Information processing system, information processing method, and program

Abstract

An information processing system, an information processing method, and a program that can output useful information from requested drawing data and are highly convenient for users are provided. [Solution] The information processing system disclosed herein is an information processing system that, when natural language is input, outputs answer data based on the input natural language and target drawing data using a learning model that has been trained to output an answer corresponding to the natural language, and is equipped with an information processing device that executes the following processes: a process of displaying pre-stored options, which are multiple standard natural language sentences, on the screen of a user terminal in a state where the user can select from them; a process of inputting the target drawing data and accepting selection input of the options from the user; and a process of referencing pre-stored past performance data, extracting information on performance data related to the target drawing data, and generating the answer data.

Description

[Technical Field]

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

[0002] Traditionally, in the business flow of a processing company that produces a wide variety of products in small quantities, determining an estimate requires taking into consideration many factors, such as material costs, equipment owned, worker skills, and the number of man-hours that workers can take on (availability), and only a limited number of members with sufficient experience are able to prepare estimates.

[0003] In response to this, in recent years, information processing devices and the like have been proposed that are capable of accurately generating processing information for new processed products without relying on the knowledge and experience of skilled workers (for example, Patent Document 1). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 2024-180232 Summary of the Invention [Problem to be solved by the invention]

[0005] According to the system of Patent Document 1, it is possible to improve inference accuracy by performing machine learning using learning data suitable for new processed products, and to accurately generate processing information for new processed products. However, even this technology still has room for improvement in terms of user convenience.

[0006] Therefore, the present disclosure has been made in consideration of the above-mentioned problems, and its purpose is to provide an information processing system, information processing method, and program that can output useful information from requested drawing data and are highly convenient for users. [Means for solving the problem]

[0007] According to the present disclosure, there is provided an information processing system that, when a natural language sentence is input, outputs answer data based on the input natural language sentence and target drawing data using a learning model that has been trained to output an answer corresponding to the natural language sentence, The information processing device A process of displaying pre-stored options including a plurality of standard natural sentences on a screen of a user terminal in a state in which the user can select from the options; A process of receiving input of the target drawing data and selection input of the option from a user; and a process of extracting information on performance data related to the target drawing data by referring to previously stored past performance data, thereby generating the response data.

[0008] According to the present disclosure, there is also provided an information processing method by an information processing device that, when a natural language sentence is input, outputs answer data based on the input natural language sentence and target drawing data using a learning model that has been trained to output an answer corresponding to the natural language sentence, comprising: The information processing device, A process of displaying pre-stored options, each of which is a set of natural sentences, on a screen of a user terminal in a state in which the user can select from the options; A process of receiving input of the target drawing data and selection input of the option from a user; and a process of extracting information on the performance data related to the target drawing data by referring to past performance data stored in advance, thereby generating the response data.

[0009] According to the present disclosure, there is also provided a program for executing an information processing method by an information processing device that, when natural language is input, outputs answer data based on input natural language and target drawing data using a learning model that has been trained to output an answer corresponding to the natural language, the program comprising: The information processing device includes: A process of displaying pre-stored options, each of which is a set of natural sentences, on a screen of a user terminal in a state in which the user can select from the options; A process of receiving input of the target drawing data and selection input of the option from a user; and a program for executing a process of referencing previously stored past performance data, extracting information on performance data related to the target drawing data, and generating the response data. [Effects of the Invention]

[0010] According to the present disclosure, it is possible to output useful information from requested drawing data, and to provide an information processing system, information processing method, and program that are highly convenient for users. [Brief explanation of the drawings]

[0011] [Figure 1] 1 is a diagram illustrating a configuration example of an information processing system according to an embodiment of the present disclosure. [Figure 2] FIG. 2 is a diagram illustrating an example of the configuration of an information processing device according to the embodiment. [Figure 3] FIG. 10 is a diagram illustrating a configuration example of a user terminal according to the embodiment. [Figure 4] FIG. 4 is a flowchart illustrating a series of controls in the system according to the embodiment. [Figure 5] FIG. 4 is a flowchart illustrating a series of controls in the system according to the embodiment. [Figure 6] FIG. 10 is a diagram showing an example of a screen according to the embodiment. [Figure 7] FIG. 10 is a diagram showing another example of the screen according to the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0012] Preferred embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings. In this specification and drawings, components having substantially the same functional configurations are designated by the same reference numerals, and redundant description will be omitted.

[0013] 1 shows an example of an information processing system 1 according to this embodiment. The system 1 includes an information processing device (server device) 10 managed by a system administrator, and a user terminal 20 (receiver's terminal) used by a receiver such as a processor that receives requests for estimates and manufacturing requests from requesters such as product manufacturers or trading companies.

[0014] The information processing device 10 and the user terminal 20 are connected to each other via a network NW and can transmit and receive various information to and from each other. There may be multiple user terminals 20. Furthermore, for example, if a requester or a recipient directly inputs or outputs information via an input / output unit of the information processing device 10, the user terminal 20 can be omitted.

[0015] This system 1 allows a user who has received drawing data (target drawing data) of a requested object from a client, such as a request for quotation or manufacturing, to easily obtain the necessary information from a pre-trained learning model (e.g., in chat format via an AI chatbot). A learning model server (e.g., a server with AI chatbot functionality) periodically obtains information from information sources such as internet sites by crawling, and when natural language is input, it provides a learning model (e.g., an AI chatbot) that has been trained to generate various responses in response to the natural language based on common sense. The AI ​​chatbot may be known by names such as ChatGPT or Marvin. This system is not limited to interactive input / output formats like AI chatbots, and can adopt any input and output format.

[0016] The information processing device constituting the information processing system of this embodiment uses a learning model trained to output a response corresponding to a natural language input, and outputs response data based on the input natural language input and target drawing data. The information processing device executes the following processes: displaying pre-stored options including multiple standard natural language inputs on a user terminal screen so that the user can select from them; receiving the target drawing data and a selection input from the user; and extracting performance data information related to the target drawing data by referencing pre-stored past performance data to generate the response data. With this configuration, the user can obtain an answer from the learning model simply by inputting the requested drawing data and selecting from pre-stored options. In other words, the user can easily obtain the necessary response information without having to think about what natural language to input and then inputting it as text. Thus, this embodiment can output useful information from the target drawing data, providing a highly user-friendly information processing system. The information processing device can then execute a process to display the information on a user terminal 20 or the like. Note that the device that outputs the data is not limited to the user terminal 20, but may be the information processing device 10 itself or another external device.

[0017] In this embodiment, one of the options may include instructions for outputting caution information based on past project data similar to the target drawing data. Each past project data included in the performance data may include caution information that the user should pay attention to. The information processing device may, for example, extract and output caution information included in past project data similar to the target drawing data. The caution information may include, for example, information indicating that the final manufacturing cost (processing cost) is higher than the estimated cost at the time of estimation, that the actual delivery date is later than the estimated delivery date at the time of estimation, that the actual processing time is longer than the estimated processing time at the time of estimation, that a contract has been lost, and information regarding the cause of the contract loss. The caution information may be pre-stored information or information generated using a learning model (including a generation AI) based on information included in the past project data.

[0018] The options (natural language) stored in the storage unit may include, for example, a request for an estimated manufacturing cost (estimated processing cost) for manufacturing the target drawing data (e.g., "Please tell me the estimated cost for the attached drawing"), a request for an estimated manufacturing period (estimated processing period) (e.g., "Please tell me the estimated manufacturing period for the attached drawing"), and a request for recommended subcontractor information (e.g., "Please tell me the recommended subcontractor"). Among the many options stored in the storage unit, the conditions for multiple options to be presented to the user are pre-stored in the storage unit. For example, predetermined options may be displayed in a predetermined order regardless of the user, or the options displayed may vary for each user. For example, options that each user has selected most frequently in the past may be displayed preferentially. The content of options selected in the past, the number of times selected, and the frequency with which they were selected may be stored in the storage unit, and the learning model (generative AI) may refer to this information to determine the options to display. In this case, the system administrator may input a prompt to the learning model to instruct it to perform the operation.

[0019] In this embodiment, one of the options may include an instruction to output a parts list based on the target drawing data. If a parts list is already included in the drawing data, the parts list can be extracted and output as text. If the target drawing data does not include a parts list, the parts shown in the target drawing data may be recognized by image recognition or text analysis, and the part names, materials, quantities, etc. may be estimated from the text of the drawing data to generate a new parts list.

[0020] In this embodiment, the options may be displayed on the screen of the user terminal, and any natural language may be input (see FIG. 6). This allows a user to request a question or request that is not included in the displayed options from the learning model.

[0021] Furthermore, in this embodiment, the screen of the user terminal (e.g., a chat screen or chat window) may be displayed adjacent to or superimposed on top of a project information viewing screen (window) that displays project information included in the performance data (see FIG. 7). This allows a user to simultaneously check, for example, past project information related to the target drawing data while engaging in a chat-style conversation with the AI ​​chatbot. In the example of FIG. 7, the chat screen 51 is displayed superimposed on top of the project information viewing screen 52. This allows the user to simultaneously check recently added past project information 36 and past project information corresponding to the conditions entered by the user while chatting with the AI ​​chatbot.

[0022] The response data is output on the user's device screen in the form of text (natural language, etc.), images (related drawing data), tables, etc. In principle, the response data is the information requested by the user, but if the AI ​​determines that the requested information cannot be obtained, it will output a message stating that the information does not exist (or cannot be generated), along with the reason why.

[0023] Furthermore, in this system, the information processing device may execute a process of displaying another option associated with the option selected by the user on the screen of the user terminal after generating the response data. For example, the control unit 11 may display the next option together with the output of the response data. Information on the next option may be pre-stored in association with each previously displayed option and automatically determined based on the option previously selected by the user. For example, if the user selects option 31 in FIG. 6 requesting information on a precaution, an option requesting information on countermeasures for that precaution, which is pre-associated with option 31, may be displayed along with the response data. It is preferable that the option information displayed the second time be related to the option information selected the first time. For example, if the user first selects option 32 requesting a parts list, the option displayed the second time may be a request for an estimated estimate based on the parts list or a request for processing method information based on the parts list. Any option can be prepared in advance.

[0024] The information processing device 10 is a server device (AI chatbot server) used by system administrators and the like when operating and managing various services, and may be, for example, a general-purpose computer such as a workstation or personal computer, or may be logically realized using cloud computing technology, or may be composed of multiple devices that communicate with each other via a network, etc.

[0025] The information processing device 10 includes a control unit 11, a storage unit 12, an input unit 13, an output unit 14, and a communication unit 15, as shown in FIG.

[0026] 3, the user terminal 20 includes a control unit 21, a storage unit 22, an input unit 23, an output unit 24, and a communication unit 25. The user terminal 20 is a computer operated by a user for inputting and outputting various information, and is configured, for example, as a smartphone, a tablet computer, or a personal computer. The user can access the information processing device 10, for example, by using an application or a web browser executed on each terminal.

[0027] When the information processing device 10 receives various commands (requests) from other information processing devices or the like via the input unit 13 or the communication unit 15, the control unit 11 executes processing according to a program, and the program processing results (e.g., images, sounds, etc.) are sent to the output unit 14 or other information processing devices (e.g., user terminal 20). Alternatively, the information processing device 10 receives various commands (requests) from other information processing devices or the like via the communication unit 15, and transmits the program processing results executed by the control unit 11 to the other information processing devices or the like. Note that part of the program may be sent to the other information processing devices and executed on the other information processing devices. In this case, the other information processing devices may be, for example, smartphones, mobile phone terminals, tablet terminals, personal computers, etc., and are connected to the information processing device 10 wirelessly or via a wire via a network such as the Internet.

[0028] The control units 11 and 21 transfer data between each unit and control the entire device, and are realized, for example, by a CPU (Central Processing Unit), MPU (Micro Processing Unit), or GPU (Graphics Processing Unit) executing a program stored in a specified memory.

[0029] The storage units 12 and 22 store various data and programs, and are non-volatile or volatile semiconductor memories such as RAM (Random Access Memory), ROM (Read Only Memory), flash memory, EPROM (Erasable Programmable Read Only Memory), and EEPROM (Electrically Erasable Programmable Read-Only Memory), as well as magnetic disks, flexible disks, optical disks, compact disks, minidisks, DVDs (Digital Versatile Discs), etc.

[0030] The input units 13 and 23 are used by users and system administrators to input various data, and are realized by, for example, a keyboard, a mouse, a touch panel, buttons, a microphone, and the like.

[0031] The output units 14 and 24 output various information generated by the control unit, etc. The output units are, for example, a liquid crystal display (LCD), a touch panel, a printer, a speaker, etc.

[0032] The communication units 15 and 25 are for communicating with other information processing devices, and have a function as a receiving unit that receives various data and signals transmitted from other information processing devices, etc., and a function as a transmitting unit that transmits various data and signals to other information processing devices, etc. in response to commands from the control unit. The communication units are realized by, for example, a NIC (Network Interface Card), an adapter for connecting to Ethernet (registered trademark), a modem for connecting to a public telephone line network, a wireless communication device for wireless communication, a USB (Universal Serial Bus) connector or an RS232C connector for serial communication, etc.

[0033] 2, the control unit 11 of the information processing device 10 can function as an information acquisition unit 111 and an information generation unit 112. The storage unit 12 can function as an element information storage unit 121 and an output information storage unit 122.

[0034] The information acquisition unit 111 acquires various information used to generate output data. The acquired information may include information input by a user or a system administrator, and information acquired from an external system via a network such as the Internet (for example, web information such as raw material unit price information and part unit price information published on the Internet).

[0035] The information generation unit 112 generates output information based on the input information and the information in the storage unit, by inputting the input information into the generative model as appropriate, and further editing the generated information output by the generative model as necessary. The generative model may be implemented in the information processing device 10 or in another server accessible via a communication network, but is not limited thereto. When the generative model is implemented in the information processing device 10, the information generation unit 112 inputs prompt information to the generative model. When the generative model is implemented on another server, the information generation unit 112 transmits the prompt information to the generative model via the network. The information generated by the information generation unit 112 may include information generated without using a generation AI. For example, information previously stored in the storage unit in association with each element information constituting the request information may be output (without the intervention of a generation AI) in accordance with various data included in the request information. The generation conditions for such output information (generated information) are previously stored in the storage unit.

[0036] The element information storage unit 121 stores various types of element information including performance data used to generate output data, etc. The element information storage unit 121 can store information input in advance by users such as a system administrator, a requester, or a recipient, information acquired from an external device or the Web, etc.

[0037] The output information storage unit 122 stores the output information generated by the control unit 11. The output information may include text information, image information, audio information, and the like.

[0038] When the information processing device 10 generates various types of information using a generative AI, the information acquisition unit 111 generates a prompt based on the acquired information (or based on a pre-stored prompt or the input prompt itself), and if the information is text information, the information generation unit 112 inputs the prompt into a text generation model (e.g., a large-scale language model such as ChatGPT), or if the information is an image or design, the information generation unit 112 inputs the prompt into an image generation model to generate the information. The generative model may, for example, receive a specific input vector or random noise as input and generate an image from that information. The generative model includes, for example, a generator. The generator converts the input information into appropriate features or patterns and converts them into text or an image. The generator is constructed using, for example, a convolutional neural network (CNN), a transformer, or other deep learning architecture, although other architectures may also be used.

[0039] As shown in FIG. 5, the control unit 11 of the present system executes a process (S1) of displaying pre-stored options including a plurality of standard natural sentences on the screen of the user terminal (for example, an arbitrary input screen, chat screen, etc.) in a state where the user can select from them.

[0040] For example, as shown in Figures 6 and 7, two option icons 31 and 32 are displayed in a selectable state on a display screen 50 (51 in Figure 7) of an AI chatbot. The screen 50 may also display an icon 33 for attaching and inputting target drawing data, an input field 34 for natural language input, and an icon 35 for sending the input drawing data and natural language input.

[0041] The control unit 11 also executes a process (S2) of receiving input of target drawing data and selection input of the option from the user. For example, when the user selects icon 33 to attach drawing data and selects the icon of option 31 (or 32), the control unit 11 receives the input information.

[0042] Furthermore, the control unit 11 executes a process (S3) of extracting information from performance data related to the target drawing data by referencing previously stored past performance data to generate the response data. The information processing device then executes a process (S4) of displaying the response data on a user terminal 20 or the like. The response data (output data) generation process can be performed, for example, by the control unit inputting prompt information to the generation AI specifying which items (element information) of performance data to reference and which information to output, thereby causing the AI ​​to output the specified information. This prompt information may be stored in advance in a storage unit (presumably by storing in advance the combination of conditions for which element information in the performance data to reference in order to obtain each piece of output information). Alternatively, the prompt information may be generated by acquiring input information from a user (such as a system administrator or recipient). For example, when a user inputs element information (such as an estimated cost or delivery date) to be output, the control unit may determine which performance information to reference in order to obtain the output information according to the condition information stored in the storage unit, and generate the prompt information.

[0043] In addition, in this embodiment, the information processing device may estimate the similarity of the past drawing data to the target drawing data by comparing feature information of the past drawing data included in the performance data with feature information of the target drawing data, and extract performance data related to the target drawing data based on the similarity information. In this way, output data is generated based on information on projects similar to the target drawing data, thereby improving the accuracy of the output information.

[0044] 5, the process may include estimating the similarity of the past drawing data to the target drawing data by comparing feature information of the past drawing data with feature information of the target drawing data (S21), extracting cases of the past drawing data that satisfy a predetermined specific condition as related cases (S22), generating output data based on the extracted case information (S23), and transmitting the data to a user terminal for display (S24).The specific condition may be, for example, the highest similarity value, a similarity equal to or greater than a predetermined specific value (threshold value), or a specific number (which can be set arbitrarily, for example, 3, 5, 10, etc.) that ranks the similarities.

[0045] The control unit 11 may store the acquired information and generated information in association with user information in the storage unit 12. The requested drawing data may include a two-dimensional (or three-dimensional) object image representing the target object, the target object name (processed product name), dimensional information (length, width, thickness, height, inner diameter, outer diameter), material information (quality), processing method information, quantity information, weight information, drawing number, requesting company name, component part information (type of part, quantity, manufacturer information, etc.), and the like.

[0046] 6, the information generating unit 112 may perform a process of estimating amount information such as an estimated amount (S23), and a process of estimating period information such as a manufacturing period and a delivery date (delivery date) (S24). In this case, the estimated information can be output as output data.

[0047] In S23, for example, the control unit can extract past similar projects whose similarity satisfies a predetermined condition, and then use monetary information such as the estimated price, material cost, and profit amount of the similar projects to cause the generation AI to output various estimated prices (the estimated price, material cost, profit amount, etc. of the current requested project). The control unit can also prompt the generation AI to calculate output data such as the estimated price and manufacturing period by referring to various information such as the quantity of the manufacturing object, processing requirements (processing information), and finishing specifications contained in the request information and performance data. Such calculation condition information is stored in the storage unit in advance and is updated according to user input.

[0048] In S24, for example, the control unit can cause the generation AI to output various period information (such as the manufacturing period and delivery date of the current requested item) using information related to the period, such as the manufacturing period of the similar item and the period information between the date of receipt of the request and the delivery date.

[0049] FIG. 7 shows an example of the output data screen S. This screen is basically displayed on the screen of the receiver terminal 30, but may also be displayed on the screen of the requester terminal 20 or the information processing device 10. When displayed on the screen of the receiver terminal 30, the person in charge at the receiver can view the screen and decide whether to accept the request, the estimated amount, the delivery date, etc. The displayed content may be restricted depending on the terminal on which it is displayed. For example, when displayed on the requester terminal 20, it is possible to prevent certain items such as the manufacturing cost and the break-even amount from being displayed. The restriction information may be set in advance and stored in a storage unit, and may be updated by a user's input operation.

[0050] 7, the items displayed are order acceptance (manufacturing availability), estimated amount (yen), manufacturing period (hours, days, weeks, months, etc.), delivery date (date, time, etc.), manufacturing cost (yen), estimated man-hours (hours), break-even amount (yen), equipment to be used (cutting machine, milling machine, etc.), person in charge of manufacturing (name), and information on similar past projects (project name, link (URL), etc.), but are not limited to these. For example, all or any part of the information contained in the performance data of projects whose similarity meets a predetermined condition may be displayed as similar past projects.

[0051] The information processing system disclosed herein includes a control unit that executes a request information acquisition process that acquires request information, including requested drawing data for a manufacturing request object, and an output data generation process that generates output data using a generation AI based on the request information and pre-stored past performance data. The output data includes information regarding the feasibility of manufacturing or accepting an order for the manufacturing request object, estimated based on the similarity of past drawing data included in the performance data to the requested drawing data and status information associated with each past drawing data. This configuration enables useful information to be provided at the quotation stage based on performance data. Furthermore, the use of a generation AI allows for quick decisions on whether to accept an order, enabling a quick response to the requester and preventing a decline in the order acceptance rate. It also reduces the burden on estimators and reduces human error in judgment.

[0052] In this system, the control unit may estimate the similarity of the previous drawing data to the requested drawing data by comparing the characteristic information of the previous drawing data with the characteristic information of the requested drawing data. This improves the accuracy of the similarity determination, thereby improving the accuracy of generating output data based on the previous drawing data.

[0053] In this system, the control unit may select one or more pieces of past drawing data whose similarity meets a predetermined condition from among a plurality of pieces of past drawing data as similar drawing data, and generate output data using information associated with the similar drawing data. This allows past cases similar to the requested case to be narrowed down and output data to be generated, thereby reducing the processing load compared to when all past data is used and increasing the accuracy of the output data.

[0054] In this system, the output data may include information indicating the probability of receiving or manufacturing past requests whose similarity satisfies a predetermined condition. This allows the probability of receiving or actually manufacturing a similar request in the past to be confirmed, which can be used as a reference when deciding whether or not to accept the request.

[0055] In this system, the output data may also include other items, such as information on the reasons for losing orders for past requests whose similarity meets a predetermined condition. In particular, if the order acceptance determination result is not "acceptable," the cause can be analyzed by also outputting information on the reasons for losing orders. Furthermore, by inputting a prompt to the generation AI to analyze the reasons for losing orders, the control unit can output the analysis results of each company's orders and losses as output data. The items included in the output data may be selected by each user (system administrator, requester, recipient, etc.). For example, the control unit may present (display on each terminal screen) options for items that can be presented as output data (items included in the performance data) to the user, allowing the user to select.

[0056] In this system, the control unit may further generate output data including information regarding whether production or order acceptance is possible based on operation information of the in-house facilities. This allows the control unit to determine whether the requested production item can be produced or whether the order can be accepted, for example, based on operation information of the in-house facilities. For example, if the in-house facilities required for production of the product included in the request information are unavailable (for example, all have been operating for a long period of time, are out of order, are undergoing maintenance, etc.), the control unit can make a decision to "decline" or "require confirmation" regardless of other conditions. Information on such judgment conditions is also stored in advance in the storage unit.

[0057] In this system, the control unit may further generate output data including information on whether production or order acceptance is possible based on the worker's work schedule information. This allows the control unit to determine whether the worker's man-hours required to manufacture the product included in the request information can be secured and, if not, to decline the request. More specifically, the control unit may estimate the man-hours required for the requested job based on man-hour information (manufacturing period information) associated with past jobs whose similarity meets a predetermined condition as the man-hours required for the requested job, and determine whether the man-hours can be secured by a predetermined time point based on the worker's work schedule information. In this case, by storing in advance in the storage unit not only the worker's work schedule but also information on the type of work (processing process) each worker can handle (e.g., cutting, surface treatment, etc.), it is possible to more accurately determine whether the worker's man-hours required to manufacture the requested product can be secured. Such various worker-related information and information on the conditions for generating the output data are also stored in the storage unit in advance. Furthermore, if a worker who was in charge of a past job in the performance data has now retired, the control unit may determine the job as "declined" or "requires confirmation" regardless of other conditions.

[0058] In this system, the output data may include estimated cost information based on the similarity of the past drawing data to the requested drawing data and the cost information associated with each past drawing data. This allows for the presentation of estimated costs based on past performance.

[0059] In this system, the output data may include time-related information regarding the delivery date or work period of the processed product estimated based on the similarity of the past drawing data to the requested drawing data and the manufacturing period information associated with each past drawing data. This makes it possible to present information regarding the delivery date or work period based on past performance.

[0060] The information generating unit 112 may then execute a determination process regarding whether or not an order can be accepted or whether or not a manufacturing process can be performed. For example, if the status information associated with the past drawing data (similar drawing data) that satisfies a predetermined condition, such as the highest similarity, is "order accepted" (or "manufacturable"), the information generating unit 112 may determine that the order can be accepted (or "manufacturable"), and if the status information is other than "order accepted" (e.g., "rejected"), the information generating unit 112 may determine that the order cannot be accepted (or "manufacturable"), and output the determination result (order accepted, order not accepted, etc.) as output data. Alternatively, if the status information associated with the similar drawing data is other than "order accepted" (e.g., "rejected"), the information generating unit 112 may determine that the order cannot be accepted (or "manufacturable"), "confirmation required," "other," etc. Note that if no past drawing data (similar drawing data) that satisfies the predetermined condition exists in the performance data, a predetermined determination result such as "order not accepted (or "manufacturable"), "confirmation required," or "other" may be displayed. Such determination conditions regarding whether or not an order can be accepted (or manufacturable) are stored in advance in the storage unit. The condition information may be updated based on input information from a system administrator, a recipient, etc. For example, the recipient may input the judgment conditions himself / herself, thereby allowing the conditions to be registered (stored) and changed (updated).

[0061] The information generation unit 112 inputs the above-mentioned various condition information and performance data into the generation AI to have it learn, and also inputs prompts to the generation AI to generate predetermined output data (data including information regarding whether the product can be manufactured or whether the order can be accepted) based on the judgment condition information, request information, and performance data, thereby obtaining information regarding whether the product can be manufactured or whether the order can be accepted, generated by the generation AI. The information can then be displayed as output data on the user terminal 20 and presented to the user. The prompt information may be input by a system administrator and stored in advance in the storage unit, or the information generation unit 112 may cause the generation AI to generate the prompt information.

[0062] In this system, the output data may include method-related information regarding the manufacturing method estimated based on the similarity of the past drawing data to the requested drawing data and the manufacturing method information associated with each past drawing data. This allows the manufacturing method of the requested object to be estimated from past projects similar to the request information. The manufacturing method information includes information on the processing process. For example, the processing process of a past project similar to the request information may be estimated as the processing process of the requested object. Alternatively, the processing process may be estimated based on the results of comparing the shape of the processed product of the similar past project with the shape of the processed product included in the requested drawing data. Specifically, if the number of holes in the requested drawing data is twice as many as the number of holes in the processed product of the similar past project, the number of drilling processes may be doubled, and the processing cost may also be calculated as twice the drilling cost. Processing costs for other processing processes can also be calculated based on the ratio of processing type information to processing amount information such as the number of holes (number of holes) and amount (e.g., laser processing distance). The processing type information may be, for example, drilling, laser processing, cutting, surface treatment, MC processing, horizontal hole drilling, welding, chamfering, deburring, heat treatment, inspection, etc. The estimated cost calculated by the control unit as output data may be, for example, an amount obtained by multiplying the sum of the estimated raw material cost and the estimated processing cost by a predetermined profit rate. Calculation condition information such as a calculation formula for such an estimated amount is stored in the storage unit in advance and can be updated as appropriate in response to user input.

[0063] In this system, the control unit may generate output data using a pre-stored template based on the requested drawing data. For example, a template used in a past project similar to the requested drawing data may be used. Templates are provided, for example, for each type of material or each type of processing, and information such as material cost information and processing method is pre-associated and stored. The control unit may use the template to estimate and fill in blank information using a generation AI based on the request information.

[0064] In this system, the output data may include estimated cost information regarding the manufacturing cost estimated based on estimated material information estimated from the request information and raw material information updated periodically. This allows the estimated cost information for manufacturing the requested item to be presented. In particular, for example, when the control unit acquires material cost information updated by user input or published on the Web in real time, appropriate estimated cost and estimated amount information can be presented in response to fluctuations in raw material costs. As a result, an appropriate estimated amount can be set, allowing the recipient to appropriately secure profits and avoid losses.

[0065] In this system, the output data may include break-even price information estimated based on estimated cost information and manufacturing man-hour information estimated from the request information. This makes it possible to present the amount of money needed to receive an order in order to make a profit. The manufacturing man-hour information estimated from the request information may be the value of the man-hour information associated with similar drawing data, or, if there are multiple similar drawing data, the average, maximum, minimum, etc. of the man-hour information.

[0066] In this system, the control unit may further collect and store output data for each manufacturer that is generated based on the performance data of multiple manufacturers. For example, performance data entered by multiple recipients from their respective recipient terminals can be received by the information processing device and stored in the storage unit. This allows performance data from multiple recipient companies, not just one company, to be stored together.

[0067] In this system, the control unit may further provide the requesting company with output data for each manufacturer, generated based on the performance data of the multiple manufacturers. This allows the requesting company to be provided with information on multiple receiving companies. More specifically, the requesting company can compare data such as order acceptance, estimated price, and delivery time (manufacturing period) of each processing company side by side.

[0068] In this system, the information processing device 10 may also have an analysis function that performs an analysis for each recipient's manufacturing plant (processing plant) based on the performance data of each company. The information processing device 10 may generate analysis results, such as the order rate (loss rate) and reasons for loss for each recipient, and output them as output data. Specifically, for example, the information processing device 10 displays pre-stored information of options on a user terminal (requester terminal 20, receiver terminal 30, or another information processing device) and allows the user to select one of the options to accept request information. The options consist of outputtable data items such as order rate and reasons for loss. When the information processing device 10 accepts request information transmitted from the user terminal, it may generate output data corresponding to the request information and output it to the user terminal. As an output method, for example, the order rate and reasons for loss for each company may be displayed on a screen in a list. With this configuration, the order rate and reasons for loss for each supplier can be easily compared and evaluated.

[0069] Here, we will explain the information of the performance data stored in advance in the storage unit 12 (or storage unit 22). The information of the performance data can be updated in response to input from at least one of the information processing device 10, the requester terminal 20, and the receiver terminal 30.

[0070] The performance data (database) includes information about past requested cases. The performance data may include, for example, information about multiple recipients, and in that case, it may be a collection of information about past requests received from each recipient (processing manufacturer, etc.). Items included in the performance data include, for example, case identification information (case number, case ID, etc.), caution information, case name, case status, estimated amount, sales amount, profit amount (loss amount), raw material price, manufacturing method (including processing process), manufacturing man-hours, manufacturing cost, company representative (name, identification information, etc.), customer name (name and identification information of the requesting company, such as a manufacturer or trading company), customer representative (name, identification information, etc.), information about reasons for lost orders, processing requirements and finishing specifications (including internal memos and remarks), case creation date, etc. The data includes the date (request receipt date), requested drawing data, quote submission date, quote creation period information, scheduled shipping date and time, shipping date and time (delivery date and time), manufacturing period information, link information (data storage location, reference URL, etc.), item identification information (item ID, etc.), template used (pre-stored template used for the project), quote status (completed, incomplete), drawing attributes, order identification information (order number), drawing identification information (drawing number), manufacturing drawing data (processing drawing data), item name, quantity (units), material cost (total amount, unit price [yen / kg]), material type (plate), material quality, material supplier name, contact information (telephone number, email address), plate thickness [mm], specific gravity, size (width, length, thickness), etc., which are stored in association with each project. Processing process information may include, for example, information on the type of processing (e.g., drilling, laser processing), hole size, number, laser processing distance (mm), and processing unit price (e.g., unit price per hole or per mm of laser processing).

[0071] Performance data information is stored in the memory by being entered by the recipient, or by being automatically updated by receiving information from other systems within the company (financial management systems, project management systems, etc.) or information from other information processing devices. Project identification information is identification information unique to each requested project, and is indicated, for example, by a project number (numbers) or a project ID (text such as letters and numbers). Project name is a name indicating the content of the project. Project status is information indicating the status of each project, and may include items such as received (order successfully placed), lost (order not successfully placed), estimate submitted, declined, and others. Lost order cause information is information on the cause of a failure to place an order, and may be, for example, equipment-related causes such as a lack of equipment, worker-related causes such as a lack of workers, or money-related causes such as an inconsistent amount.

[0072] The storage unit 12 also stores condition information for various information processes in advance. The condition information may include, for example, conditions for which items of information are to be extracted and stored from the request information in the request information acquisition process (S1). The condition information may also include, for example, condition information for which items of information included in the request information are to be input to the generation AI and which items of information in the performance data are to be referenced to generate output data in the output data generation process (S2). The condition information may also include prompt information to be input to the generation AI and condition information for generating prompt information (what kind of prompt information is to be generated based on the information input by the user).

[0073] The condition information may also include, for example, information on the estimation conditions for the similarity of past drawing data to requested drawing data, information on the judgment conditions for the judgment process regarding whether to accept an order or whether to manufacture, information on the estimation conditions for the process (S23) of estimating monetary information such as an estimated amount, information on the estimation conditions for the process (S24) of estimating period information such as the period related to manufacturing and the delivery date (delivery date), etc. In other words, it may include information on the generation conditions for generating output data for all items that can be output by this system. It is preferable that these condition information are stored with the generation conditions set for each item of output data.

[0074] Below, we will explain an example of a method for calculating the similarity for each drawing in order to search for a drawing similar to a target drawing (requested drawing data) from among multiple candidate drawings (past drawing data).The target "drawing" can be, for example, a drawing of a part used in various devices, and the target "object" can be, but is not limited to, a front view, side view, plan view, perspective view, cross-sectional view, etc. of the part.

[0075] The control unit 11 can execute a feature estimation process to estimate feature information of a target object (front view, plan view, side view, etc.) included in the target drawing data; an individual similarity calculation process to compare the feature information of the target object with the feature information of all candidate objects included in each of a plurality of candidate drawings to calculate the individual similarity for each candidate object; an overall similarity calculation process to calculate the overall similarity for each candidate drawing using the individual similarities of all candidate objects included in each candidate drawing; and an output data generation process to generate output data based on the overall similarity.

[0076] In this way, by calculating the overall similarity for each candidate drawing based on individual comparisons between the target object and all candidate objects, the accuracy of similar drawing searches can be improved.

[0077] For example, the system may accept an input specifying a specific area in the target drawing, and recognize only objects included in the specific area as target objects. The specific area may be specified by, for example, a user inputting a frame of a specific shape (e.g., rectangle, circle, etc.) that surrounds an object in the target drawing. By specifying the target object in this way, the user can improve the convenience of the similar drawing search function and reduce the processing load of similarity calculations by preventing objects unnecessary to the user from being selected as target objects. Alternatively, the user may select multiple candidate drawings from among multiple drawing data pre-stored in the storage unit via a user terminal, or the control unit may automatically select multiple candidate drawings. When the control unit automatically selects candidate drawings, it may select all drawings stored in the storage unit as candidate drawings, or it may select drawings that share common information associated with the target drawings. Specifically, the system may select drawings that share one or more items of information among the items associated with each drawing, such as "part name," "client company name," "material," "client's contact person," "recipient's contact person," "company name," "product name," and "directional attribute (front view, plan view, etc.)." In this case, each piece of drawing data (image data) may be associated with annotation information (attribute information) indicating the item.

[0078] In the feature estimation process, the control unit estimates feature information of a target object included in the target drawing. The feature information may be, for example, a feature amount (feature vector), but is not limited to this. The feature information is data that is uniquely determined according to at least the shape of the target object.

[0079] The method for calculating the feature information is not particularly limited. For example, the feature information can be calculated (inferred) by inputting data of the connected region forming the target object into a feature inference model, and the resulting data can be output. The inference model may be, for example, one that applies a neural network or the like, but any machine learning model can be used. Such an inference model is pre-stored in a storage unit or stored in an external information processing system that can communicate via a communication unit. The number of inference models is not limited to one. For example, multiple inference models with different conditions, such as differences in machine learning methods or data, may be stored and used selectively or in parallel. The feature inference model is subjected to machine learning so that, for example, the more similar the target object (target connected region) and the candidate object (candidate connected region) are, the higher the similarity when comparing the features. The similarity is defined, for example, by the distance when comparing the feature values ​​of images. For example, a distance index such as Euclidean distance or Manhattan distance, or a similarity index such as cosine similarity may be used.

[0080] The feature (feature information) is output as vector data of a fixed-length numeric array, but is not limited to vector format and may be output in other data formats. The feature may be, for example, SIFT feature, SURF feature, ORB feature, AKAZE feature, etc.

[0081] Here, the target drawing data may include or be associated in advance with connected areas and feature information of the target objects included in the target drawing, or the control unit 11 may be able to detect the target objects included in the target drawing.

[0082] When the control unit 11 detects a target object, for example, it may extract a connected area (which can be, for example, a circular or polygonal annular area, but may also be a shape with some discontinuous parts rather than a completely continuous shape) defined by multiple pixels included in each drawing whose brightness values ​​are consecutive pixels that have a brightness value equal to or greater than a predetermined value (i.e., multiple pixels that form a continuous line), and detect the connected area as an object.

[0083] In addition, in the process of detecting such a target object, the brightness values ​​of multiple pixels included in each drawing may be binarized. "Binarization" refers to the process of converting, for example, pixels in each drawing that are below a predetermined brightness threshold into white and pixels that exceed the threshold into black.

[0084] Furthermore, the control unit 11 may perform a line thickening (dilation) process on each drawing. For example, the control unit 11 may thicken lines by converting or maintaining all pixels adjacent to pixels with a luminance value of "gray to black, intermediate between white and black" (pixels whose luminance exceeds a threshold) before binarization to a luminance value of "black," or by converting or maintaining pixels adjacent to pixels with a luminance value of "black" after binarization to a luminance value of "black." Furthermore, in addition to the adjacent pixels, pixels close to pixels with a predetermined luminance value before or after binarization may be thickened to a predetermined line thickness by converting them to black. By thickening lines, it is possible to connect unintentionally broken (disconnected) lines on the drawing, for example, because the original drawing's color is too light or the lines are too thin, thereby improving the accuracy of detecting the target object (the accuracy of extracting connected regions). The control unit 11 may perform the various processes described above on either or both of the target drawing and the candidate drawing. Condition information and other information required for executing each process are stored in advance in the storage unit 12.

[0085] In the individual similarity calculation process, the control unit compares the feature information of the target object with the feature information of all candidate objects included in each of the multiple candidate drawings to calculate the individual similarity for each candidate object. The target object can be a front view of the part. The target drawing may contain only one target object, or three or more target objects. The candidate drawing includes one or more candidate objects. Note that the target drawing may contain a table element in which text is surrounded by a rectangular frame, but the control unit may not recognize the table element as a target object based on the annotation information, and may instead extract the text as request information.

[0086] The control unit compares the feature amount of the target object with the feature amounts of all target objects in all candidate drawings, and calculates the individual similarity of each target object for each target object. The individual similarity is expressed, for example, as a value between 0 and 1, with the higher the similarity being expressed as a numerical value closer to 1. The value of the individual similarity is not limited to this, and may be set between a lower limit and an upper limit so that the higher the similarity is, the closer it is to the upper limit value, or the closer it is to the lower limit value. If the number of candidate objects differs for each candidate drawing, the number of individual similarities will differ for each candidate drawing.

[0087] Then, in the overall similarity calculation process, the control unit 11 calculates the overall similarity for each candidate drawing using the individual similarities of all candidate objects included in each candidate drawing.

[0088] Information such as conditions for calculating the overall similarity is stored in advance in the storage unit. The method for calculating the overall similarity is not particularly limited as long as it uses information about individual similarities. For example, the control unit calculates the overall similarity by applying individual similarity information for all candidate objects included in each candidate drawing to a predetermined calculation formula. Specifically, the control unit 11 may calculate the average value of all individual similarities for all candidate objects included in the candidate drawing to obtain the overall similarity, or may further apply the average value to a predetermined formula to calculate an index to obtain the overall similarity. For example, the harmonic mean is preferable, but arithmetic mean, geometric mean, etc. may also be used. The overall similarity may be calculated using any similarity evaluation mechanism. The overall similarity may not necessarily be calculated using the average value of individual similarities. For example, a model for predicting similarity may be constructed by combining the complexity of the shape of the drawing or vectorized data with metadata (data about the data). More specifically, the objects in each drawing may be structured (graphed) and the structures may be compared to determine partial similarity or overall structural similarity. The overall similarity calculation process determines the overall similarity for each candidate drawing. Then, in the output data generation process, the control unit generates output data based on the overall similarity.

[0089] In this embodiment, the overall similarity calculation process may calculate the overall similarity for each candidate drawing by a harmonic mean process using the individual similarities of all candidate objects included in each candidate drawing. Using the harmonic mean can improve the accuracy of detecting drawings showing highly similar solids (e.g., components) compared to using the arithmetic mean. For example, if there are three candidate objects (Drawings 1 to 3, e.g., side views, plan views, etc.), and only one of them (e.g., Drawing 1) is similar and the others are dissimilar, using the arithmetic mean will have a greater influence (individual similarities) on the values ​​of the dissimilar drawings (e.g., Drawings 2 and 3) and will likely be determined to be dissimilar. However, using the harmonic mean will have a greater influence on the similar drawing (Drawing 1), making it more likely to be determined to be similar. This allows for accurate retrieval of drawings with high similarity in solid form (shape) even when the candidate objects include drawings (candidate objects) that are the same (or similar) as solids but viewed from different directions.

[0090] In the present embodiment, the control unit may execute a ranking process for ranking the plurality of candidate drawings based on the overall similarity, and generate the output data based on the ranking, thereby making it possible to present data related to candidate drawings with high similarity to the user in an easy-to-understand manner.

[0091] Although the preferred embodiments of the present disclosure have been described in detail above with reference to the accompanying drawings, the technical scope of the present disclosure is not limited to such examples. It is clear that a person skilled in the art of the present disclosure can conceive of various modified or altered examples within the scope of the technical idea described in the claims, and it is understood that these also naturally fall within the technical scope of the present disclosure.

[0092] The devices described in this specification may be realized as a single device, or may be realized by a plurality of devices (e.g., cloud servers) partly or entirely connected via a network. For example, the control unit 11 and the storage unit 12 of the information processing device 10 may be realized by different servers connected to each other via a network.

[0093] The series of processes performed by the device described in this specification may be realized using software, hardware, or a combination of software and hardware. A computer program for realizing each function of the device and terminal according to this embodiment may be created and installed on a PC or the like. A computer-readable recording medium on which such a computer program is stored may also be provided. Examples of the recording medium include a magnetic disk, an optical disk, a magneto-optical disk, and a flash memory. The computer program may also be distributed, for example, via a network, without using a recording medium.

[0094] Furthermore, the processes described herein using flowchart diagrams do not necessarily have to be performed in the order shown. Some process steps may be performed in parallel. Additional process steps may be employed, and some process steps may be omitted.

[0095] Furthermore, the effects described herein are merely descriptive or exemplary and are not limiting. In other words, the technology according to the present disclosure may achieve other effects that will be apparent to those skilled in the art from the description of this specification, in addition to or in place of the above-described effects.

[0096] The following configurations also fall within the technical scope of the present disclosure. (Item 1) An information processing system that, when a natural language sentence is input, outputs answer data based on the input natural language sentence and target drawing data using a learning model that has been trained to output an answer corresponding to the natural language sentence, The information processing device A process of displaying pre-stored options including a plurality of standard natural sentences on a screen of a user terminal in a state in which the user can select from the options; A process of receiving input of the target drawing data and selection input of the option from a user; and a process of extracting information on the performance data related to the target drawing data by referring to past performance data stored in advance, and generating the response data. (Item 2) Item 2. The information processing system according to item 1, wherein one of the options includes an instruction to output attention point information based on past case data similar to the target drawing data. (Item 3) 3. The information processing system according to item 1 or 2, wherein one of the options includes an instruction to output a parts list based on the target drawing data. (Item 4) 3. The information processing system according to item 1 or 2, wherein the options are displayed on the screen of the user terminal and any natural sentence can be input. (Item 5) The information processing system according to item 1 or 2, wherein the screen of the user terminal is displayed adjacent to or in front of a case information viewing screen that displays the case information included in the performance data, superimposed thereon. (Item 6) The information processing device includes: By comparing feature information of the past drawing data included in the performance data with feature information of the target drawing data, a similarity of the past drawing data to the target drawing data is estimated; 3. The information processing system according to item 1 or 2, wherein performance data related to the target drawing data is extracted based on the similarity information. (Item 7) The information processing device includes: 3. The information processing system according to item 1 or 2, wherein after the process of generating the answer data, a process of displaying another option associated with the option selected by the user on the screen of the user terminal is executed. (Item 8) 1. An information processing method for an information processing device that, when a natural language sentence is input, outputs answer data based on the input natural language sentence and target drawing data using a learning model that has been trained to output an answer corresponding to the natural language sentence, The information processing device, A process of displaying pre-stored options, each of which is a set of natural sentences, on a screen of a user terminal in a state in which the user can select from the options; A process of receiving input of the target drawing data and selection input of the option from a user; and a process of extracting information on performance data related to the target drawing data by referring to past performance data stored in advance, and generating the response data. (Item 9) A program for executing an information processing method by an information processing device that, when a natural language sentence is input, outputs answer data based on the input natural language sentence and target drawing data using a learning model that has been trained to output an answer corresponding to the natural language sentence, The information processing device includes: A process of displaying pre-stored options, each of which is a set of natural sentences, on a screen of a user terminal in a state in which the user can select from the options; A process of receiving input of the target drawing data and selection input of the option from a user; and a process of referencing previously stored past performance data, extracting information on performance data related to the target drawing data, and generating the response data. [Explanation of symbols]

[0097] 1. Information Processing Systems 10 Information processing device (server) 11 Control section 12 Storage section 20 User terminal

Claims

1. An information processing system that outputs answer data based on input natural language and target drawing data using a learning model that has been trained to output an answer corresponding to the input natural language, The information processing device A process of displaying pre-stored options including a plurality of standard natural sentences on a screen of a user terminal in a state in which the user can select from the options; A process of receiving input of the target drawing data and selection input of the option from a user; a process of extracting information on past performance data related to the target drawing data by referring to the past performance data stored in advance, and generating the response data; An information processing system in which one of the options includes instructions to output attention point information based on past case data similar to the target drawing data.

2. An information processing system that outputs answer data based on input natural language and target drawing data using a learning model that has been trained to output an answer corresponding to the input natural language, The information processing device A process of displaying pre-stored options including a plurality of standard natural sentences on a screen of a user terminal in a state in which the user can select from the options; A process of receiving input of the target drawing data and selection input of the option from a user; a process of extracting information on past performance data related to the target drawing data by referring to the past performance data stored in advance, and generating the response data; An information processing system in which one of the options includes instructions for outputting a parts list based on the target drawing data.

3. An information processing system that outputs answer data based on input natural language and target drawing data using a learning model that has been trained to output an answer corresponding to the input natural language, The information processing device A process of displaying pre-stored options including a plurality of standard natural sentences on a screen of a user terminal in a state in which the user can select from the options; A process of receiving input of the target drawing data and selection input of the option from a user; a process of extracting information on past performance data related to the target drawing data by referring to the past performance data stored in advance, and generating the response data; An information processing system in which the screen of the user terminal is displayed adjacent to or in front of a case information viewing screen that displays case information included in the performance data.

4. An information processing system that outputs answer data based on input natural language and target drawing data using a learning model that has been trained to output an answer corresponding to the input natural language, The information processing device A process of displaying pre-stored options including a plurality of standard natural sentences on a screen of a user terminal in a state in which the user can select from the options; A process of receiving input of the target drawing data and selection input of the option from a user; a process of extracting information on past performance data related to the target drawing data by referring to the past performance data stored in advance, and generating the response data; The information processing device includes: By comparing feature information of the past drawing data included in the performance data with feature information of the target drawing data, a similarity of the past drawing data to the target drawing data is estimated; An information processing system that extracts performance data related to the target drawing data based on the similarity information.

5. An information processing system that outputs answer data based on input natural language and target drawing data using a learning model that has been trained to output an answer corresponding to the input natural language, The information processing device A process of displaying pre-stored options including a plurality of standard natural sentences on a screen of a user terminal in a state in which the user can select from the options; A process of receiving input of the target drawing data and selection input of the option from a user; a process of extracting information on past performance data related to the target drawing data by referring to the past performance data stored in advance, and generating the response data; The information processing device includes: After the process of generating the answer data, the information processing system executes a process of displaying another option associated with the option selected by the user on a screen of the user terminal.

6. 1. An information processing method for an information processing device that, when a natural language sentence is input, outputs answer data based on the input natural language sentence and target drawing data using a learning model that has been trained to output an answer corresponding to the natural language sentence, The information processing device, A process of displaying pre-stored options, each of which is a set of natural sentences, on a screen of a user terminal in a state in which the user can select from the options; A process of receiving input of the target drawing data and selection input of the option from a user; a process of extracting information on past performance data related to the target drawing data by referring to the past performance data stored in advance, and generating the response data; The information processing device includes: By comparing feature information of the past drawing data included in the performance data with feature information of the target drawing data, a similarity of the past drawing data to the target drawing data is estimated; An information processing method for extracting performance data related to the target drawing data based on the similarity information.

7. A program for executing an information processing method by an information processing device that, when a natural language sentence is input, outputs answer data based on the input natural language sentence and target drawing data using a learning model that has been trained to output an answer corresponding to the natural language sentence, The information processing device includes: A process of displaying pre-stored options, each of which is a set of natural sentences, on a screen of a user terminal in a state in which the user can select from the options; A process of receiving input of the target drawing data and selection input of the option from a user; a process of referring to previously stored past performance data, extracting information on performance data related to the target drawing data, and generating the response data; The information processing device includes: By comparing feature information of the past drawing data included in the performance data with feature information of the target drawing data, a similarity of the past drawing data to the target drawing data is estimated; A program that extracts performance data related to the target drawing data based on the similarity information.

Images