Information processing device, information processing method, program, and storage medium

Abstract

Even as externally generated AI becomes more sophisticated, this information processing device allows companies to maintain control over the hiring decision-making process through gate-based judgments based on their own specific criteria. [Solution] An information processing device 11 comprising a processor 19 and an auxiliary storage device 21, wherein the processor 19 functions as an acquisition means, an extraction means, a candidate generation means, a gate determination means, a translation / normalization means, and a presentation means by executing a program stored in the auxiliary storage device 21.

Description

[[Technical Field]]

[0001] The present embodiment relates to an information processing apparatus, an information processing method, a program, and a storage medium that take communication information (work chat, meeting transcription, voice input, etc.) regarding work performed on a collaboration platform and work logs as inputs, and generate information related to organizing work issues, governing external proposals (gate determination), translating and normalizing external expressions, creating trails of selection, approval, execution and results, as well as prerequisite information on responsibility demarcation and information related to consideration calculation. [[Background Art]]

[0002] In recent years, collaboration platforms (for example, platforms integrating work chat, task management, schedule management, online meetings, etc.) have become the center of business operations, and most decision-making has come to be buried in conversation logs. Additionally, generative AI (Artificial Intelligence), external generative AI agents, external services, proposals from agencies and trading companies, etc., are mixed, creating a situation where various external proposals flow into business processes.

[0003] However, external proposals tend to contain noise such as sales promotion expressions and external terminology, which may mislead internal decision-making. Furthermore, which proposal (which AI, which external service, which agency proposal) was adopted is ambiguously determined in the flow of conversation, leading to the problem that it is impossible to explain "why it was chosen" afterward.

[0004] Furthermore, in the event of an accident or public controversy, the demarcation of responsibilities for errors from external generative AI, system presentation, user approval, operational problems, etc., becomes unclear, which becomes a factor for legal, audit, and information system departments to halt the introduction of AI. In addition, even when results are achieved, value attribution is ambiguous, making billing, settlement, and agency incentive design difficult. [[Prior Art Literature]] [[Patent Literature]]

[0005] [[Patent Literature 1]] Japanese Patent Publication No. 2024-50065 [Overview of the project] [Problems that the invention aims to solve]

[0006] This embodiment aims to govern (gate) multiple candidates, including externally generated AI, external services, and agency proposals, according to the company's specific criteria (corporate philosophy, business plan, organizational structure, roles and responsibilities, terminology definitions, purchasing rules, prohibitions, approval routes, expression standards, etc.), and to present external expressions and terminology in a standard corporate format (translation and normalization). Furthermore, it aims to link the presentation → selection → approval → execution → results as evidence, enabling consistent handling of accountability, prerequisite information for responsibility divisions, and consideration calculation (settlement). [Means for solving the problem]

[0007] This embodiment provides an information processing device that provides a communication environment in which users can communicate with each other using user terminals, and outputs business support information as input to information acquired from the communication environment, comprising at least a processor and a storage device, wherein the processor executes a program stored in the storage device to acquire communication information and business logs related to business from the communication environment; an extraction means for analyzing conversations included in the communication information to extract business issues and conditions and constraints related to said business issues; a candidate generation means for generating a plurality of candidate actions to be taken next that solve the business issues, satisfy the conditions, and clear the constraints; and corporate parameters including judgment criteria information defined for each implementing company stored in the storage device. Referencing the above, an information processing device is disclosed which functions as follows: a gate determination means that performs one of the following processes for at least a portion of the plurality of action candidates: pass, reject, or prioritize; a translation and normalization means that converts the action candidates that have been passed or prioritized by the gate determination means into structured data (itemized and attributed data structure) that can be immediately processed in the next step in which the selected action candidates are performed, in accordance with the terminology definitions, expression standards and business formats based on the corporate parameters; and a presentation means that presents the action candidates converted by the translation and normalization means to the user terminal. [Effects of the Invention]

[0008] The information processing device of this embodiment can achieve the following effects.

[0009] (1) Even if externally generated AI becomes more sophisticated, companies can retain the initiative in hiring decisions through gate decisions based on their own specific criteria.

[0010] (2) It converts (translates and normalizes) sales expressions and noise from external proposals into the company's standard language, and functions as a protocol converter for culture and terminology.

[0011] (3) The presentation of external products can be limited to addressing business challenges and meeting functional requirements, making it easier to technically avoid suspicions of advertising or stealth marketing.

[0012] (4) By assigning a responsibility token at the time of presentation and documenting the selection, approval, execution, and results, the underlying information for accountability and the division of responsibility becomes easier to clarify.

[0013] (5) Contribution and value attribution can be determined at the time of achievement and used for billing, settlement, and agent incentive design. [Brief explanation of the drawing]

[0014] [Figure 1] This is a conceptual diagram showing an example of an information processing system including an information processing device. [Figure 2] This is a conceptual diagram showing an example of an information processing device. [Figure 3] This is a conceptual diagram showing an example configuration of the first user terminal and the second user terminal. [Figure 4] This flowchart shows specific examples of information processing methods performed by information processing devices. [Figure 5] This is a conceptual diagram showing an example of a functional component realized by an information processing device. [Figure 6] This flowchart shows an overview of the information processing method implemented by the information processing device. [Figure 7] This flowchart shows other specific examples of information processing methods performed by information processing devices. [Figure 8] This diagram shows an example of a "Perfect Order" data structure used in information processing devices. [Figure 9] This is a diagram showing an example of a list of "Candidates" (potential actions) used in an information processing system. [Figure 10] This is a diagram showing an example list of "responsibility tokens" used in information processing devices. DESCRIPTION OF EMBODIMENTS OF THE INVENTION

[0015] (Summary)

[0016] (A) Gate (Governance / Checkpoint) Candidates including a plurality of AIs, external services, and agent proposals are reviewed based on company-specific judgment criteria (Should), and only candidates that are permitted to be put into business operations can be presented. When external candidates are presented, they are derived from business issues and necessary conditions extracted from conversations, and can be limited to candidates that satisfy functional requirements (this has a different structure from that of advertisement / recommendation).

[0017] (B) Translate / Normalize (Translation / Normalization) Expressions introduced from outside (externally-derived expressions), such as inflammatory phrases, external terms, exaggerated expressions, and sales pitches, are converted into a standard format consistent with the in-company term definitions and expression norms (Should) and then presented. For example, an external expression "Great deal if you order now! Super-fast delivery!" is converted into an attribute / condition expression processible for business operations, such as "Short delivery option (cost increase) / delivery: ○ days / conditions: ○○".

[0018] (C) Liability / Value (Determination of preconditions for liability demarcation / Value / Settlement) At the time of presentation or generation, each option is attached with liability division information (liability token), so that precondition information for liability demarcation is determined at the time of selection and approval. The liability herein is not intended to mechanically determine the final legal liability, but intended to attach and determine the attributes (division information) serving as preconditions for liability demarcation at the time of presentation, so that such information can be verified as an audit trail. In addition, presentation, warning, approval, external transmission, execution, and results are associated as a state transition log, so that the log can be used for accountability in the event of an accident and for value attribution and consideration calculation when outcomes are achieved. Liability demarcation means clarifying where responsibility lies. Token means words, terms, and the like.

[0019] Figure 5 shows an example of the functional components realized by the hardware and software of the information processing device 11 shown in Figures 1 and 2. The information processing device 11 implements a conversation acquisition unit / log acquisition unit 90, a context analysis / requirements extraction unit 91, a corporate parameter management unit 92, a candidate generation unit 93, a functional requirements conformance determination unit 94, a gate determination unit 95, a translation / normalization unit 96, a three-choice presentation unit 97, an external inquiry support unit 98, an audit log generation unit 99, and a responsibility boundary / value calculation unit 100. The functional components shown in Figure 5 have the function of realizing the information processing method shown in Figure 4.

[0020] (1) Conversation acquisition unit / log acquisition unit (acquisition means, conversation acquisition / log acquisition means) The conversation acquisition unit / log acquisition unit 90 acquires conversation logs (to the necessary extent) and associated work logs (to the necessary extent) linked to specific chat groups, etc., on the collaborative platform.

[0021] (2) Context analysis / requirements extraction unit (extraction means, context analysis / requirements extraction means) The context analysis and requirements extraction unit 91 analyzes the conversation context to extract and summarize issues, requirements, constraints, desires, etc., and estimates the business state (Is) in which "consultation, arrangement, and comparative review with external parties" is necessary.

[0022] (3) Enterprise parameter management unit (should storage) The Corporate Parameter Management Unit 92 stores and references corporate philosophy, business plans, organizational charts, roles and responsibilities, terminology definitions, purchasing rules, prohibitions, approval routes, and expression standards for each implementing company. It also enables the development and updating of the company's decision-making framework (Should) through interviews with managers and other means.

[0023] (4) Candidate generation unit (Next candidate generation / candidate generation means) The candidate generation unit 93 generates multiple candidate actions to be taken next based on the state (Is) and requirements. If external consultation is necessary, it generates a candidate category or direction for the candidate. It may also incorporate candidates obtained from external AI generation, external services, agency proposals, etc.

[0024] (5) Functional requirements conformance determination unit (indivisibility guarantee / functional requirements conformity determination means) The functional requirements conformity determination unit 94 determines and derives functional requirements (e.g., short delivery time, service area, standards, audit compliance, NDA compliance, etc.) that conform to the business issues (Is) and requirements extracted from the conversation. External products, external services, and agency proposals are presented as candidates only if they meet the functional requirements, and the basis for selection (conformity to functional requirements) can be logged.

[0025] (6) Gate determination unit (governance / conformity review, gate determination means) The gate determination unit 95 determines the suitability of candidates (including externally generated AI / external services / agency proposals) against the company parameters (Should), and excludes unsuitable candidates, lowers their ranking, or imposes conditions (designation of approver, information masking, mandatory logging, restriction of external transmission, etc.).

[0026] (7) Translation / Normalization Unit (Translate / Normalize, Translation and Normalization Methods) The translation and normalization unit 96 performs a process that converts unstructured information into business-processable structured data (Perfect Order), rather than simply translating language. Specifically, for external candidates that have passed the gate judgment, the translation and normalization unit 96 converts external expressions (hype, external terminology, promotional phrases) into the company's standard format before presenting them. Examples of conversion include converting advertising slogans into conditional expressions, external vocabulary into internal vocabulary, and benefit expressions into premise / constraint / cost expressions. External vocabulary may include promotional (sales-promoting) expressions.

[0027] (8) Three-choice presentation unit (UI (User Interface) / UX (User Experience), three-choice presentation means, presentation means) The three-option presentation unit 97 presents candidates as three options (for example, Should option / Status quo option / Next option). Each presented option is assigned a responsibility token and necessary approval and audit conditions. Acceptance or rejection requires explicit user action (final human approval).

[0028] (9) External Inquiry Support Unit (Draft generation, external inquiry support means) The external inquiry support unit 98, upon the user's selection of an external consultation candidate, organizes the key points of the request's background, requirements, and conditions, and generates a draft of the inquiry. External transmission is always an explicit operation by the user and is not automated. External transmission means that information is sent from the information processing device 11 to an external generation AI or an external expert AI.

[0029] The External Inquiry Support Unit 98 may impose conditions such as masking or approver verification before external transmission. Furthermore, the External Inquiry Support Unit may mask or conceal confidential information before external transmission based on information disclosure restriction rules included in the company parameters. For example, if the company parameters define a rule that "customer names will be anonymized when transmitted externally," the unit may automatically convert customer names included in the draft inquiry to an anonymous expression such as "Customer A."

[0030] (10) Audit log generation unit (evidence creation, audit log generation means) The audit log generation unit 99 stores the presented content (normalized display), the basis for compliance with functional requirements, the gate judgment result, warnings (risks), responsibility tokens, user selection, approver, transmitted content, execution status, etc., in association with each other. This makes it possible to trace "why those three options were presented."

[0031] (11) Responsibility Demarcation and Value Calculation Department (Settlement / Evaluation, Responsibility Demarcation and Value Calculation Methods) The Responsibility Boundary and Value Calculation Unit 100 has the function of determining the prerequisite information for the responsibility boundary by linking selection, approval, and results based on the responsibility tokens assigned at the time of generating the three-option presentation. Furthermore, when the execution results (success / failure / evaluation) of a business proposal are input, the Responsibility Boundary and Value Calculation Unit 100 determines which action candidates / AI / approvals were involved in (contributed to) the execution results of the business proposal, and performs processing that can be used for compensation calculation (results-based, usage-based, agency incentives, etc.).

[0032] (Overview of information processing methods implemented by information processing equipment) Figure 6 shows an overview of the information processing method implemented by the information processing device 11. The information processing method takes communication information related to business operations performed on the collaborative platform as input and outputs business support information in which the information processing device 11 outputs business support information, and includes a first step S21 of acquiring communication information and business logs, a second step S22 of extracting business issues and constraints, a third step S23 of generating multiple candidate actions, a fourth step S24 of gate-judging the candidates by referring to company parameters, a fifth step S25 of translating and normalizing the candidates that have passed or been prioritized in the gate-judgment into an in-house standard format, and a sixth step S26 of presenting the translated and normalized candidates to the user terminal.

[0033] (Structure of responsibility tokens) Responsibility tokens are metadata assigned to each option and indicate the classification information that forms the basis of responsibility boundaries. The "responsibility token / classification information" is not intended to automatically determine the final legal responsibility, but rather is a technical structure that determines and audits the classification information that forms the basis of responsibility boundaries at the time the candidate actions are presented. Responsibility tokens may include at least "Token_ID, Proposal_Type (generator)", "Responsibility_Category (category)", "Required_Approver (approval required)", "Disclosure_Restriction (disclosure restrictions)", "Accounting_Flag (value calculation target)", and "Issued_Timestamp".

[0034] (Structure of structured data (Perfect Order)) "Perfect Order" is a data record that takes information extracted from ambiguous business conversations (unstructured data) conducted on a collaborative platform and organizes it into structured data with items and attributes that can be immediately processed in the next process. "Perfect Order" is used, for example, as "processable order data" that is passed on to the next internal process (quotation, ordering, arrangement, approval, task creation, etc.) or to an external service provider.

[0035] Figure 8 shows an example of the data structure (table structure) of "Perfect Order". Figure 8 shows an example of a data record for holding structured data (Perfect Order) that is itemized and attributed in a way that allows for immediate business processing in the next step, after extracting information from unstructured business conversations conducted on the collaborative platform.

[0036] A "Perfect Order" includes at least the following items: "Order_ID," "Context_Summary," "Business_Issue," "Required_Conditions," "Functional_Requirements," "Gate_Judgement_Result," "Normalized_Request," "Candidate_List," "Responsibility_Token_List," "Approval_Status," "Execution_Result," and "Audit_Log_Reference." These items are stored in association with the audit log. Furthermore, a "Perfect Order" may also include the version of the enterprise parameter or the identifier of the applicable policy. Figure 8 shows an example of the data type and technical meaning of each item.

[0037] Figure 9 also shows a list of "Candidates" (example records). The example records for "Candidates" include items such as "Candidate_ID," "Candidate_Label," "Candidate_Content," "Match_Evidence," "Gate_Link," and "Responsibility_Token_Ref," as well as examples of the corresponding data types and technical meanings.

[0038] Furthermore, Figure 10 shows a list of responsibility tokens (example fields, minimum set). The list of responsibility tokens includes items such as "Token_ID", "Candidate_ID", "Responsibility_Class", "Assigned_By", "Assigned_At", "Basis_Reference", and "Param_Version", as well as examples of the corresponding data types and technical meanings.

[0039] (Variations of the embodiment) The collaborative platform in this embodiment is not limited to business chat, but may also include email, forms, meeting transcripts, voice input, image input, conversation and operation logs in VR (Virtual Reality) space, AR (Augmented Reality) space, etc. Furthermore, the application field of this embodiment is not limited to business operations such as manufacturing, finance, healthcare, construction, logistics, human resources, legal affairs, and general affairs.

[0040] (Embodiment: PCAD (Plan·Do·Check·Action) sequence enforcement control) In this embodiment, business processing will not proceed until the following steps 1-3 are completed. Step 1 is the determination of prerequisites (purpose, deadline, completion level, constraints) by a supervisor or authorized person. Step 2 is the confirmation and agreement of the person in charge regarding the interpretation presented by the externally generated AI. Step 3 is a control that functionally locks task registration, schedule placement, and execution processing until steps 1 and 2 are completed. Steps 1 to 3 above systematically enforce a business state transition in which Check (confirmation) is always completed before Do (execution).

[0041] (Implementation: Knowledge transfer loop with external expert AI) If the confidence level of information obtained from an external expert AI is below a predetermined value, or if it includes specific attribute tags such as manufacturing know-how, a confirmation request is sent to a pre-designated expert. The expert's response is stored as knowledge data or decision-making aid information that can be referenced in subsequent generation processes.

[0042] (Embodiment: Resource negotiation control) If a new task exceeds the assigned person's available time, instead of automatically rejecting it, a negotiation UI (User Interface) will be presented to the supervisor. Task registration will only be permitted if a trade-off is reached, such as reducing, postponing, or reallocating existing tasks.

[0043] (Implementation: Input and management of enterprise parameters) Implementing companies can input information such as corporate philosophy, business plan, organizational structure, roles and responsibilities, terminology definitions, operational decision criteria, and approval conditions as corporate parameters. These corporate parameters are maintained within each implementing company's tenant environment and are referenced for gate determination, translation, normalization, and control of business progress for candidates that include information obtained from externally generated AI or external specialized AI. The provider of the information processing device may, in principle, configure the system so that it does not refer to the contents of the corporate parameters, and the parameters may be configured to be updated only after explicit operation or approval by the implementing company.

[0044] (Implementation: Collaboration with external expert AI) External expert AI is used not as a decision-maker, but as a source of information providing specialized knowledge. The information processing device 11 may include a configuration that queries the external expert AI only after the preconditions have been determined based on the aforementioned corporate parameters and PCAD. PCAD is a concept that aims to improve the quality of management (business proposals) by circulating a hypothesis-verification process of Plan, Do, Check, and Action. Information obtained from the external expert AI is not applied to business as is, but is adopted after confirmation or approval by a human.

[0045] (Notes) This embodiment does not involve AI replacing final decision-making, but rather relies on human final approval. External transmission is conditional on explicit user operation and includes measures to prevent the system from automatically transmitting data externally. In this embodiment, decision criteria or business parameters are updated through human operation or approval, and the information processing device 11 or externally generated AI does not autonomously change such decision criteria. As a result, the order of business processes, pre-decision control, and preconditions for responsibility demarcation are always maintained under human involvement, ensuring accountability and auditability.

[0046] (Specific example) In this embodiment, several specific examples of information processing (business support) systems, information processing (business support) devices, information processing (business support) methods, programs, and storage media are described with reference to the drawings. Figure 1 shows an example of an information processing system 10. The information processing system 10 has a configuration that supports the business (operations) performed by business operator D1. The information processing system 10 builds a collaborative platform by business operators D1 and D2. Business operator D1 may be a company, corporation, organization, individual, etc. The industry of business operator D1 may be any of the following: manufacturing, information and communication, transportation, finance, real estate, accommodation, retail, etc.

[0047] The information processing system 10 may be implemented by an information processing device 11, a first user terminal 12, a second user terminal 13, and an information providing computer 50. The information processing device 11 and the first user terminal 12 may be used within business operator D1. The information processing device 11 is communicated to the first user terminal 12 via network 14, and is also communicated to the second user terminal 13 and the information providing computer 50 via network 60.

[0048] The second user terminal 13 and the information provision computer 50 are located outside the area of ​​business operator D1. The second user terminal 13 is used by user 61. The second user terminal 13 is used by business operator D2. Business operator D2 may be a company, corporation, agency, trading company, individual, etc. that provides goods to business operator D1. The goods that business operator D2 provides to business operator D1 may include goods, products, external services, external proposals, agency proposals, external merchandise, etc. User 61 may be a representative, person in charge, employee, staff, etc. belonging to business operator D2. The second user terminal 13 may acquire information from the information processing device 11, and the second user terminal 13 may transmit information to the information processing device 11. The information provision computer 50 is managed by an administrator separate from the administrator 15 of the information processing device 11.

[0049] Networks 14 and 60 are communication lines for transmitting various types of information. This information may include image information, sound information, text data, files, commands, requests, and signals. Once the information processing device 11 is installed within the operator D1, network 14 may be implemented by, for example, an intranet, a local area network, etc. Network 60 may be implemented by the internet. Networks 14 and 60 may consist of communication antennas, communication equipment (access points), repeaters, communication circuits, communication cables (copper wires, optical fibers), etc. Repeaters may include modems (devices that convert analog signals to digital signals), hubs (concentrators), routers, etc.

[0050] A communication line includes one or more means of communication, including wireless communication and wired communication. Wireless communication includes short-range wireless communication. Short-range wireless communication includes, for example, wireless LAN (Wi-Fi®) and Bluetooth®. Wireless communication includes electrical signals, optical signals, radio waves, infrared rays, satellite communication, etc. Signals used for communication include digital signals and analog signals.

[0051] The information processing device 11 is a computer managed and operated by the administrator 15 of business operator D1. The first user terminal 12 is operated and used by a user 16 who performs work at business operator D1. User 16 may be affiliated with, employed by, or employed by business operator D1. User 16 may include a supervisor 16A and a subordinate 16B. The administrator 15 may also be User 16.

[0052] User 16's duties at business operator D1 may include sales, accounting, technical development, human resources, product manufacturing, service provision, etc. User 16's duties are the specific tasks performed in User 16's duties, such as User 16's sales pitch to customers, User 16's bookkeeping, User 16's prototyping of new products, User 16's performance evaluation, User 16's assembly work on the manufacturing line, User 16's customer support, etc.

[0053] The first user terminal 12 may include a supervisor terminal 12A and a subordinate terminal 12B. The supervisor terminal 12A is used and operated by the supervisor 16A. The subordinate terminal 12B is used and operated by the subordinate 16B. The first user terminal 12 may be singular. Information can be communicated between the information processing device 11 and the first user terminal 12.

[0054] (Description of information processing device) The information processing device 11 is a computer implemented by various hardware and various software. The software includes an operating system and applications. The information processing device 11 may be implemented by a main unit (casing), ROM (Read Only Memory) 18, processor 19, main memory 20, auxiliary memory 21, media drive device 51, input device 22, output device 23, and communication device 24. The ROM 18 is a data read-only memory and the data written to it during manufacturing is not modified. The ROM 18 stores programs that are executed first when the information processing device 11 is started, such as the IPL (Initial Program Loader). The ROM 18 is a non-volatile memory device.

[0055] The processor 19 is located in the main body of the information processing device 11 and may be composed of a central processing unit (CPU) that integrates an arithmetic unit (arithmetic circuit) and a control unit (control circuit). The processor 19 is connected to the ROM 18, main memory 20, auxiliary memory 21, media drive device 51, input device 22, output device 23, and communication device 24 via a communication bus 25. The processor 19 comprehensively controls other devices and circuits located inside the main body and devices and circuits located outside the main body.

[0056] Furthermore, the processor 19 may be a multiprocessor. The processor 19 may be composed of, for example, one of the following elements: CPU, MPU (Micro Processing Unit), DSP (Digital Signal Processor), ASIC (Application Specific Integrated Circuit), PLD (Programmable Logic Device), FPGA (Field Programmable Gate Array), or GPU (Graphics Processing Unit). Alternatively, the processor 19 may be a combination of two or more elements from among the CPU, MPU, DSP, ASIC, PLD, FPGA, and GPU.

[0057] The processor 19 performs various processes by running a program. The various processes performed by the processor 19 include calculation, judgment, learning, comparison, identification, classification, analysis, extraction, creation, reasoning, regression, proposal, control, generation, etc. The various processes performed by the processor 19 may also include reading a program from the auxiliary storage device 21, running the program read from the auxiliary storage device 21 in the main memory device 20, and retrieving various information from the auxiliary storage device 21 according to the program and processing it.

[0058] Furthermore, the processing performed by the processor 19 may include processing various types of information input by the input device 22, processing the output device 23, processing various types of information obtained from the second user terminal 13 and the information providing computer 50, processing various types of information to be stored in the auxiliary storage device 21, processing various types of information obtained from the first user terminal 12, and processing various types of information to be transmitted to the first user terminal 12.

[0059] The various types of information processed by the processor 19 may include image information, sound information, text data, file data, etc. Image information includes video and still images. Sound information includes voice and music. The various types of information processed by the processor 19 may also include user information, communication information, and information read from the auxiliary storage device 21. The information stored in the auxiliary storage device 21 may include information input using the input device 22, information obtained from the second user terminal 13, and information obtained from the information providing computer 50. User information and communication information will be described later.

[0060] The main memory 20 is a volatile memory device and may be implemented by semiconductor memory such as RAM (Random Access Memory) or SRAM (Static Random Access Memory). The main memory 20 functions as a work area and buffer area for storing programs, instructions, etc., retrieved from the auxiliary memory 21 when the processor 19 processes and executes those programs, instructions, etc.

[0061] The auxiliary storage device 21 is a non-temporary storage medium. The auxiliary storage device 21 may also be understood as storage. The auxiliary storage device 21 may be located inside the main unit, or it may be located outside the main unit and connected to the communication bus 25 by wireless or wired communication. The auxiliary storage device 21 operates according to input and output commands from the processor 19. Processes for writing various types of information to the auxiliary storage device 21 and processes for reading various types of information from the auxiliary storage device 21 may be performed.

[0062] The storage medium 21A of the auxiliary storage device 21 may be implemented by, for example, a magnetic recording type, an optical recording type, or a semiconductor recording type. A magnetic recording type storage medium 21A may be implemented by an HDD (Hard Disk Drive), a floppy disk, a magnetic tape, etc. An optical recording type storage medium 21A may consist of a CD (Compact Disc), a DVD (Digital Versatile Disc), or a BD (Blu-ray® Disc). A semiconductor recording storage medium 21A may be implemented by an SSD (Solid State Drive), a USB flash drive, or an SD card (Secure Digital Card).

[0063] The auxiliary storage device 21 may have non-temporary programs pre-stored in it. Furthermore, the auxiliary storage device 21 may have non-temporary programs read by the media drive device 51, or non-temporary programs downloaded from the information providing computer 50, installed on it.

[0064] The media drive device 51 is hardware into which a non-temporary storage medium 51A is inserted and removed. The storage medium 51A is portable and can be implemented as, for example, a magnetic disk, an optical disk, a semiconductor memory, etc. The storage medium 51A may store a program that is operated by the processor 19. After the program stored in the storage medium 51A is installed into the auxiliary storage device 21 and then stored in the main memory 20, the processor 19 can execute the program. In addition, various kinds of information may be stored in the storage medium 51A of the media drive device 51. Various kinds of information stored in the storage medium 51A may be transmitted to the auxiliary storage device 21.

[0065] The processor 19 may implement various functional components shown in Figure 2 by running non-temporary programs and cooperating with various hardware components constituting the information processing device 11. For example, the processor 19 may implement a user information management unit 26, a communication environment management unit 27, a provisional proposal processing unit 52, a requirements processing unit 53, a business proposal processing unit 54, a provided information processing unit 55, various information processing units 29, an artificial intelligence unit 30, and the like.

[0066] The user information management unit 26 may process user information obtained from the first user terminal 12 and the second user terminal 13 to perform user registration, and may also store the user information and the results of user registration in the auxiliary storage device 21. The user information may include information about user 16 using the first user terminal 12 and information about user 61 using the second user terminal 13.

[0067] User information for supervisor 16A, an example of user 16, may include an identifier that identifies the supervisor's terminal 12A, supervisor 16A's name, supervisor 16A's job duties and years of work experience, etc. The identifier that identifies the supervisor's terminal 12A consists of an IP address, email address, and password, etc.

[0068] User information for subordinate 16B, an example of user 16, may include an identifier that identifies the subordinate's terminal 12B, the subordinate's name, the subordinate's job duties and years of work experience, etc. The identifier that identifies the subordinate's terminal 12B consists of an IP address, email address, and password, etc.

[0069] User information for user 61 may include an identifier that identifies the second user terminal 13, user 61's name and occupation, years of work experience, etc. The identifier that identifies the second user terminal 13 consists of an IP address, email address, and password, etc.

[0070] The communication environment management unit 27 is configured to provide a communication environment that enables communication with multiple first user terminals 12 and second user terminals 13 via the information processing device 11. The specified environment may include an environment in which a chat application is running, an environment in which a web conferencing application is running, and so on.

[0071] The communication environment management unit 27 may provide the user screen for when the chat application is running and the user screen for when the web conferencing application is running to the first user terminal 12 and the second user terminal 13, respectively. The communication environment management unit 27 may also acquire the communication information exchanged between the first user terminal 12 and the second user terminal 13 and store the acquired communication information in the auxiliary storage device 21.

[0072] The provisional proposal processing unit 52 may collaborate with the artificial intelligence unit 30 to create a provisional proposal based on communication information and various information stored in the auxiliary storage device 21, and may also store the created provisional proposal in the auxiliary storage device 21. The meaning and specific examples of a provisional proposal will be described later.

[0073] The requirements processing unit 53 may extract requirements based on communication information and various information stored in the auxiliary storage device 21, and store the extracted requirements in the auxiliary storage device 21. The meaning and specific examples of requirements will be described later.

[0074] The business proposal processing unit 54 may, in cooperation with the artificial intelligence unit 30, create a business proposal based on the provisional proposal, necessary conditions, and various information stored in the auxiliary storage device 21, and store the created business proposal in the auxiliary storage device 21. Alternatively, the business proposal processing unit 54 may, in cooperation with the artificial intelligence unit 30, extract options and business executors based on the business proposal and various information stored in the auxiliary storage device 21, and store the extracted options and business executors in the auxiliary storage device 21. The meanings and specific examples of business proposals, options, and business executors will be described later.

[0075] The information processing unit 55 may create information including business proposals, options, and business performers, and store the created information in the auxiliary storage device 21. The information processing unit 55 may transmit the information to the first user terminal 12 and the second user terminal 13. The information processing unit 55 may present the information to users 16 and 61.

[0076] The information provision processing unit 55 may, if it determines that the confidence level (reliability) of the information obtained from the externally generated AI is below a predetermined value, or if it determines that the information obtained from the externally generated AI contains a specific attribute tag, implement a "routing means" that will withhold output to present the information to users 16,61 and send a confirmation request or response request to a third party with a pre-specified specific attribute. The third party with the specific attribute is the person or manager responsible for executing the business proposal. The externally generated AI may, for example, be the second user terminal 13.

[0077] Here, the "predetermined value" for determining the confidence level of the information may, for example, be "3 pieces of information" obtained from the externally generated AI. Alternatively, the "predetermined value" for determining the confidence level of the information may be "2 documents" obtained from the externally generated AI that explain the basis of the information. Furthermore, the "predetermined value" for determining the confidence level of the information may be, for example, "3 years of work experience" for user 61 obtained from the externally generated AI. These "predetermined values" for determining the confidence level of the information may be stored in the auxiliary storage device 21. Also, the "third party" may be, for example, a supervisor 16A.

[0078] Furthermore, specific attribute tags include, for example, know-how in product manufacturing operations, including unique technologies and procedures for efficiently and high-quality product manufacturing; know-how in sales and marketing operations, including practical methods or tricks that lead to building customer relationships and closing deals; and know-how in operating and managing business operator D1, including methodologies useful for improving operational efficiency, quality, and personnel development. These "specific attribute tags" may be stored in the auxiliary storage device 21. The information provided will be described later.

[0079] The various information processing units 29 may process various information in cooperation with the artificial intelligence unit 30 and store the processing results in the auxiliary storage device 21. The various information processing units 29 may, for example, process first auxiliary information and second auxiliary information. The meanings of first auxiliary information and second auxiliary information will be described later.

[0080] The various information processing units 29 may process sound information and store the processing results in the auxiliary storage device 21. The sound information processed by the various information processing units 29 may include human speech, music, or noise. The various information processing units 29 may convert the sound information into text data by performing sampling and quantization processing.

[0081] Furthermore, the various information processing units 29 may perform filtering processing to emphasize sounds in a specific frequency band. By performing filtering processing, the various information processing units 29 can achieve things like clarifying human-generated speech and removing noise. The various information processing units 29 may also perform processing to analyze the meaning of words within the speech, processing to identify the characteristics of the speaker, and so on.

[0082] Furthermore, the various information processing units 29 may process image information and store the processing results in the auxiliary storage device 21. The various information processing units 29 may, in cooperation with the artificial intelligence unit 30, perform image modification, filtering, transformation, synthesis, etc. The various information processing units 29 may process text data and store the processing results in the auxiliary storage device 21. The various information processing units 29 may obtain information from the information provision computer 50 about other companies selling products that compete with the products sold by business operator D1, and store the obtained information in the auxiliary storage device 21.

[0083] The various information processing units 29 may process execution information obtained from the first user terminal 12 and the second user terminal 13 and store the processing results in the auxiliary storage device 21. When users 16 and 61 perform a task in accordance with the task proposal and achieve results, the various information processing units 29 may also make decisions regarding the attribution of task value and compensation, billing, settlement, incentives, etc., based on the execution information.

[0084] The various information processing units 29 may update one or more of the judgment criteria or inference parameters based on the results of the proposed work execution, the approval results of the provided information, the evaluation results of the provided information, etc., obtained by the information processing method shown in Figure 4, which will be described later. The inference parameters may be understood as an inference parameter set realized by a large number of inference parameters. Furthermore, the inference parameter set may be realized by multiple inference parameter sets of different types (technical significance).

[0085] In other words, the various information processing units 29 may function as update means that enable updating of the enterprise parameters and / or inference parameter sets based on the execution results, approval results, or evaluation results included in the audit log. The various information processing units 29 may also switch the operating criteria of the gate determination means and / or translation / normalization means based on the selected inference parameters. The execution information will be described later.

[0086] The various information processing units 29 may determine whether the time required for a new task, as contained in the communication information and business logs, exceeds the available time of the person responsible for performing the new task. If the various information processing units 29 determine that the time required for a new task exceeds the available time of the person responsible, they may offer the user 16 or user 61, who is responsible for the task, a negotiation to delete or postpone the existing task. When the various information processing units 29 receive confirmation from the first user terminal 12 or the second user terminal 13 that an operation to delete or postpone the existing task has been performed, they may permit the inclusion (registration) of the new task in the business proposal.

[0087] Various information processing units 29 may store the following information in association with each other: presented content (normalized display), basis for compliance with functional requirements, gate judgment results, warnings (risks), responsibility tokens, user selection, approver, transmitted content, execution status, etc.

[0088] When the program is run, the artificial intelligence unit 30 works in cooperation with the auxiliary storage device 21 to perform various processes. The various processes performed by the artificial intelligence unit 30 include learning processes (learning stage) and inference processes (inference stage). Various information and models used by the artificial intelligence unit 30 for processing are stored in the auxiliary storage device 21. The various information used by the artificial intelligence unit 30 for processing may include image information, sound information, text data, etc. The various models are, for example, learning models, and the learning models may include large-scale language models.

[0089] The artificial intelligence unit 30 outputs processing results (output data) by processing the input data to be learned with a learning model during learning processing, such as machine learning processing. The artificial intelligence unit 30 outputs inference results by processing the data to be inferred with a learning model during inference processing. The inference performed by the artificial intelligence unit 30 may include processes that predict future trends and characteristics, predictions of the next word to appear in text data, etc. The learning model has inference parameters, which are judgment criteria for processing input data and generating output data.

[0090] The artificial intelligence unit 30 may extract features from the input data using machine learning. The input data processed by the artificial intelligence unit 30 may include text data, sound information, image information, etc., exchanged in a communication environment. Feature extraction from the input data includes classification and regression. Classification is the process of classifying and identifying features. Features may include regularities in the input data. Regression includes understanding the overall trend of the classified features, predicting the future state of the features, etc. The learning model is stored in the auxiliary memory device 21, and the learning model is trained as the artificial intelligence unit 30 repeats machine learning.

[0091] The machine learning performed by the artificial intelligence unit 30 may include three types: supervised learning, unsupervised learning, and reinforcement learning. In supervised learning, the artificial intelligence unit 30 uses a loss function to calculate the error between the processing result and the correct data, and trains the learning model based on that error. Specifically, it adjusts weights, biases, etc., to minimize the error. The artificial intelligence unit 30 trains the learning model by comparing the output data with the correct data (correct labels).

[0092] Unsupervised learning is a method of training a model using training data that does not contain correct answers, and it involves classifying the training data into groups of data with similar features. Reinforcement learning is a method of training a model by repeatedly processing input data with a learning model to obtain output data, and training the learning model to improve its evaluation (reward) of the output data.

[0093] The artificial intelligence unit 30 can use a neural network as a learning model to realize machine learning. A neural network has an input layer, an intermediate layer (hidden layer), and an output layer. The intermediate layer is configured to make the input information useful for the output layer. Furthermore, the artificial intelligence unit 30 can perform deep learning in realizing the intermediate layer of the neural network. In deep learning, the intermediate layer is multi-layered with multiple types of stages (nodes), and weights and biases are assigned to the functions (features) contained in the information. Deep learning can reduce noise in the information and data input to the learning model because it automatically instructs the intermediate layer to focus on the functions (features).

[0094] Furthermore, part of the learning model that realizes the processing of the artificial intelligence unit 30 may be composed of large language models (LLMs). Large language models are a type of generative artificial intelligence specialized in natural language processing (NLP).

[0095] Large-scale language models are language models that achieve advanced natural language generation (NLG) by being built through deep learning of vast amounts of text data. Natural language processing combines processes such as morphological analysis, syntactic analysis, semantic analysis, contextual analysis, and intent analysis to process natural language, enabling machine translation, text summarization, speech recognition, and more.

[0096] The large-scale language model may perform tasks such as parsing and analyzing text data contained in input prompts, generating text data based on the analysis results of the prompts, automatically generating prompts based on various information, and converting speech information into text.

[0097] Large-scale language models may be trained by analyzing which word sequences have a high probability of occurrence during the training phase, using a large amount of training data. The training data consists of various types of information obtained from networks, various types of information stored in auxiliary memory, etc. The training data may also include text data, image information, sound information, book data, etc. Alternatively, large-scale language models may be trained by analyzing which answers to questions composed of text data have a high accuracy rate.

[0098] In the inference phase, a large-scale language model first divides the input prompt (text data) into processable units (tokens) that can be analyzed and parsed for meaning, if the model is required to output the sentence following the prompt as text data. The large-scale language model then performs a process to understand the context of the prompt by calculating the relationships between each token.

[0099] Furthermore, the large-scale language model may extract multiple candidate words that are likely to appear next in the understood context. The large-scale language model may also select one of these highly likely words and concatenate it to the end of the text. In this way, the large-scale language model may complete the text data corresponding to the prompt by repeatedly extracting multiple candidate words that are likely to appear next, and then selecting and concatenating the most likely word from the extracted candidate words.

[0100] Furthermore, during the inference phase, when a question is input as a prompt, a large-scale language model may refer to a large amount of text data about the question in the context it has understood, and output answers that are likely to be correct as text data.

[0101] The artificial intelligence unit 30 may, in cooperation with the communication environment management unit 27, perform the process of converting sound information into text data. First, the microphone converts the sound waves of the voice into electrical signals, and the artificial intelligence unit 30 converts the electrical signals into digital data and also converts the sound characteristics into characters using an acoustic model. Furthermore, the artificial intelligence unit 30 can convert the characters converted by the acoustic model into text data by processing them with a large-scale language model.

[0102] Large-scale language models may be implemented using, for example, ChatGPT (Chat Generative Pre-trained Transformer) (registered trademark), LLama-2 (registered trademark), etc.

[0103] The learning model used by the artificial intelligence unit 30 for processing may be one that has undergone not only pre-training by machine learning on a base model, but also fine-tuning and human feedback. Fine-tuning (additional learning) is a technique that retrains a pre-trained learning model to suit a specific task, a specific industry, a specific application, specific data, etc., and fine-tunes its parameters.

[0104] Human feedback is a form of reinforcement learning (RLHF) using human feedback. The first step in human feedback is to use model answers (correct labels) created by humans to train a large-scale language model to use natural language (supervised learning).

[0105] The second stage of human feedback involves repeatedly having humans evaluate (reward) the output data produced by the large-scale language model, thereby training the model's parameters to improve the evaluation of the output data.

[0106] The third process of human feedback is reinforcement learning optimization (RL): Based on feedback from the reward model, reinforcement learning using algorithms such as PPO (Proximal Policy Optimization) optimizes the output generated by the large-scale language model, adjusting parameters so that it can generate more favorable answers for humans.

[0107] In the artificial intelligence unit 30, during the learning phase, machine learning processing may be performed to train the learning model by using judgment criteria (company parameters), first auxiliary information, and second auxiliary information as input data. In addition, the artificial intelligence unit 30 may perform processing to fine-tune the learning model by using various information acquired by the information processing device 11 during the process of executing the information processing method, and various processing results, as input data.

[0108] When the artificial intelligence unit 30 functions as a generative artificial intelligence unit, it can perform text generation, image generation, video generation, and speech generation. Text generation is a function that, when a prompt (in text format), such as an instruction or question, is input, analyzes the content and intent of the prompt and uses the information and data stored in the auxiliary storage device 21 to automatically generate a response (in text format) to the prompt. Because a large-scale language model is used, natural and highly accurate responses can be obtained.

[0109] Image generation is a function that automatically generates an original image similar to the prompt's image by using information and data stored in the auxiliary storage device 21 in response to an input prompt (in text format). Examples of artificial intelligence that generate images include Stable Diffusion, Midjourney, and DALL·E2.

[0110] Video generation is a function that automatically generates an original video that closely resembles the image of a prompt (in text format) by using information and data stored in the auxiliary storage device 21 in response to an input prompt. Gentoo is a well-known example of artificial intelligence that generates videos.

[0111] The voice generation function automatically generates original voice data (optimal actions, responses, music, etc.) in response to input prompts (in voice or text format) using information and data stored in the auxiliary storage device 21.

[0112] The input device 22 is a device operated by the administrator 15, and the input device 22 may have devices such as a keyboard 22A, a mouse 22B, a microphone 22C, a camera 22D, a reader 22F, and a display 22E.

[0113] Microphone 22C is an electronic component that converts acquired sound data into electrical signals and outputs them. Microphone 22C is connected to the communication bus 25 by either a wireless or wired communication system. Camera 22D is connected to the communication bus 25 by either a wireless or wired communication system. Camera 22D captures images of objects and generates image information. Image information includes videos, still images, photographs, etc. The image information generated by camera 22D is processed by processor 19 and stored in auxiliary storage device 21.

[0114] The reading device 22F may be implemented by a barcode reader, a two-dimensional code reader, an RFID (Radio Frequency Identification) reader, a scanner, etc. A scanner is a device that optically reads paper documents and photographs and converts them into image information.

[0115] Furthermore, the display 22E can be implemented using, for example, a liquid crystal display or an organic electroluminescent display. The screen of the display 22E can display images, graphics, text, operation buttons, operation tabs, etc. The display 22E can also be considered a monitor.

[0116] Administrator 15 can input various types of information to the information processing device 11 by using one or more of the following configurations: keyboard 22A, mouse 22B, microphone 22C, camera 22D, reader 22F, and operation buttons and operation tabs displayed on the screen of display 22E.

[0117] The output device 23 may include devices such as a display 22E, a printer 23A, and an audio information output device 23B. The display 22E is the same as the display 22E of the input device 22. The audio information output device 23B is a device that converts electrical signals contained in information and data into audio information and outputs it. The audio information output device 23B is implemented by a speaker, headphones, earphones, etc. The printer 23A prints information and data onto recording material using toner or ink and outputs it.

[0118] Furthermore, the auxiliary storage device 21 may consist of a program storage unit 31, a user information storage unit 32, a model storage unit 33, various information storage units 34, a provisional proposal storage unit 35, a requirements storage unit 36, a business proposal storage unit 70, and a provided information storage unit 71, as shown in Figure 2. The program storage unit 31 may store an application operated by the processor 19.

[0119] The applications stored in the program storage unit 31 include non-temporary programs 31A, manuals, configuration files, data storage files, data, various libraries, etc. When the processor 19 reads and runs program 31A, the information processing device 11 implements the information processing method.

[0120] Furthermore, when the processor 19 loads and runs program 31A, it may fine-tune the learning model used by the artificial intelligence unit 30. In other words, program 31A can be said to be an operating system (business OS) for making the information processing device 11 function. An example of an information processing method and an example of fine-tuning performed by the artificial intelligence unit 30 will be described later.

[0121] The user information storage unit 32 stores user information and registration information acquired from the first user terminal 12 and the second user terminal 13. The model storage unit 33 stores learning models and large-scale language models, etc. The various information storage unit 34 may store various information acquired from the first user terminal 12, the second user terminal 13 and the information providing computer 50, various information used by the information processing device 11 to perform various processes, and various databases.

[0122] The various information storage units 34 may store, for example, judgment criteria, inference parameters, first auxiliary information, and second auxiliary information. The judgment criteria and inference parameters may be stored in the various information storage units 34 in advance, or they may be stored in the various information storage units 34 after being acquired from the media drive device 51, or they may be stored in the various information storage units 34 after being acquired from the information providing computer 50.

[0123] Inference parameters can be introduced externally, and when the inference parameters are switched, the operating criteria of the processor 19 are switched. In addition, multiple different types of decision criteria are stored, and the processor 19 may select any of the decision criteria. The meanings and specific examples of the decision criteria, first auxiliary information, and second auxiliary information will be described later. Furthermore, the results of various processes performed by the information processing device 11 may be stored in the various information storage units 34.

[0124] The provisional proposal storage unit 35 may store information including the provisional proposal created by the provisional proposal processing unit 52. The requirements storage unit 36 ​​may store information including the requirements extracted by the requirements processing unit 53. The business proposal storage unit 70 may store information including the business proposal created by the business proposal processing unit 54. The provided information storage unit 71 may store provided information created by the provided information processing unit 55.

[0125] The communication device 24 includes devices, equipment, and standards for connecting the information processing device 11 to the first user terminal 12 via the network 14 using at least one of a wireless communication system or a wired communication system, and for connecting to the second user terminal 13 and the information providing computer 50 via the network 60. The hardware constituting the communication device 24 may consist of, for example, a communication circuit, a LAN card, a network adapter, a network interface card, a communication cable, a communication antenna, a communication port, etc.

[0126] (Description of the first user terminal) The supervisor terminal 12A and the subordinate terminal 12B included in the first user terminal 12 have similar configurations, as shown in Figure 3. For convenience, the configurations of the supervisor terminal 12A and the subordinate terminal 12B will be comprehensively described as the first user terminal 12. The first user terminal 12 includes, for example, portable computers and fixed computers. Portable computers include smartphones and tablet computers. Fixed computers include desktop computers, notebook computers, etc.

[0127] The first user terminal 12 is connected to the information processing device 11 via the network 14. The first user terminal 12 includes a main unit (casing), a processor 40, main memory 41, auxiliary memory 42, input device 43, output device 44, communication device 45, etc. The processor 40 may be located inside the main unit and consist of a central processing unit (CPU) in which an arithmetic unit (arithmetic circuit) and a control device (control circuit) are integrated. The processor 40 may be a multiprocessor, similar to the processor 19.

[0128] The processor 40 is communicated via a communication bus 46 to the main memory 41, auxiliary memory 42, input device 43, output device 44, and communication device 45. The processor 40 has a configuration that comprehensively controls other devices and circuits located inside the main unit, as well as devices and circuits located outside the main unit.

[0129] The processor 40 processes various types of information by running non-temporary programs stored in the auxiliary storage device 42. The various types of information processed by the processor 40 may include information, data, signals, and commands. Specifically, the processor 40 processes sound information, image information, text data, etc. Sound information may include speech and music. Image information may include video, still images, photographs, etc. Text data may include characters, numbers, sentences, symbols, figures, etc. The processing performed by the processor 40 includes processing, calculations, judgments, comparisons, control, etc.

[0130] The processes performed by the processor 40 include processing various types of information input from the input device 43, controlling the output device 44, reading various types of information from the auxiliary storage device 42, processing various types of information acquired from the information processing device 11, storing various types of information in the auxiliary storage device 42, and sending the various types of information processed by the processor 40 to the information processing device 11.

[0131] The main memory 41 is a volatile memory device and may be implemented by semiconductor memory such as RAM (Random Access Memory) or SRAM (Static Random Access Memory). The main memory 41 functions as a work area and buffer area for storing programs, data, instructions, etc., retrieved from the auxiliary memory 42 when the processor 40 executes processing of programs, data, instructions, etc.

[0132] The auxiliary storage device 42 is a non-temporary storage medium. The auxiliary storage device 42 stores non-temporary applications operated by the processor 40. The non-temporary applications stored in the auxiliary storage device 42 include non-temporary programs, manuals, configuration files, data storage files, data, various libraries, etc. The auxiliary storage device 42 may also have a chat application installed for receiving business support services related to the information processing system 10. In addition, the auxiliary storage device 42 stores various information acquired from the information processing device 11, various information processed by the processor 40, and the processing results performed by the processor 40.

[0133] The auxiliary storage device 42 operates according to input and output commands from the processor 40. The auxiliary storage device 42 is implemented, for example, by a head and a non-temporary storage medium 42A. The head writes information and data to the storage medium 42A and reads information and data from the storage medium 42A. The storage medium 42A is implemented by a magnetic disk, optical disk, flash memory, etc. The storage medium 42A may be configured to be attachable to and detachable from the main unit.

[0134] The input device 43 is a user interface that is operated when inputting various information to the first user terminal 12, when executing various processes on the first user terminal 12, when sending various information from the first user terminal 12 to the network 14, when the first user terminal 12 acquires various information via the network 14, and so on.

[0135] The input device 43 is a device operated by the user 16, and may consist of devices such as a keyboard 43A, a mouse 43B, a microphone 43C, a camera 43D, a display 43E, and a reader 43F. The display 43E may be implemented as, for example, a liquid crystal display or an organic electroluminescent display. Images, videos, graphics, text, operation buttons, operation tabs, etc., are displayed on the screen of the display 43E. The display 43E may also be understood as a monitor.

[0136] User 16 can input various types of information and commands by using one or more of the following configurations: keyboard 43A, mouse 43B, microphone 43C, operation buttons and tabs displayed on the screen of display 43E, and reading device 43F. User 16 can input sound information via microphone 43C.

[0137] Microphone 43C is an electronic component that converts acquired sound information into an electrical signal and outputs it. Microphone 43C is connected to the communication bus 46 by either a wireless or wired communication system. Camera 43D is connected to the communication bus 46 by either a wireless or wired communication system. Camera 43D photographs an object and generates image information. The image information generated by camera 43D is processed by processor 40 and stored in auxiliary storage device 42.

[0138] User 16 can access the information processing device 11 from the first user terminal 12 and transmit information by using the input device 43, and can also process information acquired by the first user terminal 12 from the information processing device 11. The information transmitted from the first user terminal 12 to the information processing device 11 and the information acquired by the first user terminal 12 from the information processing device 11 may include information itself, data, commands, signals, etc. In addition, some of the information processed by the first user terminal 12 may be transmitted in real time from the first user terminal 12 to the information processing device 11 when the input device 43 is not being operated.

[0139] The output device 44 is a user interface that includes devices such as a display 43E, an audio information output device 44A, and a printer 44B. The display 43E is the same as the display 43E of the input device 43. The audio information output device 44A is a device that converts electrical signals contained in information and data into audio information and outputs it. The audio information output device 44A is implemented by a speaker, headphones, earphones, etc. The audio information output device 44A is connected to the communication bus 46 by wireless or wired communication. The printer 44B prints information and data onto paper using toner or ink and outputs it.

[0140] The communication device 47 includes devices, equipment, and standards for connecting the first user terminal 12 to the network 14 by at least one of the following communication systems: a wireless communication system or a wired communication system. The devices and equipment for connecting the first user terminal 12 to the network 14 may include cables, antennas, routers, communication circuits, communication ports, communication connectors, communication hubs, etc.

[0141] The first user terminal 12 may transmit user information to the information processing device 11. User 16 can communicate with user 61, who is using the second user terminal 13, via the information processing device 11 by using the first user terminal 12. User 16 can communicate with user 61, who is using the second user terminal 13, via the information processing device 11 by using the first user terminal 12. Supervisor 16A and subordinate 16B can communicate with each other via the information processing device 11 by using supervisor terminal 12A and subordinate terminal 12B.

[0142] The first user terminal 12 may obtain the provided information from the information processing device 11. The first user terminal 12 may output the provided information using the output device 44. User 16 may use the first user terminal 12 to confirm the provided information and execute the proposed tasks and options. User 16 may use the first user terminal 12 to transmit execution information, including the process and results of executing the task proposal, to the information processing device 11.

[0143] (Description of the second user terminal) The second user terminal 13 has the configuration shown in Figure 3. The second user terminal 13 includes, for example, a portable computer and a fixed computer. Portable computers include smartphones and tablet computers. Fixed computers include desktop computers, notebook computers, etc.

[0144] The second user terminal 13 is connected to the information processing device 11 via the network 60. The second user terminal 13 includes a main unit (casing), a processor 80, main memory 81, auxiliary memory 82, input device 83, output device 84, communication device 85, etc. The processor 80 may be located inside the main unit and consist of a central processing unit (CPU) in which an arithmetic unit (arithmetic circuit) and a control device (control circuit) are integrated. The processor 80 may be a multiprocessor, similar to the processor 19.

[0145] The processor 80 is communicated via a communication bus 86 to the main memory 81, auxiliary memory 82, input device 83, output device 84, and communication device 85. The processor 80 has a configuration that comprehensively controls other devices and circuits located inside the main unit, as well as devices and circuits located outside the main unit.

[0146] The processor 80 processes various types of information by running non-temporary programs stored in the auxiliary storage device 82. The various types of information processed by the processor 80 may include information, data, signals, and commands. Specifically, the processor 80 processes sound information, image information, text data, etc. Sound information may include speech and music. Image information may include videos, still images, photographs, etc. Text data may include characters, numbers, sentences, symbols, figures, etc.

[0147] The processing performed by the processor 80 includes processing, calculation, judgment, comparison, control, etc. The processor 80 may have an artificial intelligence unit similar to the artificial intelligence unit 30 of the processor 19. The auxiliary storage device 82 may have a model storage unit and a program storage unit similar to the model storage unit 33 and program storage unit 31 of the auxiliary storage device 21. The processor 80 is capable of functioning as generative artificial intelligence.

[0148] The processes executed by the processor 80 include processing various types of information input from the input device 83, controlling the output device 84, reading various types of information from the auxiliary storage device 82, processing various types of information acquired from the information processing device 11, storing various types of information in the auxiliary storage device 82, and sending the various types of information processed by the processor 80 to the information processing device 11.

[0149] The main memory 81 is a volatile memory device and may be implemented by semiconductor memory such as RAM (Random Access Memory) or SRAM (Static Random Access Memory). The main memory 81 functions as a work area and buffer area for storing programs, data, instructions, etc., retrieved from the auxiliary memory 82 when the processor 80 executes processing of programs, data, instructions, etc.

[0150] The auxiliary storage device 82 is a non-temporary storage medium. The auxiliary storage device 82 stores non-temporary applications operated by the processor 80. The non-temporary applications stored in the auxiliary storage device 82 include non-temporary programs, manuals, configuration files, data storage files, data, various libraries, etc. The auxiliary storage device 82 may also have a chat application installed for receiving business support services related to the information processing system 10. In addition, the auxiliary storage device 82 stores various information acquired from the information processing device 11, various information processed by the processor 80, and the processing results performed by the processor 80.

[0151] The auxiliary storage device 82 operates according to input and output instructions from the processor 80. The auxiliary storage device 82 is implemented, for example, by a head and a non-temporary storage medium 82A. The head writes information and data to the storage medium 82A and reads information and data from the storage medium 82A. The storage medium 82A is implemented by a magnetic disk, optical disk, flash memory, etc.

[0152] The input device 83 is a user interface that is operated when inputting various information to the second user terminal 13, when executing various processes on the second user terminal 13, when sending various information from the second user terminal 13 to the network 60, when the second user terminal 13 acquires various information via the network 60, and so on.

[0153] The input device 83 is a device operated by the user 61, and may consist of devices such as a keyboard 83A, a mouse 83B, a microphone 83C, a camera 83D, a display 83E, and a reader 83F. The display 83E may be implemented as, for example, a liquid crystal display or an organic electroluminescent display. Images, videos, graphics, text, operation buttons, operation tabs, etc., are displayed on the screen of the display 83E.

[0154] The display 43E may be considered a monitor. User 61 can input various types of information and commands by using one or more of the following: keyboard 83A, mouse 83B, microphone 83C, operation buttons and tabs displayed on the screen of display 83E, and reading device 83F. User 61 can input sound information via microphone 83C.

[0155] Microphone 83C is an electronic component that converts acquired sound information into an electrical signal and outputs it. Microphone 83C is connected to communication bus 86 by either a wireless or wired communication system. Camera 83D is connected to communication bus 86 by either a wireless or wired communication system. Camera 83D photographs an object and generates image information. The image information generated by camera 83D is processed by processor 80 and stored in auxiliary storage device 82.

[0156] User 61 can access the information processing device 11 from the second user terminal 13 and transmit information by using the input device 83, and can process information acquired by the second user terminal 13 from the information processing device 11. The information transmitted from the second user terminal 13 to the information processing device 11 and the information acquired by the second user terminal 13 from the information processing device 11 may include information itself, data, commands, signals, etc. In addition, some of the information processed by the second user terminal 13 may be transmitted in real time from the second user terminal 13 to the information processing device 11 when the input device 83 is not being operated.

[0157] The output device 84 is a user interface that includes devices such as a display 83E, an audio information output device 84A, and a printer 84B. The display 83E is the same as the display 43E of the input device 83. The audio information output device 84A is a device that converts electrical signals contained in information and data into audio information and outputs it. The audio information output device 84A is implemented by a speaker, headphones, earphones, etc. The audio information output device 84A is connected to the communication bus 86 by wireless or wired communication. The printer 84B prints information and data onto paper using toner or ink and outputs it.

[0158] The communication device 85 includes devices, equipment, and standards for connecting the second user terminal 13 to the network 60 by at least one of the following communication systems: a wireless communication system or a wired communication system. The devices and equipment for connecting the second user terminal 13 to the network 60 may include cables, antennas, routers, communication circuits, communication ports, communication connectors, communication hubs, etc.

[0159] The second user terminal 13 may transmit user information to the information processing device 11. By using the second user terminal 13, user 61 can communicate with user 16 using the first user terminal 12 via the information processing device 11.

[0160] The second user terminal 13 may obtain the provided information from the information processing device 11. The second user terminal 13 may output the provided information using the output device 84. User 61 may use the second user terminal 13 to confirm the provided information and execute the proposed tasks and options. User 61 may use the second user terminal 13 to send execution information, including the process and results of executing the task proposal, to the information processing device 11.

[0161] (An example of an information processing method) Figure 4 shows a specific example of an information processing method that can be executed by the information processing device 11. In step S10, the information processing device 11 may store the judgment criteria, first auxiliary information, and second auxiliary information in the auxiliary storage device 21. The meanings and specific examples of the judgment criteria, first auxiliary information, and second auxiliary information will be described later.

[0162] In step S11, the information processing device 11 may acquire communication information between the first user terminal 12 and the second user terminal 13, and store the acquired communication information in the auxiliary storage device 21. The communication information is various types of information generated through communication between user terminals using chat (business chat), web interviews, email, etc. The format of the various types of information may include sound information, image information, and text data. Sound information may include voice and music. Image information may include still images, videos, photographs, illustrations, etc.

[0163] The various types of information included in the communication information are related to the business of business operator D1, and may include items such as consultations related to business operator D1's business, proposals related to business operator D1's business, questions related to business operator D1's business, answers to questions, judgments related to business operator D1's business, explanations related to business operator D1's business, and conversations related to business operator D1's business. In step S12, the information processing device 11 may perform processing to align the information included in the communication information with the judgment criteria.

[0164] The criteria for judgment may be the company D1's own standards for its industry and business, and may also include public standards and company standards. Public standards are "standards" set by the government or public institutions, and their purpose is to ensure the safety, quality, and interchangeability of products. Public standards are those that company D1 should comply with in its industry, and may include, for example, Japanese Industrial Standards (JIS) set by the Japanese Standards Association (JSA), standards set by the International Organization for Standardization (ISO), standards set by the International Electrotechnical Commission (ICE), etc.

[0165] Business standards are standards and criteria independently established by business operator D1, and are intended to unify product specifications, quality standards, business execution processes, etc., within business operator D1. Business standards include, for example, one or more items specific to the industry to which business operator D1 belongs (company-specific, company-specific), such as philosophy, business plan, terminology definitions, expression standards, culture, climate, traditions, past performance, corporate philosophy, organizational information, roles and responsibilities, purchasing rules, prohibitions, approval routes, expression standards, and work formats.

[0166] The process of aligning communication information with judgment criteria may include processes that initialize (standardize) or normalize the items of various information contained in the communication information so that they do not deviate from the conditions and items included in the judgment criteria, or are permitted by the items of the judgment criteria, such as translating, changing, correcting, deleting, etc.

[0167] The processing performed by the information processing device 11 in step S12 may include processing to align various types of information included in the communication information, obtained from the second user terminal 13 used by user 61, with the specific terminology definitions or expression standards of business operator D1. Here, the various types of information may include external expressions, hype phrases, terminology, exaggerations, sales pitches, etc., used by business operator D2.

[0168] The information processing device 11 may use the communication information processed in step S12 to determine, for example, extract a provisional proposal and necessary conditions in step S13. The provisional proposal is information that is created in the preliminary stage of preparing a business proposal to be sent to one or more users, either the first user or the second user.

[0169] The provisional proposals are objectives related to the operations of business operator D1, and may include, for example, increasing sales of products sold by business operator D1, increasing the number of customers for services provided by business operator D1, completing prototypes of products sold by business operator D1, meeting delivery deadlines for products provided by business operator D2 to business operator D1, etc. The number of provisional proposals extracted by the information processing device 11 in step S13 may be one or more.

[0170] The requirements are the conditions and constraints that must be considered in order to realize the provisional proposal. These requirements may be understood as the functional requirements of the work, suitability assessment, suitability, checkpoints for passing (approving) or rejecting the work proposal, and criteria for prioritizing the work proposal. Prioritization may be based on the history of previously implemented work proposals, with proposals that received relatively higher evaluations for their implementation results being given a relatively higher priority.

[0171] The requirements can also be understood as functional requirements that the proposed external products or services or externally generated AI must satisfy. The requirements may include, for example, that there are no competing products from other companies that sell the products sold by company D1, that the costs necessary to increase the number of customers for the services provided by company D1 are within budget, that company D1 has a technical developer (user 16) with experience in completing prototypes of the products sold by company D1, and that company D2 has a system in place to meet the delivery deadlines for the products provided to company D1.

[0172] The information processing device 11 can use a preliminary proposal and first auxiliary information stored in the auxiliary storage device 21 to extract the necessary conditions. The first auxiliary information may include information on business proposals previously created by the information processing device 11, budget information for the business of business operator D1, information on technical developers at business operator D1, and the history of goods delivered by business operator D2 at the request of business operator D1 in the past.

[0173] The information processing device 11 may extract, that is, create, business proposals in step S14, following step S13. Business proposals may be identified as issues that should be solved in the business of business operator D1. The information processing device 11 may extract as business proposals those that meet the necessary conditions from among the provisional proposals extracted in step S13. Alternatively, the information processing device 11 may extract as business proposals those that meet the necessary conditions and do not deviate from the judgment criteria from among the provisional proposals extracted in step S13.

[0174] Furthermore, the information processing device 11 may create options corresponding to the extracted business proposals in step S14. The information processing device 11 may, for example, create options corresponding to the business proposals by referring to "history information of options created in the past" stored in the auxiliary storage device 21. Options represent means for realizing the business proposals.

[0175] The information processing device 11 may create options corresponding to business proposals. The options created by the information processing device 11 may include multiple use cases, such as a first option that emphasizes executing the business proposal according to the judgment criteria, a second option that emphasizes maintaining the current state of the business, a third option that emphasizes advancing the currently running business to the next stage, a fourth option that emphasizes making the state of the business approximate past performance, and a fifth option that emphasizes avoiding the risk of business failure. Each of these options has a different significance. The information processing device 11 may create a single option corresponding to a single business proposal, or it may create multiple options corresponding to a single business proposal.

[0176] The information processing device 11 may include a request for inquiry to the user 61 as an option to respond to the business proposal. In this case, the information processing device 11 may generate a draft of the inquiry statement, and its transmission to the second user terminal 13 may be performed on the condition that one or more of the following operations are performed: a transmission operation performed at the first user terminal 12, or an approval operation for transmission by a designated approver. Note that there may be multiple options, for example, three options.

[0177] Furthermore, the information processing device 11 may select a business executor capable of executing the business proposal in step S14. Alternatively, the information processing device 11 may select one or more users from among user 16 or user 61 who are capable of executing the business proposal as the business executor. Furthermore, if the information processing device 11 selects user 16 as the business executor, user 16 may be a person in a different position within business operator D1, for example, a person in charge of normally executing the business proposal, a supervisor in a position to guide the person in charge in the course of business, a developer who experimentally executes the business proposal, etc.

[0178] The information processing device 11 may refer to the second auxiliary information stored in the auxiliary storage device 21 in step S14 and select a user to be the business executor. The second auxiliary information may include the user selected in a previously created business proposal, the results of executing the business proposal, the assigned tasks of user 16 at business operator D1, the number of years of experience of user 16 in those tasks, the assigned tasks of user 61 at business operator D2, the number of years of experience of user 61 in those tasks at business operator D2, etc. The information processing device 11 may also display the created business proposal, options, and business executor information on the display 22E of the information processing device 11.

[0179] In step S15, the administrator 15 may visually check the display 22E of the information processing device 11 and confirm the provided information. Furthermore, if there is only one business proposal, the administrator 15 may approve or reject that business proposal. Additionally, if there are multiple business proposals, the administrator 15 may approve one or more business proposals. Furthermore, if there are multiple options corresponding to a business proposal, the administrator 15 may approve one or more of the options.

[0180] The administrator 15 may also input the reasons for approving the business proposal and the reasons for approving one or more options into the information processing device 11. In this way, the information processing device 11 creates the approved business proposal, the approved options, and the information to be provided, including the business executor.

[0181] In step S16, the information processing device 11 may transmit the approved information to the user terminal used by the user who is performing the task. Here, when the information processing device 11 transmits the information to the second user terminal, it is assumed that the administrator 15 operates the input device 22 to approve it. In addition, in step S16, the information processing device 11 may perform the function of a "routing means".

[0182] The information processing device 11 may, in step S17, obtain execution information, including the process and results of executing the business proposal, from either the first user terminal 12 or the second user terminal 13. The execution information may include the executed business proposal, the options executed, the person who executed the business, the date the business proposal was executed, the progress and results of the business execution, the opinions and impressions of the person who executed the business, the evaluation of the supervisor 16A when the subordinate 16B executed the business proposal, etc. The information processing device 11 may also, in step S15, provide the information to user terminals used by users who do not execute business proposals.

[0183] Furthermore, the information processing device 11 may transmit the comprehensive information to the first user terminal 12 in step S14. The user 16 may then perform the operation corresponding to step S15 at the first user terminal 12. In addition, the user 16 may transmit the results of the operation performed at the first user terminal 12 to the information processing device 11 in step S15. Subsequently, the information processing device 11 may create, in step S15, the business proposal approved at the first user terminal 12, the approved options, and the information provided, including the user who will execute the business proposal.

[0184] The information processing device 11 may store the results of the various processes performed in steps S11 to S17 in the auxiliary storage device 21. Furthermore, the information processing device 11 may use the various information and the results of the various processes acquired in steps S11 to S17 to perform a learning process to train the learning model stored in the auxiliary storage device 21.

[0185] Furthermore, the information processing device 11 may, in step S11, acquire communication information from the supervisor terminal 12A and the subordinate terminal 12B, and execute the processes from step S12 to step S17 without acquiring communication information from the second user terminal 13. Also, the multiple users utilizing the first user terminal 12 are not limited to supervisors and subordinates; they may be colleagues performing the same tasks or users performing different tasks.

[0186] Figure 7 shows another specific example of an information processing method that can be executed by the information processing device 11. In Figure 7, the information processing device 11 executes step S13 after step S11. The information processing device 11 also executes step S12 after step S14, and step S15 after step S12. In step S12 of Figure 7, the information processing device 11 may perform translation and normalization processing for each of the business proposals, options, and business implementers so that they conform to the items defined in the judgment criteria mentioned above.

[0187] In step S12 of Figure 7, the information processing device 11 may perform at least one of the following processes for each of the business proposals and options: pass, reject, or prioritize. Pass means allowing the use of the business proposal or option, and reject means prohibiting the use of the business proposal or option. Prioritization is the process of assigning a priority to each of the multiple business proposals or multiple options.

[0188] Furthermore, in the information processing method shown in Figures 4 and 7, the information processing device 11 may store the history and logs of the various processes performed in steps S11 to S17 in the auxiliary storage device 21, associating them with each business proposal, each option, and each business executor.

[0189] Furthermore, the information processing system 10 does not need to have one or more user terminals, either the first user terminal 12 or the second user terminal 13. In this case, one or more users, either users 16 or 61, may input user information, communication information, and execution information to the information processing device 11 using the input device 22. Alternatively, one or more users, either users 16 or 61, may use the output device 23 to confirm the provided information.

[0190] Furthermore, the computer constituting the information processing device 11 may be a single computer or a distributed computer composed of multiple devices. The information processing device 11 may also be composed of one or more computers, such as a server, supercomputer, mainframe, or workstation. Moreover, the information processing device 11 may be located outside of business operator D1.

[0191] (Problems with the embodiment) An example of an information processing device for processing communication information is disclosed in Patent Document 1. The communication support device (information processing device) disclosed in Patent Document 1 includes a communication unit, a dialogue unit, and a communication support unit. The communication unit is capable of communicating with an external terminal, and the dialogue unit analyzes user dialogue received from one of the external terminals via the communication unit.

[0192] Furthermore, the communication support unit disclosed in Patent Document 1 identifies keywords that may facilitate communication with the other user based on the analyzed results. The dialogue unit generates device dialogue to promote communication with the other user based on the keywords, and outputs device dialogue to one external terminal via the communication unit to promote community formation with the other user. Patent Document 1 also states that the community may be a company, a medical facility, a government office, a school, etc.

[0193] The inventors of this application recognized that the communication support device described in Patent Document 1 has the problem that it does not take into account the specific judgment criteria of the business to which the user belongs. The objective of this embodiment is to provide an information processing device, an information processing method, a program, and a storage medium that can create business proposals that conform to the judgment criteria of the business.

[0194] (Effects of the embodiment) The information processing device 11 and information processing method can create business proposals that conform to the judgment criteria of business operator D1. In other words, the judgment criteria act as a gate that functions as a governing body or checkpoint. To put it another way, it is possible to eliminate the possibility that business proposals will be created based on sponsorship, such as when business operator D2 is a sponsor of business operator D1, or on promotional conditions imposed by business operator D2. Therefore, it becomes less likely that users 16,61 who acquire business proposals will feel uncomfortable or object, and it can contribute to business governance by business operator D1.

[0195] Furthermore, the information processing device 11 and the information processing method can determine the necessary conditions for executing a business proposal and create information to be provided that includes these conditions. Therefore, users can more easily execute the proposed business, and business efficiency is improved.

[0196] Furthermore, the information processing device 11 and the information processing method make it possible to align user information obtained from a second user terminal 13 used by business operator D2, which is located outside business operator D1, with the judgment criteria. Therefore, it becomes easier to align the information provided by the information processing device 11 with the judgment criteria of business operator D1.

[0197] Furthermore, the information processing device 11 and the information processing method create information that includes the fact that one or more of the users 16 or 61 are the ones performing the tasks. Therefore, it becomes easier for the users who are performing the tasks to execute the task proposals.

[0198] Furthermore, the information processing device 11 and the information processing method create multiple options with different significances as business proposals, and determine the business execution user who is the implementing entity for each of the multiple options. Therefore, it becomes easier for the users who are business executioners to implement the business proposals.

[0199] Furthermore, since the information processing device 11 and information processing method process communication information using the judgment criteria of business operator D1, even if the functionality of business operator D2's second user terminal 13 improves, the accuracy of business proposals at business operator D1 can be maintained at a higher level than the judgment accuracy of the second user terminal 13.

[0200] Furthermore, the information processing device 11 and information processing method function as a protocol converter for converting communication information containing business expressions and noise from business operator D2 into the standard language of business operator D1. In addition, the information processing device 11 and information processing method can suppress the inclusion of information such as advertisements or stealth marketing information from business operator D2 contained in the communication information into the provided information.

[0201] Furthermore, in the information processing device 11 and information processing method, the user who will perform the work and the approver of the work proposal become clear at the stage of creating the provided information and options. Therefore, if an error occurs when the work is performed in accordance with the work proposal, the locus of responsibility (boundary of responsibility) can be clarified. Also, when the work is performed in accordance with the work proposal and results are achieved, the information processing device 11 makes it easier to make judgments on items such as the attribution of work value and consideration, billing, settlement, settlement, incentive design, evidence, responsibility tokens (words indicating the locus of responsibility), infringement judgment, etc., and to utilize these items.

[0202] Furthermore, the information processing device 11 and information processing method make it possible to avoid deviations from the audit, legal, and information system functions within business operator D1 that are included in business operator D1's judgment criteria when creating business proposals from communication information. Therefore, the information processing device 11 and information processing method can create switching costs, or in other words, barriers, to prevent other companies from entering the scope of business operator D1.

[0203] Furthermore, the information processing device 11 and information processing method reveal state transitions such as the log of the audit trail of communication information, the presentation of business proposals, the selection of business proposals and options, the approval of proposed business, the execution of business, and the results of executing the business proposal. In addition, the information processing device 11 and information processing method can store the judgment result of communication information (Gate judgment result) using judgment criteria in the auxiliary storage device 21. Furthermore, the information processing device 11 and information processing method exhibit a checkpoint function that governs and selects multiple externally generated AIs composed of the second user terminal 13 of business operator D2, external services of business operator D2, and proposals from external agents of business operator D2, based on business operator D1's unique judgment criteria.

[0204] (supplementary explanation) In this embodiment, the generation AI used for business support is not limited to that owned or developed by the provider of the information processing device, but also includes configurations that utilize external generation AI services or external specialized AI services contracted or managed by the implementing company via linking means such as APIs.

[0205] This embodiment does not protect the generated AI model itself, the learning algorithm itself, or the internal structure of said generated AI, but rather relates to an information processing device, method, and program that controls information and business progress acquired from an external generated AI or an external expert AI in accordance with the decision criteria and sequence control specific to the implementing company.

[0206] An example of the technical meaning disclosed in this embodiment is as follows: For example, the information processing device 11 is an example of an information processing device and a computer. The processor 19 is an example of a processor. The auxiliary storage device 21 and the media drive device 51 are examples of a storage device and a storage medium. The communication environment management unit 27 and the various information processing units 29 are examples of acquisition means and conversation acquisition unit / log acquisition unit. The communication environment management unit 27, the various information processing units 29 and the requirements processing unit 53 are examples of extraction means and context analysis / requirements extraction unit. The various information processing units 29 are examples of value calculation means, audit log generation means, update means and enterprise parameter management unit.

[0207] The provisional proposal processing unit 52, the business proposal processing unit 54, the requirements processing unit 53, and the artificial intelligence unit 30 are examples of candidate generation means and gate determination means. The communication environment management unit 27, the various information processing units 29, and the artificial intelligence unit 30 are examples of translation and normalization means. The provided information processing unit 55 is an example of presentation means. The processor 80 is an example of an externally generated AI or an external specialized AI. Program 31A is an example of a program.

[0208] Among the history and logs of various processes performed by the information processing device 11 from step S11 to step S17, the logs related to the business of business operator D1 are an example of business logs. Business proposals and options are examples of candidate actions. Information including judgment criteria is an example of judgment criterion information. Judgment criterion information is an example of company parameters.

[0209] The next process refers to the following processing step. In an itemized and attributed data structure, itemization means breaking down unstructured data such as communication information into predetermined items (fields), extracting values ​​corresponding to each item, and storing them as data records. Attributeization means assigning attribute values ​​such as predetermined classification labels, status, classification information, authority, or risk classification to the items or data records and storing them as metadata.

[0210] Furthermore, external specialized AI refers to artificial intelligence that specializes in a specific business area or field of expertise (e.g., manufacturing, legal affairs, accounting, etc.). The difference between externally generated AI and external specialized AI lies in the fact that while externally generated AI can perform general-purpose natural language processing or general-purpose reasoning, external specialized AI is capable of performing specialized reasoning or decision assistance based on a knowledge base, model, or processing procedure in a specific field, as the learning model used by the artificial intelligence department is fine-tuned and subjected to human feedback.

[0211] Furthermore, examples of information supporting compliance with functional requirements include specification information, pricing information, delivery date information, usage conditions, scope of application, past performance, or mapping information showing the correspondence between the functional requirements and the specification information, all of which are acquired in conjunction with the candidate. Additionally, an example of an audit log generation means is a log recording unit that records the "Order_ID," "Action Candidate ID," version of the company parameter, the functional requirements, the basis for compliance with the functional requirements, the gate judgment result, the translated and normalized presentation content, the responsibility token, and user operations (selection and approval) in chronological order.

[0212] Furthermore, "traceable" means that, at a later date, it is possible to track and reproduce, based on audit logs and reference information, who generated and presented which candidates, when, and based on what information, which candidates were selected and approved, whether or not external transmission occurred, and what the execution results were. In addition, classification information refers to classification information used to determine how candidates or processes are handled, such as responsibility classification, authority classification, disclosure classification, or risk classification.

[0213] Furthermore, information disclosure restrictions mean limiting the scope of disclosure of information subject to external transmission or external display through controls such as masking, summarization, anonymization, encryption, or transmission prohibition. In addition, risk classification means information that classifies the type or degree of risk inherent in the candidate or the process (e.g., legal risk, quality risk, delivery risk, cost overrun risk, etc.).

[0214] The term "value calculation target" means that the value (Total_Value) of the business process or transaction that is the subject of the calculation is quantified based on indicators included in the audit log (e.g., number of confirmations, number of rejections, processing time, lead time, or transaction amount), or calculated using a prescribed calculation formula.

[0215] An example of the prescribed calculation formula is as follows:

[0216] Total_Value = Σ(Contribution_Ratio_i × Base_Value_i) ···(Formula 1) Here, "Contribution_Ratio_i" ​​is a coefficient indicating the degree of contribution corresponding to "each contributing entity i", and "Base_Value_i" is the base value corresponding to the business process or transaction in which the contributing entity was involved. Note that the above calculation formula is just one example, and this embodiment is not limited to it; other calculation methods, weightings, or rule-based calculations may be used.

[0217] One example of recording responsibility tokens in association with the basis for generating options and the gate determination results is to store the responsibility tokens, the functional requirements, the basis for conforming to the functional requirements, and the gate determination results as the same audit log record or a cross-referenced record, linked to the "Order_ID" and "Action Candidate ID".

[0218] Furthermore, the information processing device 11 performs exclusive control, such as disabling or hiding the operation interface related to task registration or execution start on the first user terminal 12, until it detects that the input of the business prerequisites and the agreement (approval operation) of the person in charge (administrator or user) regarding them have been completed.

[0219] Furthermore, in the phrase "the audit log generation means stores, in association with at least the functional requirements, the basis for conforming to the functional requirements, the gate determination result, the translated and normalized presentation content, the responsibility token, the user's selection and approval, whether or not external transmission occurred, and the execution result," "to associate" means to store each piece of information in a manner that allows for cross-referencing by assigning the same identifier or reference information.

[0220] Furthermore, a concrete example of "masking or concealing candidates before sending them externally" is the process of limiting information to what is necessary for external transmission by masking, replacing, anonymizing, or encrypting personal names, company names, product names, amounts, quantities, or sensitive internal information contained in communication information or normalization requests.

[0221] Furthermore, in the context of contributing entities, "entity" refers to entities that contributed to each stage of the value calculation process, such as generation, normalization, presentation, approval, and execution (e.g., information processing devices, users, administrators, externally generated AI, external specialized AI, or external businesses), and "contribution" refers to the degree or ratio to which the processing or information provided by the entity in each stage contributed to the establishment of the value calculation object or the generation of value.

[0222] The calculation of consideration means calculating the amount of consideration (e.g., usage fee, commission, or remuneration) related to the value calculation subject, based on the value calculation result or the contribution, according to a predetermined rate, fixed amount, or step-by-step rule. The settlement information means generating the calculated consideration as invoice information, settlement instruction information, or payment / receipt history information, and settling the accounts between the parties concerned. Furthermore, "externally accessible" includes the meaning of being obtainable from the storage medium 51A and obtainable from the information provision computer 50.

[0223] Furthermore, the candidate generation means and / or the gate determination means means one or more of the candidate generation means or the gate determination means. In addition, the acquisition means for acquiring communication information and / or business logs means one or more of the acquisition means for acquiring communication information or business logs.

[0224] Business operator D1 is an example of the first business operator. Business operator D2 is an example of the second business operator. User 16 is an example of the first user and person in charge, and User 61 is an example of the second user and person in charge. Supervisor 16A is an example of an approver. First user terminal 12 is an example of the first terminal and person in charge terminal, and second user terminal 13 is an example of the second terminal and person in charge terminal. Information processing device 11 is an example of a proposal generation entity. Users 16 and 61 are examples of execution entities. In Figures 4 and 7, step S11 is an example of the first step. Step S13 is an example of the second and fourth steps. Step S14 is an example of the third and fifth steps. Step S16 is an example of the sixth step.

[0225] This embodiment discloses the following technical features:

[0226] [Technical Feature 1] An information processing device having a processor that performs processing, The aforementioned processor, A first process for acquiring communication information between a first business operator capable of performing the work and a second business operator located outside of the first business operator and capable of performing the work upon request from the first business operator, A second process involves adapting various pieces of information contained in the aforementioned communication information to the specific judgment criteria of the first business operator in its operations. A third process involves using the communication information that has undergone the second process to create provision information that includes a business proposal related to the business of the first business operator and that can be executed by one or more of the first or second business operators, An information processing device configured to perform the following actions.

[0227] [Technical Feature 2] The information processing apparatus described in technical feature 1, The processor then performs a fourth process to determine the necessary conditions for executing the business proposal (conditions for solving the problem, conditions for clearing business constraints, being in a format that can be processed within the business organization, being feasible within the budget, etc.). An information processing apparatus wherein the third process performed by the processor is configured to create the provided information including the necessary conditions.

[0228] [Technical Feature 3] The information processing apparatus described in technical feature 1, The information processing apparatus is configured such that the second process performed by the processor includes a process of aligning the information from the second business operator, among the various types of information contained in the communication information, with the judgment criteria.

[0229] [Technical Feature 4] The information processing apparatus described in technical feature 1, The processor is configured to perform a fifth process that determines which of the first or second businesses is the entity that will execute the business proposal, An information processing device wherein the third process executed by the processor is configured to create the provided information, including the business executor.

[0230] [Technical Feature 5] The information processing apparatus described in technical feature 4, The processor is configured to perform a sixth process that creates multiple options with different significance as business proposals to be provided to one or more of the first or second businesses. The information processing device is configured such that the fifth process performed by the processor determines each of the business executors corresponding to the plurality of options.

[0231] [Technical Feature 6] The information processing apparatus described in technical feature 1, An information processing device configured such that the first process executed by the processor is to acquire the communication information between a first user belonging to the first business operator and using the first terminal, and a second user belonging to the second business operator and using the second terminal, from the first terminal and the second terminal.

[0232] [Technical Feature 7] A method of information processing performed by a computer, The aforementioned computer, A first process for acquiring communication information between a first business operator capable of performing the work and a second business operator located outside of the first business operator and capable of performing the work upon request from the first business operator, A second process involves adapting various pieces of information contained in the aforementioned communication information to the specific judgment criteria of the first business operator in its operations. A third process involves using the communication information that has undergone the second process to create provision information that includes a business proposal related to the business of the first business operator and that can be executed by one or more of the first or second business operators, An information processing method configured to perform the following.

[0233] [Technical Feature 8] A non-temporary program that causes a computer to perform a process, To the aforementioned computer, A first process for acquiring communication information between a first business operator capable of performing the work and a second business operator located outside of the first business operator and capable of performing the work upon request from the first business operator, A second process involves adapting various pieces of information contained in the aforementioned communication information to the specific judgment criteria of the first business operator in its operations. A third process involves using the communication information that has undergone the second process to create provision information that includes a business proposal related to the business of the first business operator and that can be executed by one or more of the first or second business operators, A program configured to execute [something].

[0234] [Technical Feature 9] A temporary storage medium that stores a temporary program that causes a computer to perform a process, The program is sent to the computer, A first process for acquiring communication information between a first business operator capable of performing the work and a second business operator located outside of the first business operator and capable of performing the work upon request from the first business operator, A second process involves adapting various pieces of information contained in the aforementioned communication information to the specific judgment criteria of the first business operator in its operations. A third process involves using the communication information that has undergone the second process to create provision information that includes a business proposal related to the business of the first business operator and that can be executed by one or more of the first or second business operators, A storage medium configured to execute the program is disclosed. In this embodiment, the program may be understood as a program product. [Industrial applicability]

[0235] This embodiment can be used as an information processing device, an information processing method, and a program for processing communication information acquired from a user. [Explanation of symbols]

[0236] 10... Information processing system, 11... Information processing device, 12... First user terminal, 13... Second user terminal, 16, 61... User, 19... Processor, 31A... Program, D1, D2... Business operator

Claims

1. An information processing device that provides a communication environment in which users can communicate with each other using user terminals, and outputs business support information using information acquired from the communication environment as input, It comprises at least a processor and a memory device, The processor executes the program stored in the memory device, A means for acquiring communication information and business logs related to business operations from the aforementioned communication environment. An extraction means for analyzing conversations contained in the aforementioned communication information to extract business issues and conditions and constraints related to said business issues, Candidate generation means for generating multiple candidates for the next action to be taken that solve the aforementioned business problem, satisfy the aforementioned conditions, and clear the aforementioned constraints, A gate determination means that refers to company parameters, including judgment criterion information defined for each implementing company and stored in the aforementioned storage device, and performs one of the following processes for at least a portion of the multiple action candidates: pass, reject, or prioritize. Translation and normalization means that converts, among the action candidates that have passed or been prioritized by the gate determination means, externally derived expressions, vocabulary, or promotional expressions contained in information obtained from externally generated AI or external specialized AI, business chat between user terminals, or text data entered into a user terminal by a person in charge of performing a task, into structured data (itemized and attributed data structure) that can be immediately processed in the next step in which the selected action candidate is performed, in accordance with the terminology definitions, expression standards, and business format based on the corporate parameters. Presentation means for presenting the action candidates converted by the translation and normalization means to the user terminal, An information processing device that functions as such.

2. An information processing method performed by a computer that has a storage device, provides a communication environment that allows users to communicate with each other using user terminals, and outputs business support information using information acquired from the communication environment as input, The aforementioned computer, The first step is to obtain communication information and work logs related to work from the aforementioned communication environment, The second step involves analyzing the conversations contained in the aforementioned communication information to extract business challenges and the conditions and constraints related to those business challenges. A third step involves generating multiple candidates for the next course of action that solve the aforementioned business problem, satisfy the aforementioned conditions, and clear the aforementioned constraints. A fourth step involves referring to company parameters, which include judgment criteria information defined for each implementing company and stored in the aforementioned storage device, and performing one of the following processes for at least some of the multiple action candidates: pass, reject, or prioritize. Step 5 involves converting the selected action candidates, from among the action candidates that have passed through or been prioritized in Step 4, into structured data (itemized and attributed data structure) that can be immediately processed in the next step where the selected action candidates are performed, in accordance with the terminology definitions, expression standards, and business formats based on the company parameters, which include externally derived expressions, vocabulary, or promotional expressions contained in information obtained from externally generated AI or external specialized AI, business chats between user terminals, or text data entered into user terminals by personnel performing the business. A sixth step involves presenting the action candidates converted in the fifth step to the user terminal, An information processing method that performs the following.

3. A non-temporary program that provides a communication environment equipped with a storage device, enabling users to communicate with each other using user terminals, and causes a computer that outputs business support information to perform processing using information acquired from the communication environment as input, To the aforementioned computer, The first step is to obtain communication information and work logs related to work from the aforementioned communication environment, The second step involves analyzing the conversations contained in the aforementioned communication information to extract business challenges and the conditions and constraints related to those business challenges. A third step involves generating multiple candidates for the next course of action that solve the aforementioned business problem, satisfy the aforementioned conditions, and clear the aforementioned constraints. A fourth step involves referring to company parameters, which include judgment criteria information defined for each implementing company and stored in the aforementioned storage device, and performing one of the following processes for at least some of the multiple action candidates: pass, reject, or prioritize. Step 5 involves converting the selected action candidates, from among the action candidates that have passed through or been prioritized in Step 4, into structured data (itemized and attributed data structure) that can be immediately processed in the next step where the selected action candidates are performed, in accordance with the terminology definitions, expression standards, and business formats based on the company parameters, which include externally derived expressions, vocabulary, or promotional expressions contained in information obtained from externally generated AI or external specialized AI, business chats between user terminals, or text data entered into user terminals by personnel performing the business. A sixth step involves presenting the action candidates converted in the fifth step to the user terminal, A program configured to execute.

4. A non-temporary storage medium equipped with a storage device, providing a communication environment in which users can communicate with each other using user terminals, and storing a non-temporary program that causes a computer to execute processing using information acquired from the communication environment as input and outputting business support information, The program is sent to the computer, The first step is to obtain communication information and work logs related to work from the aforementioned communication environment, The second step involves analyzing the conversations contained in the aforementioned communication information to extract business challenges and the conditions and constraints related to those business challenges. A third step involves generating multiple candidates for the next course of action that solve the aforementioned business problem, satisfy the aforementioned conditions, and clear the aforementioned constraints. A fourth step involves referring to company parameters, which include judgment criteria information defined for each implementing company and stored in the aforementioned storage device, and performing one of the following processes for at least some of the multiple action candidates: pass, reject, or prioritize. Step 5 involves converting the selected action candidates, from among the action candidates that have passed through or been prioritized in Step 4, into structured data (itemized and attributed data structure) that can be immediately processed in the next step where the selected action candidates are performed, in accordance with the terminology definitions, expression standards, and business formats based on the company parameters, which include externally derived expressions, vocabulary, or promotional expressions contained in information obtained from externally generated AI or external specialized AI, business chats between user terminals, or text data entered into user terminals by personnel performing the business. A sixth step involves presenting the action candidates converted in the fifth step to the user terminal, A storage medium configured to execute [a certain action].

Images