedure, nursing care, or conference / meeting. For example, if the document is a business report, the "task" item stores business plan creation, legal procedure execution, fundraising, human resource acquisition / management, product / service development, marketing, sales, financial accounting, or risk management. For example, if the document is a research report, the "task" item stores research plan development, literature research, experiments, data collection, data analysis, result reporting, presentation / publication, or ethical considerations (protection of personal information, ensuring the rights of research participants, etc.).

[0058] The type ID is an item that stores an identifier for uniquely identifying the type of document. Document types include, for example, medical reports, business reports, and research reports. The item "type ID" stores an identifier that can identify the type of these reports. In this embodiment, information linked to all type IDs is stored together in one document table 2022, but, for example, a document table for each type ID may be stored in the storage unit 202.

[0059] Fig. 6 is a diagram showing the data structure of the tag table 2023. The tag table 2023 shown in Fig. 6 is a table having columns of tag data, task, and type ID, with the tag ID as a key.

[0060] The tag ID is an item that stores an identifier for uniquely identifying multiple tags associated with a specific task. The tag data is an item that stores multiple pieces of data related to tags associated with a specific task. The tag data is, for example, text data of words as tags.

[0061] For example, if the type of task linked to a medical report is nursing care, the "tag data" item stores evaluation, goal, intervention, intervention plan, etc. For example, if the type of task linked to a medical report is diagnosis, the "tag data" item stores diagnosis, condition, test results, treatment plan, prognosis, etc. For example, if the type of task linked to a business report is marketing, the "tag data" item stores segmentation, loyalty, awareness, targeting, positioning, customer experience, persona, insight, etc. For example, if the type of task linked to a research report is experiment, the "tag data" item stores experimental design, experimental conditions, objectives, methods, materials, measurements, results, analysis, conclusions, references, etc.

[0062] In the example of FIG. 6 , the column for the item "Tag Data" is divided into a column for the item "Technical Term" and a column for the item "General Term." The item "Technical Term" stores information about technical terms in the field to which the task belongs as information about tags. The item "General Term" stores information about general terms that can be used in fields other than the field to which the task belongs as information about tags. Note that it is not essential that the column for the item "Tag Data" be divided into a column for the item "Technical Term" and a column for the item "General Term." For example, tags of technical terms and tags of general terms may be mixed and stored together in the column for the item "Tag Data."

[0063] The "task" field stores the type of task to which multiple tags are linked. The "task" field in the tag table 2023 corresponds, for example, to the "task" field in the document table 2022. The "type ID" field stores an identifier for uniquely identifying the type of document from which multiple tags are extracted. The "type ID" field in the tag table 2023 corresponds, for example, to the "type ID" field in the document table 2022.

[0064] Fig. 7 is a diagram showing the data structure of the prompt table 2024. The prompt table 2024 shown in Fig. 7 is a table having columns of prompt data, document name, task, and type ID, with a prompt ID as a key.

[0065] The prompt ID is an item that stores an identifier for uniquely identifying an instruction. The prompt data is an item that stores data related to an instruction. Data related to an instruction is, for example, text data of an instruction that is partially blank. For example, the text data of an instruction is as follows: "Please create △△△ (blank part: enter the document name) that includes XXX (blank part: enter the word as a tag)."

[0066] The "prompt data" item may store, for example, reference information (path) for a prompt data file stored in another storage area. For example, the text data of the instruction statement stored in the "prompt data" item may include a blank section for describing the summary of the manager / executor of the task associated with the creation of the automatically created document. The manager / executor of the task may be, for example, a hospital if the automatically created document is a medical report, a company if the automatically created document is a business report, or a research institute if the automatically created document is a research report. The summary of the manager / executor of the task may include, for example, the industry, business content, organizational structure, history, number of employees, capital, sales, etc. In this case, the prompt table 2024 may store, for example, information regarding the summary of the manager / executor of the task.

[0067] The document name is an item that stores the name of the document that the instruction statement instructs to create. For example, if the task is nursing care, the "document name" item stores the name of the nursing record, nursing plan, care report, instruction record, etc. The task is an item that stores the type of task that the instruction statement is associated with. The "task" item in the prompt table 2024 corresponds, for example, to the "task" item in the document table 2022 and the tag table 2023. The type ID is an item that stores an identifier for uniquely identifying the type of automatically created document. The "type ID" item in the prompt table 2024 corresponds, for example, to the "type ID" item in the document table 2022 and the tag table 2023.

[0068] [4 Operation] The operation of the server 20 when presenting an automatically created document to a user will be described. FIG. 8 is a flowchart showing an example of the operation of the server 20 when presenting an automatically created document to a user. In the description of FIG. 8, the server 20 provides a service for creating an automatically created document as SaaS (Software as a Service). In this case, the specified task is, for example, nursing care, and the target document is, for example, a nursing record. Information regarding past nursing records (past document information) is stored, for example, in the document table 2022.

[0069] First, the service processing module 2033 stores information about the previous automatically created document edited by the user, along with the corresponding task type, in the document table 2022. The previous automatically created document is the automatically created document output from the LLM and presented to the user by the previous execution of each process shown in Figure 8. In other words, the service processing module 2033 adds the information about the previous automatically created document to the past document information stored in the document table 2022.

[0070] In step S11, the server 20 analyzes multiple documents created in the past for a specific task, extracts multiple tags for each task, and stores them (storing step). Specifically, for example, the service processing module 2033 reads past document information to which information about the immediately previous automatically created document has been added from the document table 2022. The service processing module 2033 transmits the read past document information to the transmission control module 2032. The transmission control module 2032 receives the past document information transmitted from the service processing module 2033 and transmits it to the NLP system 30.

[0071] The NLP system 30 receives the past document information transmitted from the transmission control module 2032, inputs the information into the NLP model, and causes the NLP model to output a plurality of tags. The NLP system 30 transmits the plurality of tags output from the NLP model to the server 20.

[0072] When the NLP system 30 outputs multiple tags from the NLP model, that is, when the server 20 extracts multiple tags, the server 20 may extract the multiple tags at a predetermined timing, for example.

[0073] 8 , the server 20 extracts multiple tags each time a new document is created in a specified task. Specifically, each time the previous automatically created document (a new document) linked to nursing care (a specified task) is created and information about the automatically created document is added to the tag table 2023, the service processing module 2033 extracts multiple tags using the NLP system 30. This allows the past document information stored in the tag table 2023 to be updated in a timely manner, thereby fully meeting user needs in tag specification when creating an automatically created document.

[0074] Note that the predetermined timing does not have to be every time a new document is created in a predetermined task. For example, the server 20 may extract multiple tags at a predetermined cycle. Specifically, the service processing module 2033 may read past document information from the document table 2022 at a predetermined cycle and transmit it to the NLP system 30 to extract multiple tags. The predetermined cycle can be set arbitrarily. This allows multiple tags to be extracted using updated past document information, thereby fully meeting user needs in tag specification when creating automatically created documents.

[0075] The reception control module 2031 receives the multiple tags transmitted from the NLP system 30 and transmits them to the service processing module 2033. The service processing module 2033 receives the multiple tags transmitted from the reception control module 2031, associates the past document information read by the service processing module 2033 with the type of task to which it is linked, and stores the tags in the tag table 2023.

[0076] When storing multiple tags in the tag table 2023, the server 20 may, for example, analyze the extracted multiple tags and store them in a manner according to the analysis results. For example, the server 20 may analyze the extracted multiple tags and store the multiple tags in a hierarchy based on the analysis results. This improves the user's convenience in specifying tags when creating an automatically created document.

[0077] Specifically, for example, the service processing module 2033 may analyze multiple tags received from the reception control module 2031 and classify the tags into specialized terms in the medical field (particularly nursing) and general terms that can be used in other fields. When classifying the tags, the service processing module 2033 may use, for example, a predetermined dictionary tool. Using a dictionary tool improves the accuracy of the classification, thereby further improving user convenience in specifying tags when creating automatically created documents.

[0078] The type of dictionary tool is not particularly limited, and known dictionary tools such as WordNet, GloVe (Global Vectors for Word Representation), NER (Stanford Named Entity Recognizer), and Pattern may be used. The predetermined dictionary tool may be updated as needed. Furthermore, the predetermined dictionary tool may learn new information in response to user operations.

[0079] The service processing module 2033 may store the analyzed tags in a hierarchical structure of technical terms and general terms in the tag table 2023. In this case, the tag table 2023 may have a column for the item "tag data" divided into a column for the item "technical term" and a column for the item "general term," as shown in FIG.

[0080] The hierarchical classification is not limited to the classification of technical terms and general terms, and may be any classification that allows the user to easily specify tags. Furthermore, the service processing module 2033 may store the analyzed tags in the tag table 2023 in a classification other than the hierarchical classification (e.g., grouping tags by relevance).

[0081] In step S12, the server 20 accepts the selection of a task linked to the creation of a document (selection acceptance step). In this embodiment, the server 20 accepts the selection of nursing care linked to the creation of a nursing record. Specifically, for example, the operation acceptance unit 191 accepts the selection of a task through a selection operation by the user on the input device 13. The operation acceptance unit 191 transmits information regarding the type of task input to the input device 13 to the transmission / reception unit 192. The transmission / reception unit 192 receives the information regarding the type of task transmitted from the operation acceptance unit 191 and transmits it to the server 20. The reception control module 2031 receives the information regarding the type of task transmitted from the terminal device 10. As a result, the server 20 accepts the user's selection of a task linked to the creation of a document.

[0082] 8, the server 20 executes the process of step S11 and the process of step S12 consecutively, but this is not limiting. For example, the server 20 may execute the process of step S11 separately from the processes from step S12 onward.

[0083] In step S13, the server 20 presents multiple tags associated with the selected task (multiple presentation step). Specifically, for example, the reception control module 2031 transmits information regarding the task type received from the terminal device 10 to the service processing module 2033. Based on the information regarding the task type received from the reception control module 2031, the service processing module 2033 searches the tag table 2023 for tasks of the same type as the task selected by the user, and reads out multiple tags associated with the tasks of the same type. The service processing module 2033 transmits the read out multiple tags to the presentation control module 2034.

[0084] The presentation control module 2034 receives the multiple tags transmitted from the service processing module 2033, and transmits the multiple tags and a first switching command to switch the display screen of the display 141 to the document creation screen 142 shown in Fig. 9 to the terminal device 10. The transmission / reception unit 192 receives the multiple tags and the first switching command transmitted from the presentation control module 2034 and transmits them to the presentation control unit 193. Based on the multiple tags and the first switching command received from the transmission / reception unit 192, the presentation control unit 193 switches the display screen of the display 141 to the document creation screen 142 and displays the multiple tags on the document creation screen 142.

[0085] There is no particular rule between the timing of switching to document creation screen 142 under the control of presentation control unit 193 and the timing of displaying multiple tags on document creation screen 142. However, it is preferable that the timing be such that multiple tags are displayed on document creation screen 142 at least at the time of switching to document creation screen 142.

[0086] In this embodiment, the presentation control module 2034 presents multiple tags associated with the selected task in a presentation format according to the analysis results on the document creation screen 142. Details of the document creation screen 142 will be described below with reference to Fig. 9. Fig. 9 is a schematic diagram showing an example of the display screen of the display 141 when tags are presented to the user.

[0087] 9 includes, for example, a task button 1411, a task selection box 1412, a document selection box 1413, a tag specification box 1414, and a start button 1415. In the initial state of the document creation screen 142, for example, nothing is displayed in the task selection box 1412, the document selection box 1413, and the tag specification box 1414.

[0088] The task button 1411 is a button for displaying multiple types of tasks in the task selection box 1412. When the task button 1411 is pressed, the operation reception unit 191 transmits reception information indicating that the press has been received to the server 20. The presentation control module 2034, which has received the reception information, reads out multiple types of tasks from the document table 2022 or the tag table 2023 and transmits them to the terminal device 10. The presentation control unit 193, which has received the multiple types of tasks, displays the multiple types of tasks in the task selection box 1412.

[0089] When a specific type of task is selected from the multiple types of tasks displayed in the task selection box 1412, the operation receiving unit 191 transmits task information related to the specific type of task whose selection has been received to the server 20. In the example of Fig. 9, nursing care is selected as the specific type of task.

[0090] The presentation control module 2034, which has received the task information, reads the document name of the document linked to the specific type of task from the prompt table 2024 and transmits the document name to the terminal device 10. The presentation control module 2034 also reads multiple tags associated with the specific type of task from the tag table 2023 and transmits the tags to the terminal device 10.

[0091] The presentation control unit 193, which has received the document name, displays the document name in a document selection box 1413. Furthermore, the presentation control unit 193, which has received multiple tags, displays the multiple tags in a tag specification box 1414. In the example of FIG. 9 , multiple document names linked to nursing care are displayed in the document selection box 1413, and multiple tags associated with nursing care are displayed in the tag specification box 1414.

[0092] In the example of FIG. 9 , the presentation control unit 193 provides a column for technical terms and a column for general terms in the tag specification box 1414, and displays the corresponding tags in each column. This allows the user to easily select tags. Note that the display manner for the hierarchical division of technical terms and general terms is not limited to the example of FIG. 9 . For example, technical terms may be displayed in a large size, and general terms may be displayed in a small size. Furthermore, for example, only technical terms may be displayed in bold, or technical terms may be displayed in red, and general terms may be displayed in black. Alternatively, two or more of these may be displayed in combination.

[0093] 9 , the presentation control module 2034 controls the presentation control unit 193 to display the document creation screen 142, on which multiple tags are displayed, on the display screen of the display 141, whereby the server 20 presents multiple tags associated with the task selected by the user. Note that the screen example shown in FIG. 9 is merely an example, and the document creation screen 142 may employ various screen configurations and UI elements.

[0094] In step S14, the server 20 accepts a designation for the presented tag (designation acceptance step). In this embodiment, the server 20 accepts a designation of a specific tag from among multiple tags associated with nursing care displayed on the document creation screen 142. The number of designated specific tags may be, for example, one or more. Specifically, for example, the operation acceptance unit 191 accepts a designation of a specific tag displayed on the document creation screen 142 through a designation operation by the user on the input device 13. The operation acceptance unit 191 transmits information about the specific tag whose designation has been accepted to the transmission / reception unit 192. The transmission / reception unit 192 receives the information about the specific tag transmitted from the operation acceptance unit 191 and transmits it to the server 20. The reception control module 2031 receives the information about the specific tag transmitted from the terminal device 10. As a result, the server 20 accepts a user's designation for the presented specific tag.

[0095] In the example of Figure 9, nursing records are selected from the document names displayed in the document selection box 1413, and nursing, treatment, and patient are specified from the multiple tags displayed in the tag specification box 1414.

[0096] In step S15, the server 20 inputs the specified tag word and instruction sentence into the LLM and outputs an automatically created document from the LLM (output step). Specifically, for example, the reception control module 2031 transmits information about the specific tag received from the terminal device 10 to the service processing module 2033. The service processing module 2033 reads data about the specific instruction sentence from the prompt table 2024 based on the information about the specific tag received from the reception control module 2031.

[0097] In more detail, for example, the service processing module 2033 searches the prompt table 2024 for a task of the same type as the task type associated with a specific tag. Then, the service processing module 2033 reads data on a specific instruction associated with the task of the same type from the prompt table 2024. In other words, the information on a specific tag includes information on the type of task associated with the specific tag.

[0098] The service processing module 2033 edits the data related to the read specific instruction sentence to complete the instruction sentence. In detail, for example, the service processing module 2033 completes the instruction sentence by filling in blank spaces in the read text data with specific tags and adjusting the expressions so that the sentence does not sound out of place.

[0099] The service processing module 2033 transmits to the transmission control module 2032 the specific tag received from the reception control module 2031 and the specific instruction sentence completed by editing the text data read from the prompt table 2024. The transmission control module 2032 receives the specific tag and the specific instruction sentence transmitted from the service processing module 2033 and transmits them to the NLP system 30.

[0100] The NLP system 30 receives the specific tag and the specific instruction sent from the transmission control module 2032, inputs them to the LLM, and causes the LLM to output an automatically created document. As a result, the server 20 inputs the word as the specific tag and the specific instruction to the LLM, and causes the LLM to output an automatically created document.

[0101] In the example of Figure 9, when a specific document name is selected from the document names displayed in the document selection box 1413, a specific tag is specified from the multiple tags displayed in the tag specification box 1414, and the start button 1415 is pressed, an automatically created document is output.

[0102] The start button 1415 is a button for starting the creation of an automatically created document, that is, for causing the LLM to output the automatically created document. When the start button 1415 is pressed, the operation reception unit 191 receives the selection of a specific document name, the designation of a specific tag, and the press, and transmits information related to the specific document name and the specific tag to the server 20. The service processing module 2033, which has received the information related to the specific document name and the specific tag, transmits the specific tag read from the tag table 2023 and the instruction (instructing the creation of a specific document) read from the prompt table 2024 and edited, to the NLP system 30. The service processing module 2033 then causes the LLM to output the automatically created document.

[0103] In step S16, the server 20 presents the output automatically generated document (presenting step). Specifically, for example, the NLP system 30 transmits the automatically generated document output by the LLM to the server 20. The reception control module 2031 receives the automatically generated document transmitted from the NLP system 30, associates information about the automatically generated document with a task, and stores it in the document table 2022. The reception control module 2031 also transmits the received automatically generated document to the presentation control module 2034.

[0104] The presentation control module 2034 receives the automatically created document transmitted from the service processing module 2033, and transmits the automatically created document and a second switching command to switch the display screen of the display 141 to the document confirmation screen 143 shown in Fig. 10 to the terminal device 10. The transmission / reception unit 192 receives the automatically created document and the second switching command transmitted from the presentation control module 2034, and transmits them to the presentation control unit 193. Based on the automatically created document and the second switching command received from the transmission / reception unit 192, the presentation control unit 193 switches the display screen of the display 141 to the document confirmation screen 143, and displays the automatically created document on the document confirmation screen 143.

[0105] There is no particular rule between the timing of switching to document confirmation screen 143 under the control of presentation control unit 193 and the timing of displaying the automatically created document on document confirmation screen 143. However, it is preferable that the automatically created document is displayed on document confirmation screen 143 at least at the time of switching to document confirmation screen 143.

[0106] Details of the document confirmation screen 143 will be described below with reference to Fig. 10. Fig. 10 is a schematic diagram showing an example of the display screen of the display 141 when an automatically created document is presented to the user.

[0107] On the document confirmation screen 143 shown in Fig. 10, for example, an automatically created document is displayed in area 1431. In the example of Fig. 10, as a result of a user operation on the document creation screen 142 shown in Fig. 9, a nursing record format is displayed in area 1431. Note that the format shown in Fig. 10 is merely an example, and the LLM can output automatically created documents in formats of various types and contents.

[0108] The document confirmation screen 143 has, for example, an edit start button 1433 and a print button 1434 in an area 1432. The edit start button 1433 is a button for editing an automatically created document. When the edit start button 1433 is pressed, the operation acceptance unit 191 accepts the press, and the presentation control unit 193 makes the automatically created document displayed in the area 1431 editable. This allows the user to edit the automatically created document displayed in the area 1431 to the desired content, format, etc., thereby fully meeting user needs.

[0109] The presentation control unit 193 transmits the edited automatically created document to the server 20. The service processing module 2033 stores the received edited automatically created document in the document table 2022. As a result, the service processing module 2033 updates the automatically created document stored in the document table 2022.

[0110] The print button 1434 is a button for printing out the automatically created document displayed in the area 1431. When the print button 1434 is pressed, the operation reception unit 191 receives the press, and the presentation control unit 193 transmits the automatically created document to, for example, a printer (not shown). The printer is wirelessly connected to, for example, the terminal device 10, and prints out the automatically created document received from the presentation control unit 193.

[0111] 10 , the presentation control module 2034 controls the presentation control unit 193 to display the document confirmation screen 143, which displays the automatically created document, on the display screen of the display 141, thereby allowing the server 20 to present the automatically created document output by the LLM. Note that the screen example shown in FIG. 10 is merely an example, and the document confirmation screen 143 may employ various screen configurations and UI elements.

[0112] Hereinafter, other examples of the document creation screen 142 and the document confirmation screen 143 will be described with reference to Fig. 11. Fig. 11 is a schematic diagram showing an example of the display screen of the display 141 when an automatically created document is being created. In the example shown in Fig. 11, it is assumed that the type of task to which the automatically created document is linked has been selected in advance, and the type of the automatically created document (work report) has been determined.

[0113] 11(a), the automatically generated document is in a format for inputting both pre-work and post-work report contents. However, the format of the automatically generated document is not limited to this, and it may be a format for inputting only post-work report contents, or a format for inputting each of pre-work, during-work, and post-work report contents.

[0114] When the operation receiving unit 191 receives a tap from the user on an input-enabled area, the presentation control module 2034 controls the presentation control unit 193 to display a report form for inputting report content into that area on the display screen of the display 141. Specifically, for example, when the operation receiving unit 191 receives a tap on the first editing area 1416 located below "Before work," the presentation control module 2034 displays a report form for inputting report content into the first editing area 1416.

[0115] 11(b) is a schematic diagram showing a report form for inputting report content into a predetermined area. In the example shown in FIG. 11(b), the presentation control module 2034 controls the presentation control unit 193 to display a first editing area 14161 and a second editing area 14162 on the display screen of the display 141. The first editing area 14161 is an area for attaching an image. The second editing area 14162 is an area for presenting multiple tags extracted by the server. The image attached to the first editing area 14161 may be stored in, for example, the storage unit 180 or the storage unit 202.

[0116] 11B , to attach an image to the first editing area 14161, the user taps the first editing area 14161. When the operation reception unit 191 receives the tap on the first editing area 14161, the presentation control module 2034 controls, for example, the presentation control unit 193 to display a selection window (not shown) for an image to be attached to the first editing area 14161. The user selects a specific image to be attached to the first editing area 14161 from a plurality of images displayed in a list in the selection window. When the operation reception unit 191 receives the user selection, the presentation control module 2034 reads out the specific image from the storage unit 180 or the storage unit 202 and controls the presentation control unit 193 to attach the specific image to the first editing area 14161.

[0117] 11B , to display multiple tags in the second editing area 14162, the user taps the second editing area 14162. When the operation receiving unit 191 receives the tap on the second editing area 14162, the presentation control module 2034 reads, for example, multiple tags associated with a pre-selected task type from the tag table 2023. Then, the presentation control module 2034 controls the presentation control unit 193 to display the read multiple tags in the second editing area 14162.

[0118] 11(c) is a schematic diagram showing an example of input to the report form. In the example shown in FIG. 11(c), a specific image is attached to a first editing area 14161. Furthermore, a plurality of tags extracted by the server 20 are displayed in a second editing area 14162. The user can specify a specific tag from among the plurality of tags displayed in the second editing area 14162 by tapping the specific tag from the plurality of tags and the operation receiving unit 191 receiving the tap operation. When the user has finished inputting the report content, the user presses a button 14163.

[0119] 11(d) is a schematic diagram showing an example of the display of the created report. In the example shown in FIG. 11(d), an automatically created document with both pre-work and post-work report contents entered is displayed on the display screen 141.

[0120] [5. Summary] As described above, in this embodiment, the service processing module 2033 inputs past document information into the NLP model for each task, extracts multiple tags for each task, and stores them in the tag table 2023. The reception control module 2031 accepts the selection of a task associated with document creation, accepted by the operation acceptance unit 191. The presentation control module 2034 controls the presentation control unit 193 to present multiple tags associated with the task selected by the user. The reception control module 2031 accepts the user's designation of tags, accepted by the operation acceptance unit 191. The service processing module 2033 inputs the specified tag words and instruction sentences into the LLM to create an automatically created document. The presentation control module 2034 controls the presentation control unit 193 to present the automatically created document.

[0121] This allows for the creation of automated documents for each task using a single NLP model and a single LLM, even when there are multiple types of tasks. Therefore, when there are multiple types of tasks, the burden of system construction can be reduced and users can be assisted in creating documents.

[0122] Furthermore, since information about automatically created documents is also past document information, the NLP model can be used to timely update multiple tags stored in the tag table 2023. This makes it possible to fully meet user needs in tag specification when creating a new automatically created document.

[0123] [6. Modifications] <6.1 First Modification> In the present embodiment, an example has been described in which the server 20 analyzes and hierarchically classifies the extracted tags. Here, in addition to classifying the tags into levels, the server 20 may, for example, classify the tags into groups.

[0124] Specifically, for example, the server 20 may analyze the extracted tags and store the tags in groups based on the analysis results. More specifically, the service processing module 2033 may analyze the tags received from the reception control module 2031, extract two or more tags with high relevance, and group them. "High relevance" refers to two or more tags having high relevance to each other in terms of their semantic content as words.

[0125] The service processing module 2033 may store the analyzed tags in a grouped state in addition to the hierarchical state of being divided into technical terms and general terms in the tag table 2023. In this case, the tag table 2023 may store the analyzed tags in a grouped state in addition to the column of the item "tag data" divided into the column of the item "technical terms" and the column of the item "general terms" as shown in FIG. 6, or may store the analyzed tags in a column of two or more groups, such as the column of the items "group 1, group 2, ...".

[0126] Furthermore, for example, the server 20 may display tags related to the tag specified by the user. More specifically, the operation accepting unit 191 transmits information about the specific tag specified by the user to the server 20 via the transmitting / receiving unit 192. When the reception control module 2031 receives information about the specific tag transmitted from the terminal device 10, the reception control module 2031 may transmit the information about the specific tag to the presentation control module 2034 instead of the service processing module 2033.

[0127] The presentation control module 2034 may use the information about the specific tag received from the reception control module 2031 to search the tag table 2023 for a group to which the specific tag belongs, and read out tags other than the specific tag (hereinafter referred to as related tags) from the group. The presentation control module 2034 transmits the read out related tags to the terminal device 10 via the transmission control module 2032. The presentation control unit 193 displays the related tags received via the transmission / reception unit 192 on the document creation screen 144.

[0128] The document creation screen 144 will be described in detail below with reference to Fig. 12. Fig. 12 is a schematic diagram showing an example of the display screen of the display 141 when related tags are presented to the user. The document creation screen 144 shown in Fig. 12 has the same screen configuration and content as the document creation screen 142 shown in Fig. 9, except that it has a related tag specification box 1416.

[0129] On document creation screen 144, when a specific tag is designated from among the multiple tags displayed in tag designation box 1414, related tags related to the specific tag are displayed in related tag designation box 1416. In other words, on document creation screen 144 in its initial state, nothing is displayed in related tag designation box 1416.

[0130] 12, the presentation control module 2034 controls the presentation control unit 193 to link a specific tag designated by the user with related tags associated with the specific tag and display them on the display 141. This allows the user to designate a specific tag and related tags at the same time on one screen, improving user convenience.

[0131] Note that the presentation control module 2034 may, for example, display a specific tag and related tags in close proximity and in a linked manner within a single box (not shown) on the document creation screen 144. Also, for example, displaying a specific tag and related tags in a linked manner is not essential, and the presentation control module 2034 may display related tags on a screen separate from the screen on which the specific tag is displayed. Furthermore, the operation reception unit 191 of the server 20 may receive a designation of at least one of a specific tag and related tags on the document creation screen 144 by receiving the designation through a flick operation. This improves the efficiency of tag designation, thereby improving user convenience.

[0132] 6.2 Second Modification The term "task" in this specification is a concept that includes not only an activity, work, or job aimed at a specific purpose / goal, such as nursing care, fundraising, data analysis, etc., but also documents created in the course of the activity, work, or job. Therefore, the server 20 may extract and store multiple tags for each type of document, for example.

[0133] Specifically, for example, the document table 2022 may store the document name for each document type in the item "task." The service processing module 2033 may, for example, input past document information for each document type into an NLP model and output multiple tags from the NLP model for each document type. The service processing module 2033 may, for example, store the multiple tags output from the NLP model in the tag table 2023 for each document type. In other words, the tag table 2023 may, for example, store the document name for each document type in the item "task."

[0134] 6.3 Third Modification In the present embodiment, the server 20 creates an automatically generated document by inputting a specified tag word and an instruction sentence into the LLM. However, the server 20 may use an image instead of the specified tag word when creating the automatically generated document. In this case, the image may be a still image or a video.

[0135] The server 20 may, for example, analyze multiple documents created in the past for a specific task and extract and store multiple images for each task. Specifically, for example, the service processing module 2023 may input past document information for each task into a first multimodal generation AI model and cause the first multimodal generation AI model to output multiple images for each task. The first multimodal generation AI model is a multimodal generation AI model provided in the NLP system 30 instead of an NLP model. The service processing module 2023 may, for example, store the multiple images output by the first multimodal generation AI model in an image table (not shown) for each task. The image table may, for example, be stored in the storage unit 202 instead of the tag table 2023, and may have the same table structure as the tag table 2023, except that the item "tag data" is replaced with the item "image data."

[0136] Similar to the example of this embodiment, the server 20 accepts the selection of a task linked to the creation of a document. The server 20 may, for example, present multiple images associated with the selected task. Specifically, for example, the service processing module 2033 may search the image table for tasks of the same type as the task selected by the user and read multiple images associated with the tasks of the same type. The presentation control module 2034 may, for example, transmit the multiple images received from the service processing module 2033 to the terminal device 10 and display the multiple images on the display screen of the display 141.

[0137] The server 20 may, for example, accept a designation for the presented image. Specifically, for example, the operation acceptance unit 191 may accept a designation of a specific image (which may be one or more images) displayed on the display screen of the display 141 through a designation operation by the user on the input device 13. The transmission / reception unit 192 may, for example, transmit information related to the specific image received from the operation acceptance unit 191 to the server 20. For example, the server 20 may accept a user designation for the presented specific image by the reception control module 2031 receiving information related to the specific image transmitted from the terminal device 10.

[0138] For example, the server 20 may input a specified image and an instruction statement instructing the creation of a document including content related to the image into the multimodal generative AI model, and cause the multimodal generative AI model to output a document including content related to the image. A "document including content related to the image" is an example of an automatically created document, and may be a document that includes the image itself, or a document in which the content displayed in the image is converted into text and reflected. In other words, a "document including content related to the image" does not necessarily have to include the image itself.

[0139] Specifically, for example, the service processing module 2033 may read and edit data related to a specific instruction from the prompt table 2024 based on information related to the specific image received from the reception control module 2031, to complete the instruction. For example, the service processing module 2033 may transmit to the NLP system 30 the specific image received from the reception control module 2031 and the specific instruction completed by editing the text data read from the prompt table 2024.

[0140] For example, the NLP system 30 may receive a specific image and a specific instruction sent from the transmission control module 2032, input them to the second multimodal generation AI model, and cause the second multimodal generation AI model to output an automatically created document (a document including content related to the image). The second multimodal generation AI model is a multimodal generation AI model provided in the NLP system 30 instead of the LLM, and is an example of a multimodal generation AI model according to one aspect of the present disclosure.

[0141] In the third modified example, the server 20 may create an automatically generated document using tags in addition to images. In this case, the server 20 may, for example, input past document information into the first multimodal generation AI model to extract images and tags. Alternatively, for example, the server 20 may create an automatically generated document by inputting images, tags, and instruction text into the second multimodal generation AI model.

[0142] 13 is a block diagram showing the basic hardware configuration of a computer 90. The computer 90 includes at least a processor 91, a main storage device 92, an auxiliary storage device 93, and a communication IF (interface) 99. These components are electrically connected to one another via a bus.

[0143] The processor 91 is hardware for executing an instruction set written in a program. The processor 91 is composed of an arithmetic unit, registers, peripheral circuits, etc. The main memory device 92 is memory for temporarily storing programs and data processed by the programs, etc. For example, it is volatile memory such as DRAM (Dynamic Random Access Memory). The auxiliary memory device 93 is storage for saving data and programs. For example, it is flash memory, HDD (Hard Disc Drive), magneto-optical disk, CD-ROM, DVD-ROM, semiconductor memory, etc. The communication IF 99 is an interface for inputting and outputting signals for communicating with other computers via a network using wired or wireless communication standards.

[0144] The network is composed of the Internet, a LAN, various mobile communication systems constructed by wireless base stations, etc. For example, the network includes 3G, 4G, and 5G mobile communication systems, LTE (Long Term Evolution), and wireless networks (e.g., Wi-Fi (registered trademark)) that can connect to the Internet via a predetermined access point. In the case of a wireless connection, communication protocols include, for example, Z-Wave (registered trademark), ZigBee (registered trademark), and Bluetooth (registered trademark). In the case of a wired connection, the network also includes a network that is directly connected using a USB (Universal Serial Bus) cable, etc.

[0145] It should be noted that the computer 90 can be virtually realized by distributing all or part of each hardware configuration across multiple computers 90 and interconnecting them via a network. In this way, the computer 90 is a concept that includes not only a computer in which each piece of hardware is housed in a single housing or the like, but also a virtualized computer system.

[0146] [8. Basic Functional Configuration of Computer] The functional configuration of the computer realized by the basic hardware configuration of the computer 90 will be described. The computer performs the functions of a control unit, a storage unit, and a communication unit. Note that each function of the computer 90 can be performed by distributing all or part of the functions to multiple computers 90 connected to each other via a network. In this way, the concept of the computer 90 includes not only a single computer but also a virtualized computer system.

[0147] The storage unit is realized by the main storage unit 92 and the auxiliary storage unit 93. The storage unit stores data, various programs, and various databases. The control unit is realized by the processor 91 reading out various programs stored in the auxiliary storage unit 93, expanding them in the main storage unit 92, and executing processing in accordance with the programs. The control unit can perform various information processing functions depending on the type of program. This allows the computer to be realized as an information processing device that processes information.

[0148] The control unit can also cause the processor 91 to allocate a storage area corresponding to the storage unit in the main storage unit 92 or the auxiliary storage unit 93 in accordance with the programs. Furthermore, the control unit can cause the processor 91 to add, update, and delete data stored in the storage unit in accordance with various programs.

[0149] A database refers to a relational database, which manages data sets called tables, which are structured by rows and columns, by associating them with each other. In a database, a table is called a table, a column in a table is called a column, and a row in a table is called a record. In a relational database, relationships between tables can be set and associated.

[0150] Typically, each table constituting a database has a column set as a key for uniquely identifying a record, but setting a key to a column is not essential. The control unit can cause the processor 91 to add, update, or delete records in a specific table stored in the storage unit in accordance with various programs.

[0151] The communication unit is realized by the communication IF 99. The communication unit performs the function of communicating with other computers 90 via a network. The communication unit can receive information transmitted from other computers 90 and output the information to the control unit. The control unit can cause the processor 91 to execute information processing on the received information in accordance with various programs. Furthermore, the communication unit can transmit information output from the control unit to other computers 90.

[0152] Although several embodiments of the present disclosure have been described above, these embodiments can be embodied in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and modifications are intended to be included in the scope of the inventions and their equivalents as defined in the claims, as well as in the scope and spirit of the inventions.

[0153] [9. Supplementary Notes] The matters explained in the above embodiments are supplementary notes below.

[0154] <Supplementary Note 1> A program to be executed by a computer having a processor and a memory, the program causing the processor to execute the steps of analyzing multiple documents created in the past for a specified task, extracting and storing multiple tags for each task, accepting a selection of a task to which the document creation is linked, presenting multiple tags associated with the selected task, accepting specifications for the presented tags, inputting words as the specified tags and instructions to create a document using the words into a large-scale language model, and outputting a document using the words from the large-scale language model, and presenting the document using the output words.

[0155] <Supplementary Note 2> A program as described in Supplementary Note 1, in which, in the storing step, the extracted multiple tags are analyzed and stored in a manner according to the analysis results, and, in the multiple presenting step, the tags are presented in a presentation manner according to the analysis results.

[0156] <Supplementary Note 3> The program described in Supplementary Note 2 causes a processor to execute a step of analyzing the extracted tags, classifying the extracted tags into hierarchies, grouping the extracted tags, or a combination of the two, and registering the resulting tags.

[0157] <Supplementary Note 4> The program according to Supplementary Note 3, wherein in the registration step, hierarchical classification is performed based on a predetermined dictionary tool.

[0158] <Supplementary Note 5> The program according to any one of (Supplementary Note 1) to (Supplementary Note 4), wherein in the storing step, a plurality of tags are extracted at a predetermined timing.

[0159] <Supplementary Note 6> The program according to (Supplementary Note 5), wherein the plurality of tags are extracted every time a new document is created in a predetermined task.

[0160] <Supplementary Note 7> The program according to (Supplementary Note 5), which extracts a plurality of tags at a predetermined cycle.

[0161] <Supplementary Note 8> The program according to any one of (Supplementary Note 1) to (Supplementary Note 7), which causes a processor to execute a step of displaying tags related to a specified tag.

[0162] <Supplementary Note 9> The program according to Supplementary Note 8, wherein the displaying step displays the specified tag and a tag related to the specified tag in a linked manner.

[0163] <Supplementary Note 10> The program according to (Supplementary Note 9), which causes a processor to execute a step of accepting a designation by a flick operation for a displayed tag.

[0164] <Supplementary Note 11> A program to be executed by a computer having a processor and a memory, the program causing the processor to execute the following steps: analyzing multiple documents created in the past for specified tasks, extracting and storing multiple tags for each task; accepting a selection of a task linked to the creation of a document; presenting multiple tags associated with the selected task; accepting a designation for the presented tags and a specified image, or a designation for the image; inputting the designated tags and image, or the designated image, and an instruction to create a document into a multimodal generation AI model, and causing the multimodal generation AI model to output a document including at least content related to the image; and presenting the output document.

[0165] <Supplementary Note 12> An information processing device including a control unit and a storage unit, wherein the control unit executes all steps in the program according to any one of (Supplementary Note 1) to (Supplementary Note 11).

[0166] <Supplementary Note 13> A method executed by a computer having a processor and a memory, in which the processor executes all steps in the program according to any one of (Supplementary Note 1) to (Supplementary Note 11).

[0167] <Supplementary Note 14> A system comprising means for executing all steps in the program described in any one of (Supplementary Note 1) to (Supplementary Note 11).

[0168] REFERENCE SIGNS LIST 1... System 10... Terminal device 120... Communication unit 13... Input device 131... Touch-sensitive device 14... Output device 141... Display 15... Memory 16... Storage 19... Processor 20... Server 22... Communication IF 23... Input / output IF 25... Memory 26... Storage 29... Processor 30... NLP system

Claims

1. A program to be executed by a computer having a processor and a memory, the program causing the processor to execute the following steps: analyzing multiple documents created in the past for a specified task, and extracting and storing multiple tags for each task; accepting a selection of the task to which the creation of the document is linked; presenting multiple tags associated with the selected task; accepting specifications for the presented tags; inputting a word as the specified tag and an instruction statement instructing the creation of a document using the word into a large-scale language model, and outputting a document using the word from the large-scale language model; and presenting the output document using the word.

2. The program according to claim 1, wherein in the storing step, the extracted tags are analyzed and stored in a manner according to the analysis results, and in the presenting step, the tags are presented in a manner according to the analysis results.

3. The program according to claim 2, wherein in the storing step, the extracted tags are analyzed, and the tags are stored after being classified into hierarchies, grouped, or a combination of these two.

4. The program according to claim 3, wherein the hierarchical classification is performed based on a predetermined dictionary tool.

5. The program according to claim 1, wherein the step of storing extracts the plurality of tags at a predetermined timing.

6. The program according to claim 5, wherein the plurality of tags are extracted each time a new document is created in the predetermined task.

7. The program according to claim 5, wherein the plurality of tags are extracted at a predetermined interval.

8. The program of claim 1, further comprising causing said processor to execute the step of displaying said tags associated with said specified tag.

9. The program according to claim 8, wherein in said displaying step, the designated tag and the tag related to the designated tag are displayed in a linked manner.

10. The program according to claim 9, which causes the processor to execute a step of accepting a designation by a flick operation for the displayed tag.

11. A program to be executed by a computer having a processor and a memory, the program causing the processor to execute the following steps: analyzing multiple documents created in the past for specified tasks, and extracting and storing multiple tags for each task; accepting a selection of the task to which the creation of the document is linked; presenting multiple tags associated with the selected task; accepting a specification for the presented tags and a specified image, or a specification for the image; inputting the specified tag and image, or the specified image, and an instruction to create a document into a multimodal generative AI model, and causing the multimodal generative AI model to output a document including at least content related to the image; and presenting the output document.

12. An information processing device comprising a control unit and a storage unit, wherein the control unit executes all steps in the program according to any one of claims 1 to 11.

13. A method implemented on a computer having a processor and a memory, wherein the processor executes all the steps of the program of any one of claims 1 to 11.

14. A system comprising means for executing all steps of the program according to any one of claims 1 to 11.

Images