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

Abstract

Provide appropriate answers to questions from operators regarding machine tools. [Solution] The information processing device comprises an acquisition unit that acquires question data indicating questions or instructions regarding machine tools in natural language from an operator; an extraction unit that extracts keywords from the question data using a generative AI model; a selection unit that selects a relevant information source related to the question data from a plurality of information sources that store information about machine tools by comparing the extracted keywords extracted by the extraction unit with a keyword table that stores related keywords related to the question data in association with the plurality of information sources; an answer data creation unit that creates answer data indicating a response to the question or instruction based on the related information sources and the extracted keywords; and an output unit that outputs the answer data, wherein at least one of the plurality of information sources includes a memory that stores real-time information about the machine tool.

Description

[Technical Field]

[0001] The present invention relates to a technique for responding to questions from an operator in a machine tool. [Background technology]

[0002] At machine tool work sites, it is common for machines to suddenly break down or for operators to become unaware of how to operate them. but In such cases, it is extremely time-consuming for the operator to consult the machine tool manual to find a solution. On the other hand, the performance of large-scale language models such as CHAT-GPT has improved rapidly in recent years. Therefore, technology development is underway to use large-scale language models to create appropriate responses to machine-related questions from operators.

[0003] For example, Patent Document 1 discloses an information processing device that refers to a knowledge source that aggregates various information about copiers, creates an answer or additional question corresponding to the content of a given question, and transmits additional content request information as a question response to an input device to request input of additional content necessary to narrow down the answer to the question or the answer to the question. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2022-112541 Summary of the Invention [Problem to be solved by the invention]

[0005] In machine tools, it is not enough to simply summarize and present the content to be answered from manuals, but depending on the question, there may be cases where real-time operating data of the machine tool needs to be provided as a response. Furthermore, information about machine tools is managed by various information sources, such as cloud servers managed by machine tool manufacturers and on-premise servers managed by machine tool users. Because there are various information sources for machine tools, creating an appropriate response requires selecting an information source that is suitable for the answer from the various information sources and creating the response from that information source.

[0006] In the technology of Patent Document 1, an answer sentence is created from one information source, so if the one information source contains inappropriate information, it may not be possible to create an appropriate response sentence to the question. Also, in the technology of Patent Document 1, since the information source does not include equipment, it is not possible to respond based on real-time operation data, and in this respect it is insufficient for creating an appropriate response.

[0007] An object of the present invention is to provide a technique for providing an appropriate response to a question from an operator regarding a machine tool. [Means for solving the problem]

[0008] An information processing device in one aspect of the present disclosure includes an acquisition unit that acquires question data indicating questions or instructions regarding machine tools in natural language from an operator; an extraction unit that extracts keywords from the question data using a generative AI model; a selection unit that selects a related information source related to the question data from a plurality of information sources that store information about the machine tool by comparing the extracted keywords extracted by the extraction unit with a keyword table that stores related keywords related to the question data in association with the plurality of information sources; an answer data creation unit that creates answer data indicating a response to the question or instruction based on the related information source and the extracted keywords; and an output unit that outputs the answer data, wherein at least one of the plurality of information sources includes a memory that stores real-time information of the machine tool. [Effects of the Invention]

[0009] The present invention can provide appropriate responses to questions from an operator regarding the machine tool. [Brief explanation of the drawings]

[0010] [Figure 1] 1 is a diagram showing an example of the overall configuration of a response creation system according to an embodiment of the present invention; [Figure 2] FIG. 10 is a diagram illustrating an example of data stored in a plurality of information sources. [Figure 3] 1 is a block diagram showing an example of the configuration of a response creation device according to an embodiment of the present invention. [Figure 4] FIG. 10 is a diagram illustrating an example of processing for extracting extracted keywords. [Figure 5] FIG. 10 is a diagram illustrating an example of a process for determining an answer type. [Figure 6] FIG. 10 is a diagram illustrating an example of a data configuration of a keyword table. [Figure 7] FIG. 4 is a diagram illustrating an example of a data configuration of a parameter table. [Figure 8] FIG. 10 is a diagram illustrating an example of a process for creating response data. [Figure 9] FIG. 2 is a block diagram showing an example of the configuration of a terminal device. [Figure 10] 10 is a flowchart showing the processing of the answer creation system according to the present embodiment. [Figure 11] FIG. 10 is an explanatory diagram of a process for acquiring a data set according to an answer type. [Figure 12] This is a continuation of the flowchart in FIG. [Figure 13] FIG. 10 is a diagram illustrating an example of a process for identifying an information source. [Figure 14] 13 is a continuation of the flowchart in FIG. 12. [Figure 15] This is a continuation of the flowchart in FIG. 14. DETAILED DESCRIPTION OF THE INVENTION

[0011] (Embodiment) Note that each of the embodiments described below represents a specific example of the present disclosure. The numerical values, shapes, components, steps, and order of steps shown in the following embodiments are merely examples and are not intended to limit the present disclosure. Furthermore, among the components in the following embodiments, components that are not described in the independent claims that represent the highest concept are described as optional components. Furthermore, in all of the embodiments, the respective contents can be combined.

[0012] 1 is a diagram showing an example of the overall configuration of a response creation system 1 in this embodiment. The response creation system 1 is a system that receives questions or instructions about a machine tool 60 from an operator, creates response data indicating a response to the question or instruction, and presents the created response data to a user.

[0013] The answer creation system 1 includes a terminal device 10, an answer creation device 20 (an example of an information processing device), a cloud server 30, an on-premise server 40, an external server 50, and one or more machine tools 60. The terminal devices 10 to the machine tools 60 are connected to each other via a network 70 so as to be able to communicate with each other.

[0014] The network 70 is a wide area communication network including, for example, a mobile phone communication network and an Internet communication network.

[0015] The terminal device 10 is configured, for example, as a mobile terminal such as a smartphone or a tablet computer, or as a stationary computer such as a desktop computer. The terminal device 10 may be implemented in the answer creation device 20 to configure an information processing device. Alternatively, the terminal device 10 may be a console terminal that accepts operations from an operator on the machine tool 60. The terminal device 10 acquires question data indicating a question or instruction regarding the machine tool in natural language from the operator. The terminal device 10 transmits the acquired question data to the answer creation device 20, acquires answer data indicating a response to the question or instruction from the answer creation device 20, and presents it to the operator. The operator is, for example, a manager of the machine tool 60 or an operator who actually operates the machine tool 60.

[0016] The answer creation device 20 is configured by a computer such as a cloud server, for example. The answer creation device 20 acquires question data from the terminal device 10, creates answer data indicating a response to the question or instruction, and outputs the answer data to the terminal device 10. The answer creation device 20 may be implemented in the terminal device 10 or in the machine tool 60.

[0017] Cloud server 30 is, for example, a computer managed by the manufacturer of machine tool 60. Cloud server 30 is a computer that stores data (hereinafter referred to as cloud data) related to machine tool 60 managed by the manufacturer.

[0018] The on-premise server 40 is a computer managed by the user of the machine tool 60. The on-premise server 40 stores data related to the machine tool 60 that is locally managed by the user of the machine tool 60 (hereinafter referred to as on-premise data).

[0019] The external server 50 is, for example, a computer that provides a website that makes available to the outside world data (referred to as external data) related to the machine tool 60 provided over the Internet. The external server 50 may further include a computer that provides a generative artificial intelligence (AI) model over the Internet.

[0020] The machine tool 60 is a machine tool installed at a material processing site. The machine tool 60 includes a memory for storing machine tool data. The machine tool 60 is a machine used to process materials. The machine tool 60 may be an NC machine tool that processes materials fully or semi-automatically, or may be a CNC machine tool that is controlled by a computer.

[0021] The cloud server 30, the on-premise server 40, the external server 50, and the machine tool 60 are examples of multiple information sources.

[0022] 2 is a diagram showing an example of data stored in a plurality of information sources. The data stored in the information sources includes the machine tool data, on-premise data, cloud data, and external data described above.

[0023] The machine tool data includes tool data, programs, operation information, etc. The tool data is data relating to the type of tool attached to the machine tool 60, the shape of the tool, etc. The program is a material processing program for the machine tool 60. The operation information is data including the current operation information of the machine tool 60 (an example of real-time information) and a history of alarms that have occurred in the past. The alarm history includes data such as the date and time when the alarm occurred, the content of the alarm, and the action taken in response to the alarm.

[0024] The on-premise data includes machining performance data, machine status data, tool status data, etc. Machining performance data is data that indicates what materials and shapes of processed products machined by machine tool 60 in the past. Machine status data is data that indicates what type of machine tool 60 is installed in what location within the entire factory where the machine tool 60 is installed. Tool status data is data that indicates what type of tool is installed on which machine tool 60. The tool status data further includes data that indicates what tools are stored in tool storage locations such as warehouses or shelves in the factory.

[0025] The cloud data includes service history, manual data, parts order history, and inquiry data. The service history is data indicating what kind of maintenance service was provided for which machine tool 60. Maintenance service includes, for example, installation work for the machine tool 60, repair of the machine tool 60, etc. The service history includes the date and time when the maintenance service was provided and the content of the maintenance service. The manual data is electronic data of the operation manual for the machine tool 60. The parts order history is data indicating what parts have been ordered for which machine tool 60. The parts order history includes the date and time when the parts were ordered and the type of part. The inquiry history shows the history of inquiries made by users about the machine tool 60. The inquiry history includes the date and time when the inquiry was made and the content of the inquiry.

[0026] The external data includes internet data, generative AI models, tool manufacturer processing data, and cloud data from affiliated companies. The internet data is web data that is publicly available on the internet. The generative AI models are publicly available on the internet.

[0027] A generative AI model is an artificial intelligence configured to generate sentences etc. in response to a given prompt by training a neural network with a large number of parameters (for example, billions to hundreds of billions) on a large amount of data. Examples of generative AI models include chat-gpt from OpenAI and Gemini from Google Inc.

[0028] The tool manufacturer's machining data is data related to tools that the tool manufacturer has made public on the Internet. The affiliated company's cloud data is data that the user's affiliated companies or partner companies have made public on the Internet.

[0029] 3 is a block diagram showing an example of the configuration of the answer creation device 20 in this embodiment. The answer creation device 20 includes a communication unit 21, a processor 22, and a memory 23. The communication unit 21 is a communication interface that connects the answer creation device 20 to the network 70. The communication unit 21 receives question data from the terminal device 10 and transmits answer data to the terminal device 10.

[0030] The processor 22 is configured by a central processing unit (CPU). The processor 22 includes an acquisition unit 220, an extraction unit 221, a type determination unit 222, a selection unit 223, a response data creation unit 224, and an output unit 225. The acquisition unit 220 to the output unit 225 are realized by the CPU executing an information processing program stored in the memory 23. However, this is just one example, and the acquisition unit 220 to the output unit 225 may also be configured by dedicated hardware circuits.

[0031] The acquisition unit 220 acquires question data indicating a question or instruction regarding the machine tool in natural language from an operator. In detail, the acquisition unit 220 acquires the question data transmitted from the terminal device 10 via the communication unit 21.

[0032] The extraction unit 221 extracts keywords from the question data. Hereinafter, the extracted keywords will be referred to as extracted keywords. For example, the extraction unit 221 may extract the extracted keywords by inputting the question data into a generative AI model. Note that the generative AI model may be a generative AI model provided in the external server 50, or if the answer creation device 20 has a generative AI model, the extracted keywords may be extracted by inputting the question data into this generative AI model.

[0033] Fig. 4 is a diagram showing an example of a process for extracting extracted keywords. In the example of Fig. 4, question data 1401 is acquired by the acquisition unit 220. This question data 1401 includes a message inquiring about data required to delete old tool data in a certain machine tool 60. Before inputting the question data 1401 to the generative AI, the extraction unit 221 inputs a prompt 1402 to the generative AI model. The prompt 1402 includes a command field 1402a, a keyword field 1402b, and a fuzzy keyword field 1402c.

[0034] The command field 1402a includes a command written in natural language to extract keywords shown in the keyword field 1402b from the question data 1401, including keywords with the same meaning. The keyword field 1402b includes keywords that the generative AI model should extract from the question data 1401, such as tool data, machine, operating status, repair, and processing method. The fuzzy keyword field 1402c includes extraction rules for keywords that are fuzzy when the generative AI model extracts keywords. In this example, the rules stipulate that a six-digit number indicates a machine, and that tool data or ToolData is extracted as tool data.

[0035] The extraction unit 221 inputs the question data 1401 to the generative AI following the input of the prompt 1402. This allows the extraction unit 221 to extract keywords included in the question data 1401. The extraction unit 221 can access the generative AI model by using the API (Application Programming Interface) function of the generative AI model. This allows the extraction unit 221 to access the generative AI model without using a browser. The extraction unit 221 may also extract extracted keywords based on rules without using the generative AI model.

[0036] The type determination unit 222 determines which of a plurality of answer types the question data corresponds to based on the question data. Fig. 5 is a diagram showing an example of the answer type determination process. The answer type is data indicating the format of the answer of the answer creation device 20. The answer types include "Create", "Read", "Update", and "Delete".

[0037] "Create" is a reply type for creating new data in the reply creating system 1, such as creating a new repair request or creating new tool data for the machine tool 60.

[0038] "Read" is a response type that reads out information about the machine tool, such as a question about repair methods and machining methods for the machine tool 60, and tool data for the machine tool 60.

[0039] "Update" is a response type that instructs updating of tool data and programs of the machine tool 60.

[0040] "Delete" is a response type that instructs the deletion of the program and tool data of the machine tool 60.

[0041] As shown in FIG. 5, the type determination unit 222 inputs question data 1401 to the generation AI. Before inputting a keyword, the type determination unit 222 inputs a prompt 1302 to the generation AI. The prompt 1302 includes a natural language instruction to respond with a number indicating which of the following categories the question or instruction content of the question data 1401 belongs to based on the input extracted keyword. Here, the numbers "1" are assigned to creation, "2" to reading, "3" to updating, and "4" to deleting. Therefore, the generation AI model outputs the numeric value of the category to which the question content belongs from the input question data 1401, and the type determination unit 222 obtains this numeric value from the generation AI. This allows the type determination unit 222 to obtain the answer type. Note that, like the extraction unit 221, the type determination unit 222 can access the generation AI model using the API (Application Programming Interface) function of the generation AI model.

[0042] The selection unit 223 compares the extracted keywords extracted by the extraction unit 221 with the keyword table 231 to select a relevant information source related to the question data from among a plurality of information sources that store information related to the machine tool 60.

[0043] 6 is a diagram showing an example of the data configuration of the keyword table 231. The keyword table 231 stores related keywords associated with question data in association with multiple information sources. In detail, the keyword table 231 stores related keywords categorized according to combinations of multiple answer types, multiple information sources, and multiple sub-information sources.

[0044] The keyword table 231 includes a plurality of data sets 231a. Each data set 231a has fields for "ID," "answer type," "related keyword K1" to "related keyword KX," "occurrence probability of each of the related keywords K1 to KX," "information source," and "sub-information source."

[0045] "ID" is an identifier for the dataset 231a. The "answer type" field stores the answer type to which the dataset 231a belongs. "Related keywords K1" to "Related keywords KX" are keywords related to the question data corresponding to each dataset 231a. The occurrence probability indicates the probability that each of the related keywords K1 to KX appears in each dataset 231a.

[0046] The occurrence probability is obtained by statistically analyzing the operator's past question data. For example, when a group of past question data in which the answer type was "Read," the information source was "machine tool," and the sub-information source was "operation status" was analyzed, the keywords "machine" and "operation" were found to be characteristic keywords. Furthermore, in this group of question data, the occurrence probability of "machine" was "91%," and the occurrence probability of "operation" was "91%." Therefore, in the dataset 231a in the first row, "machine" is stored as the related keyword K1 and "operation" is stored as the related keyword K2, with "91%" stored as the occurrence probability for each. Similarly to the dataset 231a in the first row, the datasets 231a in the second and subsequent rows also store related keywords and occurrence probabilities obtained by analyzing the group of past question data.

[0047] The "information source" field stores the information source that best matches the content of the question or instruction in the question data corresponding to the data set 231a.

[0048] The "sub-information source" field stores the sub-information source that best matches the content of the question or instruction in the query data corresponding to the data set 231a. A sub-information source is an information source that belongs to a lower hierarchy than the information source. For example, the information source "machine tool" includes "operating status," "program," and "tool" as "sub-information sources."

[0049] figure 3, and the selection unit 223 determines a related information source by comparing the extracted keywords with related keywords that are related to the answer type determined by the type determination unit 222. In detail, the selection unit 223 selects a related sub-information source from among a plurality of sub-information sources included in the related information source by comparing the extracted keywords with the keyword table 231. In more detail, the selection unit 223 calculates the degree of match between the extracted keywords and each of the plurality of data sets 231a, and selects a related information source by comparing the calculated degree of match with a threshold.

[0050] If there is not one data set 231a whose degree of match is equal to or greater than the threshold, the selection unit 223 creates additional question data based on the related keywords, and executes an interaction to acquire additional answer data from the operator by presenting the additional question data to the operator. The selection unit 223 recalculates the degree of match based on the additional answer data and the related keywords, and repeats the interaction until a related information source is identified.

[0051] The response data creation unit 224 creates response data indicating a response to the question or instruction based on the relevant information source selected by the selection unit 223 and the extracted keywords extracted by the extraction unit 221.

[0052] The answer data creation unit 224 acquires parameter attributes corresponding to the related information source by referring to a parameter table that stores parameter attributes required to access the information source. The answer data creation unit 224 extracts parameters having the acquired parameter attributes from the question data, and uses the extracted parameters to access the related information source.

[0053] FIG. 7 is a diagram showing an example of the data configuration of the parameter table 232. The parameter table 232 stores a plurality of parameter data sets 232a. The parameter data set 232a has fields for "ID", "information source", "sub-information source", "parameter attribute", and "sub-parameter attribute". "ID" is an identifier for the parameter data set 232a. The "information source" field stores the information source to be accessed. The "sub-information source" field stores the sub-information source to be accessed. The "parameter attribute" field stores the parameter attribute required to access the information source. A "sub-parameter" is a parameter at a lower level than the parameter. A "sub-parameter" field stores the parameter attribute required to access the information source. attribute " field stores sub-parameter attributes required to access the sub-information source. Note that the sub-parameter attributes are not essential. In the example of FIG. 7, a machine tool is shown as the information source, but the parameter table 232 also stores parameter data sets 232a for other information sources.

[0054] The response data creation unit 224 inputs to the generative AI model a prompt including a message for extracting, from the question data, a parameter having the parameter attribute identified from the parameter table 232. As a result, the response data creation unit 224 acquires, from the question data, a parameter for accessing the information source.

[0055] For example, if the selection unit 223 selects "machine tool" as the related information source and "operation information" as the related sub-information source, the response data creation unit 224 acquires the serial number of the machine tool 60 as a parameter attribute. The response data creation unit 224 extracts the serial number of the machine tool 60 from the question data.

[0056] In this case, the response data creation unit 224 may cause the generation AI model to extract the serial number from the question data. Then, the response data creation unit 224 may create the answer data by reading out the operating status from the machine tool 60 indicated by the acquired serial number. Note that if the question data does not include the serial number, the response data creation unit 224 may create additional question data to elicit the serial number from the operator and present it to the operator to acquire the serial number.

[0057] The response data creation unit 224 may cause the generative AI model to create response data as appropriate. For example, when the response type is "read" and a repair method or processing method for machine tool 60 is used as a sub-information source, the response data creation unit 224 may cause the generative AI model to create response data by transmitting the extracted keywords and access data for accessing the information source to the generative AI model and inputting a prompt including an instruction to create a summary related to the keywords from the accessed information source.

[0058] The output unit 225 outputs the answer data created by the answer data creating unit 224 to the terminal device 10. The output unit 225 also outputs additional question data to the terminal device 10.

[0059] 8 is a diagram showing an example of the process of creating answer data. In this example, the same question data 1401 as in FIG. 5 is used as the question data. The answer data creation unit 224 inputs a prompt 1502 to the generative AI model before inputting the question data 1401. In this example, the prompt 1502 is shown when the parameter "serial number of machine tool" and the sub-parameter "program name" are acquired in the parameter data set 232a in the second row of the parameter table 232.

[0060] The prompt 1502 includes a command field 1502a, a parameter field 1502b, and a sub-parameter field 1502c.

[0061] The command column 1502a contains a command statement for extracting data having the characteristics described in the parameter column 1502b and sub-parameter column 1502c from the question data 1401.

[0062] Parameter column 1502b describes the characteristics of the parameter. Here, a message indicating that the data to be extracted is the serial number of machine tool 60 and a message indicating the characteristics of the data configuration of the serial number are described.

[0063] The sub-parameter column 1502c describes the characteristics of the sub-parameters. Here, the sub-parameter column 1502c describes a message indicating that the data to be extracted is the pocket number of the machine tool 60, a message indicating the meaning of the pocket number, a message indicating that a numerical value indicating the pocket number will be output, and the like.

[0064] As a result, the generated AI model accesses the machine tool 60 with serial number "1111" in accordance with the command of prompt 1502 from question data 1401, reads all tool data stored in the accessed machine tool 60, and passes the read tool data to the response data creation unit 224.

[0065] The memory 23 is configured as a non-volatile rewritable storage device, and stores a keyword table 231 and a parameter table 232 .

[0066] 9 is a block diagram showing an example of the configuration of the terminal device 10. The terminal device 10 includes a display 11, a processor 12, an operation unit 13, a communication unit 14, and a memory 15. The display 11 is configured as a display device such as a liquid crystal panel, and displays answer data. The display 11 also displays question data input by an operator.

[0067] The processor 12 is configured by a central processing unit (CPU) and performs overall control of the terminal device 10.

[0068] The operation unit 13 is composed of input devices such as a keyboard, a mouse, a touch panel, etc. The operation unit 13 accepts input of question data from an operator.

[0069] The communication unit 14 is a communication interface that connects the terminal device 10 to the network 70. The communication unit 14 transmits question data to the answer preparation device 20 and receives answer data from the answer preparation device 20.

[0070] The memory 15 stores data necessary for the terminal device 10 to perform processes such as accepting input of question data and displaying answer data.

[0071] FIG. 10 is a flowchart showing the processing of the answer creation system 1 in this embodiment.

[0072] (Step S101) The terminal device 10 accepts question data. In this case, the terminal device 10 acquires the question data input to the operation unit 13. The terminal device 10 may acquire the question data by voice. In this case, the terminal device 10 may acquire the text data of the question data by collecting the voice of the question data with a microphone and converting the collected voice into text data using voice recognition processing.

[0073] (Step S102) The acquisition unit 220 of the answer creation device 20 acquires question data from the terminal device 10 via the communication unit 21.

[0074] (Step S103) The type determination unit 222 inputs an extraction request for an answer type to the generative AI model. Specifically, the type determination unit 222 inputs an extraction request including a prompt 1302 and question data 1401 to the generative AI model as shown in Fig. 5. As a result, the type determination unit 222 obtains the answer type of the question data 1401 from the generative AI model.

[0075] (Step S104) The selection unit 223 outputs to the keyword table 231 an instruction to acquire the data set 231a corresponding to the answer type acquired in step S103.

[0076] (Step S105) The selection unit 223 acquires a data set 231a according to the answer type from the keyword table 231. FIG. 11 is an explanatory diagram of the process of acquiring a data set 231a according to the answer type. In this example, the answer type acquired in step S103 was read "Read". Therefore, the selection unit 223 acquires a data set 231a whose answer type is read "Read" from the keyword table 231. In this example, three data sets 231a surrounded by a thick frame are extracted. The three extracted data sets 231a are collectively referred to as a data set group 231b.

[0077] FIG. 12 is a flowchart continuing from FIG.

[0078] (Step S201) The extraction unit 221 inputs an extraction request for an extracted keyword to the generative AI model. In detail, the extraction unit 221 inputs an extraction request including question data 1401 and a prompt 1402 to the generative AI model, as shown in Fig. 4. As a result, the extraction unit 221 extracts an extracted keyword from the question data.

[0079] (Step S202) The selection unit 223 calculates the degree of coincidence of each data set 231a by comparing the extracted keywords extracted in step S201 with the related keywords included in the data set group 231b acquired in step S105. Then, the selection unit 223 compares the calculated degree of coincidence with a threshold value to determine whether or not a related information source has been identified.

[0080] FIG. 13 is a diagram showing an example of processing for identifying a related information source. In this example, "machine" is extracted as the extracted keyword. The threshold is 0.9. Among the dataset group 231b, the datasets 231a containing "machine" are the dataset 231a in the first row and the dataset 231a in the second row. The appearance probability of "machine" in both the datasets 231a in the first and second rows is 91%. That is, the degree of match between the dataset 231a in the first row and the dataset 231a in the second row is both 0.91. Therefore, there are two datasets 231a that exceed the threshold. Therefore, the selection unit 223 determines that a related information source has not been identified because there is not just one dataset 231a that exceeds the threshold.

[0081] If a related information source is identified (YES in step S202), the process proceeds to step S301, and if a related information source is not identified (NO in step S202), the process proceeds to step S203.

[0082] (Step S203) 13, the selection unit 223 determines an additional keyword from among the related keywords in the dataset group 231b. First, the selection unit 223 extracts a dataset group 231c that includes the extracted keyword "machine" from the dataset group 231b. Then, the selection unit 223 determines, as an additional keyword, a related keyword that has the highest appearance probability and has not been extracted as an extracted keyword in the dataset 231a that has the smallest "ID" among the dataset group 231c.

[0083] 13, the dataset 231a in the first row has the smallest "ID" among the dataset group 231c, and in the dataset 231a in the first row, the related keyword with the highest appearance probability among the related keywords not included in the extracted keyword is "operation." Therefore, the selection unit 223 determines "operation" as the additional keyword.

[0084] (Step S204) The selection unit 223 inputs a request to create additional question data to the generative AI model and acquires the additional question data from the generative AI model. Specifically, the selection unit 223 creates a prompt indicating a request to create a question related to the additional keyword "operation" as a request to create additional question data. As a result, the generative AI model creates additional question data such as, "Is this a question about the operation of the machine?"

[0085] (Step S205) The selection unit 223 outputs the additional question data to the terminal device 10 via the communication unit 21.

[0086] (Step S206) The communication unit 14 of the terminal device 10 acquires the additional question data.

[0087] (Step S207) The display 11 of the terminal device 10 displays the additional question data.

[0088] (Step S208) The operation unit 13 of the terminal device 10 acquires input by an operator of additional answer data, which is answer data to the additional question data, and the communication unit 14 of the terminal device 10 outputs the acquired additional answer data to the answer creation device 20. The additional answer data includes, for example, a sentence agreeing with the additional question data.

[0089] (Step S209) The selection unit 223 acquires the additional response data via the communication unit 21.

[0090] (Step S210) The selection unit 223 determines whether or not the related information source has been identified. In detail, the selection unit 223 calculates the degree of coincidence between the additional keyword “operation” and the related keyword for each of the two data sets 231a included in the data set group 231c.

[0091] In the example in the upper part of FIG. 13, the dataset 231a in the first row includes the related keyword "operation," and the appearance probability of "operation" is "91%." Therefore, the selection unit 223 calculates the degree of match of the dataset 231a in the first row as 0.91. On the other hand, the dataset 231a in the second row does not include "operation" as a related keyword. Therefore, the selection unit 223 calculates the degree of match of the dataset 231a in the second row as 0. Here, the dataset 231a in the first row is the only dataset 231a that exceeds the threshold value of "0.9." Therefore, as shown in the lower part of FIG. 13, the selection unit 223 determines that the dataset 231a has been uniquely identified and therefore that a related information source has been identified.

[0092] In this case, as shown in the lower part of Fig. 13, the first row of data set 231a lists "machine tool" as the information source and "operating status" as the sub-information source. Therefore, the selection unit 223 identifies machine tool 60 as the related information source and "operating status" as the related sub-information source.

[0093] If the relevant information source can be identified (YES in step S210), the process proceeds to step S301. On the other hand, if the relevant information source cannot be identified (NO in step S210), the process proceeds to step S302. S203 Returning to step 2, the selection unit 223 again determines additional keywords and narrows down the data set 231a.

[0094] FIG. 14 is a flowchart continuing from FIG.

[0095] (Step S301) The response data creation unit 224 refers to the parameter table 232 and acquires parameter attributes using the identified related information source and related sub-information source. In the above example, "machine tool" is identified as the related information source, and "operating status" is identified as the related sub-information source. The related information source "machine tool" and the related sub-information source "operating status" correspond to the parameter data set 232a in the first row of the parameter table 232.

[0096] Therefore, the response data creation unit 224 extracts the parameter data set 232a in the first row of the parameter table 232. In this parameter data set 232a in the first row, only "serial number of machine tool" is written as the parameter attribute, and no sub-parameter attributes are written. Therefore, the response data creation unit 224 determines "serial number of machine tool" as the parameter attribute.

[0097] If "sub-parameter attributes" are described in the parameter data set 232a extracted from the parameter table 232, the response data creating unit 224 may acquire the sub-parameter attributes together with the parameter attributes.

[0098] (Step S302) The answer data creation unit 224 checks whether the question data acquired in step S102 includes a parameter having the parameter attribute identified in step S301. Specifically, the answer data creation unit 224 creates a prompt as a check request to make the generative AI model check whether the question data includes a parameter having the identified parameter attribute. If the question data includes the corresponding parameter, the generative AI model outputs the parameter to the answer data creation unit 224. On the other hand, if the corresponding parameter is not included, the generative AI model outputs a response to the answer data creation unit 224 indicating that the parameter is not included.

[0099] In addition, if a sub-parameter attribute is acquired in step S301, the answer data creation unit 224 may create a prompt as a check request to cause the generation AI model to check whether the question data contains a sub-parameter having a sub-parameter attribute.

[0100] (Step S303) If the check result in step S302 indicates that the question data includes a parameter (YES in step S303), the process proceeds to step S401, and if the check result indicates that the question data does not include a parameter (NO in step S303), the process proceeds to step S304. Note that if a sub-parameter attribute is also acquired in step S301, the answer data creation unit 224 may determine YES in step S303 if both a parameter and a sub-parameter are acquired in step S302.

[0101] (Step S304) The response data creation unit 224 inputs a request for creating a parameter acquisition statement for acquiring parameters to the generative AI model and acquires parameter acquisition statement data from the generative AI model. In detail, the response data creation unit 224 may create a prompt for creating a statement that prompts the operator to respond with parameters having parameter attributes as a request for creating a parameter acquisition statement.

[0102] For example, if the parameter attribute is the serial number of the machine tool 60, a sentence such as "Please tell me the serial number of the work machine" is created as a parameter acquisition sentence.

[0103] If the response data creating unit 224 has acquired the subparameter attribute in step S301, it may include in the parameter acquisition statement creation request a prompt for creating a statement that prompts the operator to respond with the subparameter having the subparameter attribute.

[0104] For example, if "program name" is acquired as a subparameter attribute, a sentence such as "Please tell me the program name" is created as a parameter acquisition sentence.

[0105] (Step S305) The response data creation unit 224 transmits the parameter acquisition statement data to the terminal device 10 via the communication unit 21.

[0106] (Step S306) The communication unit 14 of the terminal device 10 acquires the parameter acquisition statement data, and the display 11 displays the parameter acquisition statement data. The terminal device 10 may output the parameter acquisition statement data by voice.

[0107] (Step S307) The operation unit 13 of the terminal device 10 accepts the parameters input by the operator. The communication unit 14 of the terminal device 10 outputs the accepted parameters to the answer creation device 20. Note that, in step S305, if a statement that prompts the operator to answer with subparameters is created as parameter acquisition statement data, the terminal device 10 may also acquire the subparameters from the operator.

[0108] FIG. 15 is a flowchart continuing from FIG.

[0109] (Step S401) The response data creation unit 224 determines whether the response type identified in step S103 is other than Read. If the response type is other than Read (YES in step S401), the process proceeds to step S402. On the other hand, if the response type is Read (NO in step S401), the process proceeds to step S409.

[0110] (Step S402) The response data creation unit 224 creates confirmation message data for the operator to confirm the implementation of the operation indicated by the response type, and outputs the created confirmation message data to the terminal device 10. For example, if the response type is deletion, the related information source is the machine tool 60, and the related sub-information source is tool data, confirmation message data such as "We will delete the tool data. Are you sure you want to do this?" is created.

[0111] (Step S403) The communication unit 14 of the terminal device 10 acquires the confirmation message data, and the display 11 of the terminal device 10 displays the confirmation message data. The terminal device 10 may output the confirmation message data by voice.

[0112] (Step S404) The operation unit 13 of the terminal device 10 accepts input from the operator of an instruction to agree to the confirmation message data.

[0113] (Step S405) The response data creation unit 224 acquires the consent instruction via the communication unit 21.

[0114] (Step S406) The response data creation unit 224 outputs an operation instruction to the information source via the communication unit 21 to execute an operation corresponding to the response type on the related information source. For example, if the response type is deletion, the related information source is machine tool 60, and the sub-information source is tool data, an operation instruction to delete the tool data is output to the machine tool 60. The related information source and related sub-information source to which the operation instruction is output are the related information source and sub-information source having the parameters and sub-parameters identified in step S303. For example, the response data creation unit 224 outputs an operation instruction including the serial number of the target machine tool 60 and information specifying the tool data.

[0115] The related information source that has received the operation instruction executes the operation instruction and outputs a completion notice to the answer creating device 20.

[0116] (Step S407) The response data creating unit 224 receives the completion notification via the communication unit 21. The response data creating unit 224 outputs the completion notification to the terminal device 10 via the communication unit 21.

[0117] (Step S408) The communication unit 14 of the terminal device 10 receives the completion notification, and the completion notification is displayed on the display 11. The terminal device 10 may output the completion notification by voice.

[0118] (Step S409) The response data creation unit 224 outputs the read instruction to the information source via the communication unit 21. For example, when the information source is the machine tool 60 and the sub-information source is operation information, the response data creation unit 224 creates a read instruction to read out the operation information of the machine tool 60. The information source and sub-information source to which the read instruction is output are the related information source and related sub-information source identified in step S303. For example, the response data creation unit 224 creates a read instruction that includes the serial number of the target machine tool 60 and also reads out the operation information. The information source that has acquired the read instruction reads out the requested information in accordance with the read instruction, and outputs the read information (hereinafter referred to as read information) to the response creation device 20. Here, the read operation information includes real-time information of the machine tool 60.

[0119] (Step S410) The response data creation unit 224 inputs a response data creation request to the generative AI model to create response data from the readout information acquired in step S409. For example, if the acquired readout information includes an alarm, the response data creation unit 224 inputs a response data creation request to the generative AI model to create a response sentence based on the information that the alarm is included. In this case, the generative AI model creates response sentence data saying, "The current status is an alarm," and outputs it to the response data creation unit 224.

[0120] For example, when the information source is the cloud server 30 and the sub-information source is manual data, the response data creation unit 224 inputs a request to create response data that summarizes the manual data based on the extracted keywords to the generative AI model.

[0121] (Step S411) The response data creation unit 224 outputs the response data to the terminal device 10 via the communication unit 21.

[0122] (Step S412) The communication unit 14 of the terminal device 10 acquires the response data, and the display 11 displays the response data.

[0123] In this way, the answer creation device 20 in this embodiment extracts keywords from question data, compares the extracted keywords with the keyword table, selects relevant information sources related to the question data from among multiple information sources, and creates answer data from the selected relevant information sources. Therefore, the answer creation device 20 can provide an appropriate answer to a question from an operator about the machine tool 60. Furthermore, the answer creation device 20 uses multiple information sources, and at least one information source stores real-time information about the machine tool. Therefore, this configuration can create an appropriate answer to a question that requests real-time information about the machine tool.

[0124] The present invention can employ the following modifications.

[0125] (Variation 1) The information processing device may be the answer creation system 1, may be a terminal device 10 on which the answer creation device 20 is implemented, may be a console terminal on which the answer creation device 20 is implemented, may be the answer creation device 20, may be configured by a machine tool 60 on which the answer creation device 20 is implemented, may be a cloud server, may be an edge server, or may be an edge computer.

[0126] (Variation 2) The selection unit 223 may calculate the degree of match using Bayesian estimation. See FIG. 6. In this case, the keyword table 231 is assumed to store in advance the prior probability that each data set 231a will be selected. The prior probabilities of the data sets 231a in the first to third rows are assumed to be P(H1), P(H2), and P(H3), respectively. Assume that "machine" is extracted as the extracted keyword. In this case, the selection unit 223 calculates the degree of match of the data set 231a in the first row as 0.91*P(H0), the degree of match of the data set 231a in the second row as 0.91*P(H2), and the degree of match of the data set 231a in the third row as 0.91*P(H3).

[0127] For example, if the extracted keywords include "machine" and "operation," the selection unit 223 may calculate the degree of match for the dataset 231a in the first row as 0.91*0.91*P(H1), the degree of match for the dataset 231a in the second row as 0.91*0.00*P(H2), and the degree of match for the dataset 231a in the third row as 0.91*0.00*P(H3).

[0128] (Variation 3) The selection unit 223 may calculate the degree of match so that the value increases as the number of extracted keywords included increases. For example, if the number of extracted keywords included in the dataset 231a in the first row is 2, the number of extracted keywords included in the dataset 231a in the second row is 0, and the number of extracted keywords included in the dataset 231a in the third row is 0, the respective degrees of match may be calculated as 2, 0, 0. If the threshold is 1, the degree of match of the dataset 231a in the first row is equal to or greater than the threshold, and therefore the dataset in the first row is uniquely identified.

[0129] (Variation 4) The keyword table 231 may be updated as needed. For example, a survey may be conducted to inquire of the accuracy of the response data from the operators, and the keyword table 231 may be updated based on the survey results.

[0130] (Variation 5) In the embodiment, the related information source is uniquely identified, but multiple information sources may be identified as the related information source. Also, the related sub-information source is uniquely identified, but multiple sub-information sources may be identified as the related sub-information source.

[0131] The technical features of the present invention can be summarized as follows:

[0132] (Technology 1) An information processing device in one aspect of the present invention comprises an acquisition unit that acquires question data indicating questions or instructions regarding machine tools in natural language from an operator; an extraction unit that extracts keywords from the question data using a generative AI model; a selection unit that selects a related information source related to the question data from a plurality of information sources that store information about the machine tool by comparing the extracted keywords extracted by the extraction unit with a keyword table that stores related keywords related to the question data in association with the plurality of information sources; an answer data creation unit that creates answer data indicating a response to the question or instruction based on the related information source and the extracted keywords; and an output unit that outputs the answer data, wherein at least one of the plurality of information sources includes a memory that stores real-time information of the machine tool.

[0133] This configuration extracts keywords from question data, compares the extracted keywords with a keyword table, selects a relevant information source related to the question data from among multiple information sources, and creates answer data from the selected relevant information source. Therefore, this configuration can provide an appropriate response to a question from an operator regarding the machine tool. Furthermore, this configuration stores real-time information about the machine tool in at least one of the multiple information sources. Therefore, this configuration can create an appropriate response to a question or instruction that requires real-time information about the machine tool. Furthermore, because this configuration can create answer data from an appropriate information source, it is no longer necessary to access the information source multiple times to obtain appropriate answer data, reducing the number of processing steps and improving processing capacity.

[0134] (Technology 2) The information processing device according to the first technique may use a generative AI model to create the response data.

[0135] This configuration makes it easy to create response data using an AI model.

[0136] (Technology 3) The information processing device according to Technology 1 or 2 may further include a type determination unit that determines, based on the question data, to which of a plurality of answer types the question data corresponds, wherein the keyword table stores the related keywords classified according to combinations of the plurality of answer types and a plurality of information sources, and the selection unit may determine the related information source by comparing the related keywords related to the answer type determined by the type determination unit with the extracted keywords.

[0137] This configuration narrows down the related keywords according to the answer type, and selects a related information source by matching the narrowed down related keywords with the extracted keyword, so that a related information source suitable for the answer type can be selected.

[0138] (Technology 4) In the information processing device described in any one of techniques 1 to 3, the plurality of information sources may include sub-information sources, the keyword table may further store the related keywords in association with the sub-information sources, the selection unit may select a related sub-information source from the plurality of sub-information sources included in the related information sources by comparing the extracted keywords with the keyword table, and the response data creation unit may create the response data based on the related sub-information sources and the extracted keywords.

[0139] This configuration narrows down the related information sources to related sub-information sources appropriate for the content of the question or instruction, and creates answer data from the narrowed down related sub-information sources, so that more appropriate answer data can be created.

[0140] (Technology 5) In the information processing device described in any one of Techniques 1 to 4, the keyword table may include a plurality of data sets each including a plurality of related keywords, and the selection unit may calculate a degree of match between the extracted keyword and each of the plurality of data sets and select the related information source by comparing the degree of match with a threshold.

[0141] This configuration can accurately select relevant information sources using the degree of coincidence.

[0142] (Technology 6) In the information processing device described in Technology 5, if there is not one data set whose degree of match is equal to or greater than the threshold, the selection unit may create additional question data based on the related keywords, perform an interaction to obtain additional answer data from the operator by presenting the additional question data to the operator, recalculate the degree of match based on the additional answer data and the related keywords, and repeat the interaction until the related information source is identified.

[0143] In this configuration, interactions are repeated until a relevant information source with a matching degree exceeding a threshold is identified, thereby making it possible to select a relevant information source more appropriate for a question or instruction.

[0144] (Technology 7) In the information processing device according to any one of techniques 1 to 6, the plurality of information sources may further include the machine tool, and a server of a manufacturer of the machine tool or a server of a user of the machine tool.

[0145] The configuration can access the machine tool, the machine tool manufacturer's server, or the machine tool user's server to create an answer appropriate to the question or instruction.

[0146] (Technology 8) In the information processing device described in any one of Techniques 1 to 7, the response data creation unit may acquire parameter attributes corresponding to the related information sources by referring to a parameter table that stores parameter attributes required to access the information sources for each of the plurality of information sources, extract parameters having the acquired parameter attributes from the question data, and access the related information sources using the extracted parameters.

[0147] The arrangement obtains, through a question or instruction, the access data required to access the relevant information source, and is able to access the obtained access data.

[0148] The present disclosure can also be realized as an information processing program that causes a computer to execute each characteristic configuration included in such an information processing device, or as an information processing system operated by this information processing program. Needless to say, such an information processing program can be distributed on a computer-readable non-transitory recording medium such as a CD-ROM or via a communication network such as the Internet.

[0149] (Technology 9) An information processing method in another aspect of the present invention includes a computer acquiring question data indicating a question or instruction regarding a machine tool in natural language from an operator, extracting keywords from the question data using a generative AI model, selecting a relevant information source related to the question data from a plurality of information sources that store information about the machine tool by comparing the extracted keywords with a keyword table that stores related keywords related to the question data in association with the plurality of information sources, creating answer data indicating a response to the question or instruction based on the related information source and the extracted keywords, and outputting the answer data, wherein at least one of the plurality of information sources includes a memory that stores real-time information of the machine tool.

[0150] (Technology 10) In yet another aspect of the present invention, an information processing program causes a computer to execute the following steps: acquire question data indicating a question or instruction regarding a machine tool in natural language from an operator; extract keywords from the question data using a generative AI model; select a related information source related to the question data from a plurality of information sources storing information about the machine tool by comparing the extracted keywords with a keyword table that stores related keywords related to the question data in association with the plurality of information sources; create answer data indicating an answer to the question or instruction based on the related information sources and the extracted keywords; and output the answer data. height, At least one of the plurality of information sources includes a memory that stores real-time information of the machine tool. [Explanation of symbols]

[0151] 1: Answer creation system 10: Terminal device 11: Display 12: Processor 13:Operation section 14: Communications Department 15: Memory 20: Answer creation device 21: Communications Department 22: Processor 23: Memory 30: Cloud Server 40: On-premise server 50: External server 60: Machine tools 70: Network 220: Acquisition Department 221:Extraction part 222: Type determination section 223: Selection section 224: Response data creation department 225: Output section 231:Keyword table 232: Parameter table 232a: Parameter Data Set

Claims

1. an acquisition unit that acquires question data indicating a question or instruction regarding the machine tool in natural language from an operator; an extraction unit that extracts keywords from the question data using a generative AI model; a selection unit that selects a related information source related to the question data from the plurality of information sources that store information about the machine tool by comparing the extracted keywords extracted by the extraction unit with a keyword table that stores related keywords related to the question data in association with a plurality of information sources; a response data creation unit that creates response data indicating a response to the question or the instruction based on the related information source and the extracted keyword; an output unit that outputs the response data, At least one of the plurality of information sources includes a memory that stores real-time information of the machine tool. Information processing device.

2. A generative AI model is used to create the response data. The information processing device according to claim 1 .

3. a type determination unit that determines, based on the question data, which of a plurality of answer types the question data corresponds to; the keyword table stores the related keywords classified according to combinations of the plurality of answer types and the plurality of information sources; the selection unit determines the related information source by comparing the related keywords related to the answer type determined by the type determination unit with the extracted keywords.

3. The information processing device according to claim 1.

4. the plurality of information sources includes sub-information sources; The keyword table further stores the related keywords in association with the sub-information source; the selection unit selects a related sub-information source from among a plurality of sub-information sources included in the related information source by comparing the extracted keyword with the keyword table; the response data creation unit creates the response data based on the related sub-information source and the extracted keywords.

3. The information processing device according to claim 1.

5. the keyword table includes a plurality of data sets including a plurality of related keywords; the selection unit calculates a degree of match between the extracted keyword and each of the plurality of data sets, and selects the related information source by comparing the degree of match with a threshold.

3. The information processing device according to claim 1.

6. The selection unit If there is not one data set whose degree of match is equal to or greater than the threshold, creating additional question data based on the related keywords, and presenting the additional question data to the operator to perform an interaction to acquire additional answer data from the operator; recalculating the degree of match based on the additional response data and the related keywords, and repeating the interaction until the related information source is identified; 6. The information processing device according to claim 5.

7. The plurality of information sources further includes the machine tool, and a server of a manufacturer of the machine tool or a server of a user of the machine tool.

3. The information processing device according to claim 1.

8. The response data creation unit acquires parameter attributes corresponding to the related information source by referring to a parameter table that stores parameter attributes necessary for accessing the information source for each of the plurality of information sources; extracting parameters having the acquired parameter attributes from the question data; using the extracted parameters to access the relevant information sources; 3. The information processing device according to claim 1.

9. The computer acquiring question data indicating a question or instruction regarding the machine tool in natural language from an operator; Extracting keywords from the question data using a generative AI model; selecting a related information source related to the question data from among the plurality of information sources storing information about the machine tool by comparing the extracted keywords with a keyword table that stores related keywords related to the question data in association with a plurality of information sources; creating answer data indicating a response to the question or instruction based on the related information source and the extracted keywords; outputting the response data; At least one of the plurality of information sources includes a memory that stores real-time information of the machine tool. Information processing methods.

10. On the computer, acquiring question data indicating a question or instruction regarding the machine tool in natural language from an operator; Extracting keywords from the question data using a generative AI model; selecting a related information source related to the question data from among the plurality of information sources storing information about the machine tool by comparing the extracted keywords with a keyword table that stores related keywords related to the question data in association with a plurality of information sources; creating answer data indicating an answer to the question or instruction based on the related information source and the extracted keywords; outputting the response data; At least one of the plurality of information sources includes a memory that stores real-time information of the machine tool. Information processing program.

Images