Code verification program and code verification device

Abstract

The elements and reference numerals in the drawing and its accompanying description are checked against the description and the drawing. [Solution] The code inspection program causes the computer to perform the following steps: input a first prompt to the generating AI system 3, which includes an instruction to generate an explanatory code table 24 containing correspondences between codes and element names from an explanatory text 23 relating to a drawing, and obtain a response from the generating AI that includes a predetermined explanatory code table 24; input a second prompt to the generating AI system 3, which includes an instruction to generate a drawing code table 26 containing correspondences between codes and element names from the drawing described in the explanatory text 23, and obtain a response from the generating AI system 3 that includes a predetermined drawing code table 26; and output a code determination table 27 that determines whether there are errors in the codes from the explanatory code table 24 and the drawing code table 26.

Description

Technical Field

[0001] The present invention relates to a code inspection program and a code inspection device.

Background Art

[0002] With the recent development of IT (information technology), many legaltech technologies for supporting the creation of legal documents by IT have been announced. Also, in the creation of patent specifications, many technologies for creating specifications with the support of information technology have been announced. For example, Patent Document 1 describes a document check device capable of checking the consistency of the use of specific character strings in a document.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Patent Document 1 discloses code check means for checking the consistency between the explanatory terms and codes described in a document. Thereby, it is possible to confirm the consistency between the explanatory terms and codes in the document. However, in a patent specification, it is necessary to check not only the consistency between the explanatory terms and codes in the document but also the consistency between the explanatory terms and codes in the drawings. Patent Document 1 does not disclose taking code consistency considering both such drawings and documents.

[0005] Therefore, an object of the present invention is to inspect the explanatory text, element names, and codes of a drawing and the explanatory text of this drawing.

Means for Solving the Problems

[0006] In other words, the above-mentioned problems of the present invention are solved by the following configuration. The code inspection program of the present invention causes a computer to perform the following steps: input a first prompt to a generating AI that includes an instruction to generate a descriptive code table in which the correspondence between codes and element names is described from the descriptive text relating to a drawing, and obtain a response from the generating AI that includes a predetermined descriptive code table; input a second prompt to the generating AI that includes an instruction to generate a drawing code table in which the correspondence between codes and element names is described from the drawing described in the descriptive text, and obtain a response from the generating AI that includes a predetermined drawing code table; and output a code determination table that determines whether there are errors in the codes from the descriptive code table and the drawing code table.

[0007] The code inspection program of the present invention causes a computer to perform the following steps: input prompts to a generating AI, including instructions to generate a descriptive code table in which the correspondence between codes and element names is described from descriptive texts relating to drawings, and instructions to generate a drawing code table in which the correspondence between codes and element names is described from drawings described in the descriptive texts, and obtain a response from the generating AI that includes a predetermined descriptive code table and a predetermined drawing code table; and output a code determination table that determines whether there are errors in the codes from the descriptive code table and the drawing code table.

[0008] The code inspection device of the present invention is characterized by comprising: a description code table identification unit that receives a first prompt from a generating AI that includes an instruction to generate a description code table in which the correspondence between codes and element names is described from the description text relating to the drawing, and obtains a response that includes a predetermined description code table; a drawing code table identification unit that receives a second prompt from a generating AI that includes an instruction to generate a drawing code table in which the correspondence between codes and element names is described from the drawing described in the description text, and obtains a response that includes a predetermined drawing code table from the generating AI; and a code determination table creation unit that outputs a code determination table in which errors in the code are determined from the description code table and the drawing code table. Other configurations will be described in the embodiments section. [Effects of the Invention]

[0009] According to the present invention, it is possible to check the element names and reference numerals of the drawings and the accompanying text from the drawings and the accompanying text. [Brief explanation of the drawing]

[0010] [Figure 1] This is a logic diagram showing a code inspection device according to the first embodiment. [Figure 2] This is a hardware configuration diagram showing a code inspection device. [Figure 3] This is the screen displayed by the code inspection device. [Figure 4A] This is a flowchart of the code verification process. [Figure 4B] This is a flowchart of the code verification process. [Figure 5] This diagram shows the request sent to the generation AI system. [Figure 6] This is a diagram explaining the content of the prompt. [Figure 7] This diagram shows chat completion objects received from the generation AI system. [Figure 8] This is a diagram illustrating the content of the responses included in the chat completion object. [Figure 9] This diagram shows the request sent to the generation AI system. [Figure 10] This is a diagram explaining the content of the prompt. [Figure 11] This diagram shows chat completion objects received from the generation AI system. [Figure 12A] This is a diagram illustrating the content of the responses included in the chat completion object. [Figure 12B] This is a diagram showing the code table for the provided explanatory texts. [Figure 13] This is a diagram showing the image of the drawing to be attached to the prompt. [Figure 14] This diagram shows the request sent to the generation AI system. [Figure 15] This is a diagram explaining the content of the prompt. [Figure 16]A diagram showing a chat completion object received from a generation AI system. [Figure 17A] A diagram explaining the content of the answer included in the chat completion object. [Figure 17B] A diagram showing the answered drawing symbol table. [Figure 18] A diagram showing a request body to be sent to a generation AI system. [Figure 19] A diagram explaining the content of the prompt. [Figure 20] A diagram showing a chat completion object received from a generation AI system. [Figure 21A] A diagram explaining the content of the answer included in the chat completion object. [Figure 21B] A diagram showing the answered symbol determination table. [Figure 22] A logical configuration diagram showing a symbol inspection device according to the second embodiment. [Figure 23] A flowchart of symbol inspection processing. [Figure 24] A diagram showing a request body to be sent to a generation AI system. [Figure 25] A diagram explaining the content of the prompt. [Figure 26] A diagram showing a chat completion object received from a generation AI system. [Figure 27] A diagram explaining the content of the answer included in the chat completion object. [Figure 28] A flowchart of symbol inspection processing across multiple drawings. [Figure 29] A logical configuration diagram showing a symbol inspection device according to the third embodiment. [Figure 30] A flowchart of symbol inspection processing. [Figure 31] A diagram showing a request body to be sent to a generation AI system. [Figure 32] A diagram explaining the content of the prompt. [Figure 33]This diagram shows chat completion objects received from the generation AI system. [Figure 34] This is a diagram illustrating the content of the responses included in the chat completion object. [Figure 35] This is a logic diagram showing a code inspection device according to the fourth embodiment. [Figure 36] This is a flowchart of the code verification process. [Figure 37] This diagram shows the request sent to the generation AI system. [Figure 38] This is a diagram explaining the content of the prompt. [Figure 39] This diagram shows chat completion objects received from the generation AI system. [Figure 40] This is a diagram illustrating the content of the responses included in the chat completion object. [Figure 41] This diagram explains the contents of the response, including the explanatory code table for drawing number 1. [Figure 42] This is a diagram illustrating the contents of the response, including the drawing reference table for drawing number 1. [Figure 43] This figure shows an example of the explanatory text for drawing number 2. [Figure 44] This is a diagram that explains the contents of the answer, including the explanatory symbol table for drawing number 2. [Figure 45] This figure shows an example of a drawing image for drawing number 2. [Figure 46] This is a diagram illustrating the contents of the response, including the drawing reference table for drawing number 2. [Figure 47] This diagram explains the contents of the answer, including the code determination table. [Modes for carrying out the invention]

[0011] Hereafter, embodiments for carrying out the present invention will be described in detail with reference to the figures. Figure 1 is a logic diagram showing a code inspection device 1 according to the first embodiment. The code inspection device 1 comprises a description text identification unit 11, a description text extraction unit 12, a description text code table identification unit 13, a description text code table extraction unit 14, a drawing code table identification unit 15, a drawing code table extraction unit 16, a code determination table identification unit 17, and a code determination table extraction unit 18. The code inspection device 1 is a computer and, in cooperation with the generation AI system 3, inspects the correspondence between the codes and element names described in the description text 23 specified by the drawing number 22 included in the natural language text 21 and the codes and element names described in the drawing image 25.

[0012] When the description identification unit 11 receives the natural language text 21 and the drawing number 22 as input, it instructs the generation AI system 3, via a request body 51 including a first prompt, to identify the description text 23 relating to the drawing specified by the drawing number 22 from the natural language text 21. The natural language text 21 contains the description text 23 for the drawing specified by the drawing number 22.

[0013] This request 51 is written in JSON format. The request 51 is executed by the generation AI system 3, for example, by calling the API (Application Programming Interface) of the generation AI system 3, and a chat completion object 52 is generated. Then, the generation AI system 3 sends the chat completion object 52 in JSON format to the description identification unit 11. This chat completion object 52 contains the description 23 of the drawing specified by drawing number 22.

[0014] The description extraction unit 12 extracts the description 23 of the drawing specified by drawing number 22 from the chat completion object 52.

[0015] When an explanatory text 23 is input, the explanatory text code table identification unit 13 instructs the generation AI system 3 to identify the explanatory text code table 24, which is a combination of codes and element names described in the explanatory text 23, based on the requester 53.

[0016] The request 53 is written in JSON format. This request 53 is executed by the generation AI system 3, for example, by calling the API of the generation AI system 3, and a chat completion object 54 is generated. The generation AI system 3 then returns the chat completion object 54 in JSON format to the description code table identification unit 13. This chat completion object 54 contains the description code table 24.

[0017] The description code table extraction unit 14 extracts the description code table 24 from the chat completion object 54.

[0018] When a drawing image 25 is input, the drawing code table identification unit 15 instructs the generation AI system 3 to identify the drawing code table 26, which is a combination of symbols and element names drawn in the drawing image 25, based on the request body 55.

[0019] The request 55 is written in JSON format. This request 55 is executed by the generation AI system 3, for example, by calling the API of the generation AI system 3, and a chat completion object 56 is generated. The generation AI system 3 then returns the chat completion object 56 in JSON format to the drawing code table identification unit 15. This chat completion object 56 contains the drawing code table 26.

[0020] The drawing code table extraction unit 16 extracts the drawing code table 26 from the chat completion object 56.

[0021] When the symbol determination table identification unit 17 receives the explanatory symbol table 24 and the drawing symbol table 26 as input, it instructs the generation AI system 3, according to the request body 57, to determine if there are any errors in the symbols from the explanatory symbol table 24 and the drawing symbol table 26.

[0022] The request 57 is written in JSON format. This request 57 is executed by the generation AI system 3, for example, by calling the API of the generation AI system 3, and a chat completion object 58 is generated. The generation AI system 3 then returns the chat completion object 58 in JSON format to the code determination table identification unit 17. This chat completion object 58 contains the code determination table 27.

[0023] The code determination table extraction unit 18 extracts the code determination table 27 from the chat completion object 58.

[0024] Figure 2 is a hardware configuration diagram showing the code inspection device 1. The code inspection device 1 includes a CPU (Central Processing Unit) 101, RAM (Random Access Memory) 102, and ROM (Read Only Memory) 103.

[0025] The CPU 101 is a processing unit that executes computer programs and provides overall control over the code check device 1. The RAM 102 is a volatile read / write memory used by the CPU 101 as a workspace for computer programs. The ROM 103 is a non-volatile read-only memory that stores, for example, the BIOS (Basic I / O System).

[0026] The code inspection device 1 further includes a display unit 104, an input unit 105, a communication unit 106, and a storage unit 107. The display unit 104 is, for example, a liquid crystal display, which displays characters, figures, images, etc. The input unit 105 is, for example, a mouse, keyboard, or touch panel, which the user inputs information into. The communication unit 106 is, for example, a network interface card, which communicates information with, for example, the generation AI system 3.

[0027] The memory unit 107 is a large-capacity storage device such as a hard disk or an SSD (Solid State Drive). The code verification program 1071 is stored in this memory unit 107. When the CPU 101 executes the code verification program 1071, each of the functional units in Figure 1 is realized.

[0028] Figure 3 shows the code determination screen 61 displayed on the display unit 104 by the code inspection device 1. The code determination screen 61 consists of a drawing number combo box 611, a natural language text area 613, a drawing area 614, and a code determination table area 615. This code determination screen 61 is a screen for generating a code determination table by inputting natural language text, a drawing number, and a drawing.

[0029] The drawing number combo box 611 is a combo box for selecting a drawing number. The natural language text area 613 is an area for displaying entered natural language text. For example, a user can register natural language text in the natural language text area 613 by dragging and dropping a natural language text file into the natural language text area 613. The drawing area 614 is the area where the entered drawing is displayed.

[0030] The code determination table area 615 is an area that displays the description of the drawing written in natural language and the code determination table generated from the input drawing.

[0031] Figures 4A and 4B are flowcharts of the sign determination process. First, the description identification unit 11 accepts natural language text and a drawing number as input (step S10). Then, the description identification unit 11 creates a prompt to identify the description text by specifying the natural language text and the drawing number (step S11). The description identification unit 11 sends the request body containing the generated prompt to the generation AI system 3 via API (step S12), and receives a chat completion object from the generation AI system 3 (step S13).

[0032] Next, the description extraction unit 12 extracts the description from the chat completion object (step S14). The description code table identification unit 13 creates a prompt that specifies a description and identifies a code table consisting of combinations of codes and element names described in that description (step S15). The description code table identification unit 13 sends the request body containing the generated prompt to the generation AI system 3 via API (step S16), and receives a chat completion object from the generation AI system 3 (step S17).

[0033] Next, the description code table extraction unit 14 obtains the description code table from the chat completion object (step S18).

[0034] When the drawing code table identification unit 15 receives a drawing input (step S19), it creates a prompt to specify this drawing and identify the code table, which is a combination of symbols and element names drawn in this drawing (step S20). The drawing code table identification unit 15 sends a request containing the generated prompt to the generation AI system 3 via API (step S21), and receives a chat completion object from the generation AI system 3 (step S22).

[0035] Next, the drawing code table extraction unit 16 obtains the drawing code table from the chat completion object (step S23).

[0036] The code determination table identification unit 17 specifies the explanatory code table 24 and the drawing code table 26, and creates a prompt to identify a code determination table that combines these code tables (step S24). The code determination table identification unit 17 sends the request body containing the generated prompt to the generation AI system 3 via API (step S25), and receives a chat completion object from the generation AI system 3 (step S26).

[0037] Next, when the code determination table extraction unit 18 obtains the code determination table from the chat completion object (step S27), the processing shown in Figures 4A and 4B is completed.

[0038] Figure 5 shows a request body 51 containing a prompt to be sent to the generating AI system 3. This request 51 consists of a model item and a messages item. The model item specifies the ID of the generating AI model, followed by a semicolon. The messages item stores a list of messages that make up the conversation so far, and various message types such as text, images, and audio can be set. The messages item contains a role item and a content item in curly braces, followed by a semicolon.

[0039] For the role field, enter either system or user. For the contents field, enter text for the type field and enter prompt 31 from Figure 6 for the text field.

[0040] Figure 6 is a diagram illustrating the content of prompt 31. Prompt 31 contains the instruction text "Output the description of {drawing number} from the embodiment for carrying out the invention in natural language," and "{drawing number} = Figure 2." Prompt 31 is followed by "#natural language," and then the target natural language. In response to Prompt 31, the generating AI system 3 outputs the description of Figure 2 described in the embodiment for carrying out the invention. Note that the instruction in the prompt "from the embodiment for carrying out the invention" may be modified according to the structure of the natural language.

[0041] Figure 7 shows a chat completion object 52 received from the generation AI system 3. This chat completion object 52 consists of the following fields: id, object, created, model, choices, and usage.

[0042] The `id` field is an identifier that uniquely identifies the chat completion object 52. The `object` field indicates the object type, which is always `chat.completion`. The `created` field shows the Unix timestamp in seconds when the chat completion was created. The `model` field is the model used for the chat completion. The `choices` field is a list of choices for the chat completion, which may be multiple. The `usage` field is the usage statistics for the completion request.

[0043] The `choices` field consists of the `index` field, the `role` and `content` fields included in the `message` field, and the `finish_reason` field. The `index` field is the number of the option to complete the chat. The `role` field within the `message` field stores the `assistant`. The `content` field within the `message` field stores the response 32 generated by the AI ​​generation system.

[0044] Figure 8 is a diagram illustrating the content of the response 32 included in the chat completion object 52. This answer, number 32, contains a description of the drawing specified by the drawing number. This description is shown below. "Figure 2 shows the configuration of terminal 4." Terminal 4 comprises a communication unit 41, an input / output unit 42, a control unit 43, and a storage unit 44. The communication unit 41 is... The input / output unit 42 is... The control unit 43 is... The memory unit 44 is...

[0045] Figure 9 shows a request body 53 containing a prompt to be sent to the generating AI system 3. This request 53 consists of a model item and a messages item. The Model item specifies the ID of the generating AI model, followed by a semicolon. The Messages item stores a list of messages that make up the conversation so far, and various message types such as text, images, and audio can be set. The Messages item contains a role item and a content item in curly braces, followed by a semicolon.

[0046] For the role field, enter either system or user. For the contents field, enter text for the type field and enter prompt 33 from Figure 10 for the text field.

[0047] Figure 10 is a diagram illustrating the content of prompt 33. Prompt 33 contains the following instruction text on the first line: "Extract the codes and element names from the description and output a code table for the description. Codes are written as numbers and letters. Element names are the noun phrase immediately preceding the code. For example, for device 100, "100" is the code and "device" is the element name. For step S200, "S200" is the code and "step" is the element name. If the same code corresponds to different element names, please consider it a typographical error." The next line contains "#Description," indicating the description. Subsequent lines contain the description. As a result, the generating AI system 3 extracts the codes and element names from the description and outputs them in a table format.

[0048] Prompt 33, "The code is written with numbers and letters," teaches the Generative AI System 3 the rules for the code. "The element name is the noun phrase immediately preceding the code," teaches the Generative AI System 3 the rules for the element name.

[0049] "For example, in the case of device 100, "100" is the code and "device" is the element name. In the case of step S200, "S200" is the code and "step" is the element name." This provides the generation AI system 3 with specific examples to teach it the code and element name. "When the same code corresponds to different element names, please consider it a typographical error." This teaches the generation AI system 3 about typographical errors in the code.

[0050] Figure 11 shows a chat completion object 54 received from the generation AI system 3. This chat completion object 54 consists of the following fields: id, object, created, model, choices, and usage.

[0051] The `id` field is an identifier that uniquely identifies the chat completion object 54. The `object` field indicates the object type, which is always `chat.completion`. The `created` field shows the Unix timestamp in seconds when the chat completion was created. The `model` field is the model used for the chat completion. The `choices` field is a list of choices for the chat completion, which may be multiple. The `usage` field is the usage statistics for the completion request.

[0052] The `choices` field consists of the `index` field, the `role` and `content` fields included in the `message` field, and the `finish_reason` field. The `index` field is the number of the option to complete the chat. The `role` field within the `message` field stores the `assistant`. The `content` field within the `message` field stores the response 34 generated by the AI ​​generation system.

[0053] Figure 12A is a diagram illustrating the content of the response included in the chat completion object 54. This answer 34 contains the explanatory code table 24, written in Markdown notation. This explanatory code table 24 is a code table showing the combinations of codes and element names described in explanatory text 23. The text of explanatory code table 24 is transcribed below. | Sign | Element name | Notes | | ---- | ---- | ---- | | 4 | Device | Consistent | | 41 | Communications Department | Consistent | | 42 | Input / Output Section | Consistent | | 43 | Control Unit | Consistent | | 44 | Memory section | Consistent |

[0054] Figure 12B shows the answerable explanatory text code table 74. The explanatory code table 74 in Figure 12B is, for example, a Markdown text included in Answer 34, displayed in a Markdown viewer. This explanatory code table 74 consists of a code column, an element name column, and a remarks column. The code column stores the name of the code. The element name column stores the name of the element. The remarks column stores whether or not the use of this code is consistent.

[0055] Figure 13 shows the drawing image 25 to be attached to the prompt. This diagram image 25 is an image of the diagram corresponding to the explanatory text for Figure 2 written in the natural language text. This diagram image 25 is an image that explains the logical block diagram of terminal 4.

[0056] Figure 14 shows a request body 55 containing a prompt to be sent to the generating AI system 3. This request 55 consists of a model field and a messages field. The model field specifies the ID of the generating AI model, followed by a semicolon. The messages field stores a list of messages that make up the conversation so far, and various message types such as text, images, and audio can be set. The messages field contains a role field and a content field, enclosed in curly braces, followed by a semicolon.

[0057] For the role field, enter either system or user. For the contents field, enter text for the type field and enter prompt 35 from Figure 15 for the text field. For the contents field, enter image_url for the type field and enter BASE64-formatted image data or image URL (Uniform Resource Locator) for the image_url field.

[0058] Figure 15 is a diagram illustrating the content of prompt 35. Prompt 35 contains the following instruction text on the first line: "Extract the symbols and element names in Japanese from the attached drawing and output a symbol table for the drawing. Symbols are numbers or letters indicated by leader lines. When each step in the flowchart and sequence diagram is assigned a symbol starting with S, use "Step" as the element name. If there is no element name, leave it as "No element name". If different element names correspond to the same symbol, leave it as a typographical error." In response, the generating AI system 3 extracts the symbols and element names from the attached drawing image and outputs them in a table format.

[0059] "The symbols are numbers or letters indicated by leader lines." This describes the rules for drawing on a drawing. The statement, "When each step in a flowchart and sequence diagram is assigned a code starting with 'S', the element name should be 'Step'," describes the coding rules for flowcharts and sequence diagrams.

[0060] It should be noted that patent drawings are not limited to those where the top of the drawing is oriented upwards; there are also landscape drawings where the left side is oriented upwards. The generating AI can extract symbols and element names from landscape drawings without any special specifications. However, it is also possible to explicitly instruct the generating AI via a prompt to "rotate the drawing 90 degrees clockwise and perform the Japanese OCR (Optical Character Reader) processing again if the Japanese OCR processing of the attached drawing fails."

[0061] The statement, "If there is no element name, please leave it as 'no element name'," refers to situations where element names are not depicted, such as in mechanical drawings where symbols are assigned to the drawing. The statement, "If different element names correspond to the same symbol, please treat it as a typographical error," refers to typographical errors in symbols on drawings.

[0062] Figure 16 shows a chat completion object 56 received from the generation AI system 3. This chat completion object 56 consists of the following fields: id, object, created, model, choices, and usage.

[0063] The `id` field is an identifier that uniquely identifies the chat completion object 56. The `object` field indicates the object type, which is always `chat.completion`. The `created` field shows the Unix timestamp in seconds when the chat completion was created. The `model` field is the model used for the chat completion. The `choices` field is a list of choices for the chat completion, which may be multiple. The `usage` field is the usage statistics for the completion request.

[0064] The `choices` field consists of the `index` field, the `role` and `content` fields included in the `message` field, and the `finish_reason` field. The `index` field is the number of the option to complete the chat. The `role` field within the `message` field stores the `assistant`. The `content` field within the `message` field stores the response 36 generated by the AI ​​generation system 3.

[0065] Figure 17A is a diagram illustrating the content of the response included in the chat completion object 56. This answer 36 contains a drawing code table 26 written in Markdown notation. This drawing code table 26 is a code table showing combinations of symbols and element names depicted in the drawing image 25. The text of the drawing code table 26 is transcribed below. | Sign | Element name | Notes | | ---- | ---- | ---- | | 4 | Device | Consistent | | 41 | Communications Department | Consistent | | 42 | Input / Output Section | Consistent | | 43 | Control Unit | Consistent | | 44 | Memory section | Consistent |

[0066] Figure 17B shows the answerable explanatory text code table 76. The explanatory code table 76 in Figure 17B is, for example, a Markdown text included in Answer 36, displayed in a Markdown viewer. This explanatory code table 76 consists of a code column, an element name column, and a remarks column. The code column stores the name of the code. The element name column stores the name of the element. The remarks column stores whether or not the use of this code is consistent.

[0067] Figure 18 shows a request body 57 containing a prompt to be sent to the generating AI system 3. This request 57 consists of a model field and a messages field. The model field specifies the ID of the generating AI model, followed by a semicolon. The messages field stores a list of messages that make up the conversation so far, and various message types such as text, images, and audio can be set. The messages field contains a role field and a content field, enclosed in curly braces, followed by a semicolon.

[0068] For the role field, enter either system or user. For the contents field, enter text for the type field and enter prompt 37 from Figure 19 for the text field.

[0069] Figure 19 is a diagram illustrating the content of prompt 37. Prompt 37 contains the following instruction text on the first line: "Create a symbol determination table by combining the explanatory symbol table and the drawing symbol table. If the same symbol corresponds to different element names, consider it a typographical error. If a symbol is only listed in the drawing, consider it a typographical error. If a symbol is only listed in the explanatory text, consider it a typographical error." In response, the generating AI system 3 creates a symbol determination table by combining the explanatory symbol table and the drawing symbol table. If the same symbol corresponds to different element names, consider it a typographical error. If a symbol is only listed in the drawing, consider it a typographical error. If a symbol is only listed in the explanatory text, consider it a typographical error.

[0070] Figure 20 shows a chat completion object 58 received from the generation AI system 3. This chat completion object 58 consists of the following fields: id, object, created, model, choices, and usage.

[0071] The `id` field is an identifier that uniquely identifies the chat completion object 58. The `object` field indicates the object type, which is always `chat.completion`. The `created` field shows the Unix timestamp in seconds when the chat completion was created. The `model` field is the model used for the chat completion. The `choices` field is a list of choices for the chat completion, which may be multiple. The `usage` field is the usage statistics for the completion request.

[0072] The `choices` field consists of the `index` field, the `role` and `content` fields included in the `message` field, and the `finish_reason` field. The `index` field is the number of the option to complete the chat. The `role` field within the `message` field stores the `assistant`. The `content` field within the `message` field stores the response 38 generated by the AI ​​generation system.

[0073] Figure 21A is a diagram illustrating the content of the response included in the chat completion object 58. This answer 38 contains a symbol determination table 27 written in Markdown notation. This symbol determination table 27 is a table that determines the combinations of symbols and element names described in the explanatory text 23 and the errors in the symbols and element names depicted in the drawing image 25. The text of the symbol determination table 27 is transcribed below. | Symbol | Element name in description | Element name in drawing | Judgment | | ---- | ---- | ---- | ---- | | 4 | Terminal | Terminal | Match | | 41 | Communications Department | Communications Department | Match | | 42 | Input / output section | Input / output section | Match | | 43 | Control Unit | Control Unit | Match | | 44 | Storage | Storage | Match |

[0074] Figure 21B shows the code determination table provided in the response. The code determination table 77 in Figure 21B is, for example, a Markdown text included in answer 38, displayed in a Markdown viewer. Specifically, this code determination table 77 is displayed when you access ChatGPT4o in a browser and view the results of entering prompts.

[0075] This symbol determination table 77 consists of a symbol column, a description element name column, a drawing element name column, a remarks column, and a determination column. The symbol column stores the name of the symbol. The description element name column stores the element name extracted from the description. The drawing element name column stores the element name extracted from the drawing. The remarks column stores whether the use of this symbol is consistent or not. The determination column stores whether the use of this symbol is a typographical error or not.

[0076] Figure 22 is a logic diagram showing the code inspection device 1 according to the second embodiment. The code inspection device 1 comprises a descriptive text code table identification unit 131, a descriptive text code table extraction unit 141, a drawing code table identification unit 15, a drawing code table extraction unit 16, and a code determination table creation unit 171. The code inspection device 1 works in cooperation with the generation AI system 3 to inspect the correspondence between the codes and element names described in the descriptive text specified by the drawing number 22 included in the natural language text 21 and the codes and element names described in the drawing image 25.

[0077] When the natural language text 21 and drawing number 22 are input, the explanatory text code table identification unit 131, based on the request body 511, identifies the explanatory text specified by drawing number 22 from the natural language text 21 and instructs the generation AI system 3 to identify the explanatory text code table 24, which is a combination of codes and element names described in this explanatory text.

[0078] The request 511 is written in JSON format. This request 511 is executed by the generation AI system 3, for example, by calling the API of the generation AI system 3, and a chat completion object 521 is generated. The generation AI system 3 then returns the chat completion object 521 in JSON format to the description code table identification unit 131. This chat completion object 521 contains the description code table 24.

[0079] The description code table extraction unit 14 extracts the description code table 24 from the chat completion object 521.

[0080] When a drawing image 25 is input, the drawing code table identification unit 15 instructs the generation AI system 3 to identify the drawing code table 26, which is a combination of symbols and element names drawn in the drawing image 25, based on the request body 55.

[0081] The request 55 is written in JSON format. This request 55 is executed by the generation AI system 3, for example, by calling the API of the generation AI system 3, and a chat completion object 56 is generated. The generation AI system 3 then returns the chat completion object 56 in JSON format to the drawing code table identification unit 15. This chat completion object 56 contains the drawing code table 26.

[0082] The drawing code table extraction unit 16 extracts the drawing code table 26 from the chat completion object 56.

[0083] When the symbol determination table creation unit 171 receives the explanatory symbol table 24 and the drawing symbol table 26 as input, it generates a symbol determination table 27 from the explanatory symbol table 24 and the drawing symbol table 26 in a rule-based manner.

[0084] Figure 23 is a flowchart of the sign determination process. Initially, the explanatory text code table identification unit 131 accepts the input of natural language text and a drawing number (step S30). The explanatory text code table identification unit 131 creates a prompt that specifies the natural language text and the drawing number to identify the code table, which is a combination of symbols and element names described in the explanatory text of the drawing specified by the drawing number (step S31). The explanatory text code table identification unit 131 sends the request body containing the generated prompt to the generation AI system 3 via API (step S32), and receives a chat completion object from the generation AI system 3 (step S33).

[0085] Next, the description code table extraction unit 141 obtains the description code table from the chat completion object (step S34).

[0086] When the drawing code table identification unit 15 receives a drawing input (step S35), it creates a prompt to specify this drawing and identify the code table, which is a combination of symbols and element names drawn in this drawing (step S36). The drawing code table identification unit 15 sends the request body containing the generated prompt to the generation AI system 3 via API (step S37), and receives a chat completion object from the generation AI system 3 (step S38).

[0087] Next, the drawing code table extraction unit 16 obtains the drawing code table from the chat completion object (step S39).

[0088] The code determination table creation unit 171 creates a code determination table by combining the explanatory code table 24 and the drawing code table 26 (step S40), and the process shown in Figure 23 is completed.

[0089] Figure 24 shows a request body 511 containing a prompt to be sent to the generating AI system 3. This request 511 consists of a model field and a messages field. The model field specifies the ID of the generating AI model, followed by a semicolon. The messages field stores a list of messages that make up the conversation so far, and various message types such as text, images, and audio can be set. The messages field contains a role field and a content field, enclosed in curly braces, followed by a semicolon.

[0090] For the role field, enter either system or user. For the contents field, enter text for the type field and enter prompt 311 from Figure 25 for the text field.

[0091] Figure 25 is a diagram illustrating the content of prompt 311. Prompt 311 contains the instruction text: "Identify the description of {drawing number} from the embodiment for carrying out the invention in natural language, extract the reference numerals and element names from the identified description, and output a table of reference numerals for the description. Reference numerals are written as numbers and letters. Element names are the noun phrase immediately preceding the reference numeral. For example, for apparatus 100, "100" is the reference numeral and "apparatus" is the element name. For step S200, "S200" is the reference numeral and "step" is the element name. If the same reference numeral corresponds to different element names, please consider it a typographical error." and "{drawing number} = Figure 2". Following this, the target natural language is written after "#natural language". As a result, the generating AI system 3 outputs the description of Figure 2.

[0092] Figure 26 shows a chat completion object 521 received from the generation AI system 3. This chat completion object 521 consists of the following fields: id, object, created, model, choices, and usage.

[0093] The `id` field is an identifier that uniquely identifies the chat completion object 521. The `object` field indicates the object type, which is always `chat.completion`. The `created` field shows the Unix timestamp in seconds when the chat completion was created. The `model` field is the model used for the chat completion. The `choices` field is a list of choices for the chat completion, which may be multiple. The `usage` field is the usage statistics for the completion request.

[0094] The `choices` field consists of the `index` field, the `role` and `content` fields included in the `message` field, and the `finish_reason` field. The `index` field is the number of the option to complete the chat. The `role` field within the `message` field stores the `assistant`. The `content` field within the `message` field stores the response 321 generated by the AI ​​generation system.

[0095] Figure 27 is a diagram illustrating the content of response 321 included in chat completion object 521. This answer 321 contains a descriptive symbol table 24, which is a combination of symbols and element names listed in the descriptive text of the drawing specified by the drawing number. The text of this descriptive symbol table 24 is transcribed below. | Sign | Element name | Notes | | ---- | ---- | ---- | | 4 | Device | Consistent | | 41 | Communications Department | Consistent | | 42 | Input / Output Section | Consistent | | 43 | Control Unit | Consistent | | 44 | Memory section | Consistent |

[0096] In patent specifications, the overall configuration of the device is often shown in Figures 1 and 2, and the operation of the device is explained in subsequent drawings. In other words, the explanatory text for drawings numbered 1 through n may be based on the reference numerals indicated in drawings numbered 1 through n. Therefore, the following flowchart explains how to check drawings and explanatory text when explaining multiple drawings sequentially with explanatory text.

[0097] Figure 28 is a flowchart of the code determination process across multiple drawings. The CPU 101 of the code inspection device 1 repeats the process from step S80 to step S90 from the beginning to the end of the drawing number (step S80).

[0098] The explanatory text code table identification unit 131 specifies the natural language text and the drawing number, and creates a prompt to identify the code table, which is a combination of symbols and element names described in the explanatory text of the drawing specified by the drawing number (step S81). The explanatory text code table identification unit 131 sends the request body containing the generated prompt to the generation AI system 3 via API (step S82), and receives a chat completion object from the generation AI system 3 (step S83).

[0099] Next, the explanatory text code extraction unit 141 obtains the explanatory text code table for the drawing with the corresponding drawing number from the chat completion object (step S84).

[0100] The drawing code table identification unit 15 specifies the drawing related to the drawing number and creates a prompt to identify the code table, which is a combination of symbols and element names drawn in that drawing (step S85). The drawing code table identification unit 15 sends the request containing the generated prompt to the generation AI system 3 via API (step S86) and receives a chat completion object from the generation AI system 3 (step S87).

[0101] Next, the drawing code extraction unit 16 obtains the drawing code table for the drawing number from the chat completion object (step S88).

[0102] The code determination table creation unit 171 creates a code determination table by combining the descriptive code table 24 from the first drawing number to the current drawing number and the code tables from the drawing code table 26 of the first drawing number to the drawing code table 26 of the current drawing number (step S89).

[0103] Then, the CPU 101 of the code inspection device 1 determines whether or not the process from step S80 to step S90 has been repeated to the end of the drawing number (step S90). If these processes have not been repeated to the end, it returns to step S80 and repeats the process. If these processes have been repeated to the end, the process in Figure 28 is completed.

[0104] This makes it possible to check for errors in the correspondence between symbols and element names in drawings and their descriptions, even if the description of drawing number n refers to symbols that are included in drawings with earlier drawing numbers than drawing number n.

[0105] Figure 29 is a logic diagram showing the code inspection device 1 according to the third embodiment. The code inspection device 1 comprises a code table identification unit 19, a code table extraction unit 10, and a code determination table creation unit 171. The code inspection device 1 works in cooperation with the generation AI system 3 to inspect the correspondence between the codes and element names described in the explanatory text specified by the drawing number 22 included in the natural language text 21 and the codes and element names depicted in the drawing image 25.

[0106] When the symbol table identification unit 19 receives the natural language text 21, the drawing number 22, and the drawing image 25 as input, it instructs the generation AI system 3 to identify the explanatory text specified by the drawing number 22 from the natural language text 21, identify the explanatory text symbol table 24 which is a combination of symbols and element names described in this explanatory text, and identify the drawing symbol table 26 which is a combination of symbols and element names depicted in the drawing image 25.

[0107] The request 59 is written in JSON format. This request 59 is executed by the generation AI system 3, for example, by calling the API of the generation AI system 3, and a chat completion object 50 is generated. The generation AI system 3 then returns the chat completion object 50 in JSON format to the code table identification unit 19. This chat completion object 50 includes the description code table 24 and the drawing code table 26.

[0108] The code table extraction unit 10 extracts the explanatory code table 24 and the drawing code table 26 from the chat completion object 521.

[0109] When the symbol determination table creation unit 171 receives the explanatory symbol table 24 and the drawing symbol table 26 as input, it generates a symbol determination table 27 from the explanatory symbol table 24 and the drawing symbol table 26 in a rule-based manner.

[0110] Figure 30 is a flowchart of the sign determination process. Initially, the code table identification unit 19 accepts natural language text, a drawing number, and a drawing image as input (step S60). The code table identification unit 19 specifies the natural language text, the drawing number, and the drawing, and creates a prompt to identify the code table which is a combination of symbols and element names described in the description of the drawing specified by the drawing number in the natural language text, and to identify the code table which is a combination of symbols and element names drawn on the drawing (step S51). The code table identification unit 19 sends the request body containing the generated prompt to the generation AI system 3 via API (step S52), and receives a chat completion object from the generation AI system 3 (step S53).

[0111] Next, the code table extraction unit 10 obtains the description code table and the drawing code table from the chat completion object (step S54).

[0112] The code determination table creation unit 171 creates a code determination table by combining the explanatory code table 24 and the drawing code table 26 (step S55), at which point the process in Figure 30 is completed.

[0113] Figure 31 shows a request body 59 containing a prompt to be sent to the generating AI system 3. This request 59 consists of a model field and a messages field. The model field specifies the ID of the generating AI model, followed by a semicolon. The messages field stores a list of messages that make up the conversation so far, and various message types such as text, images, and audio can be set. The messages field contains a role field and a content field, enclosed in curly braces, followed by a semicolon.

[0114] For the role field, enter either system or user. For the contents field, enter text for the type field and enter prompt 39 from Figure 31 for the text field. For the contents field, enter image_url for the type field and enter BASE64-formatted image data or image URL (Uniform Resource Locator) for the image_url field.

[0115] Figure 32 is a diagram illustrating the content of prompt 39. Prompt 39 contains the following instruction text: "Identify the description of {drawing number} from the embodiment for carrying out the invention in natural language, extract the reference numerals and element names from the identified description, and output a table of reference numerals for the description. Reference numerals are written as numbers and letters. Element names are the noun phrase immediately preceding the reference numeral. For example, in device 100, "100" is the reference numeral and "device" is the element name. In step S200, "S200" is the reference numeral and "step" is the element name. If the same reference numeral corresponds to different element names, please consider it a typographical error." and "Extract the reference numerals and element names in Japanese from the attached drawing and output a table of reference numerals for the drawing. Reference numerals are numbers or letters indicated by leader lines. When each step in the flowchart and sequence diagram is assigned a reference numeral starting with S, please consider the element name to be "step". If there is no element name, please consider it as no element name. If the same reference numeral corresponds to different element names, please consider it a typographical error." and "{drawing number} = Figure 2". Subsequently, the target natural language text is written after "#natural language text". This allows the generation AI system 3 to output the explanatory text code table and the drawing code table for Figure 2.

[0116] Figure 33 shows a chat completion object 50 received from the generation AI system 3. This chat completion object 50 consists of the following fields: id, object, created, model, choices, and usage.

[0117] The `id` field is an identifier that uniquely identifies the chat completion object. The `object` field indicates the object type, which is always `chat.completion`. The `created` field shows the Unix timestamp in seconds when the chat completion was created. The `model` field is the model used for the chat completion. The `choices` field is a list of choices for the chat completion, which may be multiple. The `usage` field is the usage statistics for the completion request.

[0118] The `choices` field consists of the `index` field, the `role` and `content` fields included in the `message` field, and the `finish_reason` field. The `index` field is the number of the option to complete the chat. The `role` field within the `message` field stores the `assistant`. The `content` field within the `message` field stores the response 30 generated by the AI ​​generation system.

[0119] Figure 34 is a diagram illustrating the content of the response 30 included in the chat completion object 50. This answer 30 includes a descriptive symbol table 24, which contains combinations of symbols and element names listed in the descriptive text of the drawing specified by the drawing number, and a drawing symbol table 26, which contains combinations of symbols and element names depicted in the drawing image 25. The text of this descriptive symbol table 24 and drawing symbol table 26 is transcribed below. The explanatory code table is as follows: | Sign | Element name | Notes | | ---- | ---- | ---- | | 4 | Device | Consistent | | 41 | Communications Department | Consistent | | 42 | Input / Output Section | Consistent | | 43 | Control Unit | Consistent | | 44 | Memory section | Consistent | The drawing reference chart is as follows: | Sign | Element name | Notes | | ---- | ---- | ---- | | 4 | Device | Consistent | | 41 | Communications Department | Consistent | | 42 | Input / Output Section | Consistent | | 43 | Control Unit | Consistent | | 44 | Memory section | Consistent |

[0120] Figure 35 is a logic diagram showing the code inspection device 1 according to the fourth embodiment. The code inspection device 1 comprises a code determination table identification unit 172 and a code determination table extraction unit 182. The code inspection device 1 works in cooperation with the generation AI system 3 to inspect the correspondence between the codes and element names described in the explanatory text specified by the drawing number 22 included in the natural language text 21 and the codes and element names depicted in the drawing image 25.

[0121] When the code determination table identification unit 172 receives the natural language text 21, the drawing number 22, and the drawing image 25 as input, it instructs the generation AI system 3 to identify the explanatory text specified by the drawing number 22 from the natural language text 21, identify the explanatory text code table 24 which is a combination of codes and element names described in this explanatory text, identify the drawing code table 26 which is a combination of codes and element names depicted in the drawing image 25, and further identify the code determination table 27.

[0122] Request 591 is written in JSON format. This request 591 is executed by the generation AI system 3, for example by calling the API of the generation AI system 3, and a chat completion object 501 is generated. The generation AI system 3 then returns the chat completion object 501 in JSON format to the code determination table extraction unit 182. This chat completion object 501 includes the description code table 24, the drawing code table 26, and the code determination table 27.

[0123] The code determination table extraction unit 182 extracts the explanatory code table 24, the drawing code table 26, and the code determination table 27 from the chat completion object 521.

[0124] Figure 36 is a flowchart of the sign determination process. Initially, the code determination table identification unit 172 accepts natural language text, a drawing number, and a drawing image as input (step S60). The code determination table identification unit 172 specifies the natural language text, the drawing number, and the drawing, and creates prompts to identify the code table, which is a combination of codes and element names described in the description of the drawing specified by the drawing number in the natural language text, to identify the code table, which is a combination of codes and element names drawn on the drawing, and to identify the code determination table 27 (step S61). The code determination table identification unit 172 sends the request body containing the generated prompts to the generation AI system 3 via API (step S62), and receives a chat completion object from the generation AI system 3 (step S63).

[0125] Next, the code determination table extraction unit 182 obtains the description code table, drawing code table, and code determination table from the chat completion object (step S64), and the process shown in Figure 36 is completed.

[0126] Figure 37 shows a request body 591 containing a prompt to be sent to the generating AI system 3. This request 591 consists of a model field and a messages field. The model field specifies the ID of the generating AI model, followed by a semicolon. The messages field stores a list of messages that make up the conversation so far, and various message types such as text, images, and audio can be set. The messages field contains a role field and a content field, enclosed in curly braces, followed by a semicolon.

[0127] For the role field, enter either system or user. For the contents field, enter text for the type field and enter prompt 391 from Figure 37 for the text field. For the contents field, enter image_url for the type field and enter BASE64-formatted image data or image URL (Uniform Resource Locator) for the image_url field.

[0128] Figure 38 is a diagram illustrating the content of prompt 391. Prompt 391 contains the instruction text: From the natural language description of the invention, identify the description of {drawing number}, and then extract the symbols and element names from the identified description to output a symbol table for the description. Symbols are written as numbers and letters. Element names are the noun phrase immediately preceding the symbol. For example, in device 100, "100" is the symbol and "device" is the element name. In step S200, "S200" is the symbol and "step" is the element name. Also, extract the symbols and element names in Japanese from the attached drawing and output a symbol table for the drawing. Symbols are numbers or letters indicated by leader lines. When each step in the flowchart and sequence diagram is assigned a symbol starting with S, use "step" as the element name. If there is no element name, use "no element name". If the same symbol corresponds to different element names, use "error". The text includes "Then, create a code determination table by combining the two code tables. If different element names correspond to the same code, consider it a typographical error. Codes that are only listed in the drawing should be considered typographical errors. Codes that are only listed in the explanatory text should be considered typographical errors." and "{Drawing Number}=Figure 2". After that, the target natural language text is written after "#Natural Language Text". As a result, the generation AI system 3 outputs an explanatory language code table, a drawing code table, and a code determination table.

[0129] Figure 39 shows a chat completion object 501 received from the generation AI system 3. This chat completion object 501 consists of the following fields: id, object, created, model, choices, and usage.

[0130] The `id` field is an identifier that uniquely identifies the chat completion object 501. The `object` field indicates the object type, which is always `chat.completion`. The `created` field shows the Unix timestamp in seconds when the chat completion was created. The `model` field is the model used for the chat completion. The `choices` field is a list of choices for the chat completion, which may be multiple. The `usage` field is the usage statistics for the completion request.

[0131] The `choices` field consists of the `index` field, the `role` and `content` fields included in the `message` field, and the `finish_reason` field. The `index` field is the number of the option to complete the chat. The `role` field within the `message` field stores the `assistant`. The `content` field within the `message` field stores the response 301 generated by the AI ​​generation system.

[0132] Figure 40 is a diagram illustrating the content of response 301 included in chat completion object 501. This answer 301 contains a descriptive symbol table 24, which lists the combinations of symbols and element names described in the descriptive text of the drawing specified by the drawing number; a drawing symbol table 26, which lists the combinations of symbols and element names depicted in the drawing image 25; and a symbol determination table 27. The text of this descriptive symbol table 24, drawing symbol table 26, and symbol determination table 27 is transcribed below. The explanatory code table is as follows: | Sign | Element name | Notes | | ---- | ---- | ---- | | 4 | Device | Consistent | | 41 | Communications Department | Consistent | | 42 | Input / Output Section | Consistent | | 43 | Control Unit | Consistent | | 44 | Memory section | Consistent | The drawing reference chart is as follows: | Sign | Element name | Notes | | ---- | ---- | ---- | | 4 | Device | Consistent | | 41 | Communications Department | Consistent | | 42 | Input / Output Section | Consistent | | 43 | Control Unit | Consistent | | 44 | Memory section | Consistent | The code determination table is as follows: | Symbol | Element name in description | Element name in drawing | Judgment | | ---- | ---- | ---- | ---- | | 4 | Terminal | Terminal | Match | | 41 | Communications Department | Communications Department | Match | | 42 | Input / output section | Input / output section | Match | | 43 | Control Unit | Control Unit | Match | | 44 | Storage | Storage | Match |

[0133] Figure 41 is a diagram illustrating the contents of response 321, including the explanatory code table for drawing number 1. The explanatory code table for drawing number 1 included in response 321 lists additive manufacturing apparatus 1, wire W, power supply 19, material supply unit 16, laser oscillator 11, and laser beam LB. The use of all these codes has been determined to be consistent.

[0134] Figure 42 is a diagram illustrating the contents of Answer 361, including the drawing reference table for drawing number 1. The reference numeral table for drawing number 1 included in response 361 lists additive manufacturing apparatus 1, wire W, power supply 19, material supply unit 16, laser oscillator 11, and laser beam LB. The use of all these reference numerals has been determined to be consistent.

[0135] Figure 43 is a diagram showing an example of the explanatory text 232 for drawing number 2. The text of explanation 232 is shown below. "Figure 2 is a flowchart showing an example of the operation procedure in the manufacturing of an object using an additive manufacturing device." First, the additive manufacturing apparatus 1 energizes the wire W, which is the manufacturing material (step S70). The additive manufacturing apparatus 1 energizes the wire W by supplying current from the power supply 19 according to the current command. Step S70 corresponds to the energization process. Next, the additive manufacturing apparatus 1 supplies the wire W, which is the manufacturing material, to the workpiece (step S71). The manufacturing material supply unit 16 supplies the wire W at a supply speed according to the supply command. Step S71 corresponds to the manufacturing material supply process. Subsequently, the additive manufacturing apparatus 1 outputs a laser beam LB from the laser oscillator 11, irradiating the workpiece with the laser beam LB (step S72), and the process shown in Figure 8 is completed.

[0136] Figure 44 is a diagram illustrating the contents of Answer 322, including the explanatory symbol table for drawing number 2. The explanatory code table included in response 322 lists the combinations of codes and element names included in explanatory text 232 as a Markdown table. This code table lists additive manufacturing device 1, wire W, step S70, power supply 19, step S71, material supply unit 16, step S72, laser oscillator 11, and laser beam LB. The use of all these codes has been determined to be consistent.

[0137] Figure 45 shows an example of drawing image 252 of drawing number 2. Drawing image 252 is a flowchart described in explanatory text 232. By attaching this drawing image 252 and prompting the generation AI system 3 to extract the symbols and element names, the drawing symbol table described below is generated.

[0138] Figure 46 is a diagram illustrating the contents of Answer 362, including the drawing reference table for drawing number 2. The drawing code table included in response 362 lists the combinations of symbols and element names depicted in drawing image 252 as a Markdown table. This code table includes steps S70, S71, and S72. The use of these symbols has been determined to be consistent.

[0139] Figure 47 is a diagram illustrating the contents of Answer 382, ​​which includes a reference number determination table for drawings 1 and 2. The code determination table included in Answer 382 is created by combining the descriptive code table for drawings 1 and 2 with the drawing code table for drawings 1 and 2. This code determination table indicates that step S72 is not described in the descriptive text.

[0140] The configuration and effects of the present invention will be described below. [1] A procedure (steps S15-S18) in which a first prompt (prompt 33) is input to the generating AI (generating AI system 3) that includes an instruction to generate an explanatory text code table (24) which contains correspondences between symbols and element names from explanatory text (23) relating to a drawing, and a response including the predetermined explanatory text code table (24) is obtained from the generating AI (generating AI system 3), A procedure (steps S20-S23) in which a second prompt (prompt 35) is input to the generating AI (generating AI system 3) that includes an instruction to generate a drawing code table (26) that shows the correspondence between symbols and element names from the drawing (drawing image 25) described in the explanatory text (23), and a response including the predetermined drawing code table (26) is obtained from the generating AI, A procedure (steps S24 to S27) for outputting a symbol determination table (27) that determines the misprint of symbols from the explanatory symbol table (24) and the drawing symbol table (26), A code verification program that causes a computer to perform a code check.

[0141] This makes it possible to check the element names and reference numerals of the drawing and its accompanying description from the description and the drawing itself.

[0142] [2] A procedure for inputting a third prompt (prompt 31) to a generating AI (generating AI system 3) that includes an instruction to generate a descriptive text (23) about a drawing from natural language text (21) and a drawing number (22), and obtaining a response including the descriptive text (23) from the generating AI (generating AI system 3), A code check program according to claim 1 (1071) for further implementation of the above.

[0143] This allows the AI ​​to extract descriptive text related to the drawing to be evaluated from the entire natural language text.

[0144] [3] The first prompt (prompt 33) and / or the second prompt (prompt 35) include instructions to determine that a typographical error is present if the same code corresponds to multiple element names. The code check program (1071) according to claim 1.

[0145] This allows for the correction of errors when the same symbol corresponds to multiple element names in both the drawing and the natural language text.

[0146] [4] A procedure to input a fourth prompt (prompt 37) to the generating AI (generating AI system 3) which includes an instruction to output a symbol determination table (27) that determines symbol errors from the explanatory symbol table (24) and the drawing symbol table (26), and to obtain a response including a predetermined symbol determination table (27) from the generating AI (generating AI system 3), A code check program (1071) according to claim 1 for performing the following.

[0147] [5] The fourth prompt (prompt 37) includes an instruction to determine that a symbol present in the drawing symbol table (26) but not in the explanatory symbol table (24), and a symbol present in the explanatory symbol table (24) but not in the drawing symbol table (26), are typographical errors. A code check program (1071) according to claim 4 for performing the above.

[0148] This makes it possible to have the generating AI inspect the element names and symbols listed in the drawings and their descriptions.

[0149] [6] A procedure for generating a symbol determination table (27) in which symbols present in the drawing symbol table (26) but not in the explanatory text symbol table (24), and symbols present in the explanatory text symbol table (24) but not in the drawing symbol table (26) are determined to be typographical errors. A code check program (1071) according to claim 4 for performing the above.

[0150] This allows for rule-based inspection of element names and symbols listed in drawings and their descriptions.

[0151] [7] A procedure to input prompts (39) to a generating AI (generating AI system 3) including an instruction to generate an explanatory text code table (24) which contains correspondences between symbols and element names from explanatory text (23) about a drawing, and an instruction to generate a drawing code table (26) which contains correspondences between symbols and element names from the drawing described in the explanatory text (23), and to obtain a response from the generating AI (generating AI system 3) which includes a predetermined explanatory text code table (24) and a predetermined drawing code table (26), A procedure for outputting a symbol determination table (27) that determines the misprint of symbols from the explanatory symbol table (24) and the drawing symbol table (26), A code check program (1071) that causes a computer to perform this check.

[0152] This makes it possible to check the element names and symbols of the drawing and the accompanying natural language text that contains the drawing's description.

[0153] [8] A description text code table identification unit (13) receives a first prompt (prompt 33) that includes an instruction to generate a description text code table (24) containing the correspondence between symbols and element names from the description text of the drawing, and obtains a response that includes a predetermined description text code table (24). A drawing code table identification unit (15) receives a second prompt (prompt 35) from a generating AI (generating AI system 3) which includes an instruction to generate a drawing code table (26) containing the correspondence between symbols and element names from the drawings described in the above explanatory text, and obtains a response including a predetermined drawing code table (26) from the generating AI (generating AI system 3), A code determination table creation unit (171) outputs a code determination table (27) that determines errors in the symbols from the explanatory code table (24) and the drawing code table (26), A code inspection device (1) characterized by comprising the following:

[0154] This makes it possible to check the element names and reference numerals of the drawing and its accompanying description from the description and the drawing itself.

[0155] (modified version) The present invention is not limited to the embodiments described above, and can be modified without departing from the spirit of the invention, for example, (a) to (f) below.

[0156] (a) The explanatory text code table, drawing code table, and code determination table generated by the generation AI are not limited to text in Markdown notation, but may also be, for example, comma-separated text or tab-separated text. (b) The generation AI system is not limited to a cloud environment; it may also operate in an on-premises environment. (c) The descriptions of the drawings are not limited to those for patent specifications, but may also be descriptions for software specifications or user manuals. (d) The generative AI systems that send each prompt may be the same or different, and are not limited to such systems. (e) A descriptive text code table may, but is not limited to, be created from the descriptive text using a rule-based method. (f) All processes other than generating a drawing code table from drawings may be processed using a rule-based method, and are not limited to that method. [Explanation of Symbols]

[0157] 1. Code Inspection Device 11. Description section 12. Description Extraction Unit 13. Explanatory Text Code Table Identification Section 14. Description Code Table Extraction Unit 15. Drawing Reference Code Identification Section 16 Drawing Reference Code Extraction Unit 17 Sign judgment table identification part 18 Sign judgment table extraction part 3. Generative AI System 21 Natural sentences 22 Drawing number 51 Request body 52 Chat Completed Objects 23 Description 53 Request body 54 Chat Completed Objects 55 Request body 56 Chat Completed Objects 57 Request body 58 Chat Completed Objects 101 CPU 102 RAM 103 ROM 104 Display section 105 Input section 106 Communications Department 107 Storage section 1071 Code Check Program 61 Sign judgment screen 611 Drawing Number Combo Box 613 Natural sentence area 614 Drawing Area 615 Sign determination table area 31 Prompt 32 answers 33 Prompts 34 answers 35 Prompts 36 answers 37 Prompts 38 answers 24. Explanatory Text Code Table 25 Drawing Images 26 Drawing Reference Codes 27 Sign determination table 131 Explanatory Text Code Table Identification Section 141 Description Code Table Extraction Unit 171 Code Determination Table Creation Section 511 request body 521 Chat Completed Object 311 Prompt 321 answers 19 Code table identification part 10 Code table extraction part 59 Request body 50 chat completed objects 39 Prompts 30 answers 172 Sign judgment table identification part 182 Sign judgment table extraction part 591 request body 501 Chat Completed Object 391 Prompt 301 answers

Claims

1. A procedure for inputting a first prompt to a generating AI that includes an instruction to generate a descriptive text code table in which the correspondence between symbols and element names is described from the descriptive text of a drawing, and obtaining a response from the generating AI that includes a predetermined descriptive text code table. A procedure to input a second prompt to a generating AI that includes an instruction to generate a drawing code table in which the correspondence between symbols and element names is described from the drawings described in the above explanatory text, and to obtain a response from the generating AI that includes a predetermined drawing code table. A procedure for outputting a symbol determination table that determines errors in symbols from the aforementioned explanatory symbol table and the aforementioned drawing symbol table. A code verification program that causes a computer to perform a code check.

2. A procedure for inputting a third prompt to a generating AI, which includes an instruction to generate a descriptive text about a drawing from natural language and a drawing number, and obtaining a response from the generating AI that includes the descriptive text. A code check program according to claim 1 for further implementation of the above.

3. The first prompt and / or the second prompt include instructions to determine if a typographical error is present if the same code corresponds to multiple element names. The code check program according to claim 1.

4. A procedure to input a fourth prompt to the generating AI, which includes an instruction to output a code determination table that determines errors in the symbols from the explanatory text code table and the drawing code table, and to obtain a response from the generating AI that includes a predetermined code determination table. A code check program according to claim 1 for performing the following:

5. The fourth prompt includes instructions to determine that a symbol present in the drawing symbol table but not in the explanatory text symbol table, and a symbol present in the explanatory text symbol table but not in the drawing symbol table, are typographical errors. A code check program according to claim 4 for performing the following:

6. A procedure for generating a code determination table in which codes present in the drawing code table but not in the explanatory text code table, and codes present in the explanatory text code table but not in the drawing code table are determined to be typographical errors. A code check program according to claim 4 for performing the following:

7. A procedure to obtain a response from the generating AI that includes a predetermined explanatory text code table and a predetermined drawing code table, by inputting prompts that include instructions to generate an explanatory text code table containing correspondences between symbols and element names from explanatory texts about drawings, and instructions to generate a drawing code table containing correspondences between symbols and element names from drawings described in the explanatory texts. A procedure for outputting a symbol determination table that determines errors in symbols from the aforementioned explanatory symbol table and the aforementioned drawing symbol table. A code verification program that causes a computer to perform a code check.

8. A description text code table identification unit receives a first prompt from a generating AI that includes an instruction to generate a description text code table in which the correspondence between symbols and element names is described from the description text of the drawing, and obtains a response that includes a predetermined description text code table. A drawing symbol identification unit inputs a second prompt to a generating AI that includes an instruction to generate a drawing symbol table in which the correspondence between symbols and element names is described from the drawing described in the above explanatory text, and obtains a response including a predetermined drawing symbol table from the generating AI. A code determination table creation unit outputs a code determination table that determines errors in the notation of codes from the aforementioned explanatory code table and the aforementioned drawing code table. A code inspection device characterized by comprising the following:

Images