Dashboard display device, display method, and program

Abstract

To provide a dashboard display device, a display method, and a program that allow a user to easily understand information displayed on a dashboard.SOLUTION: A dashboard display device according to the present disclosure includes: input reception means that accepts an input of a user specifying one piece of content displayed on a dashboard; information acquisition means that acquires information on the specified content; statement generation means that generates, on the basis of information on the specified content, a statement to be input to a language model; explanation acquisition means that acquires explanation information of the specified content generated by a language model on the basis of the statement; and display processing means that performs processing to display explanatory information on the dashboard.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to a dashboard display device, a display method, and a program. [Background technology]

[0002] There are known techniques for providing users with explanations of materials. For example, Patent Document 1 discloses a technique for generating an explanatory text that points out grammatical errors and the like for a sentence written by a language learner to be corrected, and displaying the generated explanatory text on a display. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2020-140183 Summary of the Invention [Problem to be solved by the invention]

[0004] Dashboards are known as UIs (User Interfaces) that allow users to view various pieces of information about a specific field on a screen. Because the definitions of the information displayed on a dashboard may be technical, or the various pieces of information displayed may be related to one another, users may need a certain level of specialized knowledge to understand the information displayed on the dashboard. However, the technology disclosed in the above-mentioned document does not provide users with explanatory information about the information displayed on the dashboard. Therefore, there is a need for technology that allows users to easily understand the information displayed on the dashboard.

[0005] Therefore, one of the objectives that the embodiments disclosed in this specification aim to achieve is to provide a dashboard display device, a display method, and a program that allow a user to easily understand the information displayed on the dashboard. [Means for solving the problem]

[0006] A dashboard display device according to a first aspect of the present disclosure includes: an input receiving means for receiving a user's input specifying one of the contents displayed on the dashboard; an information acquisition means for acquiring information on the designated content; a command statement generating means for generating a command statement to be input to a language model based on information of the specified content; a comment acquisition means for acquiring comment information of the specified content, which is generated by the language model based on the command sentence; a display processing means for performing processing to display the commentary information on the dashboard; It has.

[0007] A display method according to a second aspect of the present disclosure includes: Accepts user input specifying one of the contents displayed on the dashboard, Acquire information about the specified content, Generate a command statement to be input to a language model based on the specified content information; Obtaining commentary information for the specified content, generated by the language model based on the command sentence; The explanation information is displayed on the dashboard.

[0008] A program according to a third aspect of the present disclosure includes: an input receiving step of receiving a user's input specifying one of the contents displayed on the dashboard; an information acquisition step of acquiring information on the designated content; a statement generation step of generating a statement to be input to a language model based on information of the specified content; a commentary acquisition step of acquiring commentary information of the specified content, which is generated by the language model based on the command sentence; a display processing step of performing processing to display the commentary information on the dashboard; to be executed by the computer. [Effects of the Invention]

[0009] According to the present disclosure, it is possible to provide a dashboard display device, a display method, and a program that allow a user to easily understand information displayed on a dashboard. [Brief explanation of the drawings]

[0010] [Figure 1] 1 is a block diagram showing an example of a configuration of a dashboard display device according to the present disclosure. [Figure 2] 1 is a block diagram showing an example of a configuration of a dashboard display device according to the present disclosure. [Figure 3] FIG. 10 is a schematic diagram illustrating an example of a dashboard displayed by processing of a display processing unit. [Figure 4] FIG. 10 is a diagram showing an example of a prompt generated by a command statement generating unit. [Figure 5] FIG. 10 is a schematic diagram illustrating an example of a dashboard with commentary information displayed by processing of a display processing unit. [Figure 6] FIG. 10 is a schematic diagram showing another example of a dashboard with commentary information displayed by the processing of the display processing unit. [Figure 7] 1 is a block diagram showing an example of a hardware configuration of a dashboard display device according to the present disclosure. [Figure 8] 5 is a flowchart illustrating an example of an operation of the dashboard display device according to the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0011] For clarity of explanation, the following description and drawings may be omitted or simplified as appropriate. In each drawing, identical or corresponding elements are designated by the same reference numerals, and redundant description is omitted as necessary for clarity of explanation. Furthermore, each drawing is merely an example for describing one or more embodiments. Each drawing is not related to only one particular embodiment, but may also be related to one or more other embodiments. As will be understood by those skilled in the art, various features or steps described with reference to any one drawing can be combined with features or steps shown in one or more other drawings to create, for example, an embodiment not explicitly shown or described. Not all features or steps shown in any one drawing are necessarily required to describe an exemplary embodiment, and some features or steps may be omitted. The order of steps described in any drawing may be changed as appropriate.

[0012] <First Embodiment> An example of the configuration of dashboard display device 10 will be described below with reference to Fig. 1. Fig. 1 is a block diagram showing an example of the configuration of dashboard display device 10. Dashboard display device 10 has a display processing unit 11, an input receiving unit 12, an information acquisition unit 13, a command statement generation unit 14, and an explanation acquisition unit 15. Dashboard display device 10 is, for example, any device equipped with computer resources.

[0013] The display processing unit 11 performs display processing related to the dashboard. The input receiving unit 12 receives input from a user viewing the dashboard displayed by the display processing of the display processing unit 11. In particular, the input receiving unit 12 receives input specifying one of the contents displayed on the dashboard. The user specifies the content displayed on the dashboard that they would like an explanation for.

[0014] The information acquisition unit 13 acquires information used to generate a command statement, which will be described later. In particular, the information acquisition unit 13 acquires information about content specified by a user. For example, the information acquisition unit 13 acquires information about the content that was used to display the content. Specifically, for example, the information acquisition unit 13 may acquire data to be displayed on a dashboard, or may acquire data used to derive the data to be displayed. Note that, if the command statement generation unit 14 can refer to information other than the information about the specified content in order to generate a command statement, the information acquisition unit 13 may acquire the other information.

[0015] The command statement generation unit 14 generates a command statement to be input to the language model based on the information of the content specified by the user, acquired by the information acquisition unit 13. A "language model" is a generative model trained by machine learning that receives a command statement written in a natural language that commands the generation of information such as text data as input, performs a predetermined calculation based on the command statement, and generates and outputs information. In other words, a language model is a machine learning model that provides a so-called generative AI (generative artificial intelligence) service. A language model may also be referred to as a large language model (LLM). The command statement generated by the command statement generation unit 14 may also be referred to as a prompt. The command statement generation unit 14 uses the information of the content specified by the user to generate a command statement that instructs the generation of commentary information, which is information that explains the content.

[0016] The explanation acquisition unit 15 acquires explanation information for content specified by a user, which is generated by a language model based on a command statement. That is, the explanation acquisition unit 15 inputs the command statement generated by the command statement generation unit 14 into the language model, thereby acquiring information output from the language model as explanation information. Then, the display processing unit 11 performs processing to display the explanation information acquired by the explanation acquisition unit 15 on a dashboard.

[0017] According to dashboard display device 10, a command statement to be input to a language model is generated based on content information specified by a user, commentary information generated by the language model based on the command statement is obtained, and the commentary information is displayed on the dashboard. This allows the user to refer to the commentary information, making it easier to understand the information (i.e., the content) displayed on the dashboard. Note that this effect can also be achieved by a display method including the above-described processing of dashboard display device 10, a program that performs the above-described processing of dashboard display device 10, or a non-transitory computer-readable medium on which the program is stored.

[0018] An embodiment that is a more specific version of the first embodiment will be described below. <Embodiment 2> An example of the configuration of the dashboard display device 100 will be described below with reference to Fig. 2. Fig. 2 is a block diagram showing an example of the configuration of the dashboard display device 100. The dashboard display device 100 is a device corresponding to the dashboard display device 10 described above.

[0019] Dashboard display device 100 includes a display processing unit 101, an input receiving unit 102, an information acquisition unit 103, a command statement generation unit 104, an explanation acquisition unit 105, a dashboard data storage unit 106, and a knowledge information storage unit 107. Dashboard display device 100 is, for example, a server, but is not limited to a server and may be any device that has the functionality of a computer.

[0020] The display processing unit 101 corresponds to the above-mentioned display processing unit 11, and performs display processing related to the dashboard. The display processing unit 101 performs processing to display the dashboard on an arbitrary display. For example, the dashboard is displayed on the display of a terminal device communicatively connected to the dashboard display device 100, but it may also be displayed on a display directly connected to the dashboard display device 100. The above-mentioned terminal device may be any device, such as a smartphone, a tablet terminal, or a personal computer.

[0021] The display processing unit 101 displays a dashboard constructed using data stored in the dashboard data storage unit 106 on a display. The displayed dashboard is composed of multiple display areas, and different information about a specific field is displayed in each display area. In the present disclosure, the information displayed in a display area is referred to as content.

[0022] Dashboard data storage unit 106 stores data for generating information to be displayed in each display area of ​​the dashboard. Dashboard data storage unit 106 may be configured as a database. In this embodiment, dashboard display device 100 includes dashboard data storage unit 106, but dashboard data storage unit 106 may be configured as a device separate from dashboard display device 100. Display processing unit 101, for example, arranges data stored in dashboard data storage unit 106 as content in the display area. Display processing unit 101 may also perform predetermined processing (e.g., calculation, processing, etc.) on one or more pieces of data stored in dashboard data storage unit 106 and arrange information obtained thereby as content in the display area. Display processing unit 101 may display a graph or table of data stored in dashboard data storage unit 106 or data obtained from the stored data.

[0023] The specific field targeted by the dashboard is not limited to any one field, and may be, for example, information security or marketing. In other words, the dashboard displayed on the display by the processing of the display processing unit 101 may be a security dashboard, a marketing dashboard, or any other dashboard.

[0024] FIG. 3 is a schematic diagram illustrating an example of a dashboard displayed by processing by the display processing unit 101. In the example illustrated in FIG. 3, the dashboard 50 includes display areas 51a, 51b, 51c, and 51d. The dashboard 50 is, for example, a dashboard for managing the security of a certain company's information processing system (hereinafter also referred to as the target system). Display area 51a displays, as an example, content indicating threats to a Web server constituting the target system. Specifically, display area 51a displays a graph showing monthly trends in threat index values. Display area 51a displays index values ​​calculated from data actually observed up to the present as a hatched bar graph, and future index values ​​as a dashed bar graph. Display area 51b displays, as an example, content indicating a system configuration diagram of the target system. Display area 51c displays, as an example, content indicating security topics. Specifically, display area 51c displays various topics in the security field, ranked by the popularity of each topic. In this way, display areas 51a, 51b, and 51c display content generated based on data stored in dashboard data storage unit 106. In contrast, display area 51d is a display area for accepting questions from a user viewing dashboard 50. That is, display processing unit 101 also arranges display area 51d having input field 52 for accepting questions. In the example shown in FIG. 3, the user inputs the text of a question into input field 52 as necessary.

[0025] 3 has a display area 51d with an input field 52 for accepting questions, but the dashboard does not necessarily have to have a display area with an input field for accepting questions. For example, the input field for accepting questions may be provided as a UI different from the dashboard, or if explanation information, described below, is provided without accepting input of questions, the input field may be omitted.

[0026] The input receiving unit 102 corresponds to the above-described input receiving unit 12. The input receiving unit 102 performs processing to receive input from a user. For example, the user performs input operations via an input device such as a pointing device or a keyboard provided in a terminal device communicatively connected to the dashboard display device 100, but the user may also perform input using an input device directly connected to the dashboard display device 100.

[0027] In this embodiment, the input accepting unit 102 accepts an input from a user viewing a dashboard displayed by the display processing of the display processing unit 101. For example, the input accepting unit 102 accepts an input specifying any of the contents displayed on the dashboard. The user specifies the content displayed on the dashboard for which the user wants an explanation. The user may specify one or more of the multiple display areas, or may specify a portion of the content within one display area. In the example shown in FIG. 3, the user uses the cursor 53 to select a future index value (a predicted value of the threat index value for May 2024) as the content for which the user wants an explanation. Note that the method of specifying the content does not necessarily have to be by the cursor 53, and any method capable of specifying any of the content may be adopted.

[0028] Furthermore, when a question is input from the user into the input field 52, the input accepting unit 102 accepts the input of the question from the user and acquires the question. As described above, the question does not necessarily have to be input via the input field 52, and the input accepting unit 102 may accept and acquire a question input by voice input from the user. Furthermore, when a question whose intention is unclear is accepted, such as a question that includes words unrelated to the content of the dashboard or the question, the input accepting unit 102 may output (for example, display) a message prompting the user to re-input the question.

[0029] The information acquisition unit 103 corresponds to the information acquisition unit 13 described above. The information acquisition unit 103 acquires information used to generate a prompt, which is a command statement for the LLM. When a user designates any content in the dashboard as content to be explained, the information acquisition unit 103 acquires information on the content designated by the user from the dashboard data storage unit 106. Here, the acquired information on the designated content may be data displayed as the content, or may be data used to derive the data displayed as the content. Note that when partial content in a certain display area (e.g., some data among multiple data constituting a graph) is designated, the information acquisition unit 103 may acquire information on the entire content in the display area (e.g., all data constituting a graph).

[0030] In this embodiment, the information acquiring unit 103 also acquires related information, which is information related to the content specified by the user. The related information acquired by the information acquiring unit 103 is, for example, information on other content displayed on the dashboard that is related to the content specified by the user. In other words, if there is content related to the content specified by the user, the information acquiring unit 103 also acquires information on the content related to the content specified by the user from the dashboard data storage unit 106. Hereinafter, content related to the content specified by the user will also be referred to as related content. Note that the presence or absence of a relationship between content items is defined in advance. Therefore, the information acquiring unit 103 identifies the related content items by, for example, referring to definition information that defines the relationship between content items. Note that the definition information that defines the relationship between content items may be stored in the knowledge information storage unit 107, which will be described later. The information on the related content items acquired by the information acquiring unit 103 may be data displayed on the dashboard or data used to derive the displayed data.

[0031] Furthermore, the related information acquired by the information acquisition unit 103 may be predetermined knowledge information related to the content specified by the user. The information acquisition unit 103 acquires predetermined knowledge information related to the content specified by the user from the knowledge information storage unit 107. The knowledge information storage unit 107 stores knowledge information related to the content. The knowledge information storage unit 107 may be configured as a database. In other words, the knowledge information storage unit 107 can also be called a knowledge base. In this embodiment, the dashboard display device 100 includes the knowledge information storage unit 107, but the knowledge information storage unit 107 may be configured as a device separate from the dashboard display device 100.

[0032] In this embodiment, the knowledge information is used to provide the LLM that generates the commentary information with detailed knowledge about the commentary subject. The knowledge information may be definition information for the data to be displayed, or information explaining what is indicated by the displayed data. The definition information for the displayed data may be information explaining data used to derive the displayed data, or information indicating a calculation method for the displayed data. Furthermore, the knowledge information may be information defining a normal or abnormal range for the value of the data displayed as content, or information explaining an event that occurs when the value of the data falls within an abnormal range. In this way, the knowledge information acquired by the information acquisition unit 103 is any supplementary information that can be used to explain the content.

[0033] The information acquiring unit 103 may acquire knowledge information related to related content from the knowledge information storage unit 107. That is, when acquiring information on related content, the information acquiring unit 103 may further acquire knowledge information about the related content.

[0034] The command statement generation unit 104 corresponds to the command statement generation unit 14 described above. The command statement generation unit 104 generates command statement text for instructing the LLM to generate explanation information that explains the content displayed on the dashboard, i.e., a prompt to be input to the LLM. The command statement generation unit 104 generates the prompt based on the information acquired by the information acquisition unit 103. FIG. 4 is a diagram showing an example of a prompt generated by the command statement generation unit 104. Note that FIG. 4 shows an example of a prompt generated when the user selects "threats to web servers," particularly the predicted threat index value for May 2024, which is a future index value, as the content for which an explanation is desired (see FIG. 3). The command statement generation unit 104 writes the prompt text according to a predetermined format corresponding to the LLM being used. In particular, in order to ensure that appropriate information is output from the LLM, the command statement generation unit 104 generates a prompt in which the information to be input to the LLM is organized and written by item. In Figure 4, the information to be entered into the LLM is organized into five items, and a prompt consisting of five partial texts 61a to 61e is generated. The text for each item is written as a simple bulleted list. By generating prompts that are organized and written in this way, it is possible to prevent inappropriate explanatory information from being output from the LLM.

[0035] In the example shown in FIG. 4, the partial text 61a is text indicating an instruction statement for the LLM. The command statement generation unit 104 generates text indicating an instruction statement for the LLM, for example, in accordance with the input received by the input receiving unit 102. Specifically, when content to be explained is specified, the command statement generation unit 104 generates text instructing the LLM to generate explanation information for the content. Furthermore, when a question is input, the command statement generation unit 104 generates text instructing the LLM to generate explanation information that explains the matter asked in the input question. Furthermore, when the input receiving unit 102 receives input of a question along with the specification of content to be explained, the command statement generation unit 104 generates text instructing the LLM to generate explanation information for the specified content that explains the matter asked in the question. By generating a command statement that reflects the question in this way, an explanation for the user's explicit question can be reflected in the explanation information.

[0036] In the example shown in Fig. 4, partial text 61b is text showing information about the content specified by the user as the subject of the explanation. Specifically, partial text 61b contains information about "Threats to Web Servers" as an example. More specifically, the text contains time-series data of numerical values ​​shown in a graph on the dashboard.

[0037] In the example shown in Fig. 4, partial text 61c is text indicating predetermined knowledge information related to the specified content. Specifically, partial text 61c contains information on "threat definition," which is information explaining the index value.

[0038] In the example shown in Fig. 4, partial text 61d is text that shows information about content related to the content specified by the user, i.e., information about the related content described above. In the example shown in Fig. 4, it is assumed that "Threats to Web Servers" and "Security Topics" are predefined as being related to each other. Therefore, partial text 61d contains information about "Security Topics."

[0039] In this way, in this embodiment, the command statement generator 104 can generate a prompt to be input to the LLM based on the information on the specified content and related information. As a result, various information about the subject of explanation is provided to the LLM, and it is expected that comprehensive explanation information will be obtained.

[0040] The partial text 61e is a text indicating the purpose of use of the information output from the LLM. In the example shown in Fig. 4, the partial text 61e contains, for example, a predetermined sentence indicating that the purpose is to help the user understand the content displayed on the security dashboard.

[0041] Note that when the input receiving unit 102 receives only the input of a question, i.e., when it does not receive the specification of the content to be explained, the information acquiring unit 103 may acquire information as follows. That is, in this case, the information acquiring unit 103 may acquire information assuming that all content displayed on the dashboard is specified, or may acquire information assuming that a predetermined portion of the content is specified. Therefore, even when only a question is input, the prompt may include text indicating information about the content and knowledge information about the content.

[0042] The command statement generation unit 104 may generate the above-mentioned prompt by embedding keywords according to the input received by the input receiving unit 102 into a fixed phrase. Alternatively, the command statement generation unit 104 may generate a prompt by performing predetermined well-known natural language processing on the input received by the input receiving unit 102. The command statement generation unit 104 may use the LLM not only to output explanatory information but also to obtain a prompt written in a predetermined format. That is, the command statement generation unit 104 may obtain a prompt written in a predetermined format by inputting a prompt requesting corrections to a draft prompt that is not written in a predetermined format into the LLM.

[0043] The explanation acquisition unit 105 corresponds to the explanation acquisition unit 15 described above. The explanation acquisition unit 105 inputs the prompt generated by the command statement generation unit 104 into the LLM and acquires explanation information generated by the LLM based on the prompt. The LLM used to generate the explanation information is, for example, a large-scale language model, which is a model trained by machine learning such as deep learning. While any known LLM can be used as such an LLM, it is preferable that the LLM is an LLM that has learned knowledge such as terminology about a specific field (e.g., security) that the displayed dashboard targets. Furthermore, the LLM used to generate the explanation information may be an LLM provided in the dashboard display device 100, an LLM provided in an external server that provides a generation AI service, or an LLM provided in a terminal device on which the dashboard is displayed.

[0044] When the explanation acquisition unit 105 acquires explanation information, the display processing unit 101 performs processing to display the explanation information on a dashboard. Fig. 5 is a schematic diagram showing an example of a dashboard with explanation information that is displayed by processing of the display processing unit 101. In the example shown in Fig. 5, a dashboard 50 is displayed to which a pop-up image 54 indicating the explanation information acquired by the explanation acquisition unit 105 is added.

[0045] In the example shown in FIG. 5 , the commentary information displayed in the pop-up image 54 includes, in particular, a commentary on future indicator values ​​(specifically, the predicted value of the threat indicator value for May 2024). This is in response to the user's selection of data indicating a future state as the content for which an explanation is desired. When the content specified by the user is data indicating the future state of an object whose state is managed in the dashboard, the command statement generation unit 104 generates a prompt instructing the output of commentary information explaining the predicted future state. Specifically, for example, as illustrated in partial text 61a in FIG. 4 , a prompt is generated instructing the output of commentary information explaining the predicted index value for May 2024, which is data indicating the future state. This allows even a user who is not familiar with the content displayed in the dashboard to understand the future state displayed in detail. For example, the user can understand why such a future state is predicted. In particular, in this embodiment, commentary information is generated according to a prompt including related information, making it possible to obtain information that more appropriately explains the basis for predicting the future state displayed in the dashboard and the meaning of the predicted future state. As described above, in this embodiment, the command statement generating unit 104 determines the time period for which the data of the specified content to be explained pertains, and generates a prompt appropriate for that time period.

[0046] 5, the pop-up image 54 displays only explanatory information, but the pop-up image 54 may also include an input field for accepting further questions. If a user who has viewed the explanatory information desires further explanation, the user may input a question in the input field 52 or the input field of the pop-up image 54. In this case, the question may be accepted by the input accepting unit 102, and a prompt for obtaining further explanatory information may be generated by the command statement generating unit 104.

[0047] In addition to displaying explanatory information on the dashboard, the display processing unit 101 may also perform processing to display, on the dashboard, an image obtained by modifying a portion of an image displayed on the dashboard based on the acquired explanatory information. This allows for providing information that is easier for the user to understand. This will be described with reference to FIG. 6. FIG. 6 is a schematic diagram showing another example of a dashboard with explanatory information displayed by the processing of the display processing unit 101. Here, it is assumed that the explanatory information acquired from the LLM identifies the device that is the origin of a cyber attack, the device that is the target of the attack, and the attack path. To obtain such explanatory information, the information acquisition unit 103 may, for example, acquire information indicating the system configuration as related information, and the command statement generation unit 104 may generate a prompt including that information. In the example shown in FIG. 6, the display processing unit 101 modifies and displays an image of a system configuration diagram in the display area 51b. Specifically, as an example, the display processing unit 101 processes the image of the system configuration diagram by adding a mark 55a indicating the device that is the origin of the attack, a mark 55b indicating the device that is the target of the attack, and a mark 55c indicating the attack path. To obtain an image in which a portion of an image displayed on the dashboard has been modified based on the acquired commentary information, the display processing unit 101 may perform, for example, the following processing. The display processing unit 101 may process the image so as to add a predetermined mark corresponding to a keyword to an object (e.g., a device constituting the system) described together with a predetermined keyword (e.g., "starting point of attack," "target of attack," or "path of attack") in the commentary information. The display processing unit 101 may also use an image generation AI to obtain an image that has been modified based on the commentary information. In this case, specifically, the display processing unit 101 obtains an image that has been modified based on the commentary information by inputting the commentary information and the image currently displayed on the dashboard into an image generation model that has been trained by machine learning such as deep learning.

[0048] 7 is a block diagram showing an example of a hardware configuration of the dashboard display device 100. As shown in FIG.

[0049] The input / output interface 151 is an interface for connecting to other devices or networks as needed, and may include a network interface.

[0050] The memory 152 is configured, for example, by a combination of a volatile memory and a non-volatile memory. The memory 152 is used to store software (computer programs) including one or more instructions executed by the processor 153, data used for various processes, etc. The dashboard data storage unit 106 and the knowledge information storage unit 107 may be realized by a storage device such as the memory 152.

[0051] The processor 153 reads and executes software (computer programs) from the memory 152 to perform processing of each component shown in Fig. 2. The processor 153 may be, for example, a microprocessor, an MPU (Micro Processor Unit), or a CPU (Central Processing Unit). The processor 153 may include multiple processors. In this way, the dashboard display device 100 has the functionality of a computer.

[0052] The program includes instructions (or software code) that, when loaded into a computer, cause the computer to perform one or more functions described in the embodiments. The program may be stored in a non-transitory computer-readable medium or a tangible storage medium. By way of example and not limitation, computer-readable media or tangible storage media include random-access memory (RAM), read-only memory (ROM), flash memory, solid-state drive (SSD) or other memory technologies, CD-ROM, digital versatile disc (DVD), Blu-ray® disc or other optical disk storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage devices. The program may also be transmitted on a transitory computer-readable medium or communication medium. By way of example and not limitation, transitory computer-readable media or communication media include electrical, optical, acoustic, or other forms of propagated signals.

[0053] Next, the flow of processing by each component shown in Fig. 2 will be described with reference to the flowchart of Fig. 8. Fig. 8 is a flowchart showing an example of the operation of the dashboard display device 100.

[0054] In step S100, display processing unit 101 performs processing to display a dashboard using data stored in dashboard data storage unit 106.

[0055] Next, in step S101, the input accepting unit 102 accepts input from a user viewing the dashboard displayed in step S100. In this step, the input accepting unit 102 accepts a specification of an explanation target from a user who desires an explanation about the dashboard displayed in step S100, that is, a specification of content that the user wants explained. Note that the input accepting unit 102 may accept input of a question along with or instead of the specification of an explanation target.

[0056] Next, in step S102, the information acquisition unit 103 acquires information for generating a prompt, which is a command statement for the LLM. For example, the information acquisition unit 103 acquires information on the content designated as the subject of the explanation and information related to the content from the dashboard data storage unit 106 or the knowledge information storage unit 107.

[0057] Next, in step S103, the command statement generating unit 104 generates a command statement for instructing the LLM to generate commentary information based on the information acquired in step S102.

[0058] Next, in step S104, the explanation acquisition unit 105 acquires explanation information using the LLM. That is, the explanation acquisition unit 105 inputs the command statement generated in step S103 into the LLM and acquires the explanation information output from the LLM.

[0059] Then, in step S105, the display processing unit 101 performs processing to display the guidance information acquired in step S104 on the dashboard.

[0060] Note that the input from the user and the display of the corresponding guidance information may be repeated, that is, after step S105, the process may return to step S101.

[0061] The second embodiment has been described above. According to the dashboard display device 100, a command statement (prompt) instructing the LLM to generate commentary information is generated in response to an input from a user, and the commentary information is acquired from the LLM. The commentary information is then displayed on the dashboard. This allows the user to refer to the commentary information, making it easier to understand the information (i.e., content) displayed on the dashboard. Furthermore, particularly in this embodiment, the related information acquired by the information acquisition unit 103 is also used to generate the command statement. Therefore, various information about the commentary subject is provided to the LLM, and it is expected that commentary information with substantial content will be obtained.

[0062] Although the present disclosure has been described above with reference to the embodiments, the present disclosure is not limited to the above-described embodiments. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure. Furthermore, each embodiment can be combined with other embodiments as appropriate.

[0063] Some or all of the above embodiments may also be described as, but are not limited to, the following supplementary notes. Some or all of the elements (e.g., configurations and functions) described in supplementary notes 2 to 9 that are dependent on supplementary note 1 may also be dependent on supplementary notes 10 and 11 in the same dependent relationship as supplementary notes 2 to 9. Furthermore, some or all of the elements described in any supplementary note may be applied to various hardware, software, recording means for recording software, systems, and methods.

[0064] (Appendix 1) an input receiving means for receiving a user's input specifying one of the contents displayed on the dashboard; an information acquisition means for acquiring information on the designated content; a command statement generating means for generating a command statement to be input to a language model based on information of the specified content; a comment acquisition means for acquiring comment information of the specified content, which is generated by the language model based on the command sentence; a display processing means for performing processing to display the commentary information on the dashboard; A dashboard display device comprising: (Appendix 2) the information acquisition means further acquires related information that is information related to the specified content; The command statement generating means generates the command statement to be input to the language model based on the information of the specified content and the related information. 10. The dashboard display device of claim 1. (Appendix 3) The related information includes information about other content displayed on the dashboard that is related to the specified content. 10. The dashboard display device according to claim 2. (Appendix 4) The related information includes predetermined knowledge information related to the specified content. 4. The dashboard display device according to claim 2 or 3. (Appendix 5) The related information includes predetermined knowledge information related to the other content. 4. The dashboard display device according to claim 3. (Appendix 6) The input receiving means further receives input of a question sentence from the user; The command sentence generation means generates the command sentence to be input to the language model based on the question sentence. 6. A dashboard display device according to any one of appendices 1 to 5. (Appendix 7) When the specified content is data indicating a future state, the command statement generating means generates the command statement instructing output of the commentary information that explains that the future state is predicted. 7. A dashboard display device according to any one of claims 1 to 6. (Appendix 8) The display processing means further performs processing for displaying, on the dashboard, an image obtained by correcting a part of the image displayed on the dashboard based on the commentary information. 8. A dashboard display device according to any one of claims 1 to 7. (Appendix 9) The command statement generation means generates the command statement in which information to be input to the language model is organized and described by item. 9. A dashboard display device according to any one of claims 1 to 8. (Appendix 10) The computer Accepts user input specifying one of the contents displayed on the dashboard, Acquire information about the specified content, Generate a command statement to be input to a language model based on the specified content information; Obtaining commentary information for the specified content, generated by the language model based on the command sentence; The explanation information is displayed on the dashboard. Display method. (Appendix 11) an input receiving step of receiving a user's input specifying one of the contents displayed on the dashboard; an information acquisition step of acquiring information on the designated content; a statement generation step of generating a statement to be input to a language model based on information of the specified content; a commentary acquisition step of acquiring commentary information of the specified content, which is generated by the language model based on the command sentence; a display processing step of performing processing to display the commentary information on the dashboard; A program that causes a computer to execute the following. [Explanation of symbols]

[0065] 10 Dashboard display device 11 Display processing section 12 Input reception section 13 Information acquisition department 14 Imperative sentence generator 15 Explanation Acquisition Section 50 Dashboards 51 Display Area 52 input fields 53 Cursor 54 Pop-up Images 55 marks 61 Partial Text 100 Dashboard display device 101 Display processing unit 102 Input reception unit 103 Information acquisition department 104 Imperative statement generator 105 Explanation Acquisition Section 106 Dashboard data storage unit 107 Knowledge information storage unit 151 Input / Output Interface 152 memory 153 processors

Claims

1. an input receiving means for receiving a user's input specifying one of the contents displayed on the dashboard; an information acquisition means for acquiring information on the designated content; a command statement generating means for generating a command statement to be input to a language model based on information of the specified content; a comment acquisition means for acquiring comment information of the specified content, which is generated by the language model based on the command sentence; a display processing means for performing processing to display the commentary information on the dashboard; A dashboard display device comprising:

2. the information acquisition means further acquires related information that is information related to the specified content; The command statement generating means generates the command statement to be input to the language model based on the information of the specified content and the related information. The dashboard display device of claim 1 .

3. The related information includes information about other content displayed on the dashboard that is related to the specified content. The dashboard display device of claim 2.

4. The related information includes predetermined knowledge information related to the specified content.

4. The dashboard display device according to claim 2 or 3.

5. The related information includes predetermined knowledge information related to the other content. The dashboard display device of claim 3.

6. The input receiving means further receives input of a question sentence from the user; The command sentence generation means generates the command sentence to be input to the language model based on the question sentence. The dashboard display device of claim 1 .

7. When the specified content is data indicating a future state, the command statement generating means generates the command statement instructing output of the commentary information that explains that the future state is predicted. The dashboard display device of claim 1 .

8. The display processing means further performs processing for displaying, on the dashboard, an image obtained by correcting a part of the image displayed on the dashboard based on the commentary information. The dashboard display device of claim 1 .

9. The computer Accepts user input specifying one of the contents displayed on the dashboard, Acquire information about the specified content, Generate a command statement to be input to a language model based on the specified content information; Obtaining commentary information for the specified content, generated by the language model based on the command sentence; The explanation information is displayed on the dashboard. Display method.

10. an input receiving step of receiving a user's input specifying one of the contents displayed on the dashboard; an information acquisition step of acquiring information on the designated content; a statement generation step of generating a statement to be input to a language model based on information of the specified content; a commentary acquisition step of acquiring commentary information of the specified content, which is generated by the language model based on the command sentence; a display processing step of performing processing to display the commentary information on the dashboard; A program that causes a computer to execute the following.

Images