Information processing device, information processing method, and program

Abstract

This technology generates suggested commands based on the status of a Gantt chart, allowing users to easily manipulate the Gantt chart. [Solution] An information processing device comprising a processing device and a display device, wherein the processing device displays a Gantt chart on the display device, generates command suggestions in natural language format based on the status of the Gantt chart, displays the command suggestions on the display device, accepts a command selected by the user from among the command suggestions, and modifies the configuration information of the Gantt chart based on the selected command.

Description

Technical Field

[0001] The present invention relates to an information processing apparatus, an information processing method, and a program.

Background Art

[0002] In project management, it is common to use a Gantt chart to manage tasks and progress. Various GUI technologies and work support technologies for assisting in the editing work of Gantt charts have also been proposed.

[0003] Editing a Gantt chart requires the user to manually operate objects such as tasks and activities. When operating multiple objects simultaneously, the input work becomes cumbersome. Also, even when using AI, the user needs to create appropriate command sentences in natural language, and there may be cases where it is not processed as intended due to ambiguity in expression.

[0004] Patent Document 1 describes a technique in which a generative AI collaborates with a calendar, organizes schedules and contents, automatically creates a Gantt chart, and manages the progress of a project.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0006] However, Patent Document 1 does not describe a configuration for generating proposed commands according to the types and states of objects on the Gantt chart management screen or for determining the updated content of the Gantt chart using chat history.

[0007] Therefore, the present invention aims to provide a technology that generates command suggestions based on the status of a Gantt chart, enabling users to easily manipulate the Gantt chart. [Means for solving the problem]

[0008] To solve the above problems, one representative information processing apparatus of the present invention is an information processing apparatus comprising a processing apparatus and a display device, wherein the processing apparatus displays a Gantt chart on the display device, generates command suggestions in natural language form based on the status of the Gantt chart, displays the command suggestions on the display device, accepts a command selected by the user from among the command suggestions, and modifies the configuration information of the Gantt chart based on the selected command. [Effects of the Invention]

[0009] According to the present invention, command suggestions are generated based on the status of the Gantt chart, allowing the user to easily manipulate the Gantt chart.

[0010] Other issues, configurations, and effects will be clarified by the following description of embodiments for carrying out the invention. [Brief explanation of the drawing]

[0011] [Figure 1] This figure shows an example of the functional configuration of the information processing device according to this embodiment. [Figure 2] This block diagram shows an example of the hardware configuration of a computer for realizing the information processing device according to this embodiment. [Figure 3A] This figure shows an example of the data structure of Gantt chart data in the information processing device according to this embodiment. [Figure 3B] This figure shows an example of the data structure of selected object data in the information processing device according to this embodiment. [Figure 4A] This figure shows an example of the configuration of the user operation screen in the information processing device according to this embodiment. [Figure 4B]This is a diagram showing a configuration example of a preview display section displayed on a user operation screen in the information processing apparatus according to this embodiment. [Figure 5] This is a diagram showing a configuration example of a Gantt chart automatic operation section in the information processing apparatus according to this embodiment. [Figure 6] This is a diagram showing a configuration example of an instruction proposal generation section in the information processing apparatus according to this embodiment. [Figure 7] This is a flowchart showing an example of an instruction proposal generation process executed by an instruction proposal generation section in the information processing apparatus according to this embodiment. [Figure 8] This is a diagram showing an example of a record recorded in a log database in the information processing apparatus according to this embodiment.

Mode for Carrying Out the Invention

[0012] Hereinafter, examples will be described with reference to the drawings.

[0013] FIG. 1 is a diagram showing an example of a functional configuration of an information processing apparatus 100 according to this embodiment.

[0014] The information processing apparatus 100 includes a Gantt chart display section 101, an instruction proposal generation section 102, an instruction proposal display section 103, an instruction proposal selection section 104, a Gantt chart automatic operation section 105, and a log database 106.

[0015] The Gantt chart display section 101 displays a Gantt chart for visually managing tasks and schedules of a project on a GUI. Further, the Gantt chart display section 101 displays a chat history between a user and AI on a GUI (Graphical User Interface).

[0016] The Gantt chart display section 101 inputs JSON (JavaScript Object Notation) data (hereinafter referred to as "Gantt chart data") representing the currently displayed Gantt chart to the instruction proposal generation section 102.

[0017] In addition, when the user selects an object such as a task, activity, milestone, etc. on the Gantt chart, the Gantt chart display unit 101 inputs JSON data representing the selected object (hereinafter referred to as "selected object data") to the instruction plan generation unit 102.

[0018] Furthermore, the Gantt chart display unit 101 inputs text data of the chat history (hereinafter referred to as "chat history information") to the instruction plan generation unit 102.

[0019] The instruction plan generation unit 102 obtains the Gantt chart data, selected object data, and chat history information from the Gantt chart display unit 101, and automatically generates a natural language form instruction plan to be sent to the AI based on situation judgments such as the progress status, delay possibility, presence of a person in charge, and load status of the Gantt chart. The generated instruction plan is output to the instruction plan display unit 103 and displayed in a list.

[0020] The instruction plan display unit 103 presents the instruction plan generated by the instruction plan generation unit 102 on the GUI. When multiple types of instruction plans are generated, an instruction plan list is displayed, and the user can select one or more instructions from it. The instruction plan display unit 103 also functions as a chatbot screen, and it is also possible for the user to directly input an instruction in natural language.

[0021] When the user selects one or more instructions from the instruction plans presented on the GUI by the instruction plan display unit 103 using a checkbox or the like and presses the send button, the instruction selection unit 104 inputs the selected instructions to the Gantt chart automatic operation unit 105.

[0022] The Gantt chart automated operation unit 105, based on commands entered from the command selection unit 104 or natural language instructions via a chatbot, performs actions such as adding, deleting, moving, and changing the duration of tasks, activities, and milestones on the Gantt chart data acquired from the Gantt chart display unit 101, thereby generating updated Gantt chart data. The Gantt chart display unit 101 updates the Gantt chart displayed on the GUI based on the updated Gantt chart data.

[0023] The log database 106 obtains generated instruction drafts and text data representing the status of the Gantt chart at the time of instruction draft generation (hereinafter referred to as "Gantt chart status information") from the instruction draft generation unit 102. The log database 106 also obtains the instruction selected by the user from the instruction draft selection unit 104. The log database 106 contributes to improving the accuracy of instruction draft generation by utilizing this historical information in subsequent instruction draft generation processes.

[0024] When a user selects an object on the Gantt chart display unit 101, the Gantt chart data and the selected object data are input to the instruction draft generation unit 102.

[0025] As shown in Figure 1, the configuration eliminates the need for users to accurately input natural language commands to manipulate the Gantt chart. Optimal command suggestions are automatically generated based on the status of the Gantt chart, thereby reducing the operational burden of Gantt chart management.

[0026] Figure 2 is a block diagram showing an example of the hardware configuration of a computer for realizing the information processing device 100 according to this embodiment.

[0027] The Gantt chart display unit 101, the command draft generation unit 102, the command draft display unit 103, the command draft selection unit 104, the Gantt chart automatic operation unit 105, and the log database 106 shown in Figure 1 are all configured to run or be maintained on the computer 200.

[0028] The computer 200 is comprised of a central processing unit (Central Processing Unit) 201, memory 202, storage device 203, communication device 204, input device 205, and display device 206, all interconnected via an internal communication channel 207.

[0029] The processing unit 201 consists of a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), etc., and loads programs stored in the storage device 203 into the memory 202 and executes them, thereby realizing management functions for the Gantt chart display unit 101, instruction proposal generation unit 102, instruction proposal display unit 103, instruction proposal selection unit 104, Gantt chart automatic operation unit 105, and log database 106 as software.

[0030] Memory 202 is the main memory used by the processing unit 201 for program execution, and is composed of volatile memory elements such as RAM (Random Access Memory).

[0031] The storage device 203 is a non-volatile auxiliary storage device that holds data such as Gantt chart information, user-selected object information, and chat history, as well as programs that implement each of the aforementioned functional units, and is composed of an SSD (Solid State Drive) or the like. Furthermore, the log database 106 is also built on the storage device 203.

[0032] The communication device 204 consists of a network adapter, a communication module, etc., and connects to an external network to perform data communication with AI services on the cloud and external systems.

[0033] The input device 205 consists of a user interface such as a mouse, keyboard, and touch panel, and accepts user input such as Gantt chart manipulation, command selection, and chat input.

[0034] The display device 206 consists of a display and the like, and presents the user with Gantt charts, command suggestions, operation results, and other information.

[0035] The internal communication channel 207 is a bus or internal network that exchanges data between the above-mentioned components.

[0036] With this configuration, each functional unit of the information processing device 100 operates in cooperation on the hardware of the computer 200, enabling the various processes described later to be realized.

[0037] Figure 3A shows an example of the data structure of Gantt chart data in the information processing device 100 according to this embodiment.

[0038] Gantt chart data is JSON data used to represent the state of a Gantt chart, and consists of task information, activity information, milestone information, and lightning line information.

[0039] The Gantt chart data serves to provide the Gantt chart status to the instruction generation unit 102 and the Gantt chart automatic operation unit 105.

[0040] The Gantt chart data includes, for example, "tasks" representing a list of tasks, "activities" representing a list of activities, "milestones" representing a list of milestones, and "kaminari" indicating the progress at a specific point in the Gantt chart as primary keys.

[0041] The "tasks" object includes attributes such as tasks.task_id, which uniquely identifies a task; tasks.name, which is the task name; and tasks.resource_id, which indicates the person responsible for the task.

[0042] The "activities" directory includes active.id, which uniquely identifies the activity; activities.start, which indicates the start date; activities.end, which indicates the end date; activities.shintyoku, which indicates the progress rate; and activities.task_id, which indicates the task to which the activity belongs.

[0043] The "milestones" directory includes milestones.id, which uniquely identifies each milestone, and milestones.date, which indicates the date the milestone was set.

[0044] "Kaminari" is a key that indicates the date of the lightning bolt line in a Gantt chart, and is used to understand the difference between the plan and the actual results.

[0045] With this configuration, the instruction draft generation unit 102 can obtain information such as task progress, remaining days, whether or not there is a person in charge, and activity status from Gantt chart data, enabling it to make highly accurate judgments about the overall situation of the Gantt chart, including the possibility of delays and load status.

[0046] Figure 3B shows an example of the data structure of selected object data in the information processing device 100 according to this embodiment.

[0047] The selected object data is generated based on the user's selection of objects on the Gantt chart.

[0048] The selected object data consists of, for example, "tasks," which lists the IDs of the selected tasks; "activities," which lists the IDs of the selected activities; and "milestones," which lists the IDs of the selected milestones.

[0049] These selected object data are input to the command draft generation unit 102, which, in addition to considering the overall status of the Gantt chart, prioritizes the selected objects to appropriately generate command drafts for manipulating the Gantt chart.

[0050] Figure 4A is a diagram showing an example of the configuration of the user operation screen in the information processing device 100 according to this embodiment.

[0051] The user operation screen 300 includes a Gantt chart display unit 301, a chat display unit 302, and a proposed command display unit 303, which work in coordination with each other.

[0052] The Gantt chart display unit 301 is an area that visually displays the elements that make up the Gantt chart, such as tasks, activities (duration), milestones, progress rates, and lightning bolt lines. The user can select any object using mouse operations, and the information of the selected object is input to the command draft generation unit 102.

[0053] In this example, the Gantt chart display unit 301 shows the activities of Tasks 1 to 3 along with their respective time periods. Each activity is marked with a diagonal line indicating its progress, allowing the user to visually grasp the status of the work. Task 2 and its activities are selected by the user.

[0054] The chat display unit 302 is an area that displays the history of conversations between the user and the AI, and is also used for inputting commands in natural language. The chat display unit 302 stores a history of previously performed operations and the AI's responses, allowing the user to easily confirm the intent of their previous operation.

[0055] The command proposal display unit 303 is an area that displays the command proposals automatically generated by the command proposal generation unit 102 in a list format. The command proposal display unit 303 also has a command proposal generation button 304 for manually starting the command proposal generation process, and the user can start command proposal generation by the command proposal generation unit 102 at any time by clicking this button.

[0056] The command draft generation unit 102 generates multiple command drafts based on information about the object selected by the user and the status of the Gantt chart (task progress rate, remaining days, presence or absence of assigned personnel, deviation from the lightning bolt line, etc.), and presents them to the command draft display unit 303.

[0057] The command suggestions displayed on the command suggestion display unit 303 are marked with checkboxes, allowing the user to select the desired command. The command suggestion display unit 303 also has a send button 305, which is used to send the command selected by the user to the Gantt chart automatic operation unit 105.

[0058] When the user selects an instruction and presses the send button 305, the selected instruction is sent to the Gantt chart automatic operation unit 105, and the configuration information of the Gantt chart is automatically edited. The edited Gantt chart is reflected in the Gantt chart display unit 301, and the content of the sent instruction is recorded as history in the chat display unit 302.

[0059] Figure 4B is a diagram showing an example of the configuration of the preview display unit displayed on the user operation screen 300 in the information processing device 100 according to this embodiment.

[0060] The preview display unit 306 in Figure 4B shows the state in which the AI ​​(Gantt chart automatic operation unit 105) is visually presenting the changes that it intends to apply to the Gantt chart based on the command selected by the user in Figure 4A.

[0061] In Figure 4B, a new person in charge, A, has been added to Task 2, and the activity period has also been extended based on the proposed instructions. This preview display unit 306 allows the user to check in advance what the result will be when the selected instructions are implemented.

[0062] Furthermore, the preview display section 306 is provided with a confirmation field 307 for the user to individually approve or reject each operation that the AI ​​is attempting to implement. The confirmation field 307 displays the number of objects to be operated on (e.g., "1 / 2") and confirmation buttons corresponding to each operation, such as "○" (approved), "×" (rejected), and "All ○" (approve all operations at once). This allows the user to finely control whether or not the AI's automated operations are implemented, preventing unwanted changes from being reflected in the Gantt chart.

[0063] This configuration allows users to easily obtain command suggestions based on objects selected in the Gantt chart display unit 301, and execute AI-driven Gantt chart operations simply by selecting and submitting checkboxes. As a result, the workload of Gantt chart management is reduced, and efficient project operation can be achieved.

[0064] Figure 5 shows an example of the configuration of the Gantt chart automatic operation unit 105 in the information processing device 100 according to this embodiment.

[0065] The Gantt chart automated operation unit 105 has the function of automatically updating the configuration information of the Gantt chart based on user instructions in natural language entered by the user in the chat display unit 302, or on the instruction suggestion display unit 303.

[0066] As shown in Figure 5, the Gantt chart automated operation unit 105 is mainly composed of an operation target determination unit 1051 and an object operation unit 1052.

[0067] When a user inputs a user instruction in natural language, such as "add a design task," the instruction is input to the operation target determination unit 1051. The operation target determination unit 1051 analyzes the content of the user instruction, determines what type of object (task, activity, milestone, etc.) should be operated on, and based on the determination result, calls the appropriate function in the object operation unit 1052 and inputs the instruction for the object operation. For example, if the user instruction is related to a task, the operation target determination unit 1051 determines that the task operation unit 1053 should be called.

[0068] The operation target determination unit 1051 may be configured such that a large-scale language model (LLM) selects and calls a function within the object operation unit 1052 based on user instructions. In this case, descriptions of callable functions and criteria for selection are set in advance as prompts, and for example, the tool calling mechanism of LangChain may be used.

[0069] Furthermore, the prompt may also include information about the structure of the Gantt chart data (such as the meaning of tags and corresponding attributes) and the operations to be performed (such as adding or deleting tasks, or changing task_ids). This allows the LLM to select the appropriate operation while referring to both the user instructions and the Gantt chart data.

[0070] Furthermore, the operation target determination unit 1051 may have a prompt pre-set for analyzing user input and calling an appropriate function. This prompt may include, for example, a statement such as "Please call an appropriate function according to the user input. [User Input]" or a list of functions such as "Available functions include the Task Operation Unit (task operation processing), Activity Operation Unit (activity operation processing), Milestone Operation Unit (milestone operation processing)." This enables the operation target determination unit 1051 to determine with high accuracy which operation unit to select in response to given user input.

[0071] The object manipulation unit 1052 is a functional unit that brings together various functions related to the automated operation of the Gantt chart, and includes the task manipulation unit 1053 (a group of functions that perform task operations), the activity manipulation unit 1054 (a group of functions that perform activity operations), and the milestone manipulation unit 1055 (a group of functions that perform milestone operations). The manipulation units 1053 to 1055 operate as a so-called AI agent configured using LLM, and can handle both natural language and JSON data.

[0072] The object manipulation unit 1052 may adopt a configuration in which the LLM directly edits the Gantt chart based on the Gantt chart data and user instructions. In this case, the prompt includes an explanation on how to read the Gantt chart data in JSON format, such as "The following JSON represents the Gantt chart" and "The task list and activity list are represented by the following tags."

[0073] Furthermore, the prompt may also include specific editing rules as part of the operation, such as "Add: Create a new task_id" or "Delete: Delete the corresponding task_id." This allows the operation units 1053-1055 to perform actions such as adding and deleting objects.

[0074] The object operation unit 1052 receives object operation instructions from the target determination unit 1051 and Gantt chart data from the Gantt chart display unit 101 as input. Based on the object operation instructions from the target determination unit 1051, the object operation unit 1052 identifies the target object and generates updated Gantt chart data by performing operations such as adding, deleting, moving, and changing the period on the corresponding object data.

[0075] With this configuration, the information processing device 100 automatically performs appropriate Gantt chart operations based on user instructions in natural language or selected commands, thereby reducing the burden on the user and improving the efficiency of Gantt chart editing work.

[0076] Figure 6 shows an example of the configuration of the instruction draft generation unit 102 in the information processing device 100 according to this embodiment.

[0077] The command suggestion generation unit 102 has the function of generating a command suggestion in natural language that indicates the next operation the user should perform, based on Gantt chart data, selected object data, and chat history information.

[0078] In this embodiment, selected object data and chat history information are not mandatory, and even if they are not present, it is possible to generate a draft instruction based solely on the Gantt chart data.

[0079] As shown in Figure 6, the command draft generation unit 102 is mainly composed of three functional units: a situation determination unit 1021, a chat analysis unit 1022, and a generation unit 1023.

[0080] The situation determination unit 1021, the chat analysis unit 1022, and the generation unit 1023 are all implemented using LLM, and the analysis accuracy and generation accuracy can be improved by providing prompt information to them in advance.

[0081] The status determination unit 1021 receives Gantt chart data and selected object data as input, extracts, for example, task delay status, remaining days, progress, dependency issues, etc., translates the status of the Gantt chart into natural language, and outputs it to the generation unit 1023 as Gantt chart status information in text format.

[0082] The status determination unit 1021 defines within the prompt how to view the JSON data, the scope of information to be acquired, the extraction conditions for delayed tasks (e.g., remaining days less than 5 and progress rate less than 30%), the person in charge information, the method for checking dependencies, and the output format, and is configured to allow the LLM to extract all the necessary information without any omissions or excesses. This improves the accuracy of the status determination process and enables the generation of text data in a unified format suitable for subsequent processing.

[0083] The chat analysis unit 1022 receives chat history information, which is the content of past conversations with the user, as input. From this chat history information, it analyzes the user's intended actions and objectives, and outputs them to the generation unit 1023 as operation information in text format.

[0084] The chat analysis unit 1022 is configured to extract user intent with high accuracy by including a policy within the prompt to "summarize what actions were performed and for what purpose based on the content of the most recent conversation." Furthermore, the chat analysis unit 1022 can perform behavioral estimation based on the context of the chat history, such as extracting delayed tasks for risk management purposes and performing operations to extend the duration of delayed tasks. As a result, the chat analysis unit 1022 appropriately grasps the user's intent and provides operation information that the generation unit 1023 should refer to.

[0085] The generation unit 1023 receives Gantt chart status information output from the status determination unit 1021 and operation information output from the chat analysis unit 1022 as input, and generates a draft command to be presented to the user as text data.

[0086] The generation unit 1023 is configured to generate appropriate command proposals according to the situation by specifying a two-stage generation policy within the prompt: "generating command proposals based on Gantt chart status information" and "generating command proposals based on operation information."

[0087] The prompts are designed to make it easier for the LLM to make decisions by including multiple examples of combinations of Gantt chart status information and proposed instructions (e.g., if there are personnel with a high workload, reallocate personnel; if there are tasks without assigned personnel, add personnel; if a task is likely to be delayed, extend the period and add assigned personnel, etc.).

[0088] Furthermore, the system will also suggest suggested commands based on the previous operation (e.g., moving an activity → moving a subsequent task, deepening a task → adding a subtask, etc.), allowing it to estimate the user's intent and suggest the next operation.

[0089] Furthermore, the system controls the output to be obtained by defining output priorities within the prompt, prioritizing chat history information if available, then suggesting instructions based on the selected object, and finally presenting general suggestions based on the overall Gantt chart situation.

[0090] This allows the generation unit 1023 to generate command suggestions that are appropriate to the situation and context.

[0091] With this configuration, the command generation unit 102 can propose the next operation based on both an objective evaluation of the Gantt chart status and the user's subjective intentions.

[0092] Figure 7 is a flowchart showing an example of the instruction draft generation process executed by the instruction draft generation unit 102 in the information processing device 100 according to this embodiment.

[0093] First, the command draft generation unit 102 determines, as a trigger to start processing, whether the configuration information of the Gantt chart has been changed by the Gantt chart automatic operation unit 105, or whether the user has clicked the command draft generation button 304 to start command draft generation (step S401).

[0094] If the trigger is met (YES in S401), the command draft generation unit 102 acquires Gantt chart data, selected object data, and chat history information (step S402).

[0095] In this embodiment, selected object data and chat history information are not required, and even if they are not present, the instruction draft generation unit 102 can generate instruction drafts based solely on JSON data (Gantt chart data) indicating the state of the Gantt chart.

[0096] Next, the status determination unit 1021 analyzes the start date, end date, progress rate, remaining days, person in charge, resource load, and potential delays for each task based on the acquired Gantt chart data and selected object data, and generates Gantt chart status information that shows the status of the entire Gantt chart (step S403). If selected object data exists, the status determination unit 1021 prioritizes extracting information about the selected objects.

[0097] Next, the command draft generation unit 102 determines whether or not chat history information exists (step S404). If chat history information does not exist, the process proceeds to step S406.

[0098] On the other hand, if chat history information exists, the chat analysis unit 1022 analyzes the chat history information, extracts the user's purpose and content of the operation from the most recent conversation, and outputs it as operation information (step S405). The chat analysis unit 1022 extracts information from the chat history that reflects the user's intentions, such as "I extended the activity of Task 2 for the purpose of risk management."

[0099] The generation unit 1023 uses the Gantt chart status information generated by the status determination unit 1021 and the operation information generated by the chat analysis unit 1022 to generate a command proposal in natural language format using LLM (step S406).

[0100] If chat history information exists, the generation unit 1023 generates command proposals prioritizing consistency with past operation purposes and content. For example, if an activity extension operation was performed, the generation unit 1023 prioritizes generating proposals for adjusting subsequent tasks affected by the activity extension operation.

[0101] On the other hand, if chat history information does not exist, the generation unit 1023 generates general instruction proposals based on Gantt chart status information, such as proposals for extending the duration or adding personnel to tasks that are likely to be delayed.

[0102] The generation unit 1023 outputs the generated command draft to the command draft display unit 103, and the command draft display unit 103 displays it on the display device 206, thereby presenting the command draft to the user (step S407).

[0103] Next, the command selection unit 104 determines whether the user has selected a desired command from the presented command options and pressed the send button.

[0104] When the user selects a desired command and presses the send button (YES in step S408), the command selection unit 104 inputs the selected command to the Gantt chart automatic operation unit 105. The Gantt chart automatic operation unit 105 automatically edits the Gantt chart data based on the input command and generates the Gantt chart data after the operation (step S409). The Gantt chart display unit 101 displays the updated Gantt chart on the display device 206 based on the Gantt chart data after the operation.

[0105] As described above, with the processing flow shown in Figure 7, if a chat history exists, the command proposal generation unit 102 can present a command proposal that reflects the user's intention as context, in addition to the objective situation of the Gantt chart.

[0106] Furthermore, if no chat history exists, a general command proposal based on the situation assessment in the Gantt chart will be presented.

[0107] This allows users to efficiently operate the Gantt chart simply by selecting the optimal operation method, thereby reducing the workload in project management tasks.

[0108] Furthermore, the command generation unit 102 may be configured to continuously improve its accuracy by sequentially accumulating the command generation output by the command generation unit 102, the command selected by the user, and the Gantt chart status information at the time of command generation in the log database 106.

[0109] By analyzing the correspondence between the Gantt chart status information recorded in the log database 106 and the selected commands, it is possible to identify which commands users are likely to select in each situation. The analysis method is not particularly limited, and any method can be used, such as known statistical methods, machine learning methods, or simple aggregation processes.

[0110] These analysis results are used to update the prompt descriptions that the command generation unit 102 inputs to the LLM, particularly the decision examples and priority settings. This allows the command generation unit 102 to generate command suggestions that adapt to the user's tendencies over time.

[0111] Figure 8 shows an example of a record recorded in the log database 106 in the information processing device 100 according to this embodiment.

[0112] The log database 106 stores, as a single record, multiple command proposals generated by the command proposal generation unit 102, a command selected by the user via the command proposal selection unit 104, and Gantt chart status information showing the Gantt chart status at the time of command proposal generation.

[0113] In this example, the command generation unit 102 presented three command options: "Extend the activity of Task 2 by 5 person-days," "Add a person in charge to Task 2," and "Investigate Task 2 further." The user selected "Extend the activity of Task 2 by 5 person-days."

[0114] Furthermore, the Gantt chart status information recorded at the time the draft instruction was generated included conditions such as "Task 2 has no assigned person" and "Task 2 has a high probability of delays."

[0115] This information is accessible to the instruction generation unit 102 in subsequent instruction generation processing and is used to prioritize the presentation of instruction suggestions that match the user's preferences based on past operation trends.

[0116] Furthermore, the command suggestion generation unit 102 can continuously improve the accuracy of command suggestion generation by analyzing the user's selection tendencies based on these records and updating the examples of prompts to be input to the LLM and their priorities.

[0117] According to this embodiment, command proposals are generated based on the status of the Gantt chart, allowing users to easily manipulate the Gantt chart. As a result, the workload required for manipulating the Gantt chart is reduced, and the efficiency of project management work can be improved.

[0118] It should be noted that the present invention is not limited to the embodiments described above, and various modifications are included. For example, the embodiments described above are described in detail to make the present invention easier to understand, and are not necessarily limited to those having all the configurations described. Furthermore, it is possible to replace parts of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add configurations from other embodiments to the configuration of one embodiment. In addition, it is possible to add, delete, or replace parts of the configuration of each embodiment with other configurations.

[0119] Furthermore, each of the above configurations, functions, processing units, and processing means may be implemented in hardware, either partially or entirely, by designing them as integrated circuits, for example. Alternatively, each of the above configurations and functions may be implemented in software by having the processor interpret and execute programs that implement each function. Information such as programs, tables, and files that implement each function can be stored in memory, a recording device such as a hard disk or SSD (Solid State Drive), or a recording medium such as an IC card, SD card, or DVD. [Explanation of Symbols]

[0120] 100 Information Processing Devices 101 Gantt chart display section 102 Instruction plan generation unit 103 Instruction draft display area 104 Command Selection Section 105 Gantt Chart Automated Operation Unit 106 Log Database 200 calculator 201 Processing Unit 202 memory 203 Storage device 204 Communication equipment 205 Input device 206 Display device 207 Internal communication channel

Claims

1. In an information processing apparatus comprising a processing device and a display device, The aforementioned processing apparatus is The system obtains JSON (JavaScript Object Notation) data representing a Gantt chart, and displays a Gantt chart based on that JSON data on the display device. Based on the information about tasks, activities, or milestones included in the aforementioned JSON data, extract the status of the Gantt chart, which includes at least one of the following: progress, remaining days, assigned personnel, dependencies, workload, and potential for delay. Based on the extracted status of the Gantt chart, a draft command in natural language format for manipulating the Gantt chart is generated. The proposed command is displayed on the display device, The system accepts an instruction selected by the user from the aforementioned draft instructions. The configuration information of the Gantt chart is modified by analyzing the selected instructions and performing at least one of the following actions on tasks, activities, or milestones included in the JSON data: Information processing device.

2. In the information processing apparatus according to claim 1, The aforementioned processing apparatus is The JSON data is obtained, which includes information representing a list of tasks, information representing a list of activities, information representing a list of milestones, and information showing the progress in a Gantt chart. Information processing device.

3. In the information processing apparatus described in Claim 1, The aforementioned processing apparatus is Obtain JSON data representing the object selected by the user on the aforementioned Gantt chart, When generating the aforementioned draft instructions, the information regarding the selected object is used preferentially. Information processing device.

4. In the information processing apparatus according to claim 1, The aforementioned processing apparatus is By analyzing the chat history between the user and the AI, operation information indicating the user's purpose or content of the operation is generated. Based on the status of the Gantt chart and the operation information, the command proposal is generated. Information processing device.

5. In the information processing apparatus according to claim 1, Equipped with additional storage, The aforementioned processing apparatus is The generated draft instructions, the selected instructions, and information indicating the status of the Gantt chart at the time the draft instructions were generated are associated and stored in the storage device as a log database. Based on the correspondence between the situation recorded in the log database and the selected instruction, the priority for generating subsequent instruction drafts is determined. Information processing device.

6. In the information processing apparatus according to claim 1, The aforementioned processing apparatus is After the configuration information of the Gantt chart is changed based on the selected command, the command draft is generated upon obtaining the updated Gantt chart data. Information processing device.

7. In the information processing apparatus according to claim 1, The aforementioned processing apparatus is Upon receiving input instructing the generation of the aforementioned draft instructions, the system obtains available information from the following: JSON data representing the Gantt chart, JSON data representing the objects selected by the user on the Gantt chart, and chat history information, and generates the aforementioned draft instructions. Information processing device.

8. In the information processing apparatus according to claim 1, The aforementioned processing apparatus is A preview display unit showing the Gantt chart when the aforementioned configuration information is changed is displayed on the display device. In the preview display unit, for each operation included in the change, the user can select whether or not to allow the change. Information processing device.

9. In an information processing method for an information processing apparatus comprising a processing device and a display device, The aforementioned processing apparatus is The system obtains JSON data representing a Gantt chart and displays a Gantt chart based on the JSON data on the display device. Based on the information about tasks, activities, or milestones included in the aforementioned JSON data, extract the status of the Gantt chart, which includes at least one of the following: progress, remaining days, assigned personnel, dependencies, workload, and potential for delay. Based on the extracted status of the Gantt chart, a draft command in natural language format for manipulating the Gantt chart is generated. The proposed command is displayed on the display device, The system accepts an instruction selected by the user from the aforementioned draft instructions. The configuration information of the Gantt chart is modified by analyzing the selected instructions and performing at least one of the following actions on tasks, activities, or milestones included in the JSON data: Information processing methods.

10. In a program that causes an information processing device, which includes a processing device and a display device, to perform processing, The system obtains JSON data representing a Gantt chart and displays a Gantt chart based on the JSON data on the display device. Based on the information about tasks, activities, or milestones included in the aforementioned JSON data, extract the status of the Gantt chart, which includes at least one of the following: progress, remaining days, assigned personnel, dependencies, workload, and potential for delay. Based on the extracted status of the Gantt chart, a draft command in natural language format for manipulating the Gantt chart is generated. The proposed command is displayed on the display device, The system accepts an instruction selected by the user from the aforementioned draft instructions. The configuration information of the Gantt chart is modified by analyzing the selected instructions and performing at least one of the following actions on tasks, activities, or milestones included in the JSON data: A program that causes the processing device to perform the processing.

Images