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

Abstract

Identify the relationship between patent rights and business, and appropriately manage information on the business. [Solution] The information processing device (1) includes an acquisition unit (11) that acquires patent information, business information which is information within an organization, and confidentiality management level information associated with the business information, a relevance information generation unit (121) that generates relevance information including a score of the relevance between the patent information and the business information and text indicating the relevance by inputting the patent information and the business information into a large-scale language model, and a management unit (13) that manages the relevance information in association with the confidentiality management level information.

Description

[Technical Field]

[0001] The present disclosure relates to an information processing device, an information processing method, a program, and a recording medium. [Background technology]

[0002] There are known techniques for analyzing intellectual property rights such as patents using AI etc. For example, Non-Patent Document 1 discloses a database linking patent rights with potential infringing products. [Prior art documents] [Non-patent literature]

[0003] [Non-Patent Document 1] "IP Define" [online], [Retrieved June 26, 2025], Internet<URL :https: / / ipdefine.com / > Summary of the Invention [Means for solving the problem]

[0004] An information processing device according to one aspect of the present disclosure includes an acquisition unit that acquires patent information, business information that is information within an organization, and secret management level information associated with the business information, a relevance information generation unit that generates relevance information including a score of the relevance between the patent information and the business information and text indicating the relevance by inputting the patent information and the business information into a large-scale language model, and a management unit that manages the relevance information in association with the secret management level information.

[0005] An information processing method according to one embodiment of the present disclosure includes an acquisition step in which one or more processors acquire patent information, business information which is information within an organization, and confidentiality management level information associated with the business information; an association information generation step in which the patent information and the business information are input into a large-scale language model to generate association information including a score of the association between the patent information and the business information and text indicating the association; and a management step in which the association information is managed in association with the confidentiality management level information.

[0006] The information processing device according to each aspect of the present invention may be realized by a computer. In this case, the information processing device program that causes the computer to operate as each part (software element) of the information processing device to realize the information processing device on the computer, and the computer-readable recording medium on which the program is recorded, also fall within the scope of the present invention. [Brief explanation of the drawings]

[0007] [Figure 1] FIG. 1 is a block diagram illustrating a functional configuration of an information processing device according to the present disclosure. [Figure 2] FIG. 10 is a flow diagram illustrating an example of a processing flow by an information processing device according to the present disclosure. [Figure 3] FIG. 2 is a diagram for explaining processing by an information processing device according to the present disclosure. [Figure 4] FIG. 2 is a diagram for explaining processing by an information processing device according to the present disclosure. [Figure 5] FIG. 2 is a diagram for explaining processing by an information processing device according to the present disclosure. [Figure 6] FIG. 10 is a flow diagram illustrating an example of a processing flow by an information processing device according to the present disclosure. [Figure 7] FIG. 10 is a diagram illustrating an example of information presented by an information processing device according to the present disclosure. [Figure 8] FIG. 10 is a flow diagram illustrating an example of a processing flow by an information processing device according to the present disclosure. [Figure 9]FIG. 2 is a diagram for explaining processing by an information processing device according to the present disclosure. [Figure 10] FIG. 2 is a diagram for explaining processing by an information processing device according to the present disclosure. [Figure 11] FIG. 2 is a diagram for explaining processing by an information processing device according to the present disclosure. [Figure 12] FIG. 1 is a block diagram illustrating a functional configuration of an information processing device according to the present disclosure. [Figure 13] FIG. 10 is a flow diagram illustrating an example of a processing flow by an information processing device according to the present disclosure. [Figure 14] FIG. 2 is a diagram for explaining processing by an information processing device according to the present disclosure. [Figure 15] FIG. 2 is a diagram for explaining processing by an information processing device according to the present disclosure. [Figure 16] FIG. 2 is a diagram for explaining processing by an information processing device according to the present disclosure. [Figure 17] FIG. 10 is a diagram illustrating an example of information presented by an information processing device according to the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0008] [Embodiment 1] Hereinafter, an embodiment of the present disclosure will be described in detail with reference to the drawings. For ease of understanding, the background and problems of the present disclosure will be described first, and then the details of the present disclosure will be described.

[0009] <Background and Issues> There are known technologies that use AI and other technologies to analyze intellectual property rights, such as patents. This is due to the widespread need for intellectual property analysis. For example, a patent application often includes inventions related to one or more businesses within a company. However, due to business turnover, the relationships between patents and businesses can change over time. Furthermore, as the number of patents and businesses increases, the total number of combinations between patents and businesses becomes enormous. For this reason, it has generally been difficult to comprehensively identify the relationships between multiple patents and businesses. Furthermore, because some businesses are kept confidential, identifying the relationships between patents and businesses requires appropriate management of the information about those businesses.

[0010] One aspect of the present disclosure aims to provide a technology that can identify the relationship between a patent right and a business and can appropriately manage information about the business.

[0011] <Overview of information processing device 1> As will be described below, the information processing device 1 according to this embodiment: an acquisition unit that acquires patent information, business information that is information within an organization, and secret management level information associated with the business information; a relevance information generation unit that generates relevance information including a score of the relevance between the patent information and the business information and text indicating the relevance by inputting the patent information and the business information into a large-scale language model; a management unit that manages the association information in association with the secret management level information; Equipped with

[0012] In this way, the information processing device 1 according to this embodiment inputs the patent information and the business information into a large-scale language model to generate relevance information including a score of the relevance between the patent information and the business information and text indicating the relevance, and manages the generated relevance information in association with the confidentiality management level information. Therefore, the information processing device 1 according to this embodiment can identify the relationship between a patent right and a business, and can perform information management of the business in an appropriate manner.

[0013] (Configuration of information processing system 100) The configuration of an information processing system 100 according to this embodiment, including an information processing device 1, will be described with reference to FIG. 1. FIG. 1 is a block diagram showing the configuration of the information processing system 100 according to this embodiment. As shown in FIG. 1, the information processing system 100 includes the information processing device 1, one or more business servers (in the example of FIG. 1, a first business server 51, a second business server 52, etc.) communicatively connected to the information processing device 1 via a network N, and a generative model server 60. The information processing device 1, the first business server 51, the second business server 52, and the generative model server 60 may be realized as an integrated configuration as a single information processing device, but will be described below as being separate devices. The network N may be a mobile communication system such as 3G, 4G, 5G, or 6G, LTE (Long Term Evolution), Wi-Fi (registered trademark), or an in-house LAN. The information processing device 1, the first business server 51, the second business server 52, and the generative model server 60 may be devices belonging to different local area networks, as long as they are able to communicate with each other via a wide area network.

[0014] (First business server 51) The first business server 51 (also simply referred to as the first server device 51) manages data related to one or more businesses. As an example, as shown in FIG. 1, the first business server 51 stores business information BI1, which is information related to each of the multiple businesses included in the first business group, in a storage unit (not shown). Here, the business information BI1 may include text information, images, audio, source code, etc. related to the business.

[0015] For example, the business information BI1 may be composed of content data and metadata. The content data may include information about the business content expressed in text, images, or audio. The metadata may include at least one of importance, creator, creation department, date and time, update history, and number of updates. However, the business information BI1 stored in the first business server 51 may not include metadata. In this case, as described below, when the business information BI is acquired by the acquisition unit 11 of the information processing device 1, the content data is structured and metadata is added. In this specification, business information may be more generally referred to as BI. Business information may be information held internally by an organization and used in the activities of the organization. An organization may be a company. An organization may also be an organization such as a department within a company. An organization may also be an organization such as a corporate group to which a company belongs.

[0016] 1, the first business server 51 stores a plurality of pieces of business information BI1 together with a secret management level SL1 (also referred to as secret management level information SL1) indicating the level of secret management of each piece of business information BI1. A specific example of how to assign the secret management level SL1 is not limited to this embodiment, but as an example, Confidentiality level C: Information that is publicly available but is merely kept within the company Confidentiality level B: Confidential information, but information that can be viewed by anyone within the company Confidentiality level A: Confidential information that can only be viewed by a select few within the company. It may be classified as follows.

[0017] When there is a secret management level associated with the business information, such as the business information BI1 stored in the first business server 51, the secret management level is also acquired together with the business information by the information processing device 1. Note that in this specification, the secret management level (secret management level information) may be more generally referred to using the symbol SL.

[0018] (Second business server 52) The second business server 52 (also simply referred to as the second server device 52), like the first business server 51, manages data related to one or more businesses. As an example, as shown in FIG. 1, the second business server 52 stores business information BI2, which is information related to each of multiple businesses included in a second business group, in a storage unit (not shown). Here, the second business group may include businesses different from the first business group described above. Furthermore, like the business information BI1, the business information BI2 may include text information, images, audio, source code, etc. related to the business. An exemplary data structure of the business information BI2 is similar to that of the business information BI1 described above, and therefore a redundant description will be omitted.

[0019] 1, in the second business server 52, no secret management level is associated with each piece of business information BI2, and instead, a secret management level SL2 is set for the second business server 52 itself. In other words, the business information BI2 is not directly associated with a secret management level, but is indirectly associated with the secret management level SL2 of the second business server 52 that stores the business information BI2. Note that a specific example of how to assign the secret management level SL2 does not limit this embodiment, but as an example, similar to the secret management level SL1, Confidentiality Management Level C: Business servers that store information that is only publicly available and kept internally Confidentiality Management Level B: A business server that stores confidential information that can be viewed by anyone within the company. Confidentiality Management Level A: A business server that stores confidential information that can only be viewed by a select few within the company. It may be classified as follows.

[0020] In the case where there is no secret management level directly associated with the business information, such as business information BI2 stored in the second business server 52, but there is an indirectly associated secret management level, as an example, the indirectly associated secret management level is also acquired by the information processing device 1 along with the business information. As another example, the information processing device 1 may be configured to acquire the business information without acquiring the indirectly associated secret management level, and assign a secret management level to the business information by referring to the secret management level information of each server held by the information processing device 1. This processing by the information processing device 1 will be described later.

[0021] (Generative Model Server 60) The generative model server 60 (also referred to as the generative model server device 60) is generally equipped with a generative model GM, and is configured to input model input data IN received from the information processing device 1 into the generative model GM and provide the output of the generative model GM to the information processing device 1 as model output data OUT.

[0022] As shown in FIG. 1 , the generative model server 60 includes a control unit 61, a storage unit 62, and a communication unit 63. The communication unit 63 communicates with devices external to the generative model server 60. As an example, the communication unit 63 communicates with an information processing device 1. The communication unit 63 transmits data supplied from the control unit 61 to the information processing device 1, and supplies data received from the information processing device 1 to the control unit 61. Note that the data received by the communication unit 63 from the information processing device 1 may include model input data IN generated by the information processing device 1. Furthermore, the data provided by the communication unit 63 to the information processing device 1 may include model output data OUT output by the generative model GM based on the model input data IN. Note that the generative model GM is, as an example, a multimodal generative model, and the model input data IN that the generative model GM can accept is not limited to text data but may include image data, audio data, source code, etc.

[0023] The storage unit 62 stores a generative model GM. As an example, the storage unit 62 stores a plurality of parameters that define the generative model GM. As an example, these parameters are parameters that have been learned in advance by machine learning (parameters that have undergone an update process by machine learning), but this does not limit the present embodiment. One or more language models (LMs) can be used as the generative model GM. Furthermore, one or more large-scale language models (LLMs) can be used as the one or more language models, but this does not limit the present embodiment.

[0024] The control unit 61 uses the generative model GM to acquire information generated by the generative model GM. As an example, the control unit 61 acquires model output data OUT output by the generative model GM based on model input data IN received from the information processing device 1. The control unit 61 also provides the model output data OUT to the information processing device 1 via the communication unit 63. Specific processing by the generative model GM will be described later.

[0025] In this embodiment, the first business server 51, the second business server 52, and the generative model server 60 are illustrated as devices separate from the information processing device 1, but this does not limit this embodiment. For example, the control unit 61 provided in the generative model server 60 or the function of the generative model execution unit in the control unit 61 may be configured to be provided in the control unit of the information processing device 1. Furthermore, the generative model GM stored in the memory unit 62 provided in the generative model server 60 may be stored in the memory unit of the information processing device 1, and the generative model GM may be executed by the information processing device 1 itself.

[0026] In addition, in the example shown in Figure 1, two business servers, a first business server 51 and a second business server 52, are exemplified, but this does not limit this embodiment, and the configuration may include three or more business servers, or may include one business server.

[0027] (Information processing device 1) As shown in FIG. 1, the information processing device 1 includes a control unit 10, a storage unit 20, a communication unit 30, and an input / output unit 40.

[0028] (Communication unit 30) The communication unit 30 is an interface for transmitting and receiving data via the network N. As an example, the communication unit 30 receives each piece of business information BI1 and a secret management level SL1 associated with each piece of business information BI1 from the first business server 51. The communication unit 30 also receives each piece of business information BI2 from the second business server 52. The communication unit 30 may also receive a secret management level SL2 from the second business server 52.

[0029] The communication unit 30 also transmits model input data IN generated by the generation unit 12 (described later) to the generative model server 60. The communication unit 30 also receives model output data OUT output by the generative model GM included in the generative model server 60.

[0030] (Input / output section 40) The input / output unit 40 is an interface with an input device that accepts input of data and an output device that outputs data. Examples of input devices include, but are not limited to, a microphone, a camera, an eye-gaze input device, a keyboard, and a touchpad. Examples of output devices include, but are not limited to, a speaker and a liquid crystal display. The input / output unit 40 may also be referred to as a presentation unit.

[0031] (Storage unit 20) The storage unit 20 stores various data referenced by the control unit 10 and various data generated by the control unit 10. At least a part of the various data stored in the storage unit 20 may be organized into a database. In other words, at least a part of the various data stored in the storage unit 20 constitutes a database according to this embodiment. As an example, the storage unit 20 stores the following data as shown in FIG. 1: One or more patent information PIs ·Multiple business information BI · Confidentiality management level SL associated with each business information BI Model input data IN Model output data OUT One or more Related Information RIs The secrecy control level SL associated with each relevant information RI ·Presentation information PI etc. are stored.

[0032] The patent information PI contains information on patent applications or patent rights that are the subject of generating the related information RI, which will be described later. For example, each patent information PI may contain: -Patent application specifications, claims, abstracts, and drawings - Bibliographic information such as inventor, applicant, and filing date of the patent application - Examination history information such as amendments and opinions for the patent application The registered claims and registration date of the patent application etc. This patent information PI may be information managed in the information processing device 1 after the filing of each patent application, or may be information obtained from a patent information server (not shown) or the like. Specific examples of patent information PI will be described later. Note that patent information PI is not limited to the above examples, and may also include information on utility models and designs.

[0033] As an example, the business information BI corresponds to either the business information BI1 or the business information BI2 described above. The acquisition unit 11, which will be described later, receives the business information BI1 and the business information BI2 from the first business server 51 and the second business server 52 via the communication unit 30 and stores them in the memory unit 20. Note that data duplication may occur in the business information. For this reason, the acquisition unit 11 may be configured to eliminate duplication of business information when collecting business information or when registering the collected business information in a database. As an example, when creating a database, if there is data (business information) with matching content, the acquisition unit 11 may be configured to skip registration of the data as duplicate data. Other duplicate explanations of the business information BI will be omitted.

[0034] Each secret management level SL corresponds to either the secret management level SL1 or the secret management level SL2, as an example. The acquisition unit 11, which will be described later, acquires either the secret management level SL1 or the secret management level SL2 from either the first business server 51 or the second business server 52 via the communication unit 30. Alternatively, the acquisition unit 11 refers to the secret management level information of each business server and assigns a secret management level to the business information. Other overlapping explanations regarding the secret management level SL will be omitted.

[0035] The model input data IN is data generated by the generation unit 12, which will be described later, and is data input to the generative model GM provided in the generative model server 60. The model input data IN is not limited to text data, and may include image data, audio data, source code, etc. Specific examples of the model input data IN will be described later. The model output data OUT is data output by the generative model GM provided in the generative model server 60 based on the model input data IN. Specific examples of the model output data OUT will be described later.

[0036] The relevance information RI is information generated by a relevance information generation unit (for example, the first relevance information generation unit 121A) described later, and is information indicating the relevance between the patent information PI and the business information BI. The relevance information RI may include, for example, a score of the relevance between the patent information PI and the business information BI and text indicating the relevance. Specific examples of the relevance information RI will be described later. The presentation information PI is information generated by the provision unit 15 described later, and is information provided to a user of the information processing system 100 via the input / output unit 40. Specific examples of the presentation information PI will be described later.

[0037] (Control unit 10) As shown in FIG. 1, the control unit 10 includes an acquisition unit 11, a generation unit 12, a management unit 13, a filtering unit 14, and a provision unit 15.

[0038] (Acquisition part 11) The acquisition unit 11 acquires patent information PI, business information BI, and secret management level information SL associated with the business information BI. Here, the business information BI is, as an example, information within a certain organization, as described above. As an example, the acquisition unit 11 acquires this information from the memory unit 20. Explanations of the information acquired by the acquisition unit 11 that overlap with the explanations described above will be omitted. Specific examples of this information will be described later.

[0039] (Generation unit 12) The generation unit 12 generates relevance information RI and standard information SI using a generation model GM (for example, a large-scale language model). As an example, the generation unit 12 includes a first relevance information generation unit 121A and a second relevance information generation unit 121B, as shown in FIG. 1. The first relevance information generation unit 121A and the second relevance information generation unit 121B may be collectively referred to simply as a relevance information generation unit. The first relevance information generation unit 121A and the second relevance information generation unit 121B may be realized as different components (components) or as the same component.

[0040] As described above, the generative model GM is provided by the generative model server 60 as an example, but this does not limit the present embodiment. The generative model GM may be provided inside the information processing device 1 (e.g., in the storage unit 20), or may be provided by the generative model server 60 or another external information processing device (e.g., a cloud-based service, etc.). The generative model provided by the external information processing device may be a large language model (LLM) accessible via a communication interface such as an API. Examples of available protocols include, but are not limited to, ChatGPT provided by OpenAI, Claude by Anthropic, and Gemini by Google. In the present embodiment, input and output to and from the generative model GM are performed via API requests and responses, but are not limited to these, and other communication protocols may also be used.

[0041] (First relevance information generation unit 121A) The first relevance information generation unit 121A generates relevance information RI indicating the relevance between the patent information PI and the business information BI acquired by the acquisition unit 11, using a generative model GM (large-scale language model). As an example, the first relevance information generation unit 121A inputs the patent information PI and the business information BI into the generative model GM (large-scale language model) to generate relevance information RI including a score of the relevance between the patent information PI and the business information BI and text indicating the relevance. For example, the first relevance information generation unit 121A inputs model input data IN (also referred to as a prompt) including the patent information PI and the business information BI into the generative model GM via the communication unit 30, and generates the relevance information RI by referring to model output data OUT, which is the output of the generative model GM.

[0042] More specifically, the first relevance information generation unit 121A generates relevance information between one piece of patent information and one piece of business information by inputting a prompt that combines only one piece of patent information included in the patent information PI and only one piece of business information included in the business information PI into the large-scale language model. Specific examples of the model input data IN generated by the first relevance information generation unit 121A, the model output data OUT acquired by the first relevance information generation unit 121A, and the relevance information RI generated by the first relevance information generation unit 121A will be described later.

[0043] (Management Department 13) The management unit 13 manages the relevance information RI generated by the first relevance information generation unit 121A in association with the secret management level SL. The relevance information RI generated by the first relevance information generation unit 121A, The secret management level SL associated with the business information BI acquired by the acquisition unit 11 and referenced when generating the relevance information RI; These are managed in association with each other. Here, the "management" process includes, for example, "storing" to the storage unit 20, "reading" from the storage unit 20, and "updating" in the storage unit 20. Therefore, for example, the management unit 13 may be expressed as being configured to store the relevance information RI generated by the first relevance information generation unit 121A in the storage unit 20 in association with the secret management level SL.

[0044] (Filtering unit 14) The filtering unit 14 identifies business information BI (also referred to as target business information BI) for which relevance information RI is to be generated by filtering a plurality of business information BI stored in the memory unit 20. The identified target business information BI is acquired by the acquisition unit 11, for example, and provided to the above-mentioned first relevance information generation unit 121A. Therefore, the filtering unit 14 may be expressed as filtering the target business information acquired by the acquisition unit 11. The first relevance information generation unit 121A generates relevance information RI indicating the relevance between the patent information PI and the target business information BI using the generative model GM.

[0045] More specifically, the filtering unit 14 The target business information BI is identified by performing filtering with reference to first organization information associated with a plurality of business information BIs stored in the storage unit 20. The target business information BI is identified by performing filtering with reference to second organization information associated with the patent information PI acquired by the acquisition unit 11. Clustering the plurality of business information BIs stored in the storage unit 20 for each technical element, and identifying the business information BI corresponding to the technical element having a higher relevance to the patent information PI as the target business information BI. The filtering unit 14 may perform the following processing. By filtering the business information BI, the filtering unit 14 can appropriately narrow down the business information to be input to the generative model GM, which leads to reduction in cost and processing time. Specific details of the processing by the filtering unit 14 will be described later.

[0046] (Provider part 15) The providing unit 15 provides information including at least a part of the relevance information RI generated by the generating unit 12 and managed by the managing unit 13 to one or more users. The providing unit 15 may provide the information via the communication unit 30 or via the input / output unit 40. As an example, the providing unit 15 generates presentation information PI including at least a part of the relevance information RI with reference to the relevance information RI, and visually presents the presentation information PI to the user via a display provided in the input / output unit 40. For this reason, the providing unit 15 may also be referred to as the presentation unit 15.

[0047] Furthermore, the providing unit 15 changes the content of the presentation information PI in accordance with the secret management level SL associated with the relevance information RI. Acquires authority information of a user (target user) via an acquisition unit 11; According to the secret management level information SL associated with the relevance information RI and the authority information of the target user, at least a part of the relevance information RI included in the presentation information PI to be provided to the target user is deleted or changed. The following process is performed.

[0048] By performing the above processing, the providing unit 15 generates presentation information PI according to the secret management level SL associated with the business information BI and the user's authority, thereby enabling appropriate management according to the secret management level SL of the business information BI.

[0049] (Effects of information processing device 1) As described above, the information processing device 1 An acquisition unit 11 that acquires patent information PI, business information BI, which is information within an organization, and secret management level information SL associated with the business information BI; a relevance information generation unit (for example, a first relevance information generation unit 121A) that generates relevance information RI including a score of the relevance between the patent information PI and the business information BI and text indicating the relevance by inputting the patent information and the business information into a large-scale language model; a management unit 13 that manages the relevance information RI in association with the secret management level information SL; It is equipped with:

[0050] In this way, the information processing device 1 according to this embodiment inputs the patent information and the business information into a large-scale language model to generate relevance information RI including a score of the relevance between the patent information PI and the business information BI and text indicating the relevance, and manages the generated relevance information RI in association with the confidentiality management level information SL. Therefore, the information processing device 1 according to this embodiment can identify the relationship between a patent right and a business, and can perform information management of the business in an appropriate manner.

[0051] (Example 1-1 of processing flow by information processing device 1) Next, a specific example of the flow of processing by the information processing device 1 will be described. Fig. 2 is a flow diagram showing Example 1-1 of the flow of processing by the information processing device 1. This example shows the flow of processing from when the information processing device 1 generates relevance information RI and associates it with a secret management level SL to when it stores it in the storage unit 20.

[0052] (Step S11) In step S11, the acquisition unit 11 acquires patent information PI, business information BI, which is information within the organization, and secret management level information SL associated with the business information BI. The upper part of Fig. 3 shows an example of patent information PI acquired by the acquisition unit 11 in this step. In the example shown in the upper part of Fig. 3, the patent information PI related to a certain patent application is Bibliographic information such as the application number, application date, publication number, control number, applicant, and inventor of the patent application - The specification, claims, abstract, and drawings of the patent application The patent information PI also includes: - Examination history information such as amendments and opinions for the patent application The registered claims and registration date of the patent application It may also include the following:

[0053] On the other hand, the lower part of Fig. 3 shows an example of the business information BI acquired by the acquisition unit 11 in this step and the secret management level SL associated with the business information BI. In the example shown in the lower part of Fig. 3, the business information BI related to a certain business is -Business ID of the business Natural language description of business activities More specifically, the business description includes the following: Business Description: We developed technology to integrate technology A into a carrier-grade network and successfully completed the integration. This resulted in an improvement in network performance, which in turn increased profitability through ~~. Contains:

[0054] Furthermore, the business information BI may be configured to include structured information (structured data) along with or instead of the description in the natural language. Here, the structured information may be generated or configured by the first business server 51 or the second business server 52 and acquired via the communication unit 30, or may be generated or configured by the acquisition unit 11. The acquisition unit 11 performs the following process, for example, to generate the structured information of the business information BI.

[0055] (Step S11-1) The acquisition unit 11 acquires a list of files contained in a predetermined folder and files contained in folders below the predetermined folder from the first business server 51 or the second business server 52. As an example, the acquisition unit 11 acquires a list of paths of these files.

[0056] (Step S11-2) The acquisition unit 11 then converts the file with the PDF (Portable Document Format) extension into a blob (Binary Large Object) and stores it in the storage unit 20. Note that the processing in this step is not limited to such direct blob conversion and storage. For example, in addition to or instead of the blob conversion, processing may be performed using a PDF structuring tool (such as Microsoft Azure's Document Intelligence) with an OCR function and a layout analysis function, or another commercial OCR tool.

[0057] (Step S11-3) Then, the acquiring unit 11 generates structured data from the blobbed data. As an example, the acquiring unit 11 may provide the blobbed data together with a request to a predetermined server that performs structuring processing, and acquire the structured data generated by the server.

[0058] (Step S11-4) Then, the acquisition unit 11 stores the structured data (structured information) in the storage unit 20 as part of the business information BI.

[0059] 3, a secret management level SL (level A in the example shown in the lower part of FIG. 3) is associated with the business information BI. The secret management level SL may be associated with the business information BI in advance in the first business server 51 or the second business server 52, or may be associated with the business information BI after it is acquired from the first business server 51 or the second business server 52.

[0060] As an example, when the acquisition unit 11 acquires business information BI from the second business server 52, the business information BI is not directly associated with a secret management level SL. As an example, the acquisition unit 11 may be configured to refer to server secret management information indicating the secret management level of each business server stored in the memory unit 20, identify the secret management level of the second business server 52, and associate the identified secret management level with the business information BI.

[0061] The process of associating the secrecy management level SL with the business information BI by the acquisition unit 11 is not limited to the above example. For example, - Persons to whom the business information BI is delivered Location of the folder where the business information BI is stored -Header or footer of the file related to the business information BI The confidentiality management level of the business information BI may be specified depending on the business information BI, and the specified confidentiality management level may be associated with the business information BI. Also, the business information BI itself may be configured to include the confidentiality management level SL.

[0062] Furthermore, the method of acquiring business information BI is not limited to the above example, and the inputs each user makes regarding each business in a designated form such as a questionnaire form may be stored in the memory unit 20 as business information BI, and the acquisition unit 11 may acquire the business information BI.

[0063] (Step S121) Next, in step S121, the first relevance information generation unit 121A generates relevance information RI indicating the relevance between the patent information PI and the business information BI acquired by the acquisition unit 11, using the generative model GM. As an example, the first relevance information generation unit 121A inputs the patent information PI and the business information BI into a large-scale language model, thereby generating relevance information RI including a score of the relevance between the patent information PI and the business information BI and text indicating the relevance. In this example, as shown in FIG. 2, step S121 includes steps S1211 to S1213.

[0064] (Step S1211) In step S1211, the first relevance information generation unit 121A generates model input data IN (also called a prompt) including the patent information PI and the business information BI. As an example, the first relevance information generation unit 121A generates model input data IN (prompt) that combines only one piece of patent information included in the patent information PI and only one piece of business information included in the business information BI. The upper part of Figure 4 shows an example of the model input data IN generated by the first relevance information generation unit 121A in this step.

[0065] In the example shown in the upper part of Figure 4, the model input data IN has the following directive ID11: Please evaluate whether the following <Business Information> and <Patent Information> are relevant. When evaluating, please follow the <Notes> below. The model input data IN also includes the following as <Notes> (IND12 in the upper part of Figure 4): The evaluation results will be expressed as a number ranging from 0 (no relevance at all) to 100 (close relevance). The reason for the rating should be explained in approximately 200 characters. Evaluation will be based on the following <Evaluation Criteria>. Contains:

[0066] The model input data IN is the evaluation criterion (IND13 in the upper part of Figure 4), <Evaluation criteria> a) Evaluate the relevance of technology to show whether the technical field of the <patent information> and the technical field of the <business information> are related. b) Evaluate whether the process by which the invention described in the <Patent Information> solves the problem contributes to the explanation described in the <Business Information>. c) Evaluate the applicability of the applied technology, which indicates whether the technology applying the invention described in the <Patent Information> can be implemented in the business described in the <Business Information> in the future. d) The final evaluation will be determined by comprehensively considering the results of evaluations a), b), and c). Contains:

[0067] Furthermore, the model input data IN includes business information BI and patent information PI, as shown in the upper part of Fig. 4. The business information BI and patent information PI have been described above, so a duplicated explanation will be omitted here.

[0068] In this way, the model input data IN (prompt) generated by the first relevance information generating unit 121A includes, for example, Instructions to evaluate the relevance between the technical fields included in the patent information PI and the technical fields included in the business information BI - Instructions to evaluate whether the means for solving the problem contained in the patent information PI contribute to the explanation contained in the business information BI - Instructions to evaluate whether the technology obtained by applying the inventions included in the patent information PI can be implemented in the future in the businesses included in the business information BI By including the above specific instructions in the model input data IN, it is possible to generate suitable relevance information RI using the generative model GM.

[0069] (Step S1212) Next, in step S1212, the first relevance information generating unit 121A The model input data IN is input to the generative model GM (large-scale language model) of the generative model server 60 via the communication unit 30, The model output data OUT output by the generation model server 60 to which the model input data IN has been input is obtained.

[0070] The lower part of Fig. 4 shows an example of model output data OUT output by the generation model server 60 to which the above model input data IN has been input. As shown in the lower part of Fig. 4, the model output data OUT includes: Numerical value showing the evaluation result (relevance score SC in the lower part of Figure 4) Text explaining the reason for the evaluation (the symbol RE in the bottom row of Figure 4) Contains:

[0071] (Step S1213) Next, the first relevance information generation unit 121A generates relevance information RI indicating the relevance between the patent information PI and the business information BI by referring to the model output data OUT acquired in step S1212. As an example, the first relevance information generation unit 121A generates relevance information RI indicating the relevance between one piece of patent information included in the patent information PI and one piece of business information included in the business information BI. FIG. 5 shows an example of the relevance information RI generated by the first relevance information generation unit 121A in this step. As shown in FIG. 5, the relevance information RI includes, for example, a patent ID, a business ID, a relevance score SC, and text RE indicating the evaluation reason. Here, the text RE may be expressed as text indicating the relevance between the patent information PI and the business information BI. Also, as shown in FIG. 5, the relevance information RI generated by the first relevance information generation unit 121A is associated with the confidentiality management level SL of the business information BI. More specifically, the relevance information RI shown in FIG. 5 is associated with level A shown in the lower part of FIG. 3 as the confidentiality management level SL.

[0072] 5 does not limit the present embodiment, and the relevance information RI may be configured to include at least one of the application number, publication number, and patent number of the patent instead of or together with the patent ID. Also, the relevance information RI may be configured to include the business name (project name) or business division name of the business instead of or together with the business ID.

[0073] (Step S13) Then, in step S13, the management unit 13 manages the relevance information RI generated by the first relevance information generation unit 121A in step S121 in association with the secret management level SL. As an example, the management unit 13 stores the relevance information RI generated by the first relevance information generation unit 121A in association with the secret management level SL in the storage unit 20. The processing by the management unit 13 has been described above, so a duplicated description will be omitted.

[0074] (Example 1-2 of processing flow by information processing device 1) Next, a description will be given of another specific example of the processing flow by the information processing device 1. Fig. 6 is a flow diagram showing example 1-2 of the processing flow by the information processing device 1. This example shows the processing flow in which the information processing device 1 outputs presentation information PI based on relevance information RI.

[0075] (Step S21) In step S21, the acquisition unit 11 acquires a request from a user (a request to provide the relevance information RI or the presentation information PI). The request may be acquired via the communication unit 30 or the input / output unit 40. The request in this step does not have to be explicit. As an example, when the user attempts to access the relevance information RI or the presentation information PI in the information processing system 100, this action can be considered as an implicit request.

[0076] (Step S22) Subsequently, in step S22, the providing unit 15 identifies the authority (access authority, viewing authority) of the user. As an example, in step S21, the user may be identified, and in this step, the authority of the user may be identified by referring to authority information indicating the authority for each user. Although specific examples of the authority of each user do not limit this embodiment, as an example, Permission level 1: Cannot access (view) information at Confidentiality Management Level A, but can access (view) information at Confidentiality Management Level B or C Permission level 2: Can access (view) any information of secret management levels A, B, and C. It may be classified as follows.

[0077] (Step S23) Subsequently, in step S23, the providing unit 15 generates and outputs presentation information PI according to the authority of the user identified in step S22 and the confidentiality management level SL of the related information RI for which provision has been requested.

[0078] The upper part of Fig. 7 shows an example of the presented information PI generated by the providing unit 15 and presented via the input / output unit 40 when the user has authority level 2, that is, when the user is a user who can access information of secret management level A. In the example shown in the upper part of Fig. 7, the following information is provided as information about a business related to patent ID: P001: ·Business ID:B001 Relevance score indicating relevance to the patent (ID: P001): 75 Reasons for evaluation of the project ·Business ID:B004 Relevance score indicating relevance to the patent (ID: P001): 55 Reasons for evaluation of the project The above evaluation reasons include a description of the content of the project.

[0079] The lower part of Fig. 7 shows an example of the presented information PI generated by the providing unit 15 and presented via the input / output unit 40 when the user has authority level 1, that is, when the user is a user who cannot access information at secret management level A. In the example shown in the lower part of Fig. 7, the following information is provided as information about a business related to patent ID: P001: Scores 75 and 55 However, there is no description of the business. - You do not have permission to view business information It contains a notice to that effect.

[0080] In this way, the providing unit 15 generates the presentation information PI for a user with authority level 1 by deleting the business ID and the evaluation reason from the presentation information PI for a user with authority level 2 or the relevance information RI included in the presentation information PI. This can also be expressed as a process of deleting or changing at least a part of the relevance information RI to be provided to the target user in accordance with the secret management level information SL associated with the relevance information RI and the authority information of the target user.

[0081] By performing the above processing, the providing unit 15 generates presentation information PI according to the secret management level SL associated with the business information BI and the user's authority, thereby enabling appropriate management according to the secret management level SL of the business information BI.

[0082] (Example 1-3 of the processing flow by the information processing device 1) Next, another specific example of the processing flow by the information processing device 1 will be described. Fig. 8 is a flow diagram showing Example 1-3 of the processing flow by the information processing device 1. Similar to the processing flow example 1-1 described above, this example shows the processing flow up to when the information processing device 1 generates relevance information RI, associates it with the secret management level SL, and stores it in the storage unit 20. In the following, the description of matters similar to the processing flow example 1-1 will be omitted as appropriate, and the description will focus on the points that are different from the processing flow example 1-1.

[0083] (Step S11) In step S11, the acquisition unit 11 acquires patent information PI, business information BI, and secret management level information SL associated with the business information BI. Here, in this example, the acquisition unit 11 may acquire multiple different business information BI and secret management level information SL associated with each business information BI. Note that the timing at which the acquisition unit 11 acquires the multiple business information BI is not limited to this example, and the business information BI may be acquired at approximately the same time, or at different times, such as on different days. Specific examples of patent information PI and each business information BI have been described above, so redundant explanations will be omitted here.

[0084] (Step S14) Next, in step S14, the filtering unit 14 identifies the target business information BI by filtering the multiple business information BIs acquired in step S11, and provides the identified target business information BI to the first relationship information generation unit 121A.

[0085] (Another example of the processing in steps S11 and S14) The processes in steps S11 and S14 are not limited to the above example. As an example, the following processes may be performed.

[0086] (Step S11-1) In step S11, the acquisition unit 11 acquires the patent information PI.

[0087] (Step S14) Next, in step S14, the filtering unit 14 identifies the target business information BI by filtering the multiple business information BI stored in the memory unit 20. Here, the filtering may be performed by referring to the patent information PI or other information acquired in step S11-1. The processing in this step may be expressed as processing of filtering the target business information BI acquired by the acquisition unit 11.

[0088] (Step S11-2) Subsequently, in step S11-2, the acquisition unit 11 acquires the target business information BI identified by the filtering process in step S14. The acquired business information BI is provided to the first relevance information generation unit 121A together with the above-mentioned patent information PI.

[0089] Note that specific filtering processes performed by the filtering unit 14 are not limited to this embodiment, but examples are as follows: Note that the filtering unit 14 may perform the following filtering processes in combination with each other.

[0090] (Filtering processing example 1) As a first example, the acquisition unit 11 may first acquire information on the organization that manages the business indicated by the business information (also referred to as first organizational information DI1) in association with each of a plurality of business information BI. As one example, such information can be acquired from the first business server 51 or the second business server 52. Then, the filtering unit 14 may be configured to identify the target business information BI by performing filtering with reference to the first organizational information DI1 associated with each business information BI.

[0091] The upper part of Fig. 9 shows an example of business information BI and first organizational information DI1 associated with the business information BI. In the example shown in the upper part of Fig. 9, the business information BI "Technology A has been improved..." is associated with "AA Headquarters BB Department," which is the organization in charge of the business, as the department name (first organizational information DI1).

[0092] Furthermore, the acquisition unit 11 may acquire task allocation information associated with the first organizational information DI1. The task allocation information is information relating to task allocation (role allocation), and may be acquired, for example, from the first business server 51 or the second business server 52. Each piece of task allocation information is associated with the first organizational information DI1, and therefore is also information associated with each piece of business information BI. The filtering unit 14 may be configured to identify the target business information BI by performing filtering with reference to the task allocation information associated with each piece of business information BI.

[0093] The lower part of Fig. 9 shows an example of task allocation information BR associated with the first organization information DI1. In the example shown in the lower part of Fig. 9, task allocation information BR "Plans and develops carrier networks and network services" is associated with "AA Headquarters" included in the first organization information DI1.

[0094] (Filtering processing example 2) As a second example, the acquisition unit 11 may acquire organizational information (also referred to as second organizational information DI2) related to the patent indicated by the patent information, in association with the patent information PI, from management data such as invention notifications related to the patent. These management data may be stored in the storage unit 20 separately from each patent information PI, or may be stored in the storage unit 20 as part of each patent information PI. Then, the filtering unit 14 may be configured to identify the target business information BI by performing filtering with reference to the second organizational information DI2 associated with the patent information PI.

[0095] 10 shows an example of second organization information DI2 included in invention notification information associated with patent information PI. In the example shown in Fig. 10, the second organization information DI2 "Affiliation: AA Headquarters EE Department" is included in the invention notification information associated with patent information identified by patent ID: P001 and application number 2025-xxxxxx. Also, the second organization information DI2 "Affiliation: CC Headquarters FF Department" is included in the invention notification information associated with patent information identified by patent ID: P002 and application number 2024-yyyyyy.

[0096] (Filtering processing example 3) As a third example, the filtering unit 14 may be configured to identify the target business information BI by performing filtering with reference to both the first organizational information DI1 and the second organizational information DI2. For example, the filtering unit 14 may be configured to identify, as the target business information BI, business information BI associated with the first organizational information DI1, which includes at least a portion of the second organizational information DI2 associated with the patent information PI. As an example, if the second organizational information DI2 is "AA Headquarters EE Department" and the first organizational information DI1 is "AA Headquarters BB Department," the first organizational information DI1 includes "AA Headquarters," which is part of the second organizational information DI2. In this case, the filtering unit 14 identifies, as the target business information BI, business information BI associated with the first organizational information DI1.

[0097] The filtering unit 14 may also be configured to score the relevance between each piece of first organizational information DI1 and each piece of second organizational information DI2, and identify the target business information BI based on the scoring results. For example, the filtering unit 14 may score multiple pieces of first organizational information DI1 so that the higher the relevance with the second organizational information DI2 "AA Headquarters EE Department" associated with the patent information PI, the higher the score, and identify the business information BI associated with the first organizational information DI1 with the highest score as the target business information BI. Alternatively, the filtering unit 14 may identify the business information BI associated with each of the multiple pieces of first organizational information DI1 with the highest scores as the target business information BI. Note that the scoring process may be performed, for example, using a rule-based algorithm, a predetermined scoring algorithm, or the like.

[0098] As an example, the filtering unit 14 generating an organization relevance score that evaluates the business relevance between the first organization and the second organization by inputting information on the first organization identified by the first organization information DI1 and information on the second organization identified by the second organization information DI2 into a relevance information generation unit (for example, the second relevance information generation unit 121B); Identifying the target business information based on the generated organization relevance score Here, the information on the first organization may be information including a part of the first organization information DI1, or may be information derived from the first organization information DI1. Furthermore, the information on the second organization may be information including a part of the second organization information DI2, or may be information derived from the second organization information DI2.

[0099] Furthermore, as one example, the second relevance information generation unit 121B may be configured to generate the organization relevance score by inputting information about the first organization and information about the second organization into a generative model GM (large-scale language model).

[0100] The information of the first organization and the information of the second organization are input into an encoder model (such as, but not limited to, OpenAI Embeddings API, RoBERTa, or XLNet) corresponding to each natural language to generate a feature vector V1 and a feature vector V2 representing the information of each organization; The inter-organizational relevance score is calculated by comparing the generated feature vectors V1 and V2. Here, the comparison between the feature vector V1 and the feature vector V2 may be performed using cosine similarity, Euclidean distance, Pearson correlation, or the like.

[0101] (Filtering processing example 4) As a fourth example, the filtering unit 14 may be configured to cluster (classify) the multiple business information BIs by technology element and identify the business information BI corresponding to the technology element that is more highly related to the patent information PI as the target business information BI. The upper part of FIG. 11 shows an example of business information BIs classified into at least one of multiple technology elements by the clustering performed by the filtering unit 14. In the example shown in the upper part of FIG. 11, business information B002 and B005 are classified only as element technology X, business information B003 and B004 are classified as element technology Y, and business information B001 is classified as both element technology X and element technology Y. As an example, if the patent information PI includes both technology elements X and Y, the filtering unit 14 identifies business information B001 as the target business information BI.

[0102] The clustering (classification) may be performed in units of classification symbols such as the patent classification FI (File Index) or IPC (International Patent Classification). The clustering (classification) process may be performed using, for example, a K-means algorithm or a hierarchical clustering process. The clustering (classification) process may use not only an unsupervised classification model (clustering model), but also a supervised classification model that has undergone learning, or a generative model such as a large-scale language model.

[0103] (Filtering processing example 5) As a fifth example, the filtering unit 14 may generate (embed) a feature vector BIV by inputting each of the plurality of business information BI into an encoder model corresponding to a natural language (for example, but not limited to, OpenAI Embeddings API, RoBERTa, or XLNet). Similarly, the filtering unit 14 may generate the feature information PI as a feature vector PIV. In other words, the filtering unit 14 may build a database including feature vectors for each of the business information BI and the patent information PI.

[0104] Then, the filtering unit 14 may be configured to identify the target business information BI by comparing a feature vector BIV representing each of the plurality of business information BI with a feature vector PIV representing the patent information PI. The lower part of Fig. 11 schematically shows the end points of the feature vectors for each of the plurality of patent information P001 and P002 and the plurality of business information B001 and B002 in the feature space, which are configured by the filtering unit 14.

[0105] As a more specific process, the filtering unit 14 may identify, as the target business information BI, the business information BI represented by a feature vector BIV whose cosine similarity with the feature vector PIV representing the patent information PI is greater than a predetermined threshold. Here, Euclidean distance, Pearson correlation, etc. may be used together with or instead of the cosine similarity. Note that the feature vectorization of the business information BI and the patent information PI can be performed using a predetermined embedding model.

[0106] Furthermore, when the filtering unit 14 creates a database of the business information BI and the patent information PI, a structured database such as a knowledge graph may be constructed in addition to or instead of the above-mentioned feature vectors (in other words, distributed representations). Furthermore, when converting the patent information PI into a feature vector, processing may be performed according to the structured format of the patent information PI. As an example, feature vectorization may be performed for each item (topic unit) such as "prior art," "problem," and "embodiment." In this configuration, a certain patent information PI is represented by a collection of feature vectors for each of the "prior art," "problem," "embodiment," etc., contained in the patent information PI.

[0107] (Filtering processing example 6) As a sixth example, the acquisition unit 11 may acquire importance information indicating the importance of the business indicated by each of the plurality of business information BIs, in association with the plurality of business information BIs. As one example, such information can be acquired from the first business server 51 or the second business server 52. Then, the filtering unit 14 may be configured to refer to the importance information indicating the importance of each of the plurality of business information BIs, and identify business information having a higher importance as the target business information.

[0108] The importance information may be pre-associated with each business information BI, or may be configured to collect a history of communication related to the business information BI, and assign a higher importance (or treat the information as more important) to information that includes a higher-ranking official in the destination of the communication (email To or Cc, chat room member). For example, such communication history can be acquired from the first business server 51 or the second business server 52. Assigning importance may be performed by the filtering unit 14, or may be performed by another information processing device, etc.

[0109] The filtering unit 14 may also be configured to acquire information on the date and time when the business information BI was updated or viewed, and assign a higher importance to information that has been updated or viewed most recently (or treat it as information with a higher importance). Such information on the date and time when the business information BI was updated or viewed can also be acquired from the first business server 51 or the second business server 52, for example.

[0110] As described above, in this step, the filtering unit 14 filters the business information BI, thereby enabling the business information BI to be input to the generative model GM to be suitably narrowed down, which leads to reduced costs and processing time.

[0111] (Steps S1211, S1212, S1213, S13) Steps S1211, S1212, S1213, and S13 are the same as those in the above-described example 1-1 of the processing flow, and therefore redundant explanations will be omitted.

[0112] (Additional Note 1 Regarding the Relevance Information Generation Unit) In the above filtering process example 5, the filtering unit 14 performs processing by referring to the feature vector BIV representing each piece of business information BI and the feature vector PIV representing the patent information PI, but the use of the feature space and feature vectors is not limited to the above example.

[0113] As an example, the first relevance information generating unit 121A The patent information PI and the business information BI are input into an encoder model (such as, but not limited to, OpenAI Embeddings API, RoBERTa, or XLNet) corresponding to each natural language, thereby generating a feature vector PIV representing each of the patent information PI and a feature vector BIV representing each of the business information BI; By comparing the generated feature vector PIV with the feature vector BIV, a score of the relevance between the patent information PI and the business information BI is calculated. The following processing may be performed.

[0114] (Additional Note 2 Regarding the Relevance Information Generation Unit) In the above description, the filtering unit 14 filters the business information BI, but this does not limit the present embodiment. A relevance score (also called the first score) between the patent information PI and the business information BI; A relevance score (second score) between the second organization information DI2 associated with the patent information PI and the first organization information DI1 associated with the business information BI; and then calculate an overall score of the first score and the second score. Here, the first score and the second score may be calculated using the generative model GM (large-scale language model) as described above, or may refer to a feature vector. Furthermore, the overall score calculated in this manner may be used as relevance information RI.

[0115] As an example, the first relevance information generating unit 121A · Relevance score between patent information PI and business information BI (first score) = 85 Relevance score between the first organizational information DI1 and the second organizational information DI2 (second score) = 70 The first score is calculated as follows, and the weighted average of the second score is calculated as follows: The total score of the first score and the second score above = 80 The overall score may also be calculated using a generative model GM (large-scale language model).

[0116] As an example, the first relevance information generation unit 121A can be configured to calculate the above-mentioned overall score for all combinations of patent information PI, business information BI, first organizational information DI, and second organizational information DI2.

[0117] [Embodiment 2] Other embodiments of the present disclosure will be described below. For ease of explanation, the same reference numerals will be used to designate components having the same functions as those described in the above embodiments, and redundant descriptions will be omitted as appropriate.

[0118] FIG. 12 is a block diagram showing the configuration of an information processing system 100A according to this embodiment. As shown in FIG. 12, the information processing device 1A included in the information processing system 100A according to this embodiment differs from the information processing device 1 according to the first embodiment in that the generation unit 12 includes a standard specification identification unit 122 and a correspondence information generation unit 123 in addition to the components included in the information processing device 1 according to the first embodiment. Furthermore, the information processing device 1A according to this embodiment stores standard information SI in the storage unit 20. The standard information SI includes information generated by at least one of the standard specification identification unit 122 and the correspondence information generation unit 123, which will be described later. Specific examples of the standard information SI will be described later. The standard information SI may be a document related to standard technology that is disclosed on the website of a standardization organization or the like.

[0119] (Standard specification section 122) The standard specification unit 122 uses a generative model GM (large-scale language model) or other models to specify a standard ST related to the patent information PI acquired by the acquisition unit 11. Here, the standard ST (or information specifying the standard ST) is an example of information included in the standard information SI. In addition, the standard ST may include, for example: ·ITU-T Recommendation (International Telecommunication Union - Telecommunication Standardization Sector) ·IEEE Standard(Institute of Electrical and Electronics Engineers) ·3GPP(3rd Generation Partnership Project) IETF Standards (Internet Engineering Task Force) ·ETSI(European Telecommunications Standards Institute) Examples of the processing performed by the standard specification unit 122 are described below.

[0120] (Correspondence information generation unit 123) The correspondence information generation unit 123 generates correspondence information CC (also called a claim chart) that indicates the correspondence between the standard ST identified by the standard identification unit 122 and each element of one or more claims included in the patent information PI, using a generative model GM (large-scale language model) or other model.

[0121] (Example of processing flow by information processing device 1A) Next, a specific example of the flow of processing by the information processing device 1A will be described. Fig. 13 is a flow diagram showing an example (example 2-1) of the flow of processing by the information processing device 1A. This example shows the flow of processing up to when the information processing device 1A generates the correspondence information CC.

[0122] (Steps S11 to S13) The processes in steps S11 to S13 shown in FIG. 13 are the same as those in example 1-1 of the process flow described in the first embodiment, and therefore will not be described again.

[0123] (Step S1221) After step S13, in step S1221, the standard specification identification unit 122 determines whether the relevance score included in the relevance information stored in step S13 is within a predetermined range. As an example, the standard specification identification unit 122 determines whether the relevance score is equal to or greater than a predetermined threshold. If the relevance score is equal to or greater than the predetermined threshold (within the predetermined range), the process proceeds to step S1222; if not (No in step S1221), the process returns to step S11.

[0124] (Step S1222) In step S1222, the standard identification unit 122 identifies the standard related to the patent information PI acquired in step S11 using the generative model GM. The upper part of Fig. 14 shows an example of model input data IN generated by the standard identification unit 122 in this step. The model input data IN is supplied to the generative model GM of the generative model server 60 via the communication unit 30.

[0125] In the example shown in the upper part of FIG. 14, the model input data IN is -Description of the role of the generative model GM (IND21 in the top row of Figure 14) - Description of requests to the generative model GM (IND22, IND23 in the top row of Figure 14) Here, the first request item (IND22) to the generative model GM includes the application number (JP Patent Publication No. 20xx-xxxxxx) of the patent information PI acquired in step S11 and the storage location (https: / / www.zzzzzz) of the specification etc. of the patent application identified by the patent information PI. The specification etc. may be data stored in the memory unit 20 or may be data published on the homepage etc. of the patent office of each country.

[0126] In addition, the second request to the generative model GM (IND23) is about the standard ST related to patent information PI, - Name of the standard Standards document number Standards document version It contains instructions to output the

[0127] The bottom part of Fig. 14 shows an example of model output data OUT output by the generative model GM to which the above-mentioned model input data IN is input. In the example shown in the bottom part of Fig. 14, the model output data OUT contains the following information about the standard ST related to the patent information PI: - Name of the standard: AAA Standards document number: SD-001 Standard document version 2.1 These pieces of information are sometimes called standard specification information. As an example, the standard specification unit 122 stores the standard specification information in the storage unit 20 as part of the standard information SI.

[0128] (Step S123) Next, in step S123, the correspondence information generation unit 123 generates correspondence information (claim chart) CC indicating the correspondence between the patent information PI acquired in step S11 and the standard ST identified in step S1222, using the generative model GM. Here, the claim chart includes, as an example, information regarding the correspondence between the standard ST and each element of one or more claims included in the patent information PI.

[0129] 15 shows an example of the model input data IN generated by the correspondence information generating unit 123 in this step. As shown in FIG. 15, the model input data IN is -Description of the role of the generative model GM (IND31 in the top row of Figure 15) - Description of requests to the generative model GM (IND32 in Figure 15) Instructions for claim clauses (IND33 in Figure 15) Instructions on how to output the judgment results (IND34 in Figure 15) Contains:

[0130] The request item (IND32) includes, as a first item, information specifying the standard document (SD-001_draft-ver2.1) and the storage location (https: / / www.yyyyyy) of the standard document (specification document).

[0131] In addition, the above request items (IND32) include, as a second item, information specifying the patent application indicated by the patent information PI (JP Patent Publication No. 20xx-xxxxxx) and information specifying the claims of the patent application (procedural amendment submitted on xx / yy / 2025), as well as the storage location (https: / / www.zzzzzz).

[0132] Furthermore, the above-mentioned request item (IND32) includes, as the third item, a more specific instruction regarding the correspondence: "Does the invention described in claim 1 fall under the target standard (if a third party were to implement the target standard, would it meet the requirements for patent infringement of the target patent)?"

[0133] In addition, the instructions regarding the clauses of the claims (IND33) include a clause of the target claim, and the instructions regarding the method of outputting the judgment results (IND34) include the following instructions regarding how to output the judgment results: "For each constituent element, please explain which statement in the target standard applies. Please clearly indicate which section of the target standard the quotation is from." along with specific examples.

[0134] 16 shows an example of model output data OUT output by the generative model GM to which the above-mentioned model input data IN is input. In the example shown in FIG. 16, the model output data OUT includes the following as correspondence information (claim chart) CC: Evaluation of the "controller" component A of claim 1, the relevant descriptions in the specifications related to said component, and the correspondence (comparison and evaluation) The same applies to constituent elements B and C.

[0135] The correspondence information CC (claim chart) generated by the correspondence information generating unit 123 is presented to the user via a display provided in the input / output unit 40, for example, and is stored in the storage unit 20 in association with the patent information PI.

[0136] As described above, according to the information processing device 1A of this embodiment, Identifying standards ST related to patent information PI using generative models GM (large-scale language models) or other models; Generate correspondence information CC (claim chart) showing the correspondence between the identified standard ST and each element of one or more claims included in the patent information PI using a generative model GM (large-scale language model) or other model. Therefore, the information processing device 1A can suitably generate, with respect to patent information PI relating to a target patent, correspondence information CC indicating the correspondence between a standard ST related to the patent information PI and each element of one or more claims included in the patent information PI.

[0137] In the above example, a determination is made in step S1221 as to whether the relevance score indicating the relevance between the patent information PI and the business information BI is within a predetermined range, and if the relevance score is within the predetermined range, steps S1222 and S123 are performed to identify the standard ST and generate a claim chart. However, this does not limit the present embodiment. As an example, step S1221 may be omitted, and the standard ST may be identified and a claim chart may be generated in steps S1222 and S123 regardless of whether the relevance score is within the predetermined range.

[0138] (Display of transition information) The following describes the presentation information PI generated by the providing unit 15 included in the information processing device 1A according to this embodiment. The standard ST identified by the standard identification unit 122, and the standard ST (the name of the standard, the standard document number, and the version of the standard document) and may present the presentation information PI including the above to the user via a display provided in the input / output unit 40. The response information CC (complaint chart) generated by the response information generation unit 123 and presenting the information to the user via a display provided in the input / output unit 40.

[0139] The acquisition unit 11 may be configured to acquire at least one of past business information BI and previously generated relevance information RI. Generate transition information indicating the transition of a business related to the patent information PI by referring to at least one of past business information BI or previously generated related information RI; - Providing the generated transition information to the target user As an example, the providing unit 15 may perform the following process: Identifying one or more business information BIs that were previously related to the target patent information PI by referring to at least one of past business information BIs or previously generated related information RIs; Generate transition information including one or more identified business information BIs (or their business names) in chronological order Here, the transition information generated by the providing unit 15 is provided, for example, in the form of being presented to the user via a display provided in the input / output unit 40. As another example, the transition information may be provided in the form of being transmitted to an external information processing device via the communication unit 30.

[0140] 17 is a diagram showing an example of transition information (presentation information PI) generated by the providing unit 15 and presented via a display provided in the input / output unit 40. In the example shown in FIG. 17, the transition information is Patent Information PI (Patent: 2015-zzzzzz, Invention title: Communication device, communication method) The transition of business related to the patent application identified by is shown. More specifically, the transition of business related to the patent application is as follows: ·IoT business ·Financial business Advertising business Autonomous driving business Space communications business The transition information provided by the providing unit 15 is shown in chronological order. The transition information provided by the providing unit 15 allows the user to visually recognize how the business related to the patent application has changed over time.

[0141] If the transition information generated by the providing unit 15 includes business information BI or its business name that the user does not have permission to view, the providing unit 15 may be configured to delete the business information BI and its business name from the transition information before presenting it to the user. The transition information described above can also be generated by the providing unit 15 included in the information processing device 1 according to the first embodiment.

[0142] [Software implementation example] The functions of the information processing device 1, 1A (hereinafter referred to as the "device") may be realized by a program for causing a computer to function as the device, and by a program for causing a computer to function as each control block of the device (particularly each part included in the control unit 10).

[0143] In this case, the device includes a computer having at least one control device (e.g., a processor) and at least one storage device (e.g., a memory) as hardware for executing the program. The control device and storage device execute the program, thereby realizing the functions described in each of the above embodiments.

[0144] The program may be non-transitory and may be recorded on one or more computer-readable recording media. The recording media may or may not be included in the device. In the latter case, the program may be supplied to the device via any wired or wireless transmission medium.

[0145] In addition, some or all of the functions of each of the control blocks can be realized by logic circuits. For example, integrated circuits in which logic circuits that function as each of the control blocks are formed are also included in the scope of the present disclosure. In addition, the functions of each of the control blocks can also be realized by, for example, a quantum computer.

[0146] Furthermore, each process described in each of the above embodiments may be executed by AI (Artificial Intelligence). In this case, the AI ​​may run on the control device or on another device (for example, an edge computer or a cloud server).

[0147] <Summary> The present disclosure describes at least the following aspects.

[0148] (Configuration 1) an acquisition unit that acquires patent information, business information that is information within an organization, and secret management level information associated with the business information; a relevance information generation unit that generates relevance information including a score of the relevance between the patent information and the business information and text indicating the relevance by inputting the patent information and the business information into a large-scale language model; a management unit that manages the association information in association with the secret management level information; An information processing device comprising:

[0149] (Configuration 2) The relevance information generation unit generates relevance information between one piece of patent information and one piece of business information by inputting a prompt that combines only one piece of patent information included in the patent information and only one piece of business information included in the business information into the large-scale language model. 2. The information processing device according to configuration 1.

[0150] (Configuration 3) The system further includes a filtering unit that filters the business information to be acquired by the acquisition unit. 2. The information processing device according to configuration 1.

[0151] (Configuration 4) The filtering unit The target business information is identified by performing filtering with reference to first organizational information associated with the business information and second organizational information associated with the patent information. 4. The information processing device according to configuration 3.

[0152] (Configuration 5) The filtering unit By inputting information on a first organization specified by the first organization information and information on a second organization specified by the second organization information into the relevance information generation unit, an organization relevance score is generated that evaluates the business relevance between the first organization and the second organization, and the target business information is identified based on the organization relevance score. 5. The information processing device according to configuration 3 or 4.

[0153] (Configuration 6) The filtering unit Clustering a plurality of pieces of business information by technical element, and identifying business information corresponding to technical elements that are more highly related to the patent information as the target business information. 4. The information processing device according to any one of configurations 3.

[0154] (Configuration 7) The filtering unit The target business information is identified by comparing a feature vector representing each of the plurality of business information with a feature vector representing the patent information. 4. The information processing device according to any one of configurations 3.

[0155] (Configuration 8) The relevance information generation unit The patent information and the business information are input into an encoder model to generate a feature vector representing each of the patent information and a feature vector representing each of the business information, and the generated feature vectors are compared to generate the score. 8. The information processing device according to any one of configurations 1 to 7.

[0156] (Configuration 9) The filtering unit By referring to importance information indicating the importance of each of the plurality of pieces of business information, the business information having a higher importance is identified as the target business information. 4. The information processing device according to configuration 3.

[0157] (Configuration 10) The system further includes a standard identification unit that identifies standards related to the patent information using a large-scale language model. 10. The information processing device according to any one of configurations 1 to 9.

[0158] (Configuration 11) The system further includes a correspondence information generation unit that generates correspondence information indicating correspondence between the standard identified by the standard identification unit and each element of one or more claims included in the patent information using a large-scale language model. 11. The information processing device according to configuration 10.

[0159] (Configuration 12) a providing unit that provides information including at least a part of the relationship information managed by the managing unit to one or more users; The providing unit deletes or modifies at least a part of the relevance information to be provided to the target user in accordance with secret management level information associated with the relevance information and authority information of the target user. 12. The information processing device according to any one of configurations 1 to 11.

[0160] (Configuration 13) The providing unit By referring to at least one of past business information and previously generated related information, transition information indicating transitions in a business related to the patent information is generated, and the generated transition information is provided to the target user. 13. The information processing device according to configuration 12.

[0161] (Configuration 14) The prompt may include: The instruction to evaluate the relevance of the technical field contained in the patent information with the technical field contained in the business information is included. 3. The information processing device according to configuration 2. (Configuration 15) The prompt may include: An instruction to evaluate whether the means for solving the problem contained in the patent information contributes to the explanation contained in the business information is included. 3. The information processing device according to configuration 2.

[0162] (Configuration 16) The prompt may include: The patent information includes instructions to evaluate whether the technology obtained by applying the invention contained in the patent information can be implemented in the business contained in the business information in the future. 3. The information processing device according to configuration 2.

[0163] (Configuration 17) one or more processors an acquisition step of acquiring patent information, business information that is information within an organization, and secret management level information associated with the business information; a relevance information generating step of generating relevance information including a score of the relevance between the patent information and the business information and text indicating the relevance by inputting the patent information and the business information into a large-scale language model; a management step of managing the association information in association with the secret management level information; An information processing method comprising:

[0164] (Configuration 18) A program for causing a computer to function as the information processing device according to configuration 1, the program causing a computer to function as the acquisition unit, the relevance information generation unit, and the management unit. (Configuration 19) A computer-readable recording medium having the program according to configuration 18 recorded thereon.

[0165] The present disclosure is not limited to the above-described embodiments, and various modifications are possible within the scope of the claims. Embodiments obtained by appropriately combining the technical means disclosed in different embodiments are also included in the technical scope of the present disclosure.

[0166] Furthermore, this disclosure identifies the relationship between patent rights and business, and enables appropriate information management of the business. Therefore, this disclosure can contribute to achieving Goal 9 of the Sustainable Development Goals (SDGs), "Build resilient infrastructure, promote inclusive and sustainable industrialization, and promote innovation and resilience." [Explanation of symbols]

[0167] 100,100A ···Information Processing System 1,1A Information processing device 10 Control section 11...Acquisition part 12...Generation section 121A: First relevance information generation unit 121B Second relevance information generation unit 122...Standard Specification Department 123 Correspondence information generation unit 13...Management Department 14 Filtering section 15...Providing Department

Claims

1. an acquisition unit that acquires patent information, business information that is information within an organization, and secret management level information associated with the business information; a relevance information generation unit that generates relevance information including a score of the relevance between the patent information and the business information and text indicating the relevance by inputting the patent information and the business information into a large-scale language model; a management unit that manages the association information in association with the secret management level information; An information processing device comprising:

2. The relevance information generation unit generates relevance information between one piece of patent information and one piece of business information by inputting a prompt that combines only one piece of patent information included in the patent information and only one piece of business information included in the business information into the large-scale language model. The information processing device according to claim 1 .

3. The system further includes a filtering unit that filters the business information to be acquired by the acquisition unit, The filtering unit The target business information is identified by performing filtering with reference to first organizational information associated with the business information and second organizational information associated with the patent information. The information processing device according to claim 1 .

4. The filtering unit By inputting information on a first organization specified by the first organization information and information on a second organization specified by the second organization information into the relevance information generation unit, an organization relevance score is generated that evaluates the business relevance between the first organization and the second organization, and the target business information is identified based on the organization relevance score. The information processing device according to claim 3 .

5. The system further includes a filtering unit that filters the business information to be acquired by the acquisition unit, The filtering unit Clustering a plurality of pieces of business information by technical element, and identifying business information corresponding to technical elements that are more highly related to the patent information as the target business information. The information processing device according to claim 1 .

6. The system further includes a filtering unit that filters the business information to be acquired by the acquisition unit, The filtering unit The target business information is identified by comparing a feature vector representing each of the plurality of business information with a feature vector representing the patent information. The information processing device according to claim 1 .

7. The relevance information generation unit The patent information and the business information are input into an encoder model to generate a feature vector representing each of the patent information and a feature vector representing each of the business information, and the generated feature vectors are compared to generate the score. The information processing device according to claim 1 .

8. The system further includes a filtering unit that filters the business information to be acquired by the acquisition unit, The filtering unit By referring to importance information indicating the importance of each of the plurality of pieces of business information, the business information having a higher importance is identified as the target business information. The information processing device according to claim 3 .

9. The system further includes a standard identification unit that identifies standards related to the patent information using a large-scale language model. The information processing device according to claim 1 .

10. The system further includes a correspondence information generation unit that generates correspondence information indicating correspondence between the standard identified by the standard identification unit and each element of one or more claims included in the patent information using a large-scale language model. The information processing device according to claim 9 .

11. a providing unit that provides information including at least a part of the relationship information managed by the managing unit to one or more users; The providing unit deletes or modifies at least a part of the relevance information to be provided to the target user in accordance with secret management level information associated with the relevance information and authority information of the target user. The information processing device according to claim 1 .

12. The providing unit By referring to at least one of past business information and previously generated related information, transition information indicating transitions in a business related to the patent information is generated, and the generated transition information is provided to the target user. The information processing device according to claim 11.

13. The prompt may include: The instruction to evaluate the relevance of the technical field contained in the patent information with the technical field contained in the business information is included. The information processing device according to claim 2 .

14. The prompt may include: An instruction to evaluate whether the means for solving the problem contained in the patent information contributes to the explanation contained in the business information is included. The information processing device according to claim 2 .

15. The prompt may include: The patent information includes instructions to evaluate whether the technology obtained by applying the invention contained in the patent information can be implemented in the business contained in the business information in the future. The information processing device according to claim 2 .

16. one or more processors an acquisition step of acquiring patent information, business information that is information within an organization, and secret management level information associated with the business information; a relevance information generating step of generating relevance information including a score of the relevance between the patent information and the business information and text indicating the relevance by inputting the patent information and the business information into a large-scale language model; a management step of managing the association information in association with the secret management level information; An information processing method comprising:

17. 2. A program for causing a computer to function as the information processing device according to claim 1, the program causing a computer to function as the acquisition unit, the relevance information generation unit, and the management unit.

18. A computer-readable recording medium on which the program according to claim 17 is recorded.

Images