Information processing systems, information processing methods

Abstract

To provide a novel information processing system that is superior in convenience, usefulness, or reliability. [Solution] This is an information processing system consisting of three components. The first component has the function of receiving and transmitting knowledge graphs and receiving and providing documents. The second component has the function of performing processing using a large-scale language model and transmitting documents according to a prompt chain. The third component has the function of receiving and sharing knowledge graphs, executing prompt chains, and receiving and transmitting documents. In this system, subcomponents explore paths between nodes and create instruction sentences. The large-scale language model generates documents according to the instruction sentences. This makes it possible to generate documents that describe the claims or the relationships between elements.

Description

Technical Field

[0001] One aspect of the present invention relates to an information processing system, an information processing method, or a semiconductor device.

[0002] Note that one aspect of the present invention is not limited to the above technical field. The technical field of one aspect of the invention disclosed in this specification, etc. relates to a thing, a method, or a manufacturing method. Or, one aspect of the present invention relates to a process, a machine, a manufacture, or a composition of matter. Therefore, more specifically, as the technical field of one aspect of the present invention disclosed in this specification, an information processing device, a semiconductor device, a storage device, their driving methods, or their manufacturing methods can be cited as an example.

Background Art

[0003] In recent years, the development of language models using neural networks has been actively carried out, and in particular, large language models (LLMs) have attracted attention. A large language model is a natural language processing model learned using a large amount of data. By using a large language model, for example, a dialogue model that answers user instructions can be realized. In Non-Patent Document 1, GPT-4 (Generative Pre-trained Transformer 4) (registered trademark) is disclosed as a large language model, and ChatGPT is disclosed as a dialogue model.

[0004] By using a large language model, the capabilities of natural language processing models have been greatly improved. On the other hand, due to the enlargement of language models, it is difficult to build and operate a language model on one's own from the aspects of equipment and cost. Therefore, using an external service that provides a language model has become one form of using a language model.

[0005] Furthermore, a document search system has been proposed that takes into account the concept of a document (Patent Document 1). This document search system has a processing unit, and a search graph is created in the processing unit from the search text. The search graph has first to m (m is an integer of 1 or more) search local graphs, and each search local graph consists of two nodes and one edge. The processing unit also performs a search of the first to m sentences against the reference document. The i (i is an integer between 1 and m) sentence has one of the two nodes in the i-th search local graph, or a related word or subordinate word of one of the two nodes, the other of the two nodes in the i-th search local graph, or a related word or subordinate word of the other of the two nodes, and an edge in the i-th search local graph, or a related word or subordinate word of the edge. Then, a score is assigned to the reference document based on the number of sentences from the first to m that are included in the reference document. [Prior art documents] [Patent Documents]

[0006] [Patent Document 1] International Publication No. 2021 / 140406 Brochure [Non-patent literature]

[0007] [Non-Patent Document 1] Summary of ChatGPT / GPT-4 Research and Perspective Towards the Future of Large Language Models, Yiheng Liu et al. (Submitted on 4 Apr 2023, [online], Internet)<URL:https: / / arxiv.org / abs / 2304.01852> [Overview of the project] [Problems that the invention aims to solve]

[0008] One aspect of the present invention aims to provide a novel information processing system that is superior in convenience, usefulness, or reliability. Alternatively, it aims to provide a novel information processing method that is superior in convenience, usefulness, or reliability. Alternatively, it aims to provide a novel information processing system, a novel information processing method, or a novel semiconductor device.

[0009] Furthermore, the description of these problems does not preclude the existence of other problems. Moreover, one aspect of the present invention does not need to solve all of these problems. Other problems will naturally become apparent from the description in the specification, drawings, and claims, and it is possible to extract other problems from the description in the specification, drawings, and claims. [Means for solving the problem]

[0010] (1) One aspect of the present invention is an information processing system having a first component, a second component, and a third component.

[0011] The first component includes the function of receiving a first knowledge graph and transmitting it to a third component, and the function of receiving and providing a first document. The first knowledge graph comprises a first group of nodes, each of which comprises a first field and a second field. The first field stores attributes indicating the scope of the claims. The second field stores a first phrase that has been identified as a component of the claims. The first document is a document that describes the relationship between a first element and a second element, and both the first element and the second element are first phrases.

[0012] The second component has the function of sending the first document to the third component in response to a prompt chain, and the function of performing processing using a large-scale language model. The large-scale language model has the function of generating the first document according to the first instruction.

[0013] The third component includes the functions of receiving the first knowledge graph and sharing it within the third component, executing a prompt chain, and receiving the first document and sending it to the first component. The prompt chain includes the first instruction statement. The third component also includes a first subcomponent and a second subcomponent.

[0014] The first subcomponent has the function of obtaining a first node and a second node from a first group of nodes. The first node stores a first element in a second field, and the second node stores a second element in a second field. The first subcomponent also has the function of searching for a first path between the first node and the second node, and the function of sharing the first path within the third component.

[0015] The second subcomponent has the function of creating the first instruction. The first instruction includes a first instruction and a first path, the first instruction including a procedure to use the first path to generate a first document that describes the relationship between a first element and a second element.

[0016] This allows, for example, the retrieval of a pair of nodes from the first knowledge graph, including a first node that stores a first element in a second field, and a second node that stores a second element in a second field. It also allows for the exploration of paths connecting the first and second nodes. Furthermore, it enables the generation of a first document describing the relationships between the elements constituting the invention described in the claims. Additionally, the first document can be generated via graph data extracted from the first knowledge graph, allowing for a more accurate explanation of the relationships between elements. The first knowledge graph can also be used to explain the basis for the first document. As a result, a novel information processing system with superior convenience, usefulness, and reliability can be provided.

[0017] (2) Another aspect of the present invention is the above-described information processing system, wherein the first component has the function of receiving a second knowledge graph and transmitting it to a third component, and the function of receiving a second document and providing it.

[0018] The second knowledge graph comprises a second group of nodes, each of which comprises a third field, a fourth field, and a fifth field. The third field stores an attribute indicating prior art literature. The fourth field stores a second phrase identified as an element. The fifth field stores either a first phrase or false. The first phrase is a phrase determined to correspond to the second phrase. The second document is a document that explains the relationship between the third and fourth elements, and both the third and fourth elements are second phrases.

[0019] The second component has the capability to send the second document to the third component in response to the prompt chain. The large-scale language model also has the capability to generate the second document according to the second instruction.

[0020] The third component has the functionality to receive the second knowledge graph and share it within the third component, and the functionality to receive the second document and send it to the first component. The prompt chain includes the second instruction.

[0021] The first subcomponent has the function of obtaining the third and fourth nodes from the second group of nodes. The third node stores the third element in the fourth field and the first element in the fifth field. The fourth node also stores the fourth element in the fourth field and the second element in the fifth field. The first subcomponent also has the function of searching for a second path between the third and fourth nodes, and the function of sharing the second path within the third component.

[0022] The second sub-component has a function of creating a second instruction statement. The second instruction statement includes a second instruction and a second path, and the second instruction includes a procedure for generating a second document that explains the relationship between a third element and a fourth element using the second path.

[0023] Thereby, for example, a pair of nodes including a third node that stores a first element in a fifth field and a fourth node that stores a second element in the fifth field can be searched from the second knowledge graph. Also, for example, a path connecting the third node and the fourth node can be searched. Also, a second document describing the relationship between elements described in the prior art document can be generated. Also, the second document can be generated via the graph data retrieved from the second knowledge graph, and the relationship between elements can be described more accurately. Also, the basis of the second document can be explained using the second knowledge graph. As a result, a novel information processing system excellent in convenience, usefulness, or reliability can be provided.

[0024] (3) Also, one aspect of the present invention is the above information processing system in which the first component has a function of receiving and providing a comparison document. The comparison document is a document that explains the difference between the relationship between the first element and the second element and the relationship between the third element and the fourth element.

[0025] The second component has a function of transmitting the comparison document to the third component in response to the prompt chain. Also, the large language model has a function of generating the comparison document according to the third instruction statement.

[0026] The third component has a function of receiving the comparison document and transmitting it to the first component. The prompt chain includes the third instruction statement.

[0027] The second subcomponent has the function of creating a third instruction. The third instruction includes a third instruction, a first document, and a second document, and the third instruction includes a procedure to compare the first document and the second document to generate a comparison document that explains the differences in the relationship between the first and second elements and the relationship between the third and fourth elements.

[0028] This allows us to use the fifth field to associate nodes in the second knowledge graph with nodes in the first knowledge graph. Furthermore, we can use the fifth field to select a node from a second group of nodes in the second knowledge graph that is associated with any of the first group of nodes in the first knowledge graph. We can also select a node from a second group of nodes in the second knowledge graph that is associated with any of the first group of nodes in the first knowledge graph to obtain a pair of nodes. Additionally, we can find a pair of nodes in the first knowledge graph that corresponds to a pair of nodes in the second knowledge graph. For example, we can obtain a pair of nodes from the second knowledge graph containing a third node that stores the first element in the fifth field and a fourth node that stores the second element in the fifth field, and then find a pair of nodes from the first knowledge graph containing a first node that stores the first element in the fourth field and a second node that stores the second element in the fourth field. Furthermore, for example, it is possible to generate a comparative document that explains the difference between a path connecting the first node and the second node and a path connecting the third node and the fourth node. It is also possible to compare the relationships between the elements constituting the invention described in the claims with the relationships between the elements described in the prior art documents. Additionally, by comparing the first knowledge graph and the second knowledge graph, it is possible to generate a comparative document that compares the configuration described in the claims with the configuration described in the prior art documents. As a result, it is possible to provide a novel information processing system that is superior in convenience, usefulness, and reliability.

[0029] (4) Another aspect of the present invention is the above-described information processing system, wherein the first component has the function of receiving a third document and transmitting it to the third component, and the function of receiving and providing a first knowledge graph. The third document is a document describing the claims, and the first knowledge graph is the third document converted into a graph format.

[0030] The second component has the function of receiving the fourth instruction and sending the first inference result to the third component. The large-scale language model also has the function of generating the first inference result according to the fourth instruction.

[0031] The third component includes the function of receiving the third document and sending the fourth instruction to the second component, and the function of receiving the first inference result and sending the first knowledge graph to the first component. The third component also includes a third subcomponent.

[0032] The third subcomponent includes functions for natural language processing, creating a first element list, and sharing the first element list within the third component. The first element list stores the first words and phrases that have been identified as elements in the third document through natural language processing.

[0033] The second subcomponent includes a function to select a first pair of elements sequentially from the first element list, and a function to create a fourth instruction. The fourth instruction includes a fourth instruction, the first pair of elements, and a third document, and the fourth instruction includes a procedure for generating a first inference result from the third document. The first inference result includes an expression that describes a first relationship between one of the first pair of elements.

[0034] The first subcomponent includes the functions of creating first graph data from first inference results, adding the first graph data to a first knowledge graph, and sharing the first knowledge graph within the third component. The first graph data includes a fifth node, a sixth node, and a first edge. The fifth node stores attributes indicating the claims in a first field and one of a first pair of elements in a second field. The sixth node stores attributes indicating the claims in a first field and the other of a first pair of elements in a second field. The first edge includes a sixth field, which stores a representation describing a first relationship.

[0035] This allows, for example, the first element, the second element, and the second relationship between the first and second elements described in the claims to be stored in the second graph data. Furthermore, the second graph data can be added to the first knowledge graph. Additionally, the elements constituting the invention described in the claims, and the relationships between those elements, can be represented in the first knowledge graph. As a result, a novel information processing system with superior convenience, usefulness, and reliability can be provided.

[0036] (5) Another aspect of the present invention is an information processing system comprising a first component having the function of receiving a fourth document and a correspondence list and transmitting them to a third component, and a function of receiving and providing a second knowledge graph. The fourth document is a prior art document, and the correspondence list stores a second term that is determined to be in a correspondence relationship with a first term, associating it with the first term. The second knowledge graph is the fourth document converted into a graph format.

[0037] The second component has the function of receiving the fifth instruction and sending the second inference result to the third component, and the large-scale language model has the function of generating the second inference result according to the fifth instruction.

[0038] The third component includes the function of receiving the fourth document and correspondence list and sending the fifth instruction to the second component, and the function of receiving the second inference result and sending the second knowledge graph to the first component. The third component also includes a third subcomponent.

[0039] The third subcomponent includes functions for natural language processing, creating a second element list, and sharing the second element list within the third component. The second element list stores the second words and phrases that have been identified as elements in the fourth document through natural language processing.

[0040] The second subcomponent includes the function of selecting a second pair of elements sequentially from a second element list, and the function of creating a fifth instruction. The fifth instruction includes a fifth instruction, a second pair of elements, and a fourth document, and the fifth instruction includes a procedure for generating a second inference result from the fourth document. The second inference result includes an expression that describes a third relationship between one of the second pair of elements.

[0041] The first subcomponent has the functionality to create third graph data from the second inference result, to add the third graph data to the second knowledge graph, and to share the second knowledge graph within the third component. The third graph data includes a seventh node, an eighth node, and a second edge. The seventh node stores an attribute indicating prior art literature in the third field and one of the second pair of elements in the fourth field. Also, when one of the second pair of elements is associated with the fifth element in the correspondence list, the fifth element is stored in the fifth field, and when one of the second pair of elements is not associated with any element in the correspondence list, false is stored in the fifth field. The eighth node stores an attribute indicating prior art literature in the third field and the other of the second pair of elements in the fourth field. Furthermore, when the other element of the second pair is associated with the sixth element in the correspondence list, the sixth element is stored in the fifth field, and when the other element of the second pair is not associated with any element in the correspondence list, false is stored in the fifth field. The second edge has a seventh field, which stores a representation that describes the third relationship.

[0042] This allows, for example, the storage of representations describing the third element, the seventh element, and the fourth relationship between the third and seventh elements, as described in the prior art document, in the fourth graph data. Furthermore, the fourth graph data can be added to the second knowledge graph. Additionally, in the third node where the third element is stored in the fourth field, the first element can be stored in the fifth field based on the correspondence list. Furthermore, when the seventh element is not associated with any element in the correspondence list, false can be stored in the fifth field at the ninth node where the seventh element is stored in the fourth field. Additionally, the elements constituting the technology described in the prior art document, and the relationships between those elements, can be represented in the second knowledge graph. As a result, a novel information processing system with superior convenience, usefulness, and reliability can be provided.

[0043] (6) One aspect of the present invention is an information processing system having a first component, a second component, and a third component.

[0044] The first component has the function of receiving a first knowledge graph and transmitting it to a third component, and the function of receiving and providing a first document. The first knowledge graph comprises a first group of nodes, each of which comprises a first field, a second field, and a third field. The first field stores attributes that indicate the specification. The second field stores a first phrase that has been identified as an element, and the third field stores a second phrase or false. The second phrase is a phrase that has been determined to correspond to the first phrase. The first document is a document that describes the relationship between the first element and the second element, and both the first element and the second element are first phrases.

[0045] The second component includes the function of sending the first document to the third component in response to a prompt chain, and the function of performing processing using a large-scale language model. The large-scale language model includes the function of generating the first document according to the first instruction.

[0046] The third component includes the function of receiving the first knowledge graph and sharing it within the third component, the function of executing a prompt chain, and the function of receiving the first document and sending it to the first component. The prompt chain includes the first instruction statement, and the third component includes the first subcomponent and the second subcomponent.

[0047] The first subcomponent has the function of obtaining the first node and the second node from the first group of nodes. The first node stores the first element in the second field and the third element in the third field. The second node stores the second element in the second field and the fourth element in the third field. The first subcomponent also has the function of searching for the first path between the first node and the second node, and the function of sharing the first path within the third component.

[0048] The second subcomponent has the function of creating the first instruction. The first instruction includes a first instruction and a first path, the first instruction including a procedure to use the first path to generate a first document that describes the relationship between a first element and a second element.

[0049] This allows, for example, the retrieval of a pair of nodes from the first knowledge graph, including a first node that stores a third element in a third field, and a second node that stores a fourth element in a third field. It also allows for the exploration of paths connecting the first and second nodes. Furthermore, it enables the generation of a first document describing the relationships between elements described in a specification detailing the invention. Additionally, the first document can be generated via graph data extracted from the first knowledge graph, allowing for a more accurate explanation of the relationships between elements. The first knowledge graph can also be used to explain the basis for the first document. As a result, a novel information processing system with superior convenience, usefulness, and reliability can be provided.

[0050] (7) Another aspect of the present invention is the above-described information processing system, wherein the first component has the function of receiving a second knowledge graph and transmitting it to a third component, and the function of receiving and providing a second document.

[0051] The second knowledge graph comprises a second group of nodes, each of which comprises a fourth field, a fifth field, and a sixth field. The fourth field stores an attribute indicating prior art literature. The fifth field stores a third phrase identified as an element. The sixth field stores a second phrase or false. The second phrase is a phrase determined to correspond to the third phrase. The second document is a document that explains the relationship between the fifth and sixth elements, and both the fifth and sixth elements are third phrases.

[0052] The second component has the capability to send the second document to the third component in response to the prompt chain. The large-scale language model also has the capability to generate the second document according to the second instruction.

[0053] The third component includes the functions of receiving the second knowledge graph and sharing it within the third component, executing a prompt chain, and receiving the second document and sending it to the first component. The prompt chain includes the second instruction statement.

[0054] The first subcomponent has the function of obtaining the third and fourth nodes from the second group of nodes. The third node stores the fifth element in the fifth field and the third element in the sixth field. The fourth node stores the sixth element in the fifth field and the fourth element in the sixth field. The first subcomponent also has the function of searching for a second path between the third and fourth nodes, and the function of sharing the second path within the third component.

[0055] The second subcomponent has the function of creating a second instruction. The second instruction includes a second instruction and a second path, and the second instruction includes a procedure to generate a second document that describes the relationship between the fifth element and the sixth element using the second path.

[0056] This allows, for example, the retrieval of a pair of nodes from the second knowledge graph, including a third node that stores a third element in a sixth field, and a fourth node that stores a fourth element in a sixth field. It also allows for the exploration of paths connecting the third and fourth nodes. Furthermore, it enables the generation of a second document describing the relationships between elements described in a specification detailing the invention. Additionally, the second document can be generated via graph data extracted from the second knowledge graph, allowing for a more accurate explanation of the relationships between elements. The second knowledge graph can also be used to explain the basis for the second document. As a result, a novel information processing system with superior convenience, usefulness, and reliability can be provided.

[0057] (8) Another aspect of the present invention is the above-described information processing system, wherein the first component has a function of receiving and providing a comparison document. The comparison document is a document that explains the difference between the relationship between the first element and the second element, and the relationship between the fifth element and the sixth element.

[0058] The second component has the functionality to send the comparison document to the third component in response to the prompt chain. The large-scale language model also has the functionality to generate the comparison document according to the third instruction.

[0059] The third component has the function of executing a prompt chain and the function of receiving a comparison document and sending it to the first component. The prompt chain includes the third instruction statement.

[0060] The second subcomponent has the function of creating a third instruction. The third instruction includes a third instruction, a first document, and a second document, and the third instruction includes a procedure to compare the first document and the second document to generate a comparison document that explains the differences in the relationship between the first element and the second element, and the relationship between the fifth element and the sixth element.

[0061] This allows us to use the sixth field to associate nodes in the second knowledge graph with nodes in the first knowledge graph. Furthermore, we can use the sixth field to select a node from a second group of nodes in the second knowledge graph that is associated with any of the first group of nodes in the first knowledge graph. We can also select a node from a second group of nodes in the second knowledge graph that is associated with any of the first group of nodes in the first knowledge graph to obtain a pair of nodes. Additionally, we can find a pair of nodes in the first knowledge graph that corresponds to a pair of nodes in the second knowledge graph. For example, we can obtain a pair of nodes from the second knowledge graph containing a third node that stores a third element in the sixth field and a fourth node that stores a fourth element in the sixth field, and then find a pair of nodes from the first knowledge graph containing a first node that stores a third element in the third field and a second node that stores a fourth element in the third field. Furthermore, for example, it is possible to generate a comparative document that explains the difference between a path connecting the first node and the second node and a path connecting the third node and the fourth node. It is also possible to compare the relationships between elements described in the specification describing the content of the invention with the relationships between elements described in the prior art documents. In addition, by comparing the first knowledge graph and the second knowledge graph, it is possible to generate a comparative document that compares the configuration described in the specification describing the content of the invention with the configuration described in the prior art documents. As a result, it is possible to provide a novel information processing system that is superior in convenience, usefulness, and reliability.

[0062] (9) Another aspect of the present invention is the above-described information processing system, wherein the first component has the function of receiving a third document and a correspondence list and transmitting them to the third component, and the function of receiving and providing a first knowledge graph. The third document is a specification describing the content of the invention, the correspondence list stores a first phrase that is determined to be in a correspondence relationship with a second phrase, associating it with the second phrase, and the first knowledge graph is the third document converted into a graph format.

[0063] The second component has the function of receiving the fourth instruction and sending the inference result to the third component. The large-scale language model also has the function of generating the inference result according to the fourth instruction.

[0064] The third component includes the function of receiving the third document and sending the fourth instruction to the second component, and the function of receiving the inference result and sending the first knowledge graph to the first component. The third component also includes a third subcomponent.

[0065] The third subcomponent includes functions for natural language processing, creating an element list, and sharing the element list within the third component. The element list stores the first words that have been identified as elements in the third document through natural language processing.

[0066] The second subcomponent includes a function to select a pair of elements sequentially from an element list, and a function to create a fourth instruction. The fourth instruction includes a fourth instruction, a pair of elements, and a third document, and the fourth instruction includes a procedure for generating an inference result from the third document. The inference result includes an expression that describes a first relationship between one of the pair of elements.

[0067] The first subcomponent has the functionality to create first graph data from inference results, to add the first graph data to the first knowledge graph, and to share the first knowledge graph within the third component. The first graph data includes a fifth node, a sixth node, and edges. The fifth node stores attributes indicating the specification in the first field and one of a pair of elements in the second field. Furthermore, if one of the pair of elements is associated with the seventh element in the correspondence list, it stores the seventh element in the third field, and if one of the pair of elements is not associated with any element in the correspondence list, it stores false in the third field. The sixth node stores attributes indicating the specification in the first field and the other of a pair of elements in the second field. Furthermore, when the other element of a pair is associated with an eighth element in the correspondence list, the eighth element is stored in the third field, and when the other element of a pair is not associated with any element in the correspondence list, false is stored in the third field. The edge has a seventh field, which stores a representation that describes the first relationship.

[0068] This allows, for example, the first element, the ninth element, and the second relationship between the first and ninth elements described in the specification containing the details of the invention to be stored in the second graph data. Furthermore, the second graph data can be added to the third knowledge graph. Also, at the first node where the first element is stored in the second field, the third element can be stored in the third field based on the correspondence list. Furthermore, if the ninth element is not associated with any element in the correspondence list, false can be stored in the third field at the seventh node where the ninth element is stored in the second field. Additionally, the elements constituting the technology described in the specification, and the relationships between those elements, can be represented in the first knowledge graph. As a result, a novel information processing system with superior convenience, usefulness, and reliability can be provided.

[0069] (10) One aspect of the present invention is an information processing method having a first phase. The first phase comprises a first to tenth step.

[0070] In the first step of the first phase, the first component receives the first knowledge graph and the second knowledge graph and sends them to the second component.

[0071] The first knowledge graph comprises a first group of nodes, each of which comprises a first field and a second field. The first field stores attributes indicating the scope of the claims, and the second field stores a first phrase that has been identified as a component of the claims.

[0072] Furthermore, the second knowledge graph comprises a second group of nodes, each of which comprises a third field, a fourth field, and a fifth field. The third field stores attributes indicating prior art documents, the fourth field stores a second phrase identified as an element, and the fifth field stores a first phrase or false. The first phrase is a phrase determined to correspond to the second phrase.

[0073] In the second step of the first phase, the second component receives the first knowledge graph and the second knowledge graph and shares them within the second component. The second component comprises the first subcomponent and the second subcomponent.

[0074] In the third step of the first phase, the first subcomponent obtains the first and second nodes from the second group of nodes, and obtains the third and fourth nodes from the first group of nodes. The first node stores the first element in the fourth field and the second element in the fifth field. The second node stores the third element in the fourth field and the fourth element in the fifth field. The third node stores the second element in the second field. The fourth node stores the fourth element in the second field.

[0075] In the fourth step of the first phase, the first subcomponent searches for a first path between the third node and the fourth node.

[0076] In the fifth step of the first phase, the first subcomponent searches for a second path between the first node and the second node.

[0077] In the sixth step of the first phase, the first subcomponent shares the first and second routes within the second component.

[0078] In the seventh step of the first phase, the second component executes the first prompt chain, which includes the first instruction, the second instruction, and the third instruction.

[0079] The first instruction includes a first instruction and a first path, the first instruction including a procedure to use the first path to generate a first document describing the relationship between the second element and the fourth element.

[0080] The second instruction includes a second instruction and a second path, the second instruction including a procedure to use the second path to generate a second document that describes the relationship between the first element and the third element.

[0081] The third instruction includes the third instruction, the first document, and the second document, the third instruction includes a procedure for comparing the first document and the second document to generate a first comparison document that explains the differences between the relationship between the second element and the fourth element, and between the relationship between the first element and the third element.

[0082] In the eighth step of the first phase, the third component sends the first document, the second document, and the first comparison document to the second component in response to the first prompt chain.

[0083] In the ninth step of the first phase, the second component receives the first document, the second document, and the first comparison document and sends them to the first component.

[0084] In the tenth step of the first phase, the first component receives and provides the first document, the second document, and the first comparison document.

[0085] This allows us to use the fifth field to associate nodes in the second knowledge graph with nodes in the first knowledge graph. Furthermore, we can use the fifth field to select a node from a second group of nodes in the second knowledge graph that is associated with any of the first group of nodes in the first knowledge graph. We can also select a node from a second group of nodes in the second knowledge graph that is associated with any of the first group of nodes in the first knowledge graph to obtain a pair of nodes. Additionally, we can find a pair of nodes in the first knowledge graph that corresponds to a pair of nodes in the second knowledge graph. For example, we can obtain a pair of nodes from the second knowledge graph containing a first node that stores the second element in the fifth field and a second node that stores the fourth element in the fifth field, and then find a pair of nodes from the first knowledge graph containing a third node that stores the second element in the second field and a fourth node that stores the fourth element in the second field. Furthermore, for example, a first comparative document can be generated that explains the difference between a path connecting a third node and a fourth node and a path connecting a first node and a second node. It can also compare the relationships between elements constituting the invention described in the claims with the relationships between elements described in the prior art documents. Additionally, by comparing a first knowledge graph with a second knowledge graph, a first comparative document can be generated that compares the configuration described in the claims with the configuration described in the prior art documents. As a result, a novel information processing method with superior convenience, usefulness, and reliability can be provided.

[0086] (11) One aspect of the present invention is an information processing method having a first phase, wherein the first phase comprises a first to tenth step.

[0087] In the first step of the first phase, the first component receives the third knowledge graph and the second knowledge graph and sends them to the second component.

[0088] The third knowledge graph comprises a third group of nodes, each of which comprises a sixth field, a seventh field, and an eighth field. The sixth field stores attributes indicating the specification, the seventh field stores a third phrase identified as an element, and the eighth field stores a first phrase or false. The first phrase is a phrase determined to correspond to the third phrase.

[0089] The second knowledge graph comprises a second group of nodes, each of which comprises a third field, a fourth field, and a fifth field. The third field stores attributes indicating prior art documents, the fourth field stores a second phrase identified as an element, and the fifth field stores a first phrase or false. The first phrase is a phrase determined to correspond to the second phrase.

[0090] In the second step of the first phase, the second component receives the third knowledge graph and the second knowledge graph and shares them within the second component. The second component comprises the first subcomponent and the second subcomponent.

[0091] In the third step of the first phase, the first subcomponent obtains the first node and the second node from the second group of nodes, and obtains the fifth node and the sixth node from the third group of nodes. The first node stores the first element in the fourth field and the second element in the fifth field. The second node stores the third element in the fourth field and the fourth element in the fifth field. The fifth node stores the fifth element in the seventh field and the second element in the eighth field. The sixth node stores the sixth element in the seventh field and the fourth element in the eighth field.

[0092] In the fourth step of the first phase, the first subcomponent searches for a third path between the fifth and sixth nodes.

[0093] In the fifth step of the first phase, the first subcomponent searches for a second path between the first node and the second node.

[0094] In the sixth step of the first phase, the first subcomponent shares the third and second routes within the second component.

[0095] In the seventh step of the first phase, the second component executes the second prompt chain, which includes the fourth instruction, the second instruction, and the fifth instruction.

[0096] The fourth instruction includes the fourth instruction and the third path, and the fourth instruction includes a procedure for using the third path to generate a third document that describes the relationship between the fifth element and the sixth element.

[0097] The second instruction includes a second instruction and a second path, the second instruction including a procedure to use the second path to generate a second document that describes the relationship between the first element and the third element.

[0098] The fifth instruction includes the fifth instruction, the third document, and the second document, and the fifth instruction includes a procedure to compare the third document and the second document to generate a second comparison document that explains the difference between the relationship between the fifth element and the sixth element and the relationship between the first element and the third element.

[0099] In the eighth step of the first phase, the third component, in response to the second prompt chain, sends the third document, the second document, and the second comparison document to the second component.

[0100] In the ninth step of the first phase, the second component receives the third document, the second document, and the second comparison document and sends them to the first component.

[0101] In the tenth step of the first phase, the first component receives and provides the third document, the second document, and the second comparison document.

[0102] This allows us to use the fifth field to associate nodes in the second knowledge graph with nodes in the third knowledge graph. Furthermore, we can use the fifth field to select a node from a second group of nodes in the second knowledge graph that is associated with any of the third group of nodes in the third knowledge graph. We can also select a node from a second group of nodes in the second knowledge graph that is associated with any of the third group of nodes in the third knowledge graph to obtain a pair of nodes. Additionally, we can find a pair of nodes in the third knowledge graph that corresponds to a pair of nodes in the second knowledge graph. For example, we can obtain a pair of nodes from the second knowledge graph containing a first node that stores the second element in the fifth field and a second node that stores the fourth element in the fifth field, and then find a pair of nodes from the third knowledge graph containing a fifth node that stores the second element in the eighth field and a sixth node that stores the fourth element in the eighth field. Furthermore, for example, a second comparative document can be generated that explains the difference between the path connecting the fifth node and the sixth node and the path connecting the first node and the second node. It can also compare the relationships between elements described in the specification describing the invention with the relationships between elements described in the prior art documents. Additionally, by comparing the third knowledge graph with the second knowledge graph, a second comparative document can be generated that compares the configuration described in the specification describing the invention with the configuration described in the prior art documents. As a result, a novel information processing method with superior convenience, usefulness, and reliability can be provided.

[0103] (12) Another aspect of the present invention is an information processing method having the first phase and the second phase described above. The first phase follows the second phase, and the second phase comprises the first to tenth steps.

[0104] In the first step of the second phase, the first component receives the fourth document and transmits it to the second component. The fourth document is a document containing the claims.

[0105] In the second step of the second phase, the second component receives the fourth document and shares it within the second component. The second component also includes a third subcomponent.

[0106] In the third step of the second phase, the third subcomponent creates the first element list and shares it within the second component. The first element list stores the first words that have been identified as elements in the fourth document through natural language processing.

[0107] In the fourth step of the second phase, the second subcomponent selects the first pair of elements sequentially from the first element list.

[0108] In the fifth step of the second phase, the second subcomponent creates a sixth instruction and sends it to the third component. The sixth instruction includes a sixth instruction, a first pair of elements, and a fourth document, the sixth instruction including a procedure for generating a first inference result from the fourth document. The first inference result includes an expression describing a first relationship between one of the first pair of elements.

[0109] In the sixth step of the second phase, the third component receives the sixth instruction, uses a large-scale language model to generate the first inference result, and sends it to the second component.

[0110] In the seventh step of the second phase, the first subcomponent creates first graph data from the first inference result. The first graph data includes a seventh node, an eighth node, and a first edge. The seventh node stores attributes indicating the claims in a first field and one of a first pair of elements in a second field. The eighth node stores attributes indicating the claims in a first field and the other of a first pair of elements in a second field. The first edge comprises a ninth field, which stores a representation describing the first relationship.

[0111] In the eighth step of the second phase, the first subcomponent adds the first graph data to the first knowledge graph, and shares the first knowledge graph within the second component.

[0112] In the ninth step of the second phase, the second component sends the first knowledge graph to the first component.

[0113] In the tenth step of the second phase, the first component receives and provides the first knowledge graph.

[0114] This allows, for example, the second element, the fourth element, and the second relationship between the second and fourth elements described in the claims to be stored in the second graph data. Furthermore, the second graph data can be added to the first knowledge graph. Additionally, the elements constituting the invention described in the claims, and the relationships between these elements, can be represented in the first knowledge graph. As a result, a novel information processing method with superior convenience, usefulness, and reliability can be provided.

[0115] (13) Another aspect of the present invention is an information processing method having the first phase and the second phase described above. The first phase follows the second phase, and the second phase comprises the first to tenth steps.

[0116] In the first step of the second phase, the first component receives the fifth document and the correspondence list and transmits them to the second component. The fifth document is the specification describing the content of the invention. The correspondence list stores a third term that is determined to be in a corresponding relationship with a first term, associating it with the first term.

[0117] In the second step of the second phase, the second component receives the fifth document and correspondence list and shares them within the second component. The second component also comprises a third subcomponent.

[0118] In the third step of the second phase, the third subcomponent creates a second element list and shares it within the second component. The second element list stores the third words that have been identified as elements in the fifth document through natural language processing.

[0119] In the fourth step of the second phase, the second subcomponent selects the second pair of elements sequentially from the second element list.

[0120] In the fifth step of the second phase, the second subcomponent creates a seventh instruction and sends it to the third component. The seventh instruction includes the seventh instruction, the second pair of elements, and the fifth document, and the seventh instruction includes a procedure for generating the second inference result from the fifth document. The second inference result includes an expression that describes the third relationship between one of the second pair of elements.

[0121] In the sixth step of the second phase, the third component receives the seventh instruction, uses a large-scale language model to generate the second inference result, and sends it to the second component.

[0122] In the seventh step of the second phase, the first subcomponent creates third graph data from the second inference result. The third graph data includes the ninth node, the tenth node, and the second edge. The ninth node stores the specification attribute in the sixth field and one of the second pair of elements in the seventh field. It also stores the seventh element in the eighth field when one of the second pair of elements is associated with the seventh element in the correspondence list, and stores false in the eighth field when one of the second pair of elements is not associated with any element in the correspondence list. The tenth node stores the specification attribute in the sixth field and the other of the second pair of elements in the seventh field. Furthermore, when the other element of the second pair is associated with the eighth element in the correspondence list, the eighth element is stored in the eighth field, and when the other element of the second pair is not associated with any element in the correspondence list, false is stored in the eighth field. The second edge has a tenth field, which stores a representation that describes the third relationship.

[0123] In the eighth step of the second phase, the first subcomponent adds the third graph data to the third knowledge graph, and shares the third knowledge graph within the second component.

[0124] In the ninth step of the second phase, the second component sends the third knowledge graph to the first component.

[0125] In the tenth step of the second phase, the first component receives and provides the third knowledge graph.

[0126] This allows, for example, the fifth element, the ninth element, and the expression describing the fourth relationship between the fifth and ninth elements, as described in the specification detailing the invention, to be stored in the fourth graph data. Furthermore, the fourth graph data can be added to the fourth knowledge graph. Additionally, in the fifth node where the fifth element is stored in the seventh field, the second element can be stored in the eighth field based on the correspondence list. Furthermore, if the ninth element is not associated with any element in the correspondence list, false can be stored in the eighth field in the eleventh node where the ninth element is stored in the seventh field. Additionally, the elements constituting the technology described in the specification, and the relationships between those elements, can be represented in the third knowledge graph. As a result, a novel information processing method with superior convenience, usefulness, and reliability can be provided.

[0127] (14) Another aspect of the present invention is an information processing method having the first to third phases described above. The first phase follows the third phase, the third phase follows the second phase, and the third phase comprises the first to tenth steps.

[0128] In the first step of the third phase, the first component receives the sixth document and the correspondence list and transmits them to the second component. The sixth document is a prior art document, and the correspondence list stores a second term that is determined to be in a correspondence relationship with a first term, associating it with the first term.

[0129] In the second step of the third phase, the second component receives the sixth document and correspondence list and shares them within the second component. The second component also comprises a third subcomponent.

[0130] In the third step of the third phase, the third subcomponent creates a third element list and shares it within the second component. The third element list stores the second word or phrase that has been identified as an element in the sixth document through natural language processing.

[0131] In the fourth step of the third phase, the second subcomponent selects a third pair of elements sequentially from the third element list.

[0132] In the fifth step of the third phase, the second subcomponent creates an eighth instruction and sends it to the third component. The eighth instruction includes an eighth instruction, a third pair of elements, and a sixth document, the eighth instruction including a procedure for generating a third inference result from the sixth document. The third inference result includes an expression describing the fifth relationship between one of the third pair of elements.

[0133] In the sixth step of the third phase, the third component receives the eighth instruction, uses a large-scale language model to generate the third inference result, and sends it to the second component.

[0134] In the seventh step of the third phase, the first subcomponent creates the fifth graph data from the third inference result. The fifth graph data includes the twelfth node, the thirteenth node, and the third edge. The twelfth node stores an attribute indicating prior art literature in the third field and one of the third pair of elements in the fourth field. Furthermore, if one of the third pair of elements is associated with the seventh element in the correspondence list, the seventh element is stored in the fifth field; if one of the third pair of elements is not associated with any element in the correspondence list, false is stored in the fifth field. The thirteenth node stores an attribute indicating prior art literature in the third field and the other of the third pair of elements in the fourth field. Furthermore, when the other element of the third pair is associated with the eighth element in the correspondence list, the eighth element is stored in the fifth field, and when the other element of the third pair is not associated with any element in the correspondence list, false is stored in the fifth field. The third edge has an eleventh field, which stores a representation that describes the fifth relationship.

[0135] In the eighth step of the third phase, the first subcomponent adds the fifth graph data to the second knowledge graph, thereby sharing the second knowledge graph within the second component.

[0136] In the ninth step of the third phase, the second component sends the second knowledge graph to the first component.

[0137] In the tenth step of the third phase, the first component receives and provides the second knowledge graph.

[0138] This allows, for example, the first element, the tenth element, and the sixth relationship between the first and tenth elements described in the prior art document to be stored in the sixth graph data. Furthermore, the sixth graph data can be added to the second knowledge graph. Also, in the first node where the first element is stored in the fourth field, the second element can be stored in the fifth field based on the correspondence list. Furthermore, if the tenth element is not associated with any element in the correspondence list, false can be stored in the fifth field in the fourteenth node where the tenth element is stored in the fourth field. Additionally, the elements constituting the technology described in the prior art document, and the relationships between those elements, can be represented in the second knowledge graph. As a result, a novel information processing method with superior convenience, usefulness, and reliability can be provided. [Effects of the Invention]

[0139] One aspect of the present invention can provide a novel information processing system that is superior in convenience, usefulness, or reliability. Alternatively, it can provide a novel information processing method that is superior in convenience, usefulness, or reliability. Alternatively, it can provide a novel information processing system, a novel information processing method, or a novel semiconductor device.

[0140] Furthermore, the description of these effects does not preclude the existence of other effects. Moreover, one aspect of the present invention does not necessarily have to possess all of these effects. Other effects will naturally become apparent from the description in the specification, drawings, and claims, and it is possible to extract other effects from the description in the specification, drawings, and claims. [Brief explanation of the drawing]

[0141] [Figure 1] Figure 1 is a diagram illustrating the configuration of an information processing system according to an embodiment. [Figure 2] Figures 2(A) and 2(B) illustrate the configuration of the knowledge graph according to the embodiment. [Figure 3] Figures 3(A) and 3(B) illustrate the configuration of information used in the information processing system according to the embodiment. [Figure 4] Figures 4(A) to 4(D) illustrate the configuration of information used in the information processing system according to the embodiment. [Figure 5] Figure 5 is a diagram illustrating the configuration of components used in the information processing system according to the embodiment. [Figure 6] Figures 6(A) and 6(B) illustrate the configuration of the knowledge graph according to the embodiment. [Figure 7] Figures 7(A) and 7(B) illustrate the configuration of information used in the information processing system according to the embodiment. [Figure 8] Figure 8 is a diagram illustrating the configuration of an information processing system according to an embodiment. [Figure 9] Figure 9 is a diagram illustrating the configuration of components used in the information processing system according to the embodiment. [Figure 10] Figures 10(A) and 10(B) illustrate the configuration of information used in the information processing system according to the embodiment. [Figure 11] Figure 11 is a diagram illustrating the configuration of information used in the information processing system according to the embodiment. [Figure 12] Figures 12(A) and 12(B) illustrate the configuration of information used in the information processing system according to the embodiment. [Figure 13] Figures 13(A) and 13(B) illustrate the configuration of the knowledge graph according to the embodiment. [Figure 14] Figures 14(A) and 14(B) illustrate the configuration of information used in the information processing system according to the embodiment. [Figure 15] Figures 15(A) to 15(D) illustrate the configuration of information used in the information processing system according to the embodiment. [Figure 16] Figure 16 is a diagram illustrating the configuration of components used in the information processing system according to the embodiment. [Figure 17] Figure 17 is a diagram illustrating the configuration of an information processing system according to an embodiment. [Figure 18] Figures 18(A) and 18(B) illustrate the configuration of information used in the information processing system according to the embodiment. [Figure 19] Figure 19 is a diagram illustrating the configuration of an information processing device used in an information processing system according to an embodiment. [Figure 20] Figure 20 is a diagram illustrating an information processing method according to an embodiment. [Figure 21] Figure 21 is a diagram illustrating an information processing method according to an embodiment. [Figure 22] Figure 22 is a diagram illustrating an information processing method according to an embodiment. [Modes for carrying out the invention]

[0142] An information processing system according to one aspect of the present invention comprises a first component, a second component, and a third component. The first component includes a function to receive a first knowledge graph and transmit it to the third component, and a function to receive and provide a first document. The first knowledge graph comprises a first group of nodes, each of which comprises a first field and a second field. The first field stores attributes indicating the scope of the claims, and the second field stores a first phrase that has been identified as a component of the claims. The first document is a document that describes the relationship between a first element and a second element, and both the first element and the second element are first phrases. The second component includes a function to transmit the first document to the third component in response to a prompt chain, and a function to perform processing using a large-scale language model. The large-scale language model includes a function to generate the first document according to a first instruction. The third component has the functionality to receive the first knowledge graph and share it within the third component, execute a prompt chain, and receive the first document and send it to the first component. The prompt chain includes the first instruction. The third component has a first subcomponent and a second subcomponent, the first subcomponent having the functionality to obtain the first node and the second node from the first group of nodes. The first node stores the first element in the second field, and the second node stores the second element in the second field. The first subcomponent also has the functionality to search for the first path between the first node and the second node, and to share the first path within the third component. The second subcomponent has the functionality to create the first instruction. The first instruction includes a first instruction and a first path, the first instruction including a procedure to use the first path to generate a first document describing the relationship between a first element and a second element.

[0143] This allows, for example, the retrieval of a pair of nodes from the first knowledge graph, including a first node that stores a first element in a second field, and a second node that stores a second element in a second field. It also allows for the exploration of paths connecting the first and second nodes. Furthermore, it enables the generation of a first document describing the relationships between the elements constituting the invention described in the claims. Additionally, the first document can be generated via graph data extracted from the first knowledge graph, allowing for a more accurate explanation of the relationships between elements. The first knowledge graph can also be used to explain the basis for the first document. As a result, a novel information processing system with superior convenience, usefulness, and reliability can be provided.

[0144] Embodiments will be described in detail with reference to the drawings. However, it will be readily apparent to those skilled in the art that the present invention is not limited to the following description, and that its form and details can be modified in various ways without departing from the spirit and scope of the present invention. Accordingly, the present invention is not to be interpreted as being limited to the contents of the embodiments shown below. In the configuration of the invention described below, the same reference numerals are used in common across different drawings for the same parts or parts having similar functions, and repeated descriptions are omitted.

[0145] In this specification, ordinal numbers such as "first," "second," etc., are used to avoid confusion of components and do not limit the number of components or the order of components (e.g., process order or layering order). Furthermore, even if a term does not have an ordinal number in this specification, an ordinal number may be added in the claims to avoid confusion of components. Even if a term has an ordinal number in this specification, a different ordinal number may be added in the claims. Even if a term has an ordinal number in this specification, an ordinal number may be omitted in the claims.

[0146] In the drawings attached to this specification, components are classified by function and shown as independent blocks in block diagrams. However, in reality, it is difficult to completely separate components by function, and a single component may be involved in multiple functions.

[0147] (Embodiment 1) In this embodiment, an information processing system according to one aspect of the present invention will be described with reference to Figures 1 to 19.

[0148] Figure 1 is a diagram illustrating the configuration of an information processing system according to one embodiment of the present invention.

[0149] Figure 2(A) is a diagram illustrating the configuration of a knowledge graph that can be used in an information processing system according to one embodiment of the present invention, and Figure 2(B) is a diagram illustrating a part of Figure 2(A).

[0150] Figure 3(A) is a diagram illustrating the structure of information transmitted and received within an information processing system according to one embodiment of the present invention, and Figure 3(B) is a diagram illustrating the structure of an instruction statement.

[0151] Figure 4(A) is a diagram illustrating the configuration of a prompt chain transmitted and received within an information processing system according to one embodiment of the present invention, and Figures 4(B) to 4(D) are diagrams illustrating the configuration of instruction statements.

[0152] Figure 5 is a diagram illustrating the configuration of components used in an information processing system according to one embodiment of the present invention.

[0153] Figure 6(A) is a diagram illustrating the configuration of a knowledge graph that can be used in an information processing system according to one embodiment of the present invention, and Figure 6(B) is a diagram illustrating a part of Figure 6(A).

[0154] Figure 7(A) is a diagram illustrating the structure of information transmitted and received within an information processing system according to one embodiment of the present invention, and Figure 7(B) is a diagram illustrating the structure of an instruction statement.

[0155] Figure 8 is a diagram illustrating the configuration of an information processing system according to one embodiment of the present invention.

[0156] Figure 9 is a diagram illustrating the configuration of components used in an information processing system according to one embodiment of the present invention.

[0157] Figure 10(A) is a diagram illustrating the structure of inference results transmitted and received within an information processing system according to one embodiment of the present invention, and Figure 10(B) is a diagram illustrating the structure of graph data.

[0158] Figure 11 is a diagram illustrating the configuration of information transmitted and received within an information processing system according to one embodiment of the present invention.

[0159] Figure 12(A) is a diagram illustrating the configuration of inference results transmitted and received within an information processing system according to one embodiment of the present invention, and Figure 12(B) is a diagram illustrating the configuration of graph data.

[0160] Figure 13(A) is a diagram illustrating the configuration of a knowledge graph that can be used in an information processing system according to one embodiment of the present invention, and Figure 13(B) is a diagram illustrating a part of Figure 13(A).

[0161] Figure 14(A) is a diagram illustrating the structure of information transmitted and received within an information processing system according to one embodiment of the present invention, and Figure 14(B) is a diagram illustrating the structure of an instruction statement.

[0162] Figure 15(A) is a diagram illustrating the configuration of a prompt chain transmitted and received within an information processing system according to one embodiment of the present invention, and Figures 15(B) to 15(D) are diagrams illustrating the configuration of instruction statements.

[0163] Figure 16 is a diagram illustrating the configuration of components used in an information processing system according to one embodiment of the present invention.

[0164] Figure 17 is a diagram illustrating the configuration of an information processing system according to one embodiment of the present invention.

[0165] Figure 18(A) is a diagram illustrating the structure of inference results transmitted and received within an information processing system according to one embodiment of the present invention, and Figure 18(B) is a diagram illustrating the structure of graph data.

[0166] Figure 19 is a block diagram illustrating the configuration of an information processing device that can be used in an information processing system according to one embodiment of the present invention.

[0167] <Example of an information processing system configuration 1> The information processing system described in this embodiment includes component 110, component 130, and component 120 (see Figure 1).

[0168] For example, the information processing devices that perform the functions of component 110, component 130, and component 120 each include a computing device and a communication device. Furthermore, these communication devices can be connected to each other, for example, via a network 51, to constitute an information processing system according to one embodiment of the present invention.

[0169] <Component 110 Configuration Example 1> Component 110 has the function of receiving the knowledge graph KG0 and sending it to component 120, and the function of receiving the document Doc(AB) and providing it to, for example, the user 99 of the information processing system. Specifically, it provides it to the user 99 of the information processing system using output devices such as a display device, speaker, printer, and storage device.

[0170] For example, a user 99 of the information processing system inputs the knowledge graph KG0 into component 110. Alternatively, for example, a user inputs a command to select and send the knowledge graph KG0 stored in the memory device into component 110. Specifically, the user 99 of the information processing system inputs into component 110 using an input device such as a keyboard, mouse, or eye-tracking device.

[0171] 《Example of Knowledge Graph KG0 Configuration》 A knowledge graph is a representation using a graph format or graph structure that can represent various types of information in a network-like manner. Knowledge graph KG0 comprises a group of nodes Nd0 (see Figure 2(A)). The group of nodes Nd0 includes, for example, node Nd0(A) and node Nd0(B). Each of the group of nodes Nd0 comprises fields Fld00 and Fld01 (see Figure 2(B)).

[0172] Field Fld00 stores, for example, "Claim" as an attribute of the document. This allows identification that the information stored in node Nd0 is information described in the patent claims.

[0173] Field Fld01 stores, for example, a word or phrase Wd0 that has been identified as a component of the invention described in the claims.

[0174] Document Doc(AB) is a document that describes the relationship between elements Elem(A) and Elem(B). Both elements Elem(A) and Elem(B) are phrases Wd0 (see Figure 3(A)). For example, a document that describes the positional relationship between elements Elem(A) and Elem(B) can be used in document Doc(AB). Specifically, the sentence "Element Elem(A) is located above element Elem(B)" can be used in document Doc(AB). Also, a document that describes the function of element Elem(B) relative to element Elem(A) can be used in document Doc(AB). Specifically, the sentence "Element Elem(B) has the function of dividing element Elem(A)" can be used in document Doc(AB).

[0175] <Component 130 Configuration Example 1> Component 130 includes the function of sending document Doc(AB) to component 120 in response to prompt chain PC(02), and the function of performing processing using the large-scale language model LLM.

[0176] 《Example Configuration of a Large-Scale Language Model (LLM) 1》 The large-scale language model LLM has the function of generating document Doc(AB) according to instruction Pt30(0).

[0177] For example, large-scale language models such as GPT-3®, GPT-3.5, GPT-4®, LaMDA, Llama2, or Llama3 can be used in large-scale language model LLMs.

[0178] <Component 120 Configuration Example 1> Component 120 has the functionality to accept the knowledge graph KG0 and share it within Component 120.

[0179] Furthermore, component 120 has the function of executing a prompt chain PC(02) and the function of receiving a document Doc(AB) and sending it to component 110. The prompt chain PC(02) also includes an instruction statement Pt30(0) (see Figure 4(A)). The prompt chain is a function that adds the response to the previous instruction statement to part of the next instruction statement and sends the next instruction statement.

[0180] Furthermore, component 120 includes subcomponent 120A and subcomponent 120B (see Figure 5). For the purposes of this specification, a configuration having one or more functions is referred to as a component or subcomponent.

[0181] 《Example Configuration of Subcomponent 120A》 Subcomponent 120A has the function of acquiring a pair of nodes PoN0, for example, node Nd0(A) and node Nd0(B), from a group of nodes Nd0 (see Figure 2(A)). Nodes Nd0(A) and Nd0(B) are connected by an edge Edg0(AB). The graph data GD0(AB) consists of node Nd0(A), node Nd0(B), and edge Edg0(AB).

[0182] For example, a user 99 of the information processing system inputs information to component 110 to select nodes Nd0(A) and Nd0(B) from the knowledge graph KG0. Subcomponent 120A can receive this information via component 110 and obtain nodes Nd0(A) and Nd0(B). Alternatively, subcomponent 120A can select nodes Nd0(A) and Nd0(B) from the knowledge graph KG0 by selecting words from the correspondence list CL described later. For example, it can select two words from word Wd0 in the correspondence list CL, and select all combinations in order.

[0183] Furthermore, for example, if nodes Nd2(G) and Nd2(H) of the knowledge graph KG2 described later correspond to nodes Nd0(A) and Nd0(B), then, for example, a user 99 of the information processing system inputs information to component 110 to select nodes Nd2(G) and Nd2(H) from the knowledge graph KG2. Subcomponent 120A receives this information via component 110 and can use this information in a search query to obtain nodes Nd0(A) and Nd0(B). Alternatively, subcomponent 120A can select terms from the correspondence list CL described later to select nodes Nd2(G) and Nd2(H) from the knowledge graph KG2. For example, it can select two terms from the term Wd2 in the correspondence list CL, and select all combinations in order.

[0184] Node Nd0(A) stores element Elem(A) in field Fld01 (see Figure 2(B)). Similarly, node Nd0(B) stores element Elem(B) in field Fld01.

[0185] Furthermore, subcomponent 120A has the functionality to search for the path Pth0(AB) between node Nd0(A) and node Nd0(B), and the functionality to share the path Pth0(AB) within component 120. For example, by using the information stored in the nodes and edges involved in the path Pth0(AB), the relationship between element Elem(A) and element Elem(B) can be described in various ways. When using the programming language Python, the functionality to search for the path Pth0(AB) can be implemented in subcomponent 120A using libraries such as NetworkX.

[0186] 《Example Configuration of Subcomponent 120B》 Subcomponent 120B has the function of creating instruction statement Pt30(0).

[0187] [Example of the structure of instruction Pt30(0)] Instruction Pt30(0) includes instruction g30(0) and path Pth0(AB) (see Figure 4(B)). Instruction g30(0) includes a procedure to generate document Doc(AB) that describes the relationship between element Elem(A) and element Elem(B) using path Pth0(AB). For example, the document in the following paragraph can be used in instruction Pt30(0).

[0188] ### Input Information Knowledge Graph KG0: {Knowledge Graph KG0} Route Pth0(AB): {Route Pth0(AB)} ###Instructions Please explain the relationship between nodes Nd0(A) and Nd0(B) in the knowledge graph KG0, taking into account the path Pth0(AB).

[0189] This allows, for example, the retrieval of a pair of nodes from the knowledge graph KG0, including node Nd0(A) which stores element Elem(A) in field Fld01, and node Nd0(B) which stores element Elem(B) in field Fld01. Furthermore, it allows for the exploration of a path connecting node Nd0(A) and node Nd0(B). It also allows for the generation of document Doc(AB), which describes the relationships between the elements constituting the invention described in the claims "Claim". Additionally, document Doc(AB) can be generated via graph data extracted from the knowledge graph KG0, allowing for a more accurate explanation of the relationships between elements. Moreover, the knowledge graph KG0 can be used to explain the basis for document Doc(AB). As a result, a novel information processing system with superior convenience, usefulness, and reliability can be provided.

[0190] <Component 110 Configuration Example 2> Component 110 has the function of receiving a knowledge graph KG2 and sending it to component 120, and the function of receiving a document Doc(GH) and providing it, for example, to a user 99 of the information processing system (see Figure 1). Specifically, it provides it to the user 99 of the information processing system using output devices such as a display device, speaker, printer, and storage device.

[0191] For example, a user 99 of the information processing system inputs the knowledge graph KG2 into component 110. Alternatively, for example, a user inputs a command to select and send the knowledge graph KG2 stored in the memory device into component 110. Specifically, the user 99 of the information processing system inputs into component 110 using an input device such as a keyboard, mouse, or eye-tracking device.

[0192] 《Example of Knowledge Graph KG2 Configuration 1》 The knowledge graph KG2 comprises a group of nodes Nd2 (see Figure 6(A)). This group of nodes Nd2 includes, for example, nodes Nd2(G), Nd2(H), and Nd2(I). Each of these nodes Nd2 comprises fields Fld20, Fld21, and Fld22 (see Figure 6(B)).

[0193] Field FLD20 stores, for example, "Prior Art Document ("Ref")" as a document attribute.

[0194] The field Fld21 stores a single phrase Wd2 that has been identified as an element.

[0195] The field Fld22 stores either a single phrase Wd0 or a false value False. A single phrase Wd0 is a phrase determined to correspond to a single phrase Wd2. In other words, any phrase determined to correspond to the phrase stored in field Fld21 is stored in field Fld22. If no phrase is determined to correspond to the phrase stored in field Fld21, then false False is stored in field Fld22.

[0196] Document Doc(GH) is a document that explains the relationship between elements Elem(G) and Elem(H). Both elements Elem(G) and Elem(H) are words Wd2 (see Figure 7(A)).

[0197] <Component 130 Configuration Example 2> Component 130 includes the function of sending document Doc(GH) to component 120 in response to prompt chain PC(02), and the function of performing processing using the large-scale language model LLM.

[0198] 《Example of a Large-Scale Language Model (LLM) Configuration 2》 The large-scale language model LLM has the function of generating document Doc(GH) according to instruction Pt30(2).

[0199] <Component 120 Configuration Example 2> Component 120 has the functionality to accept the knowledge graph KG2 and share it within Component 120.

[0200] Furthermore, component 120 has the function of executing the prompt chain PC(02) and the function of receiving the document Doc(GH) and sending it to component 110. The prompt chain PC(02) includes the instruction statement Pt30(2) (see Figure 4(A)).

[0201] 《Example Configuration of Subcomponent 120A 2》 Subcomponent 120A has the function of acquiring a pair of nodes PoN2, for example, node Nd2(G) and node Nd2(H), from a group of nodes Nd2 (see Figure 6(A)). Subcomponent 120A also has the function of acquiring, for example, node Nd2(G) and node Nd2(I).

[0202] Furthermore, nodes Nd2(G) and Nd2(H) are connected by an edge Edg2(GH). The graph data GD2(GH) is composed of nodes Nd2(G), Nd2(H), and the edge Edg2(GH).

[0203] Furthermore, nodes Nd2(G) and Nd2(I) are connected by an edge Edg2(GI). The graph data GD2(GI) is composed of nodes Nd2(G), Nd2(I), and the edge Edg2(GI).

[0204] For example, a user 99 of the information processing system inputs information to component 110 to select nodes Nd2(G) and Nd2(H) from the knowledge graph KG2. Subcomponent 120A can receive this information via component 110 and obtain nodes Nd2(G) and Nd2(H).

[0205] Furthermore, for example, if nodes Nd0(A) and Nd0(B) of the knowledge graph KG0 correspond to nodes Nd2(G) and Nd2(H), then, for example, a user 99 of the information processing system inputs information to component 110 to select nodes Nd0(A) and Nd0(B) from the knowledge graph KG0. Subcomponent 120A receives this information via component 110 and can use this information in a search query to obtain nodes Nd2(G) and Nd2(H).

[0206] Node Nd2(G) stores element Elem(G) in field Fld21 and element Elem(A) in field Fld22 (see Figure 6(B)). Similarly, node Nd2(H) stores element Elem(H) in field Fld21 and element Elem(B) in field Fld22.

[0207] Furthermore, subcomponent 120A has the function of searching for the path Pth2(GH) between node Nd2(G) and node Nd2(H), and the function of sharing the path Pth2(GH) within component 120.

[0208] 《Example Configuration of Subcomponent 120B 2》 Subcomponent 120B has the function of creating instruction statement Pt30(2).

[0209] [Example of the structure of instruction Pt30(2)] Instruction Pt30(2) includes instruction g30(2) and path Pth2(GH) (see Figure 4(C)). Instruction g30(2) includes a procedure to generate a document Doc(GH) that describes the relationship between element Elem(G) and element Elem(H) using path Pth2(GH). For example, the document in the following paragraph can be used in instruction Pt30(2).

[0210] ### Input Information Knowledge Graph KG2: {Knowledge Graph KG2} Route Pth2(GH): {Route Pth2(GH)} ###Instructions Please explain the relationship between node Nd2(G) and node Nd2(H) in the knowledge graph KG2, taking into account the path Pth2(GH).

[0211] This allows us to find, for example, a pair of nodes in the knowledge graph KG2 that include node Nd2(G) which stores element Elem(A) in field Fld22, and node Nd2(H) which stores element Elem(B) in field Fld22. Furthermore, it allows us to explore paths connecting node Nd2(G) and node Nd2(H). It also allows us to generate document Doc(GH) describing the relationships between elements described in prior art document "Ref". Additionally, document Doc(GH) can be generated via graph data extracted from the knowledge graph KG2, allowing for a more accurate explanation of the relationships between elements. Furthermore, the knowledge graph KG2 can be used to explain the basis for document Doc(GH). As a result, we can provide a novel information processing system that is superior in convenience, usefulness, and reliability.

[0212] <Component 110 Configuration Example 3> Component 110 has the function of receiving a comparison document Doc(02) and providing it, for example, to a user 99 of the information processing system (see Figure 1). Specifically, it provides it to the user 99 of the information processing system using output devices such as a display device, speaker, printer, and storage device.

[0213] The comparison document Doc(02) explains the difference between the relationship between elements Elem(A) and Elem(B), and the relationship between elements Elem(G) and Elem(H).

[0214] <Component 130 Configuration Example 3> Component 130 includes the function of sending the comparison document Doc(02) to component 120 in response to the prompt chain PC(02), and the function of performing processing using the large-scale language model LLM.

[0215] 《Example 3 of the configuration of a large-scale language model (LLM)》 The large-scale language model LLM has the function of generating a comparison document Doc(02) according to the instruction Pt31(02).

[0216] <Component 120 Configuration Example 3> Component 120 has the function of executing a prompt chain PC(02) and the function of receiving a comparison document Doc(02) and sending it to component 110. The prompt chain PC(02) includes an instruction statement Pt31(02), and the instruction statement Pt31(02) includes a response to the previous instruction statement (see Figure 4(A)).

[0217] 《Example Configuration of Subcomponent 120B 3》 Subcomponent 120B has the function of creating instruction statement Pt31(02).

[0218] [Example of the structure of instruction statement Pt31(02)] Instruction Pt31(02) includes instruction g31(02), document Doc(AB), and document Doc(GH) (see Figure 4(D)).

[0219] Instruction g31(02) includes a procedure to generate a comparison document Doc(02) that compares document Doc(AB) and document Doc(GH) and explains the difference between the relationship between element Elem(A) and element Elem(B) and the relationship between element Elem(G) and element Elem(H). For example, the document in the following paragraph can be used in instruction Pt31(02).

[0220] ### Input Information Document Doc(AB):{Document Doc(AB)} Document Doc(GH):{Document Doc(GH)} ###Instructions Please explain the differences in the relationships between each element based on documents Doc(AB) and Doc(GH), which describe graph paths.

[0221] This allows us to associate nodes in knowledge graph KG2 with nodes in knowledge graph KG0 using the field Fld22. Furthermore, we can use the field Fld22 to select nodes from a group of nodes Nd2 in knowledge graph KG2 that are associated with any of the groups of nodes Nd0 in knowledge graph KG0. We can also select nodes from a group of nodes Nd2 in knowledge graph KG2 that are associated with any of the groups of nodes Nd0 in knowledge graph KG0 to obtain a pair of nodes. Additionally, we can find a pair of nodes in knowledge graph KG0 that corresponds to a pair of nodes in knowledge graph KG2. Furthermore, for example, a pair of nodes including node Nd2(G) which stores element Elem(A) in field Fld22 and node Nd2(H) which stores element Elem(B) in field Fld22 can be obtained from knowledge graph KG2, and a pair of nodes including node Nd0(A) which stores element Elem(A) in field Fld21 and node Nd0(B) which stores element Elem(B) in field Fld21 can be found from knowledge graph KG0. Also, for example, a comparison document Doc(02) can be generated that explains the difference between the path connecting node Nd0(A) and node Nd0(B) and the path connecting node Nd2(G) and node Nd2(H). In addition, the relationships between the elements constituting the invention described in the claims "Claim" can be compared with the relationships between the elements described in the prior art document "Ref". Furthermore, by comparing knowledge graph KG0 and knowledge graph KG2, a comparison document Doc(02) can be generated that compares the configuration described in the patent claims "Claim" with the configuration described in the prior art document "Ref". As a result, a novel information processing system with superior convenience, usefulness, and reliability can be provided.

[0222] <Component 110 Configuration Example 4> Component 110 has the function of receiving a document Doc(0) and sending it to component 120, and the function of receiving a knowledge graph KG0 and providing it, for example, to a user 99 of the information processing system (see Figure 8). Specifically, it provides it to the user 99 of the information processing system using output devices such as a display device, speaker, printer, and storage device.

[0223] Document Doc(0) is a document containing the patent claims "Claim". Knowledge graph KG0 is document Doc(0) converted into a graph format. In other words, information written in document Doc(0) using natural language is represented in knowledge graph KG0 using a graph format.

[0224] <Component 130 Configuration Example 4> Component 130 includes the function of receiving an instruction statement Pt0 and sending the inference result IR0 to component 120, and the function of performing processing using the large-scale language model LLM.

[0225] 《Example 4 of the configuration of a large-scale language model (LLM)》 The large-scale language model (LLM) has the capability to generate inference results (IR0) according to the instruction Pt0.

[0226] <Component 120 Configuration Example 4> Component 120 has the function of receiving document Doc(0) and sending instruction statement Pt0 to component 130, and the function of receiving inference result IR0 and sending knowledge graph KG0 to component 110.

[0227] Furthermore, component 120 includes subcomponent 120C (see Figure 9).

[0228] 《Example Configuration of Subcomponent 120C》 Subcomponent 120C has the functions of performing natural language processing, creating an element list EL0, and sharing the element list EL0 within component 120. The element list EL0 stores the phrases Wd0 that have been identified as elements in document Doc(0) through natural language processing (see Figure 3(A)). In other words, subcomponent 120C decomposes document Doc(0), removes unnecessary phrases, and identifies single or multiple phrases as phrases Wd0.

[0229] Furthermore, subcomponent 120C can, for example, decompose document Doc(0) into morphemes using a morphological analyzer, or decompose document Doc(0) into words. Subcomponent 120C can also normalize phrases. Additionally, subcomponent 120C can remove stop words.

[0230] 《Example Configuration of Subcomponent 120B 4》 Subcomponent 120B has the function of selecting a pair of elements PoE0 in order from the element list EL0, and the function of creating an instruction statement Pt0.

[0231] For example, when the element list EL0 contains n words (where n is 2 or greater), there are n × (n-1) ÷ 2 possible combinations for a pair of elements PoE0. Subcomponent 120B can sequentially select one from these combinations.

[0232] [Example of the structure of instruction statement Pt0] Instruction Pt0 includes instruction g0(), a pair of elements PoE0, and document Doc(0) (see Figure 3(B)).

[0233] The instruction g0() includes a procedure for generating the inference result IR0 from the document Doc(0). The inference result IR0 includes an expression EDR0(XY) that describes the relationship between one of a pair of elements PoE0 (see Figure 10(A)). For example, the document in the following paragraph can be used as the instruction Pt0.

[0234] ### Input Information Document Doc(0):{Document Doc(0)} ###Instructions Extract elements Elem(X) and Elem(Y) from document Doc(0) and set them as PoE0. The relationship of PoE0 is explained and designated as EDR0(XY). We define PoE0 as a node, EDR0(XY) as an edge, and the inference result IR0.

[0235] 《Example Configuration of Subcomponent 120A 3》 Subcomponent 120A has the function of creating graph data GD0(XY) from the inference result IR0 and the function of adding the graph data GD0(XY) to the knowledge graph KG0. It also has the function of sharing the knowledge graph KG0 within component 120.

[0236] [Graph data GD0(XY)] The graph data GD0(XY) includes nodes Nd0(X), Nd0(Y), and edges Edg0(XY) (see Figure 10(B)).

[0237] Node Nd0(X) stores the "Claim" as a document attribute in field Fld00 and one of the pair of elements PoE0 in field Fld01. In other words, node Nd0(X) is a node that represents the element Elem(X) described in the "Claim", and element Elem(X) is a phrase recognized as an element.

[0238] Furthermore, node Nd0(Y) stores the "Claim" as an attribute of the document in field Fld00, and stores the other half of the pair of elements PoE0 in field Fld01. In other words, node Nd0(Y) is a node that represents the element Elem(Y) described in the "Claim", and element Elem(Y) is a phrase recognized as an element.

[0239] Furthermore, the edge Edg0(XY) includes a field Fld05, which stores the representation EDR0(XY) that describes the relationship. In other words, the edge Edg0(XY) is an edge that represents the relationship between element Elem(X) and element Elem(Y) described in the claim "Claim", and stores the representation EDR0(XY) that describes the relationship.

[0240] This allows, for example, when elements Elem(A) and Elem(B) are selected as a pair of elements, the representation EDR0(AB) describing the relationship between elements Elem(A), Elem(B), and Elem(A) and Elem(B) as described in the patent claim "Claim" to be stored in the graph data GD0(AB). Furthermore, the graph data GD0(AB) can be added to the knowledge graph KG0. In addition, the elements constituting the invention described in the patent claim "Claim," and the relationships between the elements constituting the invention, can be represented in the knowledge graph KG0. As a result, a novel information processing system with superior convenience, usefulness, and reliability can be provided.

[0241] <Component 110 Configuration Example 5> Component 110 has the function of receiving document Doc(2) and correspondence list CL and sending them to component 120, and the function of receiving knowledge graph KG2 and providing it, for example, to the user 99 of the information processing system (see Figure 8). Specifically, it provides it to the user 99 of the information processing system using output devices such as display devices, speakers, printers, and storage devices.

[0242] Document Doc(2) is a prior art document "Ref". Knowledge graph KG2 is document Doc(2) converted into a graph format. In other words, information described in document Doc(2) using natural language is represented in knowledge graph KG2 using a graph format.

[0243] 《Example of a correspondence list CL configuration》 The correspondence list CL selects a term Wd2 that is determined to be in a correspondence relationship with a term Wd0 from among the terms Wd2 and stores it in association with term Wd0 (see Figure 11). For example, in the examination process of a filed patent, the examiner infers that the terms used in the claims are in a correspondence relationship with the terms used in the prior art documents. The correspondence list CL also selects a term Wd1 that is in a correspondence relationship with a term Wd0 from among the terms Wd1 and stores it in association with term Wd0. For example, the terms used in the claims are in a correspondence relationship with the terms used in the specification.

[0244] <Component 130 Configuration Example 5> Component 130 includes the function of receiving an instruction statement Pt2 and sending the inference result IR2 to component 120, and the function of performing processing using the large-scale language model LLM.

[0245] 《Example 5 of the configuration of a large-scale language model (LLM)》 The large-scale language model LLM has the capability to generate inference results IR2 according to the instruction Pt2.

[0246] <Component 120 Configuration Example 5> Component 120 has the function of receiving document Doc(2) and correspondence list CL and sending instruction statement Pt2 to component 130, and the function of receiving inference result IR2 and sending knowledge graph KG2 to component 110.

[0247] Furthermore, component 120 includes subcomponent 120C (see Figure 9).

[0248] 《Example Configuration of Subcomponent 120C 2》 Subcomponent 120C has the functions of natural language processing, creating element list EL2, and sharing element list EL2 within component 120. Element list EL2 stores the phrases Wd2 identified as elements in document Doc(2) through natural language processing (see Figure 7(A)). In other words, subcomponent 120C decomposes document Doc(2) into single or multiple phrases. It also removes unnecessary phrases and then identifies the remaining single or multiple phrases as phrase Wd2.

[0249] For example, a morphological analyzer can be used in subcomponent 120C. The document Doc(2) can be decomposed into morphemes.

[0250] 《Example Configuration of Subcomponent 120B 5》 Subcomponent 120B has the function of selecting a pair of elements PoE2 in order from the element list EL2, and the function of creating an instruction statement Pt2.

[0251] For example, when element list EL2 contains n words (where n is 2 or more), there are n × (n-1) ÷ 2 possible combinations for a pair of elements PoE2. Subcomponent 120B can sequentially select one from these combinations.

[0252] [Example of the structure of instruction statement Pt2] Instruction Pt2 includes instruction g2(), a pair of elements PoE2, and document Doc(2) (see Figure 7(B)). Instruction g2() includes a procedure for generating inference result IR2 from document Doc(2). The inference result IR2 includes a representation EDR2(XY) that describes the relationship between one of the pair of elements PoE2 (see Figure 12(A)). For example, the document in the following paragraph can be used as instruction Pt2.

[0253] ### Input Information Document Doc(2):{Document Doc(2)} ###Instructions Extract the element Elem(X) and the element Elem(Y) from the document Doc(2) and set it as PoE2. Describe the relationship of PoE2 and set it as EDR2(XY). Using PoE2 as the node and EDR2(XY) as the edge, set it as the inference result IR2.

[0254] 《Configuration Example 4 of Sub-component 120A》 The sub-component 120A has a function of creating graph data GD2(XY) from the inference result IR2, a function of adding the graph data GD2(XY) to the knowledge graph KG2, and a function of sharing the knowledge graph KG2 within the component 120.

[0255] [Graph Data GD2(XY)] The graph data GD2(XY) includes a node Nd2(X), a node Nd2(Y), and an edge Edg2(XY) (see Fig. 12(B)).

[0256] As an attribute of the document, the node Nd2(X) stores "Prior Art Document (\"Ref\")" in the field Fld20 and stores one of the pair of elements PoE2 in the field Fld21. In other words, the node Nd2(X) is a node representing the element Elem(X) described in the prior art document "Ref", and the element Elem(X) is a recognized phrase as an element.

[0257] Also, when one of the pair of elements PoE2 is associated with the element Elem(P) in the correspondence list CL, store the element Elem(P) in the field Fld22. Also, when one of the pair of elements PoE2 is not associated with any element in the correspondence list CL, store False in the field Fld22. In other words, the node Nd2(X) is a node representing whether it is associated with the element Elem(P) described in the claims "Claim", and when it is associated with the element Elem(P) described in the claims "Claim", it is a node representing the associated element.

[0258] Node Nd2(Y) stores the "prior art document ("Ref")" as a document attribute in field Fld20, and the other half of the pair of elements PoE2 in field Fld21. In other words, node Nd2(Y) is a node that represents element Elem(Y) described in the prior art document "Ref", and element Elem(Y) is a phrase recognized as an element.

[0259] Furthermore, when the other element of the pair PoE2 is associated with element Elem(Q) in the correspondence list CL, element Elem(Q) is stored in the field Fld22. Also, when the other element of the pair PoE2 is not associated with any element in the correspondence list CL, false is stored in the field Fld22. In other words, node Nd2(Y) is a node that indicates whether or not it is associated with element Elem(Q) described in the claim "Claim", and if it is associated with element Elem(Q) described in the claim "Claim", it is a node that indicates the associated element.

[0260] The edge Edg2(XY) includes a field Fld25. The field Fld25 stores a representation EDR2(XY) that describes the relationship. In other words, the edge Edg2(XY) is an edge that represents the relationship between element Elem(X) and element Elem(Y) described in the prior art document "Ref", and stores a representation EDR2(XY) that describes the relationship.

[0261] This allows, for example, when elements Elem(G) and Elem(I) are selected as a pair of elements, the representation EDR2(GI) describing the relationship between elements Elem(G), Elem(I), and Elem(G) and Elem(I) as described in the prior art document "Ref" to be stored in the graph data GD2(GI). The graph data GD2(GI) can then be added to the knowledge graph KG2. Furthermore, in node Nd2(G) where element Elem(G) is stored in field Fld21, element Elem(A) can be stored in field Fld22 based on the correspondence list CL. Also, when element Elem(I) is not associated with any element in the correspondence list CL, the false value False can be stored in field Fld22 in node Nd2(I) where element Elem(I) is stored in field Fld21. In addition, the elements constituting the technology described in the prior art document "Ref", and the relationships between elements can be represented in the knowledge graph KG2. As a result, it becomes possible to provide a novel information processing system that is superior in terms of convenience, usefulness, or reliability.

[0262] <Example of an information processing system configuration 2> The information processing system described in this embodiment includes component 110, component 130, and component 120 (see Figure 17). Note that Information Processing System Configuration Example 2 differs from Information Processing System Configuration Example 1 in that it uses knowledge graph KG1 instead of knowledge graph KG0.

[0263] <Component 110 Configuration Example 6> Component 110 has the function of receiving the knowledge graph KG1 and sending it to component 120, and the function of receiving the document Doc(ab) and providing it, for example, to the user 99 of the information processing system. Specifically, it provides it to the user 99 of the information processing system using output devices such as a display device, speaker, printer, and storage device.

[0264] For example, a user 99 of the information processing system inputs the knowledge graph KG1 into component 110. Alternatively, for example, a user inputs a command to select and send the knowledge graph KG1 stored in the memory device into component 110. Specifically, the user 99 of the information processing system inputs into component 110 using an input device such as a keyboard, mouse, or eye-tracking device.

[0265] 《Example of Knowledge Graph KG1 Configuration》 The knowledge graph KG1 comprises a group of nodes Nd1 (see Figure 13(A)). This group of nodes Nd1 includes, for example, nodes Nd1(a), Nd1(b), and Nd1(c). Each of these nodes Nd1 comprises fields Fld10, Fld11, and Fld12 (see Figure 13(B)).

[0266] Field FLD10 stores a document attribute, such as "Specification ("Spec")".

[0267] The field Fld11 stores a single word or phrase Wd1 that has been identified as an element.

[0268] Field Fld12 has the function of storing either a single phrase Wd0 or a false value False. A single phrase Wd0 is a phrase determined to correspond to a single phrase Wd1. In other words, any phrase determined to correspond to the phrase stored in field Fld11 is stored in field Fld12. If no phrase is determined to correspond to the phrase stored in field Fld11, then false False is stored in field Fld12.

[0269] Document Doc(ab) is a document that describes the relationship between elements Elem(a) and Elem(b). Both elements Elem(a) and Elem(b) are the phrase Wd1 (see Figure 14(A)).

[0270] <Component 130 Configuration Example 6> Component 130 has functions of sending document Doc(ab) to Component 120 in response to prompt chain PC(12) and performing processing using large language model LLM.

[0271] 《Configuration Example 6 of Large Language Model LLM》 Large language model LLM has a function of generating document Doc(ab) according to instruction text Pt30(1).

[0272] 〈Configuration Example 6 of Component 120〉 Component 120 has a function of receiving knowledge graph KG1 and sharing it within Component 120.

[0273] Also, Component 120 has functions of executing prompt chain PC(12) and receiving document Doc(ab) and sending it to Component 110. Note that prompt chain PC(12) includes instruction text Pt30(1) (see Fig. 15(A)).

[0274] Also, Component 120 includes sub-components 120A and 120B (see Fig. 16).

[0275] 《Configuration Example 5 of Sub-component 120A》 Sub-component 120A has functions of obtaining a pair of nodes PoN1, for example, nodes Nd1(a) and Nd1(b) from a group of nodes Nd1 (see Fig. 13(A)). Also, it has a function of obtaining, for example, nodes Nd1(a) and Nd1(c).

[0276] Note that nodes Nd1(a) and Nd1(b) are connected by edge Edg1(ab). Also, graph data GD1(ab) is composed of nodes Nd1(a), Nd1(b), and edge Edg1(ab).

[0277] Furthermore, nodes Nd1(a) and Nd1(c) are connected by an edge Edg1(ac). The graph data GD1(ac) is composed of nodes Nd1(a), Nd1(c), and the edge Edg1(ac).

[0278] For example, a user 99 of the information processing system inputs information to component 110 to select nodes Nd1(a) and Nd1(b) from the knowledge graph KG1. Subcomponent 120A can receive this information via component 110 and obtain nodes Nd1(a) and Nd1(b).

[0279] Furthermore, for example, if nodes Nd2(G) and Nd2(H) of the knowledge graph KG2 (described later) correspond to nodes Nd1(a) and Nd1(b), then, for example, a user 99 of the information processing system inputs information to component 110 to select nodes Nd2(G) and Nd2(H) from the knowledge graph KG2. Subcomponent 120A receives this information via component 110 and can use this information in a search query to obtain nodes Nd1(a) and Nd1(b).

[0280] Node Nd1(a) stores element Elem(a) in field Fld11 and element Elem(A) in field Fld12 (see Figure 13(B)). Similarly, node Nd1(b) stores element Elem(b) in field Fld11 and element Elem(B) in field Fld12.

[0281] Furthermore, subcomponent 120A has the function of searching for a path Pth1(ab) between node Nd1(a) and node Nd1(b), and the function of sharing the path Pth1(ab) within component 120.

[0282] 《Example Configuration of Subcomponent 120B 6》 Subcomponent 120B has the function of creating instruction statement Pt30(1).

[0283] [Example of the structure of instruction Pt30(1)] Instruction Pt30(1) includes instruction g30(1) and path Pth1(ab) (see Figure 15(B)). Instruction g30(1) includes a procedure to generate a document Doc(ab) that describes the relationship between element Elem(a) and element Elem(b) using path Pth1(ab). For example, the document in the following paragraph can be used in instruction Pt30(1).

[0284] ### Input Information Knowledge Graph KG1: {Knowledge Graph KG1} Route Pth1(ab): {Route Pth1(ab)} ###Instructions Please explain the relationship between nodes Nd1(a) and Nd1(b) in the knowledge graph KG1, taking into account the path Pth1(ab).

[0285] This allows, for example, the retrieval of a pair of nodes from the knowledge graph KG1, including node Nd1(a) which stores element Elem(A) in field Fld12, and node Nd1(b) which stores element Elem(B) in field Fld12. Furthermore, it allows for the exploration of a path connecting node Nd1(a) and node Nd1(b). It also allows for the generation of document Doc(ab), which describes the relationships between elements described in the specification "Spec" containing the details of the invention. Additionally, document Doc(ab) can be generated via graph data extracted from the knowledge graph KG1, allowing for a more accurate explanation of the relationships between elements. Moreover, the knowledge graph KG1 can be used to explain the basis for document Doc(ab). As a result, a novel information processing system with superior convenience, usefulness, and reliability can be provided.

[0286] <Component 110 Configuration Example 7> Component 110 has the function of receiving a knowledge graph KG2 and transmitting it to component 120, and the function of receiving a document Doc(GH) and providing it, for example, to a user 99 of the information processing system (see Figure 17). Specifically, it provides it to the user 99 of the information processing system using output devices such as a display device, speaker, printer, and storage device.

[0287] 《Example 2 of the structure of the knowledge graph KG2》 The knowledge graph KG2 comprises a group of nodes Nd2 (see Figure 6(A)). Each of the Nd2 nodes comprises fields Fld20, Fld21, and Fld22 (see Figure 6(B)).

[0288] Field FLD20 stores, for example, "Prior Art Document ("Ref")" as a document attribute.

[0289] The field Fld21 stores a single word Wd2 that has been identified as an element.

[0290] The field Fld22 stores either a phrase Wd0 or a false value False. The phrase Wd0 is determined to be a phrase corresponding to the phrase Wd2.

[0291] Document Doc(GH) is a document that explains the relationship between elements Elem(G) and Elem(H). Both elements Elem(G) and Elem(H) are words Wd2 (see Figure 7(A)).

[0292] <Component 130 Configuration Example 7> Component 130 includes the function of sending a document Doc(GH) to component 120 in response to a prompt chain PC(12), and the function of performing processing using a large-scale language model LLM.

[0293] 《Example 7 of the configuration of a large-scale language model (LLM)》 The large-scale language model LLM has the function of generating document Doc(GH) according to instruction Pt30(2).

[0294] <Component 120 Configuration Example 7> Component 120 has the functionality to accept the knowledge graph KG2 and share it within Component 120.

[0295] Furthermore, component 120 has the function of executing the prompt chain PC(12) and the function of receiving the document Doc(GH) and sending it to component 110. The prompt chain PC(12) includes the instruction statement Pt30(2) (see Figure 15(A)).

[0296] 《Example Configuration of Subcomponent 120A 6》 Subcomponent 120A has the function of acquiring a pair of nodes, for example, node Nd2(G) and node Nd2(H), from a group of nodes Nd2.

[0297] Node Nd2(G) stores element Elem(G) in field Fld21 and element Elem(A) in field Fld22 (see Figure 6(B)). Similarly, node Nd2(H) stores element Elem(H) in field Fld21 and element Elem(B) in field Fld22.

[0298] Furthermore, subcomponent 120A has the function of searching for the path Pth2(GH) between node Nd2(G) and node Nd2(H), and the function of sharing the path Pth2(GH) within component 120.

[0299] 《Example Configuration of Subcomponent 120B 7》 Subcomponent 120B has the function of creating instruction statement Pt30(2).

[0300] [Example of the structure of instruction Pt30(2)] Instruction Pt30(2) includes instruction g30(2) and path Pth2(GH) (see Figure 15(C)). Instruction g30(2) includes a procedure to generate a document Doc(GH) that describes the relationship between element Elem(G) and element Elem(H) using path Pth2(GH). For example, the document in the following paragraph can be used in instruction Pt30(2).

[0301] ### Input Information Knowledge Graph KG2: {Knowledge Graph KG2} Route Pth2(GH): {Route Pth2(GH)} ###Instructions Please explain the relationship between node Nd2(G) and node Nd2(H) in the knowledge graph KG2, taking into account the path Pth2(GH).

[0302] This allows, for example, the retrieval of a pair of nodes from the knowledge graph KG2, including node Nd2(G) which stores element Elem(A) in field Fld22, and node Nd2(H) which stores element Elem(B) in field Fld22. Furthermore, it allows for the exploration of paths connecting node Nd2(G) and node Nd2(H). It also enables the generation of a document Doc(GH) describing the relationships between elements described in the specification "Spec" which outlines the invention. Additionally, the document Doc(GH) can be generated via graph data extracted from the knowledge graph KG2, allowing for a more accurate explanation of the relationships between elements. Moreover, the knowledge graph KG2 can be used to explain the basis for document Doc(GH). As a result, a novel information processing system with superior convenience, usefulness, and reliability can be provided.

[0303] <Component 110 Configuration Example 8> Component 110 includes a function to receive a comparison document Doc(12) and provide it, for example, to a user 99 of the information processing system (see Figure 17). Specifically, it provides it to the user 99 of the information processing system using output devices such as a display device, speaker, printer, and storage device.

[0304] The comparison document Doc(12) explains the difference between the relationship between elements Elem(a) and Elem(b), and the relationship between elements Elem(G) and Elem(H).

[0305] <Component 130 Configuration Example 8> Component 130 includes the function of sending a comparison document Doc(12) to component 120 in response to a prompt chain PC(12), and the function of performing processing using a large-scale language model LLM.

[0306] 《Example 8 of a Large-Scale Language Model (LLM) Configuration》 The large-scale language model LLM has the function of generating a comparison document Doc(12) according to the instruction Pt31(12).

[0307] <Component 120 Configuration Example 8> Component 120 has the function of executing a prompt chain PC(12) and the function of receiving a comparison document Doc(12) and sending it to component 110. The prompt chain PC(12) includes an instruction statement Pt31(12), and the instruction statement Pt31(12) includes a response to the previous instruction statement (see Figure 15(A)).

[0308] 《Example Configuration of Subcomponent 120B 8》 Subcomponent 120B has the function of creating instruction statement Pt31(12).

[0309] [Example of the structure of instruction Pt31(12)] Instruction Pt31(12) includes instruction g31(12), document Doc(ab), and document Doc(GH) (see Figure 15(D)).

[0310] Instruction g31(12) includes a procedure to generate a comparison document Doc(12) that compares document Doc(ab) and document Doc(GH) and explains the difference between the relationship between elements Elem(a) and Elem(b) and the relationship between elements Elem(G) and Elem(H). For example, the document in the following paragraph can be used in instruction Pt31(12).

[0311] ### Input Information Document Doc(ab):{Document Doc(ab)} Document Doc(GH):{Document Doc(GH)} ###Instructions Please explain the differences in the relationships between elements based on documents Doc(ab) and Doc(GH), which describe graph paths.

[0312] This allows us to associate nodes in knowledge graph KG2 with nodes in knowledge graph KG1 using the field Fld22. Furthermore, we can use the field Fld22 to select nodes from a group of nodes Nd2 in knowledge graph KG2 that are associated with any of the groups of nodes Nd1 in knowledge graph KG1. We can also select nodes from a group of nodes Nd2 in knowledge graph KG2 that are associated with any of the groups of nodes Nd1 in knowledge graph KG1 to obtain a pair of nodes. Additionally, we can find a pair of nodes in knowledge graph KG1 that corresponds to a pair of nodes in knowledge graph KG2. Furthermore, for example, a pair of nodes including node Nd2(G) which stores element Elem(A) in field Fld22 and node Nd2(H) which stores element Elem(B) in field Fld22 can be obtained from knowledge graph KG2, and a pair of nodes including node Nd1(a) which stores element Elem(A) in field Fld12 and node Nd1(b) which stores element Elem(B) in field Fld12 can be found from knowledge graph KG1. Also, for example, a comparison document Doc(12) can be generated that explains the difference between the path connecting node Nd1(a) and node Nd1(b) and the path connecting node Nd2(G) and node Nd2(H). In addition, the relationships between elements described in the specification "Spec" which describes the content of the invention can be compared with the relationships between elements described in the prior art document "Ref". Furthermore, by comparing knowledge graph KG1 and knowledge graph KG2, a comparison document Doc(12) can be generated that compares the configuration described in the specification "Spec" which describes the content of the invention with the configuration described in the prior art document "Ref". As a result, a novel information processing system with superior convenience, usefulness, and reliability can be provided.

[0313] <Component 110 Configuration Example 9> Component 110 has the function of receiving document Doc(1) and correspondence list CL and sending them to component 120, and the function of receiving knowledge graph KG1 and providing it, for example, to the user 99 of the information processing system (see Figure 8). Specifically, it provides it to the user 99 of the information processing system using output devices such as display devices, speakers, printers, and storage devices.

[0314] Document Doc(1) is the specification "Spec" which describes the content of the invention. The correspondence list CL stores a term Wd1 that is determined to be in a corresponding relationship with a term Wd0, associating it with the term Wd0. Knowledge graph KG1 is document Doc(1) converted into a graph format. In other words, the information written in document Doc(1) using natural language is represented in knowledge graph KG1 using a graph format.

[0315] <Component 130 Configuration Example 9> Component 130 includes the function of receiving an instruction statement Pt1 and sending the inference result IR1 to component 120, and the function of performing processing using the large-scale language model LLM.

[0316] 《Example of a Large-Scale Language Model (LLM) Configuration 9》 The large-scale language model (LLM) has the capability to generate inference results (IR1) according to the instruction (Pt1).

[0317] <Component 120 Configuration Example 9> Component 120 has the function of receiving document Doc(1) and sending instruction statement Pt1 to component 130, and the function of receiving inference result IR1 and sending knowledge graph KG1 to component 110.

[0318] Component 120 includes subcomponent 120C (see Figure 9).

[0319] 《Example Configuration of Subcomponent 120C 7》 Subcomponent 120C has the functions of natural language processing, creating element list EL1, and sharing element list EL1 within component 120. Element list EL1 stores the phrases Wd1 that have been identified as elements in document Doc(1) through natural language processing (see Figure 14(A)). In other words, subcomponent 120C decomposes document Doc(1), removes unnecessary phrases, and identifies single or multiple phrases as phrases Wd1.

[0320] Furthermore, subcomponent 120C can, for example, decompose document Doc(1) into morphemes using a morphological analyzer, or decompose document Doc(1) into words. Subcomponent 120C can also normalize phrases and remove stop words.

[0321] 《Example Configuration of Subcomponent 120B 9》 Subcomponent 120B has the function of selecting a pair of elements PoE1 in order from the element list EL1, and the function of creating an instruction statement Pt1.

[0322] For example, when element list EL1 contains n words (where n is 2 or more), there are n × (n-1) ÷ 2 possible combinations for a pair of elements PoE1. Subcomponent 120B can sequentially select one from these combinations.

[0323] [Example of the structure of instruction statement Pt1] Instruction Pt1 includes instruction g1(), a pair of elements PoE1 and document Doc(1) (see Figure 14(B)).

[0324] Instruction g1() includes a procedure for generating the inference result IR1 from document Doc(1). The inference result IR1 includes an expression EDR1(XY) that describes the relationship between one of a pair of elements PoE1 (see Figure 18(A)). For example, the document in the following paragraph can be used as instruction Pt1.

[0325] ### Input Information Document Doc(1):{Document Doc(1)} ###Instructions Extract elements Elem(X) and Elem(Y) from document Doc(1) and name them PoE1. The relationship of PoE1 is explained and referred to as EDR2(XY). We'll use PoE1 as the node, EDR2(XY) as the edge, and the inference result IR1.

[0326] 《Example Configuration of Subcomponent 120A 8》 Subcomponent 120A includes the function of creating graph data GD1(XY) from the inference result IR1, the function of adding the graph data GD1(XY) to the knowledge graph KG1, and the function of sharing the knowledge graph KG1 within component 120.

[0327] [Graph Data GD1(XY)] The graph data GD1(XY) includes node Nd1(X), node Nd1(Y), and edge Edg1(XY) (see Figure 18(B)).

[0328] Node Nd1(X) stores the document attribute "Spec" in field Fld10 and one of the pair of elements PoE1 in field Fld11. In other words, node Nd1(X) is a node that represents the element Elem(X) described in the Spec, and element Elem(X) is a word or phrase recognized as an element.

[0329] Furthermore, when one of the pair of elements PoE1 is associated with element Elem(P) in the correspondence list CL, element Elem(P) is stored in the field Fld12. Also, when one of the pair of elements PoE1 is not associated with any element in the correspondence list CL, false is stored in the field Fld12. In other words, node Nd1(X) is a node that indicates whether or not it is associated with element Elem(P) described in the claim "Claim", and if it is associated with element Elem(P) described in the claim "Claim", it is a node that indicates the associated element.

[0330] Node Nd1(Y) stores the document attribute "Spec" in field Fld10 and the other half of the pair of elements PoE1 in field Fld11. In other words, node Nd1(Y) is a node that represents element Elem(Y) described in the Spec, and element Elem(Y) is a phrase recognized as an element.

[0331] Furthermore, when the other element of the pair PoE1 is associated with element Elem(Q) in the correspondence list CL, element Elem(Q) is stored in the field Fld12. Also, when the other element of the pair PoE1 is not associated with any element in the correspondence list CL, false is stored in the field Fld12. In other words, node Nd1(Y) is a node that indicates whether or not it is associated with element Elem(Q) described in the claim "Claim", and if it is associated with element Elem(Q) described in the claim "Claim", it is a node that indicates the associated element.

[0332] Edge Edg1(XY) includes field Fld15. Field Fld15 stores the representation EDR1(XY) that describes the relationship. In other words, edge Edg1(XY) is an edge that represents the relationship between element Elem(X) and element Elem(Y) described in specification "Spec", and stores the representation EDR1(XY) that describes the relationship.

[0333] This allows, for example, when elements Elem(a) and Elem(c) are selected as a pair, the expression EDR1(ac) describing the relationship between elements Elem(a) and Elem(c) as described in the specification "Spec" containing the details of the invention, to be stored in the graph data GD1(ac). Furthermore, the graph data GD1(ac) can be added to the knowledge graph KG1(ac). Additionally, in node Nd1(a) where element Elem(a) is stored in field Fld11, element Elem(A) can be stored in field Fld12 based on the correspondence list CL. Furthermore, when element Elem(c) is not associated with any element in the correspondence list CL, the false value False can be stored in field Fld12 in node Nd1(c) where element Elem(c) is stored in field Fld11. Finally, the elements constituting the technology described in the specification "Spec," and the relationships between elements, can be represented in the knowledge graph KG1. As a result, it becomes possible to provide a novel information processing system that is superior in terms of convenience, usefulness, or reliability.

[0334] <Example of information processing device configuration> An information processing device 20 that can be used in an information processing system according to one embodiment of the present invention includes, for example, an input unit 21, a storage unit 22, a processing unit 23, an output unit 24, and a transmission line 25 (see Figure 19).

[0335] In the drawings attached to this specification, the components are classified by function and shown as independent blocks in the block diagram. However, in reality, it is difficult to completely separate the components by function, and one component may be involved in multiple functions. For example, a part of the processing unit 23 may function as the input unit 21. Also, one function may be involved in multiple components. For example, the processing performed by the processing unit 23 may be executed by different information processing devices depending on the processing.

[0336] Input section 21 The input unit 21 can receive data from outside the information processing device. For example, the input unit 21 can receive data via the network 51. Specifically, a device such as a personal computer equipped with a communication port or communication function can be used.

[0337] The input unit 21 supplies the received data to either or both of the storage unit 22 and the processing unit 23 via the transmission line 25.

[0338] 《Storage section 22》 The memory unit 22 has the function of storing the program executed by the processing unit 23. The memory unit 22 may also have the function of storing data generated by the processing unit 23 (for example, calculation results, analysis results, inference results), data received by the input unit 21, etc.

[0339] The storage unit 22 may have a database. The information processing device may also have a database separate from the storage unit 22. The information processing device may have the function to retrieve data from a database located outside the storage unit 22, outside the information processing device itself, or outside the information processing system. Furthermore, the information processing device may have the function to retrieve data from both its own database and an external database.

[0340] Either or both of the storage and / or file server can be used in the storage unit 22. Furthermore, a database recording the paths of files stored on the file server can be used in the storage unit 22.

[0341] The storage unit 22 includes at least one of volatile memory and non-volatile memory. Examples of volatile memory include DRAM (Dynamic Random Access Memory) and SRAM (Static Random Access Memory). Examples of non-volatile memory include ReRAM (Resistive Random Access Memory), PRAM (Phase Change Random Access Memory), FeRAM (Ferroelectric Random Access Memory), MRAM (Magnetoresistive Random Access Memory), and flash memory. The storage unit 22 may also include at least one of NOSRAM (registered trademark) and DOSRAM (registered trademark). The storage unit 22 may also include a recording media drive. Examples of recording media drives include hard disk drives (HDD) and solid state drives (SSD).

[0342] NOSRAM is an abbreviation for "Nonvolatile Oxide Semiconductor Random Access Memory (RAM)". NOSRAM is a type of memory where the memory cell is a 2-transistor (2T) or 3-transistor (3T) gain cell, and the transistors are transistors that use metal oxide in the channel formation region (also called OS transistors). OS transistors have an extremely small current flowing between the source and drain when off, i.e., a leakage current. By utilizing the characteristic of extremely low leakage current, NOSRAM can be used as a non-volatile memory by holding charge corresponding to the data within the memory cell. In particular, NOSRAM can read the stored data without destroying it (non-destructive read), making it suitable for computational processing that involves repeating data read operations a large amount. Because the data capacity of NOSRAM can be increased by stacking it, it can be used as a large-scale cache memory, main memory, or storage memory to improve the performance of semiconductor devices.

[0343] DOSRAM is an abbreviation for "Dynamic Oxide Semiconductor RAM," and refers to RAM with a 1T (transistor) 1C (capacitance) type memory cell. DOSRAM is a type of DRAM formed using OS transistors, and it is a memory that temporarily stores information sent from an external source. DOSRAM is a memory that takes advantage of the low off-current of OS transistors.

[0344] In this specification, "metal oxide" refers to an oxide of a metal in a broad sense. Metal oxides are classified into oxide insulators, oxide conductors (including transparent oxide conductors), oxide semiconductors (also called oxide semiconductors or simply OS), etc. For example, when a metal oxide is used in the semiconductor layer of a transistor, that metal oxide may be referred to as an oxide semiconductor.

[0345] The metal oxide in the channel-forming region preferably contains indium (In). When the metal oxide in the channel-forming region contains indium, the carrier mobility (electron mobility) of the OS transistor is increased. For example, indium oxide (InOx) or indium gallium zinc oxide (In-Ga-Zn oxide, also written as "IGZO") can be used in the channel-forming region. Furthermore, the metal oxide in the channel-forming region is preferably an oxide semiconductor containing element M. Element M is preferably at least one of aluminum (Al), gallium (Ga), and tin (Sn). Other elements applicable to element M include boron (B), silicon (Si), titanium (Ti), iron (Fe), nickel (Ni), germanium (Ge), yttrium (Y), zirconium (Zr), molybdenum (Mo), lanthanum (La), cerium (Ce), neodymium (Nd), hafnium (Hf), tantalum (Ta), and tungsten (W). However, in some cases, element M may be a combination of multiple elements as mentioned above. Element M is, for example, an element with a high bond energy with oxygen. For example, an element with a higher bond energy with oxygen than indium. Furthermore, the metal oxide containing the channel-forming region is preferably a metal oxide containing zinc (Zn). Metal oxides containing zinc may be more prone to crystallization.

[0346] The metal oxides present in the channel-forming regions are not limited to indium-containing metal oxides. For example, the metal oxides present in the channel-forming regions may be zinc-tin oxides, gallium-tin oxides, or other metal oxides that do not contain indium but contain zinc, gallium, or tin.

[0347] Processing Unit 23 The processing unit 23 has the function of performing calculations, analyses, and inferences using data supplied from either or both of the input unit 21 and the storage unit 22. The processing unit 23 can supply the generated data (e.g., calculation results, analysis results, inference results) to either or both of the storage unit 22 and the output unit 24.

[0348] The processing unit 23 has the function of acquiring data from the storage unit 22. The processing unit 23 may also have the function of recording or registering data in the storage unit 22.

[0349] The processing unit 23 may, for example, have an arithmetic circuit. The processing unit 23 may, for example, have a central processing unit (CPU). The processing unit 23 may also have a graphics processing unit (GPU). The processing unit 23 may also have a neural processing unit (NPU / neural network processing unit).

[0350] The processing unit 23 may have a microprocessor such as a DSP (Digital Signal Processor). The microprocessor can be implemented using a PLD (Programmable Logic Device) such as an FPGA (Field Programmable Gate Array) or FPAA (Field Programmable Analog Array). The processing unit 23 may also have a quantum processor. The processing unit 23 can perform various data processing and program control by interpreting and executing instructions from various programs via the processor. Programs that can be executed by the processor are stored in at least one of the processor's memory area and the storage unit 22.

[0351] The processing unit 23 may have main memory. The main memory may include at least one of volatile memory such as RAM and non-volatile memory such as ROM (Read Only Memory). Furthermore, the main memory may include at least one of the above-mentioned NOSRAM and DOSRAM.

[0352] For RAM, for example, DRAM or SRAM is used, and a virtual memory space is allocated and used as the workspace for the processing unit 23. The operating system, application programs, program modules, program data, and lookup tables stored in the storage unit 22 are loaded into RAM for execution. These data, programs, and program modules loaded into RAM are each directly accessed and manipulated by the processing unit 23.

[0353] ROM can store BIOS (Basic Input / Output System) and firmware, etc., which do not require rewriting. Examples of ROM include mask ROM, OTPROM (One Time Programmable Read Only Memory), and EPROM (Erasable Programmable Read Only Memory). Examples of EPROM include UV-EPROM (Ultra-Violet Erasable Programmable Read Only Memory), which allows data to be erased by ultraviolet irradiation, EEPROM (Electrically Erasable Programmable Read Only Memory), and flash memory.

[0354] The processing unit 23 may have either or both an OS transistor and a transistor having silicon in its channel formation region (Si transistor).

[0355] The processing unit 23 preferably has an OS transistor. Because the OS transistor has an extremely small off-current, using the OS transistor as a switch to hold the charge (data) that has flowed into a capacitive element that functions as a memory element ensures that the data can be retained for a long period of time. By using this characteristic in at least one of the registers and cache memory of the processing unit, the processing unit can be operated only when necessary, and at other times the information of the previous processing is saved to the memory element, thereby turning off the processing unit. In other words, normally-off computing becomes possible, and the power consumption of the information processing system can be reduced.

[0356] It is preferable for information processing devices to use AI for at least some of their processing.

[0357] Information processing devices preferably utilize artificial neural networks (ANNs, also simply referred to as neural networks). Neural networks are implemented using circuits (hardware) or programs (software).

[0358] In this specification, the term "neural network" refers to any model that mimics the neural network of living organisms, determines the strength of connections between neurons through learning, and possesses problem-solving capabilities. A neural network has an input layer, an intermediate layer (hidden layer), and an output layer.

[0359] In this specification and other documents, when discussing neural networks, the process of determining the connection strength (also called weight coefficient) between neurons from existing information is sometimes referred to as "learning."

[0360] In this specification and other documents, the process of constructing a neural network using connection strengths obtained through learning and deriving new conclusions from it may be referred to as "inference."

[0361] Output section 24 The output unit 24 can output at least one of the calculation results, analysis results, and inference results from the processing unit 23 to the outside of the information processing device. For example, the output unit 24 can transmit data via the network 51. Specifically, a device such as a personal computer equipped with a communication port or communication function can be used. Alternatively, a device equipped with a communication function may be used for both the input unit 21 and the output unit 24.

[0362] Transmission line 25 The transmission line 25 has the function of transmitting data. Data can be transmitted and received between the input unit 21, the storage unit 22, the processing unit 23, and the output unit 24 via the transmission line 25. Specifically, an external bus, LAN, or the internet can be used as the transmission line 25.

[0363] This embodiment can be appropriately combined with other embodiments shown in this specification.

[0364] (Embodiment 2) In this embodiment, an information processing method according to one aspect of the present invention will be described with reference to Figures 20 to 22.

[0365] Figure 20 is a flowchart illustrating an information processing method according to one embodiment of the present invention.

[0366] Figure 21 is a flowchart illustrating an information processing method according to one embodiment of the present invention.

[0367] Figure 22 is a flowchart illustrating an information processing method according to one embodiment of the present invention.

[0368] <Example of information processing method 1> One aspect of the present invention is an information processing method having a phase PH1 (see Figure 20).

[0369] <Phase PH1> Phase PH1 comprises steps S1 to S10.

[0370] Step S1 In step S1, component 110 receives knowledge graph KG0 and knowledge graph KG2 and transmits them to component 120.

[0371] The knowledge graph KG0 comprises a group of nodes Nd0, each of which comprises fields Fld00 and Fld01. Field Fld00 stores "Claim" as a document attribute, and field Fld01 stores a word Wd0 that has been identified as an element.

[0372] Furthermore, the knowledge graph KG2 comprises a group of nodes Nd2, each of which comprises fields Fld20, Fld21, and Fld22. Field Fld20 stores "Prior Art Document ("Ref")" as a document attribute, field Fld21 stores a phrase Wd2 identified as an element, and field Fld22 has the function of storing a phrase Wd0 or false. Note that a phrase Wd0 is a phrase determined to be in a corresponding relationship with a phrase Wd2.

[0373] Step S2 In step S2, component 120 receives knowledge graphs KG0 and KG2 and shares them within component 120. Component 120 also includes subcomponents 120A and 120B.

[0374] Step S3 In step S3, subcomponent 120A obtains nodes Nd2(G) and Nd2(H) from the group of nodes Nd2. It also obtains nodes Nd0(A) and Nd0(B) from the group of nodes Nd0.

[0375] Node Nd2(G) is a node that stores element Elem(G) in field Fld21 and element Elem(A) in field Fld22.

[0376] Node Nd2(H) is a node that stores element Elem(H) in field Fld21 and element Elem(B) in field Fld22.

[0377] Node Nd0(A) is a node that stores the element Elem(A) in the field Fld01.

[0378] Node Nd0(B) is a node that stores the element Elem(B) in the field Fld01.

[0379] Step S4 In step S4, subcomponent 120A searches for the path Pth0(AB) between node Nd0(A) and node Nd0(B).

[0380] Step S5 In step S5, subcomponent 120A searches for a path Pth2(GH) between node Nd2(G) and node Nd2(H).

[0381] Step S6 In step S6, subcomponent 120A shares routes Pth0(AB) and Pth2(GH) within component 120.

[0382] Step S7 In step S7, component 120 executes prompt chain PC(02). Prompt chain PC(02) includes instruction Pt30(0), instruction Pt30(2), and instruction Pt31(02).

[0383] Instruction Pt30(0) includes instruction g30(0) and path Pth0(AB). Instruction g30(0) includes a procedure to generate a document Doc(AB) that describes the relationship between element Elem(A) and element Elem(B) using path Pth0(AB).

[0384] Instruction Pt30(2) includes instruction g30(2) and path Pth2(GH). Instruction g30(2) includes a procedure to generate a document Doc(GH) that describes the relationship between element Elem(G) and element Elem(H) using path Pth2(GH).

[0385] Instruction Pt31(02) includes instruction g31(02), document Doc(AB), and document Doc(GH). Instruction g31(02) includes a procedure to compare document Doc(AB) and document Doc(GH) to generate a comparison document Doc(02) that explains the difference between the relationship between element Elem(A) and element Elem(B) and the relationship between element Elem(G) and element Elem(H).

[0386] Step S8 In step S8, component 130 responds to prompt chain PC(02) by sending document Doc(AB), document Doc(GH), and comparison document Doc(02) to component 120.

[0387] Step S9 In step S9, component 120 receives document Doc(AB), document Doc(GH), and comparison document Doc(02) and sends them to component 110.

[0388] Step S10 In step S10, component 110 receives document Doc(AB), document Doc(GH), and comparison document Doc(02) and provides them, for example, to a user 99 of the information processing system.

[0389] This allows us to associate nodes in knowledge graph KG2 with nodes in knowledge graph KG0 using the field Fld22. Furthermore, we can use the field Fld22 to select nodes from a group of nodes Nd2 in knowledge graph KG2 that are associated with any of the groups of nodes Nd0 in knowledge graph KG0. We can also select nodes from a group of nodes Nd2 in knowledge graph KG2 that are associated with any of the groups of nodes Nd0 in knowledge graph KG0 to obtain a pair of nodes. Additionally, we can find a pair of nodes in knowledge graph KG0 that corresponds to a pair of nodes in knowledge graph KG2. Furthermore, for example, a pair of nodes including node Nd2(G) which stores element Elem(A) in field Fld22 and node Nd2(H) which stores element Elem(B) in field Fld22 can be obtained from knowledge graph KG2, and a pair of nodes including node Nd0(A) which stores element Elem(A) in field Fld21 and node Nd0(B) which stores element Elem(B) in field Fld21 can be found from knowledge graph KG0. Also, for example, a comparison document Doc(02) can be generated that explains the difference between the path connecting node Nd0(A) and node Nd0(B) and the path connecting node Nd2(G) and node Nd2(H). In addition, the relationships between the elements constituting the invention described in the claims "Claim" can be compared with the relationships between the elements described in the prior art document "Ref". Furthermore, by comparing knowledge graph KG0 and knowledge graph KG2, a comparison document Doc(02) can be generated that compares the configuration described in the patent claims "Claim" with the configuration described in the prior art document "Ref". As a result, a novel information processing method with superior convenience, usefulness, and reliability can be provided.

[0390] <Example of information processing method 2> One aspect of the present invention is an information processing method having a phase PH1 (see Figure 20).

[0391] <Phase PH1> Phase PH1 comprises steps S1 to S10.

[0392] Step S1 In step S1, component 110 receives knowledge graph KG1 and knowledge graph KG2 and transmits them to component 120.

[0393] The knowledge graph KG1 comprises a group of nodes Nd1, each of which comprises fields Fld10, Fld11, and Fld12. Field Fld10 stores the document attribute "Spec", field Fld11 stores a phrase Wd1 identified as an element, and field Fld12 has the function of storing a phrase Wd0 or false False. The phrase Wd0 is a phrase determined to be in a corresponding relationship with the phrase Wd1.

[0394] The knowledge graph KG2 comprises a group of nodes Nd2, each of which contains fields Fld20, Fld21, and Fld22. Field Fld20 stores "Prior Art Document ("Ref")" as a document attribute, field Fld21 stores a phrase Wd2 identified as an element, and field Fld22 has the function of storing a phrase Wd0 or false. Note that a phrase Wd0 is a phrase determined to be in a corresponding relationship with a phrase Wd2.

[0395] Step S2 In step S2, component 120 receives knowledge graphs KG1 and KG2 and shares them within component 120. Component 120 also includes subcomponents 120A and 120B.

[0396] Step S3 In step S3, subcomponent 120A obtains nodes Nd2(G) and Nd2(H) from a group of nodes Nd2. It also obtains nodes Nd1(a) and Nd1(b) from a group of nodes Nd1.

[0397] Node Nd2(G) is a node that stores element Elem(G) in field Fld21 and element Elem(A) in field Fld22.

[0398] Node Nd2(H) is a node that stores element Elem(H) in field Fld21 and element Elem(B) in field Fld22.

[0399] Node Nd1(a) is a node that stores element Elem(a) in field Fld11 and element Elem(A) in field Fld12.

[0400] Node Nd1(b) is a node that stores element Elem(b) in field Fld11 and element Elem(B) in field Fld12.

[0401] Step S4 In step S4, subcomponent 120A searches for a path Pth1(ab) between node Nd1(a) and node Nd1(b).

[0402] Step S5 In step S5, subcomponent 120A searches for a path Pth2(GH) between node Nd2(G) and node Nd2(H).

[0403] Step S6 In step S6, subcomponent 120A shares paths Pth1(ab) and Pth2(GH) within component 120.

[0404] Step S7 In step S7, component 120 executes prompt chain PC(12). Prompt chain PC(12) includes instruction statement Pt30(1), instruction statement Pt30(2), and instruction statement Pt31(12).

[0405] Instruction Pt30(1) includes instruction g30(1) and path Pth1(ab). Instruction g30(1) includes a procedure to generate a document Doc(ab) that describes the relationship between element Elem(a) and element Elem(b) using path Pth1(ab).

[0406] Instruction Pt30(2) includes instruction g30(2) and path Pth2(GH). Instruction g30(2) includes a procedure to generate a document Doc(GH) that describes the relationship between element Elem(G) and element Elem(H) using path Pth2(GH).

[0407] Instruction Pt31(12) includes instruction g31(12), document Doc(ab), and document Doc(GH). Instruction g31(12) includes a procedure to compare document Doc(ab) and document Doc(GH) to generate a comparison document Doc(12) that explains the difference between the relationship between elements Elem(a) and Elem(b) and the relationship between elements Elem(G) and Elem(H).

[0408] Step S8 In step S8, component 130 responds to prompt chain PC(12) by sending document Doc(ab), document Doc(GH), and comparison document Doc(12) to component 120.

[0409] Step S9 In step S9, component 120 receives document Doc(ab), document Doc(GH), and comparison document Doc(12) and sends them to component 110.

[0410] Step S10 In step S10, component 110 receives document Doc(ab), document Doc(GH), and comparison document Doc(12) and provides them to, for example, a user 99 of the information processing system.

[0411] This allows us to associate nodes in knowledge graph KG2 with nodes in knowledge graph KG1 using the field Fld22. Furthermore, we can use the field Fld22 to select nodes from a group of nodes Nd2 in knowledge graph KG2 that are associated with any of the groups of nodes Nd1 in knowledge graph KG1. We can also select nodes from a group of nodes Nd2 in knowledge graph KG2 that are associated with any of the groups of nodes Nd1 in knowledge graph KG1 to obtain a pair of nodes. Additionally, we can find a pair of nodes in knowledge graph KG1 that corresponds to a pair of nodes in knowledge graph KG2. Furthermore, for example, a pair of nodes including node Nd2(G) which stores element Elem(A) in field Fld22 and node Nd2(H) which stores element Elem(B) in field Fld22 can be obtained from knowledge graph KG2, and a pair of nodes including node Nd1(a) which stores element Elem(A) in field Fld12 and node Nd1(b) which stores element Elem(B) in field Fld12 can be found from knowledge graph KG1. Also, for example, a comparison document Doc(12) can be generated that explains the difference between the path connecting node Nd1(a) and node Nd1(b) and the path connecting node Nd2(G) and node Nd2(H). In addition, the relationships between elements described in the specification "Spec" which describes the content of the invention can be compared with the relationships between elements described in the prior art document "Ref". Furthermore, by comparing knowledge graph KG1 and knowledge graph KG2, a comparison document Doc(12) can be generated that compares the configuration described in the specification "Spec" which describes the content of the invention with the configuration described in the prior art document "Ref". As a result, a novel information processing method with superior convenience, usefulness, and reliability can be provided.

[0412] <Example 3 of information processing methods> One aspect of the present invention is an information processing method having phase PH1 and phase PH2 (see Figure 21).

[0413] <Phase PH1> Phase PH1 described in Example 1 of the above information processing method follows Phase PH2 below.

[0414] <Phase PH2> Phase PH2 comprises steps S1 to S10.

[0415] Step S1 In step S1, component 110 receives document Doc(0) and transmits it to component 120. Document Doc(0) is a document containing the claims "Claim".

[0416] Step S2 In step S2, component 120 receives document Doc(0) and shares it within component 120. Component 120 includes subcomponent 120C.

[0417] Step S3 In step S3, subcomponent 120C creates element list EL0 and shares it within component 120. Element list EL0 is a list that stores words Wd0, which have been identified as elements in document Doc(0) through natural language processing.

[0418] Step S4 In step S4, subcomponent 120B selects a pair of elements PoE0 sequentially from the element list EL0.

[0419] Step S5 In step S5, subcomponent 120B creates instruction statement Pt0 and sends it to component 130.

[0420] Instruction Pt0 includes instruction g0(), a pair of elements PoE0 and document Doc(0), where instruction g0() includes a procedure for generating inference result IR0 from document Doc(0). Inference result IR0 also includes representation EDR0(XY) which describes the relationship between one of the elements PoE0 and the other.

[0421] Step S6 In step S6, component 130 receives instruction Pt0, generates inference result IR0 using the large-scale language model LLM, and sends it to component 120.

[0422] Step S7 In step S7, subcomponent 120A creates graph data GD0(XY) from the inference result IR0. Note that graph data GD0(XY) includes nodes Nd0(X), Nd0(Y), and edges Edg0(XY).

[0423] Node Nd0(X) stores "Claim" as a document attribute in field Fld00 and stores one of the pair of elements PoE0 in field Fld01.

[0424] Node Nd0(Y) stores "Claim" as a document attribute in field Fld00 and stores the other of the pair of elements PoE0 in field Fld01.

[0425] Edge Edg0(XY) has a field Fld05. Field Fld05 stores the representation EDR0(XY) that describes the relationship.

[0426] Step S8 In step S8, subcomponent 120A adds graph data GD0(XY) to knowledge graph KG0, and shares knowledge graph KG0 within component 120.

[0427] Step S9 In step S9, component 120 sends the knowledge graph KG0 to component 110.

[0428] Step S10 In step S10, component 110 receives the knowledge graph KG0 and provides it, for example, to a user 99 of the information processing system.

[0429] This allows, for example, when elements Elem(A) and Elem(B) are selected as a pair of elements, the representation EDR0(AB) describing the relationship between elements Elem(A), Elem(B), and Elem(A) and Elem(B) as described in the patent claim "Claim" to be stored in the graph data GD0(AB). Furthermore, the graph data GD0(AB) can be added to the knowledge graph KG0. In addition, the elements constituting the invention described in the patent claim "Claim" and the relationships between the elements constituting the invention can be represented in the knowledge graph KG0. As a result, a novel information processing method with superior convenience, usefulness, and reliability can be provided.

[0430] <Example of information processing method 4> One aspect of the present invention is an information processing method having phase PH1 and phase PH2 (see Figure 21).

[0431] <Phase PH1> Phase PH1 described in Example 2 of the above information processing method follows Phase PH2 below.

[0432] <Phase PH2> Phase PH2 comprises steps S1 to S10.

[0433] Step S1 In step S1, component 110 receives document Doc(1) and correspondence list CL and transmits them to component 120. Document Doc(1) is the specification "Spec" which describes the content of the invention. Correspondence list CL is a list that stores a phrase Wd1 which is determined to be in a corresponding relationship with a phrase Wd0, associating it with the phrase Wd0.

[0434] Step S2 In step S2, component 120 receives the document Doc(1) and the correspondence list CL and shares them within component 120. Component 120 also includes subcomponent 120C.

[0435] Step S3 In step S3, subcomponent 120C creates element list EL1 and shares it within component 120. Element list EL1 is a list that stores words Wd1, which have been identified as elements in document Doc(1) through natural language processing.

[0436] Step S4 In step S4, subcomponent 120B selects a pair of elements PoE1 sequentially from element list EL1.

[0437] Step S5 In step S5, subcomponent 120B creates instruction statement Pt1 and sends it to component 130.

[0438] Instruction Pt1 includes instruction g1(), a pair of elements PoE1 and document Doc(1), where instruction g1() includes a procedure for generating inference result IR1 from document Doc(1). Inference result IR1 also includes representation EDR1(XY) that describes the relationship between one of the pair of elements PoE1 and the other.

[0439] Step S6 In step S6, component 130 receives instruction Pt1, generates inference result IR1 using the large-scale language model LLM, and sends it to component 120.

[0440] Step S7 In step S7, subcomponent 120A creates graph data GD1(XY) from the inference result IR1. Note that graph data GD1(XY) includes node Nd1(X), node Nd1(Y), and edge Edg1(XY).

[0441] Node Nd1(X) stores the "Spec" attribute of the document in field Fld10 and one of the pair of elements PoE1 in field Fld11.

[0442] When one element of a pair of elements PoE1 is associated with element Elem(P) in the correspondence list CL, element Elem(P) is stored in the field Fld12. If one element of a pair of elements PoE1 is not associated with any element in the correspondence list CL, false is stored in the field Fld12.

[0443] Node Nd1(Y) stores the document attribute "Spec" in field Fld10 and the other half of the pair of elements PoE1 in field Fld11.

[0444] When the other element of a pair of elements PoE1 is associated with element Elem(Q) in the correspondence list CL, element Elem(Q) is stored in the field Fld12. If the other element of a pair of elements PoE1 is not associated with any element in the correspondence list CL, false is stored in the field Fld12.

[0445] Edge Edg1(XY) includes a field Fld15. Field Fld15 stores a representation EDR1(XY) that describes the relationship.

[0446] Step S8 In step S8, subcomponent 120A adds graph data GD1(XY) to knowledge graph KG1, and shares knowledge graph KG1 within component 120.

[0447] Step S9 In step S9, component 120 sends the knowledge graph KG1 to component 110.

[0448] Step S10 In step S10, component 110 receives the knowledge graph KG1 and provides it, for example, to a user 99 of the information processing system.

[0449] This allows, for example, when elements Elem(a) and Elem(c) are selected as a pair, the expression EDR1(ac) describing the relationship between elements Elem(a) and Elem(c) as described in the specification "Spec" containing the details of the invention, to be stored in the graph data GD1(ac). Furthermore, the graph data GD1(ac) can be added to the knowledge graph KG1(ac). Additionally, in node Nd1(a) where element Elem(a) is stored in field Fld11, element Elem(A) can be stored in field Fld12 based on the correspondence list CL. Furthermore, when element Elem(c) is not associated with any element in the correspondence list CL, the false value False can be stored in field Fld12 in node Nd1(c) where element Elem(c) is stored in field Fld11. Finally, the elements constituting the technology described in the specification "Spec," and the relationships between elements, can be represented in the knowledge graph KG1. As a result, it is possible to provide a novel information processing method that is superior in terms of convenience, usefulness, or reliability.

[0450] <Example 5 of information processing methods> One aspect of the present invention is an information processing method having phases PH1 to PH3 (see Figure 22).

[0451] <Phase PH1> Phase PH1 described in Example 1 or Example 2 of the above information processing method follows Phase PH3.

[0452] <Phase PH3> Phase PH3 follows Phase PH2 as described in Example 3 or Example 4 of the above information processing method, and Phase PH3 comprises steps S1 to S10.

[0453] Step S1 In step S1, component 110 receives document Doc(2) and correspondence list CL and sends them to component 120. Document Doc(2) is a prior art document "Ref". Correspondence list CL is a list that stores a term Wd2 that is determined to be in a correspondence relationship with a term Wd0, associating it with the term Wd0.

[0454] Step S2 In step S2, component 120 receives the document Doc(2) and the correspondence list CL and shares them within component 120. Component 120 also includes subcomponent 120C.

[0455] Step S3 In step S3, subcomponent 120C creates element list EL2 and shares it within component 120. Element list EL2 is a list that stores words Wd2, which have been identified as elements in document Doc(2) through natural language processing.

[0456] Step S4 In step S4, subcomponent 120B selects a pair of elements PoE2 sequentially from element list EL2.

[0457] Step S5 In step S5, subcomponent 120B creates instruction statement Pt2 and sends it to component 130.

[0458] Instruction Pt2 includes instruction g2(), a pair of elements PoE2 and document Doc(2), where instruction g2() includes a procedure for generating inference result IR2 from document Doc(2). Inference result IR2 also includes representation EDR2(XY) which describes the relationship between one of the pair of elements PoE2.

[0459] Step S6 In step S6, component 130 receives instruction Pt2, generates inference result IR2 using the large-scale language model LLM, and sends it to component 120.

[0460] Step S7 In step S7, subcomponent 120A creates graph data GD2(XY) from the inference result IR2. The graph data GD2(XY) includes nodes Nd2(X), nodes Nd2(Y), and edges Edg2(XY).

[0461] Node Nd2(X) stores "Prior Art Document ("Ref")" as a document attribute in field Fld20 and stores one of the pair of elements PoE2 in field Fld21.

[0462] When one element of a pair of elements PoE2 is associated with element Elem(P) in the correspondence list CL, element Elem(P) is stored in the field Fld22. If one element of a pair of elements PoE2 is not associated with any element in the correspondence list CL, false is stored in the field Fld22.

[0463] Node Nd2(Y) stores "Prior Art Document ("Ref")" as a document attribute in field Fld20 and the other half of the pair of elements PoE2 in field Fld21.

[0464] When the other element of a pair of elements PoE2 is associated with element Elem(Q) in the correspondence list CL, element Elem(Q) is stored in the field Fld22. If the other element of a pair of elements PoE2 is not associated with any element in the correspondence list CL, false is stored in the field Fld22.

[0465] The edge Edg2(XY) has a field Fld25. The field Fld25 stores the representation EDR2(XY) that describes the relationship.

[0466] Step S8 In step S8, subcomponent 120A adds graph data GD2(XY) to knowledge graph KG2, and shares knowledge graph KG2 within component 120.

[0467] Step S9 In step S9, component 120 sends the knowledge graph KG2 to component 110.

[0468] Step S10 In step S10, component 110 receives the knowledge graph KG2 and provides it, for example, to a user 99 of the information processing system.

[0469] This allows, for example, when elements Elem(G) and Elem(I) are selected as a pair of elements, the representation EDR2(GI) describing the relationship between elements Elem(G), Elem(I), and Elem(G) and Elem(I) as described in the prior art document "Ref" to be stored in the graph data GD2(GI). The graph data GD2(GI) can then be added to the knowledge graph KG2. Furthermore, in node Nd2(G) where element Elem(G) is stored in field Fld21, element Elem(A) can be stored in field Fld22 based on the correspondence list CL. Also, when element Elem(I) is not associated with any element in the correspondence list CL, the false value False can be stored in field Fld22 in node Nd2(I) where element Elem(I) is stored in field Fld21. In addition, the elements constituting the technology described in the prior art document "Ref", and the relationships between elements can be represented in the knowledge graph KG2. As a result, it is possible to provide a novel information processing method that is superior in terms of convenience, usefulness, or reliability.

[0470] This embodiment can be appropriately combined with other embodiments shown in this specification. [Explanation of Symbols]

[0471] CL Compatibility List Doc document Elem element False Fld10 Field Fld11 Field Fld12 Field Fld15 Field Fld20 Field Fld21 Field Fld22 Field Fld25 Field g31 instructions Large-Scale Language Model (LLM) Pt31 instructions 20 Information Processing Devices 21 Input section 22 Memory section 23 Processing Unit 24 Output section 25 Transmission lines 51 Network 99 User 110 components 120 components 120A Subcomponent 120B Subcomponent 120C Subcomponents 130 components

Claims

1. The first component and The second component and It has a third component, The first component includes a function for receiving a first knowledge graph and transmitting it to the third component, and a function for receiving and providing a first document. The first knowledge graph comprises a first group of nodes, Each of the first group of nodes comprises a first field and a second field, The first field stores attributes indicating the scope of the claims, The second field stores a first phrase that has been identified as a component of the claims, The first document described above is a document that explains the relationship between the first element and the second element, The first element and the second element are both the first phrase, The second component includes a function to send the first document to the third component in response to a prompt chain, and a function to perform processing using a large-scale language model. The large-scale language model has a function to generate the first document according to the first instruction, The third component includes a function to receive the first knowledge graph and share it within the third component, a function to execute the prompt chain, and a function to receive the first document and send it to the first component. The prompt chain includes the first instruction statement, The third component comprises a first subcomponent and a second subcomponent, The first subcomponent has the function of acquiring a first node and a second node from the first group of nodes, The first node stores the first element in the second field, The second node stores the second element in the second field, The first subcomponent includes a function for searching a first path between the first node and the second node, and a function for sharing the first path within the third component, The second subcomponent has a function for creating the first instruction statement, The first instruction statement includes the first instruction and the first path, An information processing system in which the first instruction includes a procedure for generating a first document describing the relationship between the first element and the second element using the first path.

2. The first component includes a function to receive a second knowledge graph and transmit it to the third component, and a function to receive and provide a second document. The aforementioned second knowledge graph comprises a second group of nodes, The second group of nodes each comprises a third field, a fourth field, and a fifth field, The third field stores attributes indicating prior art documents, The fourth field stores a second phrase that has been identified as an element, The fifth field stores one of the first phrases or false, The first phrase in the above text is a phrase that has been determined to be in a corresponding relationship with the second phrase in the above text. The second document described above is a document that explains the relationship between the third element and the fourth element. The third and fourth elements are both the second phrase, The second component has the function of sending the second document to the third component in response to the prompt chain. The aforementioned large-scale language model includes a function to generate the second document according to the second instruction, The third component includes a function to receive the second knowledge graph and share it within the third component, and a function to receive the second document and transmit it to the first component. The prompt chain includes the second instruction statement, The first subcomponent has the function of acquiring a third node and a fourth node from the second group of nodes, The third node stores the third element in the fourth field and the first element in the fifth field. The fourth node stores the fourth element in the fourth field and the second element in the fifth field. The first subcomponent includes a function for searching for a second path between the third node and the fourth node, and a function for sharing the second path within the third component. The second subcomponent has the function of creating the second instruction statement, The second instruction statement includes the second instruction and the second path, The information processing system according to claim 1, wherein the second instruction includes a step of generating the second document describing the relationship between the third element and the fourth element using the second path.

3. The first component described above has the function of receiving and providing comparison documents, The aforementioned comparison document is a document that explains the difference between the relationship between the first element and the second element, and the relationship between the third element and the fourth element. The second component has the function of sending the comparison document to the third component in response to the prompt chain. The large-scale language model has a function to generate the comparison document according to the third instruction, The third component includes a function to receive the comparison document and transmit it to the first component, The prompt chain includes the third instruction statement, The second subcomponent has the function of creating the third instruction statement, The third instruction statement includes the third instruction, the first document, and the second document, The information processing system according to claim 2, wherein the third instruction includes a step of comparing the first document with the second document to generate a comparison document that explains the differences between the relationship between the first element and the second element and the relationship between the third element and the fourth element.

4. The first component includes a function for receiving a third document and transmitting it to the third component, and a function for receiving and providing the first knowledge graph. The third document is a document describing the claims, The first knowledge graph is the third document converted into a graph format. The second component has the function of receiving a fourth instruction and transmitting the first inference result to the third component. The large-scale language model has a function to generate the first inference result according to the fourth instruction, The third component includes a function to receive the third document and transmit the fourth instruction to the second component, and a function to receive the first inference result and transmit the first knowledge graph to the first component, The third component comprises a third subcomponent, The third subcomponent comprises a function for natural language processing, a function for creating a first element list, and a function for sharing the first element list within the third component. The first element list stores the first phrases that have been identified as elements in the third document by the natural language processing, The second subcomponent comprises a function for selecting a first pair of elements in order from the first element list, and a function for creating the fourth instruction statement. The fourth instruction statement includes a fourth instruction, the first pair of elements, and the third document, and the fourth instruction includes a procedure for generating the first inference result from the third document. The first inference result includes an expression that describes the first relationship between one and the other of the first pair of elements, The first subcomponent includes a function to create first graph data from the first inference result, a function to add the first graph data to the first knowledge graph, and a function to share the first knowledge graph within the third component. The first graph data includes a fifth node, a sixth node, and a first edge. The fifth node stores the attribute indicating the claims in the first field, and stores one of the first pair of elements in the second field. The sixth node stores the attribute indicating the claims in the first field, and stores the other of the first pair of elements in the second field. The first edge comprises a sixth field, The information processing system according to claim 1, wherein the sixth field stores an expression describing the first relationship.

5. The first component includes a function to receive a fourth document and correspondence list and transmit them to the third component, and a function to receive and provide the second knowledge graph. The fourth document mentioned above is the prior art document, The correspondence list stores a second term that is determined to be in a corresponding relationship with a first term, associating it with the first term. The second knowledge graph is the fourth document converted into a graph format. The second component has the function of receiving a fifth instruction and transmitting the second inference result to the third component. The large-scale language model has a function to generate the second inference result according to the fifth instruction, The third component includes a function to receive the fourth document and the correspondence list and to send the fifth instruction to the second component, and a function to receive the second inference result and to send the second knowledge graph to the first component, The third component comprises a third subcomponent, The third subcomponent comprises a function for performing natural language processing, a function for creating a second element list, and a function for sharing the second element list within the third component. The second element list stores the second phrases identified by the natural language processing as elements in the fourth document. The second subcomponent comprises a function to select a second pair of elements in order from the second element list, and a function to create the fifth instruction statement. The fifth instruction statement includes the fifth instruction, the second pair of elements, and the fourth document. The fifth instruction includes a procedure for generating the second inference result from the fourth document, The second inference result includes an expression that describes a third relationship between one and the other of the second pair of elements, The first subcomponent includes a function to create third graph data from the second inference result, a function to add the third graph data to the second knowledge graph, and a function to share the second knowledge graph within the third component. The third graph data includes a seventh node, an eighth node, and a second edge, The seventh node stores the attribute indicating the prior art document in the third field, and stores one of the second pair of elements in the fourth field. When one of the second pair of elements is associated with a fifth element in the correspondence list, the fifth element is stored in the fifth field. If one of the second pair of elements is not associated with any element in the correspondence list, then false is stored in the fifth field. The eighth node stores the attribute indicating the prior art document in the third field, and stores the other of the second pair of elements in the fourth field. When the other of the second pair of elements is associated with the sixth element in the correspondence list, the sixth element is stored in the fifth field. If the other of the second pair of elements is not associated with any element in the correspondence list, then false is stored in the fifth field. The second edge comprises a seventh field, The information processing system according to claim 2, wherein the seventh field stores an expression describing the third relationship.

6. The first component and The second component and It has a third component, The first component includes a function for receiving a first knowledge graph and transmitting it to the third component, and a function for receiving and providing a first document. The first knowledge graph comprises a first group of nodes, Each of the first group of nodes comprises a first field, a second field, and a third field. The first field stores attributes indicating the specification, The second field stores a first phrase that has been identified as an element, The third field stores one second word or false, The second phrase in item 1 is a phrase that has been determined to be in a corresponding relationship with the first phrase in item 1. The first document described above is a document that explains the relationship between the first element and the second element, The first element and the second element are both the first phrase, The second component includes a function to send the first document to the third component in response to a prompt chain, and a function to perform processing using a large-scale language model. The large-scale language model has a function to generate the first document according to the first instruction, The third component includes a function to receive the first knowledge graph and share it within the third component, a function to execute the prompt chain, and a function to receive the first document and send it to the first component. The prompt chain includes the first instruction statement, The third component comprises a first subcomponent and a second subcomponent, The first subcomponent has the function of acquiring a first node and a second node from the first group of nodes, The first node stores the first element in the second field and the third element in the third field. The second node stores the second element in the second field and the fourth element in the third field. The first subcomponent includes a function for searching a first path between the first node and the second node, and a function for sharing the first path within the third component, The second subcomponent has a function for creating the first instruction statement, The first instruction statement includes the first instruction and the first path, An information processing system in which the first instruction includes a procedure for generating a first document describing the relationship between the first element and the second element using the first path.

7. The first component includes a function to receive a second knowledge graph and transmit it to the third component, and a function to receive and provide a second document. The aforementioned second knowledge graph comprises a second group of nodes, The second group of nodes each comprises a fourth field, a fifth field, and a sixth field, The fourth field stores attributes indicating prior art documents, The fifth field stores a third phrase that has been identified as an element, The sixth field stores one of the second phrases or the false value, The second phrase in item 1 is a phrase that has been determined to be in a corresponding relationship with the third phrase in item 1. The second document described above is a document that explains the relationship between the fifth element and the sixth element, The fifth and sixth elements are both the third phrase, The second component has the function of sending the second document to the third component in response to the prompt chain. The aforementioned large-scale language model includes a function to generate the second document according to the second instruction, The third component includes a function to receive the second knowledge graph and share it within the third component, a function to execute the prompt chain, and a function to receive the second document and send it to the first component. The prompt chain includes the second instruction statement, The first subcomponent has the function of acquiring a third node and a fourth node from the second group of nodes, The third node stores the fifth element in the fifth field and the third element in the sixth field. The fourth node stores the sixth element in the fifth field, and stores the fourth element in the sixth field, The first subcomponent includes a function for searching for a second path between the third node and the fourth node, and a function for sharing the second path within the third component. The second subcomponent has the function of creating the second instruction statement, The second instruction statement includes the second instruction and the second path, The information processing system according to claim 6, wherein the second instruction includes a step of generating the second document describing the relationship between the fifth element and the sixth element using the second path.

8. The first component described above has the function of receiving and providing comparison documents, The aforementioned comparison document is a document that explains the difference between the relationship between the first element and the second element, and the relationship between the fifth element and the sixth element. The second component has the function of sending the comparison document to the third component in response to the prompt chain. The large-scale language model has a function to generate the comparison document according to the third instruction, The third component includes a function to execute the prompt chain and a function to receive the comparison document and transmit it to the first component, The prompt chain includes the third instruction statement, The second subcomponent has the function of creating the third instruction statement, The third instruction statement includes the third instruction, the first document, and the second document, The third instruction involves comparing the first document with the second document, The information processing system according to claim 7, comprising a step of generating a comparison document that explains the difference between the relationship between the first element and the second element and the relationship between the fifth element and the sixth element.

9. The first component includes a function to receive a third document and correspondence list and transmit them to the third component, and a function to receive and provide the first knowledge graph. The third document mentioned above is the specification which describes the content of the invention. The aforementioned correspondence list stores a first term that is determined to be in a corresponding relationship with a second term, associating it with the second term. The first knowledge graph is the third document converted into a graph format. The second component has the function of receiving a fourth instruction and transmitting the inference result to the third component. The large-scale language model has a function to generate the inference result according to the fourth instruction, The third component includes a function to receive the third document and transmit the fourth instruction to the second component, and a function to receive the inference result and transmit the first knowledge graph to the first component, The third component comprises a third subcomponent, The third subcomponent comprises a function for natural language processing, a function for creating an element list, and a function for sharing the element list within the third component. The element list stores the first phrases that have been identified as elements in the third document by the natural language processing, The second subcomponent comprises a function for selecting a pair of elements sequentially from the element list, and a function for creating the fourth instruction statement. The fourth instruction statement includes the fourth instruction, the pair of elements, and the third document. The fourth instruction includes a procedure for generating the inference result from the third document, The inference result includes an expression that describes the first relationship between one and the other of the pair of elements, The first subcomponent includes a function for creating first graph data from the inference results, a function for adding the first graph data to the first knowledge graph, and a function for sharing the first knowledge graph within the third component. The first graph data includes a fifth node, a sixth node, and an edge. The fifth node stores the attribute representing the specification in the first field and stores one of the pair of elements in the second field. When one of the pair of elements is associated with the seventh element in the correspondence list, the seventh element is stored in the third field. If one of the pair of elements is not associated with any element in the correspondence list, then false is stored in the third field. The sixth node stores the attribute representing the specification in the first field and the other of the pair of elements in the second field. When the other of the pair of elements is associated with the eighth element in the correspondence list, the eighth element is stored in the third field. If the other of the pair of elements is not associated with any element in the correspondence list, then false is stored in the third field. The aforementioned edge comprises a seventh field, The information processing system according to claim 6, wherein the seventh field stores an expression describing the first relationship.

10. An information processing method having a first phase, The first phase comprises a first to a tenth step, In the first step of the first phase, the first component receives the first knowledge graph and the second knowledge graph and transmits them to the second component. The first knowledge graph comprises a first group of nodes, Each of the first group of nodes comprises a first field and a second field, The first field stores attributes indicating the scope of the claims, The second field stores a first phrase that has been identified as a component of the claims, The aforementioned second knowledge graph comprises a second group of nodes, The second group of nodes each comprises a third field, a fourth field, and a fifth field, The third field stores attributes indicating prior art documents, The fourth field stores a second phrase that has been identified as an element, The fifth field stores one of the first phrases or false, The first phrase in the above text is a phrase that has been determined to be in a corresponding relationship with the second phrase in the above text. In the second step of the first phase, the second component receives the first knowledge graph and the second knowledge graph and shares them within the second component. The second component comprises a first subcomponent and a second subcomponent, In the third step of the first phase, the first subcomponent obtains the first node and the second node from the second group of nodes, and obtains the third node and the fourth node from the first group of nodes, The first node stores the first element in the fourth field and the second element in the fifth field. The second node stores the third element in the fourth field, and the fourth element in the fifth field. The third node stores the second element in the second field, The fourth node stores the fourth element in the second field, In the fourth step of the first phase, the first subcomponent searches for a first path between the third node and the fourth node, In the fifth step of the first phase, the first subcomponent searches for a second path between the first node and the second node. In the sixth step of the first phase, the first subcomponent shares the first path and the second path within the second component, In the seventh step of the first phase, the second component executes the first prompt chain, The first prompt chain includes a first instruction, a second instruction, and a third instruction. The first instruction statement includes the first instruction and the first path, The first instruction includes a procedure for generating a first document describing the relationship between the second element and the fourth element using the first path, The second instruction statement includes the second instruction and the second path, The second instruction includes a procedure for generating a second document describing the relationship between the first element and the third element using the second path, The third instruction statement includes the third instruction, the first document, and the second document, The third instruction includes a procedure for comparing the first document with the second document to generate a first comparison document that explains the differences between the relationship between the second element and the fourth element and the relationship between the first element and the third element, In the eighth step of the first phase, the third component transmits the first document, the second document, and the first comparison document to the second component in response to the first prompt chain. In the ninth step of the first phase, the second component receives the first document, the second document, and the first comparison document and transmits them to the first component. An information processing method in which, in the tenth step of the first phase, the first component receives and provides the first document, the second document, and the first comparison document.

11. An information processing method having a first phase, The first phase comprises a first to a tenth step, In the first step of the first phase, the first component receives the third knowledge graph and the second knowledge graph and transmits them to the second component. The third knowledge graph comprises a third group of nodes, The third group of nodes each comprises a sixth field, a seventh field, and an eighth field, The sixth field stores attributes indicating the specification, The seventh field stores a third word or phrase that has been identified as an element, The eighth field stores one first word or false, The first phrase in item 1 is a phrase that has been determined to be in a corresponding relationship with the third phrase in item 1. The aforementioned second knowledge graph comprises a second group of nodes, The second group of nodes each comprises a third field, a fourth field, and a fifth field, The third field stores attributes indicating prior art documents, The fourth field stores a second phrase that has been identified as an element, The fifth field stores one of the first phrases or the false value, The first phrase in the above text is a phrase that has been determined to be in a corresponding relationship with the second phrase in the above text. In the second step of the first phase, the second component receives the third knowledge graph and the second knowledge graph and shares them within the second component. The second component comprises a first subcomponent and a second subcomponent, In the third step of the first phase, the first subcomponent obtains the first node and the second node from the second group of nodes, and obtains the fifth node and the sixth node from the third group of nodes. The first node stores the first element in the fourth field and the second element in the fifth field. The second node stores the third element in the fourth field, and the fourth element in the fifth field. The fifth node stores the fifth element in the seventh field and the second element in the eighth field. The sixth node stores the sixth element in the seventh field and the fourth element in the eighth field. In the fourth step of the first phase, the first subcomponent searches for a third path between the fifth node and the sixth node, In the fifth step of the first phase, the first subcomponent searches for a second path between the first node and the second node. In the sixth step of the first phase, the first subcomponent shares the third path and the second path within the second component, In the seventh step of the first phase, the second component executes the second prompt chain, The second prompt chain includes a fourth instruction, a second instruction, and a fifth instruction. The fourth instruction statement includes the fourth instruction and the third path, The fourth instruction includes a procedure for generating a third document describing the relationship between the fifth element and the sixth element using the third path, The second instruction statement includes the second instruction and the second path, The second instruction includes a procedure for generating a second document describing the relationship between the first element and the third element using the second path, The fifth instruction statement includes the fifth instruction, the third document, and the second document, The fifth instruction includes a procedure for comparing the third document with the second document to generate a second comparison document that explains the differences between the relationship between the fifth element and the sixth element and the relationship between the first element and the third element, In the eighth step of the first phase, the third component transmits the third document, the second document, and the second comparison document to the second component in response to the second prompt chain. In the ninth step of the first phase, the second component receives the third document, the second document, and the second comparison document and transmits them to the first component. An information processing method in which, in the tenth step of the first phase, the first component receives and provides the third document, the second document, and the second comparison document.

12. An information processing method having the first phase and the second phase, The first phase follows the second phase, The second phase comprises the first to tenth steps, In the first step of the second phase, the first component receives the fourth document and transmits it to the second component. The fourth document is a document that describes the claims, In the second step of the second phase, the second component receives the fourth document and shares it within the second component. The second component comprises a third subcomponent, In the third step of the second phase, the third subcomponent creates a first element list and shares it within the second component. The first element list stores the first words and phrases that have been identified as elements in the fourth document through natural language processing. In the fourth step of the second phase, the second subcomponent selects a first pair of elements in order from the first element list. In the fifth step of the second phase, the second subcomponent creates a sixth instruction and transmits it to the third component. The sixth instruction statement includes the sixth instruction, the first pair of elements, and the fourth document. The sixth instruction includes a procedure for generating a first inference result from the fourth document, The first inference result includes an expression that describes the first relationship between one and the other of the first pair of elements, In the sixth step of the second phase, the third component receives the sixth instruction, generates the first inference result using a large-scale language model, and transmits it to the second component. In the seventh step of the second phase, the first subcomponent creates first graph data from the first inference result, The first graph data includes a seventh node, an eighth node, and a first edge. The seventh node stores the attribute indicating the claims in a first field, and stores one of the first pair of elements in a second field. The eighth node stores the attribute indicating the claims in the first field, and stores the other of the first pair of elements in the second field. The first edge comprises a ninth field, The ninth field stores an expression that describes the first relationship, In the eighth step of the second phase, the first subcomponent adds the first graph data to the first knowledge graph and shares the first knowledge graph within the second component. In the ninth step of the second phase, the second component transmits the first knowledge graph to the first component. The information processing method according to claim 10, wherein in the tenth step of the second phase, the first component receives and provides the first knowledge graph.

13. An information processing method having the first phase and the second phase, The first phase follows the second phase, The second phase comprises the first to tenth steps, In the first step of the second phase, the first component receives the fifth document and correspondence list and transmits them to the second component. The fifth document mentioned above is the specification which describes the content of the invention. The correspondence list stores a third term that is determined to be in a corresponding relationship with a first term, associating it with the first term. In the second step of the second phase, the second component receives the fifth document and the correspondence list and shares them within the second component. The second component comprises a third subcomponent, In the third step of the second phase, the third subcomponent creates a second element list and shares it within the second component. The second element list stores the third phrases that have been identified as elements in the fifth document through natural language processing. In the fourth step of the second phase, the second subcomponent selects a second pair of elements in order from the second element list. In the fifth step of the second phase, the second subcomponent creates a seventh instruction and transmits it to the third component. The seventh instruction statement includes the seventh instruction, the second pair of elements, and the fifth document. The seventh instruction includes a procedure for generating a second inference result from the fifth document, The second inference result includes an expression that describes a third relationship between one and the other of the second pair of elements, In the sixth step of the second phase, the third component receives the seventh instruction, generates the second inference result using a large-scale language model, and transmits it to the second component. In the seventh step of the second phase, the first subcomponent creates third graph data from the second inference result, The third graph data includes a ninth node, a tenth node, and a second edge, The ninth node stores the attribute representing the specification in the sixth field, and stores one of the second pair of elements in the seventh field. When one of the second pair of elements is associated with the seventh element in the correspondence list, the seventh element is stored in the eighth field. If one of the second pair of elements is not associated with any element in the correspondence list, then false is stored in the eighth field. The tenth node stores the attribute representing the specification in the sixth field, and stores the other of the second pair of elements in the seventh field. When the other of the second pair of elements is associated with the eighth element in the correspondence list, the eighth element is stored in the eighth field. If the other of the second pair of elements is not associated with any element in the correspondence list, then false is stored in the eighth field. The second edge comprises a tenth field, The tenth field stores an expression that describes the third relationship, In the eighth step of the second phase, the first subcomponent adds the third graph data to the third knowledge graph and shares the third knowledge graph within the second component. In the ninth step of the second phase, the second component transmits the third knowledge graph to the first component. The information processing method according to claim 11, wherein in the tenth step of the second phase, the first component receives and provides the third knowledge graph.

14. An information processing method having the first to third phases, The first phase follows the third phase, The third phase follows the second phase, The third phase comprises the first to tenth steps, In the first step of the third phase, the first component receives the sixth document and the correspondence list and transmits them to the second component. The sixth document mentioned above is the prior art document, The correspondence list stores a second term that is determined to be in a corresponding relationship with a first term, associating it with the first term. In the second step of the third phase, the second component receives the sixth document and the correspondence list and shares them within the second component. The second component comprises the third subcomponent, In the third step of the third phase, the third subcomponent creates a third element list which is shared within the second component. The third element list stores the second phrases identified by the natural language processing as elements in the sixth document. In the fourth step of the third phase, the second subcomponent selects a third pair of elements in order from the third element list. In the fifth step of the third phase, the second subcomponent creates an eighth instruction and transmits it to the third component. The eighth instruction statement includes the eighth instruction, the third pair of elements, and the sixth document. The eighth instruction includes a procedure for generating a third inference result from the sixth document, The third inference result includes an expression that describes a fifth relationship between one and the other of the third pair of elements, In the sixth step of the third phase, the third component receives the eighth instruction, uses the large-scale language model to generate the third inference result, and transmits it to the second component. In the seventh step of the third phase, the first subcomponent creates fifth graph data from the third inference result, The fifth graph data includes a twelfth node, a thirteenth node, and a third edge, The 12th node stores an attribute indicating the prior art document in the third field, and stores one of the third pair of elements in the fourth field. When one of the third pair of elements is associated with the seventh element in the correspondence list, the seventh element is stored in the fifth field. If one of the third pair of elements is not associated with any element in the correspondence list, then false is stored in the fifth field. The 13th node stores the attribute indicating the prior art document in the third field, and stores the other of the third pair of elements in the fourth field. When the other of the third pair of elements is associated with the eighth element in the correspondence list, the eighth element is stored in the fifth field. If the other of the third pair of elements is not associated with any element in the correspondence list, then false is stored in the fifth field. The third edge comprises an eleventh field, The 11th field stores an expression that describes the 5th relationship, In the eighth step of the third phase, the first subcomponent adds the fifth graph data to the second knowledge graph, and shares the second knowledge graph within the second component. In the ninth step of the third phase, the second component transmits the second knowledge graph to the first component. The information processing method according to claim 13, wherein in the tenth step of the third phase, the first component receives and provides the second knowledge graph.

Images