Work decomposing system

Abstract

To enable a user to decompose a work with high responsiveness and precision while generating the details subjected to the work decomposition even the user does not understand the work details subjected to the work decomposition.SOLUTION: A work decomposing processing system 1 includes: an input unit 10 that receives an input from a user; a work generating unit 20 that generates the work details subjected to decomposition; a work detail determining unit 30 that determines whether the work is an already-known work or an unknown work; a work decomposition result generating unit 40 that generates a work decomposition result; a result improving unit 50 that improves the work decomposition result; and a result output unit 60 that outputs the result to a user. This system enables a work decomposition with high responsiveness and precision by merely an answer to a question from the system.SELECTED DRAWING: Figure 1

Description

Detailed Description of the Invention [Technical Field]

[0001] This invention relates to a task decomposition system that utilizes generative artificial intelligence (generative AI). For example, it relates to a system that breaks down work items (hereinafter simply referred to as "tasks") included in various tasks and supports workers in bringing the tasks into a state where they can be executed.

[0002] In particular, the technology focuses on improving the response speed when subdividing tasks and improving the quality of the task content output as a result of the subdivision. [Background technology]

[0003] In recent years, generative AI technology has developed significantly, and by utilizing tools that use generative AI technology, such as ChatGPT, it is now possible to efficiently perform tasks that require time and effort, such as research, creating and posting marketing content, and clerical work. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2016-62178 Summary of the Invention [Problem to be solved by the invention]

[0005] However, in the above technology, in order for multiple workers to handle multiple tasks, it is necessary to first break down the work content into multiple tasks, but there is a problem in that it is difficult to break down the work content into tasks.

[0006] Furthermore, as a result of having a large number of tasks, there is the issue of it being difficult to clearly define the work that should be broken down into tasks.

[0007] Furthermore, when the work content that needs to be broken down into tasks is clear, generative AI can be used to break it down to a certain extent. However, using generative AI technology to break down work involves computationally intensive processes, which means it takes time to respond. This creates an intolerable user experience (UX) for modern people who are accustomed to the seamless UX found in many recent systems and applications, making it less practical.

[0008] Furthermore, whether or not the generative AI can provide the desired answer depends heavily on the input, known as a prompt, to the generative AI, which presents the challenge of not being able to obtain a stable output.

[0009] The present invention aims to solve the above problems by enabling even users who are not clear about the tasks they want to break down to be able to break down tasks, and to provide a highly responsive and seamless UX and highly accurate output of task content.

[0010] [Means for solving the problem]

[0011] A task decomposition system according to one embodiment of the present invention comprises a task questioning unit that asks a user about task content to be decomposed, a task generation unit that generates tasks to be decomposed from the results of the questioning, a task content determination unit that matches the tasks with information held by the system and determines whether they are known tasks or unknown tasks, a task decomposition result creation unit that creates task decomposition results based on the determination results, a result output unit that displays the task decomposition creation results, a result evaluation unit that evaluates the decomposition results, and an AI task decomposition unit that performs task decomposition using a generation AI in the case of unknown tasks.

[0012] According to the work decomposition system of the present invention, even if a user has not organized the work he or she is currently doing and does not know what work should be decomposed to improve work efficiency, the user can decompose the work into multiple tasks that can be executed with high accuracy simply by answering questions from the system.

[0013] Furthermore, the task content determination unit has a task content database, which allows it to determine whether the task content to be broken down is known or unknown by comparing it with the information held in the database.If the information is known, task breakdown information for that task can be provided instantly, making it possible to provide a highly responsive and seamless UX.

[0014] Furthermore, the information in this business content database may be configured such that the business content determination unit determines that the business is unknown, and the business decomposition content generated using the AI ​​business decomposition unit is saved as a known business.

[0015] As a result, the more a user breaks down a task, the more known task content is accumulated in the task content database, and the more users and the number of times they use the system increases, the more cases there are where they can experience a seamless UX.

[0016] Furthermore, the results output by the result output unit may be evaluated by a user, and the evaluation results may be used as reinforcement learning data for the business content database.

[0017] This allows known work content that has received poor user evaluations to be reviewed or re-decomposed by the AI ​​work decomposition unit, making it possible to produce more accurate information that meets users' needs. [Effects of the Invention]

[0018] According to the present invention, even those who are not clear about the tasks they want to break down can perform them, and the response speed of the task breakdown is significantly improved, ensuring the consistency and accuracy of the output tasks. As a result, users can enjoy a seamless UX and work efficiently and effectively based on highly consistent and accurate task breakdown information. [Brief explanation of the drawings]

[0019] [Figure 1] 1 is a diagram illustrating a system configuration of a business decomposition system according to an embodiment. [Figure 2] FIG. 2 is a diagram illustrating a system configuration of a task generation unit according to the embodiment. [Figure 3] FIG. 4 is a diagram illustrating an example of information stored in a question content database according to the embodiment. [Figure 4] FIG. 1 is a diagram illustrating an example of a business process creation framework according to an embodiment. [Figure 6] FIG. 2 is a diagram illustrating a system configuration of a task decomposition result generation unit according to the embodiment. [Figure 7] FIG. 4 is a diagram illustrating an example of information stored in a business content database according to the embodiment. [Figure 8] FIG. 2 is a diagram showing the relationship between input and output to a business content database according to the embodiment. [Figure 9] FIG. 2 is a diagram illustrating the configuration of an input unit and a result output unit according to the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0020] Hereinafter, a task decomposition system according to the present invention will be described with reference to the drawings.

[0021] Fig. 1 is a diagram showing the system configuration of a task decomposition system according to this embodiment. As shown in Fig. 1, the task decomposition system 1 includes an input unit 10, a task generation unit 20, a task content determination unit 30, a task result generation unit 40, a result improvement unit 50, and a result output unit 60.

[0022] (1) Business generation department 20 2 is a diagram showing the system configuration of the task generation unit 20 according to this embodiment. As shown in FIG. 2, the task generation unit 20 includes a task question unit 21, a task content creation unit 22, a question content database 23, and a task content framework 24.

[0023] The business question unit 21 refers to the question content database 23 shown in Fig. 3 and asks a question about the business content to the result output unit 60. After checking the content, the user inputs an answer to the input unit 10, and the business content creation unit 22 applies the input answer to the business content creation framework 24 shown in Fig. 4 to create a business content and sends it to the business content determination unit 25. At this time, the method of inputting the user's answer does not matter, such as free description or multiple choice.

[0024] This allows a user to create the content of work to be decomposed even when the user has not yet organized the work he or she is currently doing and does not know what work needs to be decomposed in order to improve work efficiency. Furthermore, the task content to be input to the task content determination unit may be input directly to the input unit 10 by the user without going through the task generation unit 20 .

[0025] (2) Business content judgment department 30 FIG. 5 is a diagram showing the system configuration of the business content determination unit 30 according to this embodiment. As shown in FIG. 5, the search execution unit 31 executes a search by comparing the task content received from the task creation unit 20 with the information in the task content database 42 .

[0026] As a search method, for example, language processing AI is used to set a threshold and perform an ambiguous search. The presence or absence of search results is sent to the determination result generation unit 32, which determines whether the information is known or unknown, and sends the determination result to the task decomposition result generation unit 40.

[0027] This makes it possible to determine the business content even if the business content created based on the user's input does not completely match the business content held in the business content database 42.

[0028] The search method performed by the search execution unit 31 is not limited to fuzzy search, and any method is acceptable.

[0029] (3) Business breakdown result generation unit 40 6 is a diagram showing the system configuration of the task decomposition result generation unit 40 according to this embodiment. As shown in FIG. 6, the system includes a task content database 41 and an AI task decomposition unit 42.

[0030] The task content determination unit 30 matches the task content sent from the task generation unit 20 with the task content held in the task content database 41 shown in Figure 7, determines whether the task content is known or unknown, and sends the determination result to the task decomposition generation unit 40.

[0031] If the information is known, the task decomposition generation unit 40 sends the task information linked to the task content held in the task content database 41 to the result output unit 60 as a decomposed task list.

[0032] If the information is unknown, it is sent to the AI ​​task decomposition unit 42, which creates a decomposed task list using a generation AI tool such as ChatGPT, and sends it to the result output unit. At this time, the generation AI tool is not limited to ChatGPT.

[0033] This means that if the information is already known, the system will provide task breakdown results from information held by the system without going through the generation AI, allowing users to get answers instantly and enjoy a seamless system UX.

[0034] Furthermore, even if the information is unknown, it is possible to improve the UX to some extent by utilizing streaming functions, for example.

[0035] (4) Business Content Database 41 As shown in FIG. 8, the task content database 41 sends known task content information to the task content determination unit 30, and sends decomposed task list information linked to the known task content to the result output unit 60.

[0036] It also has the function of learning the business content created by the business decomposition result generation unit 40 using a new generation AI as an unknown result and storing it as known business content.

[0037] This means that the amount of known information increases as the number of users and the number of times they use the system increases, increasing the number of cases in which users can be provided with a list of tasks broken down into known business operations, and increasing the number of cases in which users can obtain answers instantly.

[0038] It also has a function to accumulate feedback information from users, such as evaluations of Good or Bad, for the task list information output to the result output unit 60, and a result improvement unit 50 that improves the information held in the business database 41 based on the results.

[0039] This improves the quality of the task list provided to the user when the task content is known, and makes it possible to perform task decomposition with the higher accuracy desired by the user.

[0040] The method for improving the data in the result improvement unit 50 is not important, and may be manual improvement, fine tuning using AI, or any other method.

[0041] FIG. 9 is a block diagram showing an example of a hardware configuration of the input unit 10 and the result output unit 60. As shown in FIG.

[0042] The task processing device 41 has a hardware configuration similar to that of a general computer. For example, as shown in FIG. 6, the task processing device 41 has a CPU (Central Processing Unit) 101, a ROM (Read Only Memory) 102, a RAM (Random Access Memory) 103, an HDD (Hard Disk Drive) 105, a communication unit 106, and an input / output interface (input / output I / F) 107.

[0043] The CPU 101, ROM 102, RAM 103, HDD 105, communication unit 106, and input / output I / F 107 are interconnected via a bus line 104. The input / output I / F 107 is connected to a cursor control device 108, a keyboard 109, and a display 110.

[0044] The CPU 101 includes a processor and controls the overall operation of the task processing device 41 . In a storage unit such as the ROM 102, the RAM 103, or the HDD 104, an information processing program for the task processing device 41 is stored.

[0045] The storage portion of HDD 104 may be one or more high-speed storage devices configured to store instructions and information executed or used by CPU 101 to perform certain functions, methods, and processes related to embodiments of the present invention.

[0046] The specific operation of the task decomposition system is as follows.

[0047] First, the business question unit 21 asks the user a question retrieved from the question content database 23, and the question is displayed on the result output unit 60. After checking the question, the user inputs an answer into the input unit 10. This process is repeated until all questions have been answered, and the business content creation unit 22 applies the results to the information in the business content creation framework 24 to create business content.

[0048] The created business content is sent to the business content determination unit 30, where a search is performed on the business content information held in the business content database 41 using the search execution unit 31, and a determination is made in the determination result generation unit 32 as to whether the information is known or unknown.

[0049] If the judgment result is known, the task decomposition result generation unit 40 again matches the information in the task content database 41, and a list of tasks linked to the task content is instantly sent to the result output unit 60.

[0050] If the judgment result is unknown, the AI ​​task decomposition unit 42 in the task decomposition result generation unit 40 generates a task list linked to the task content and sends it to the result output unit 60. At the same time, this generated result is registered in the task content database 41 as known task content.

[0051] The user checks the results displayed in the result output section, and if there are no problems, they can use it as is, or rate it Good before using it.If there are problems, they rate it Bad and perform the task decomposition again.

[0052] This user's behavior is used by the result improvement unit to improve the business content database, making it possible to improve the quality of the output results.

[0053] Furthermore, if the user wishes to further break down some of the tasks in the task list output to the results display section, the user can obtain a new task list for that task by using the system again in the same procedure, treating that task as a job.

[0054] As described above, the task decomposition system according to the present invention makes it possible to perform task decomposition with high user responsiveness and accuracy.

[0055] Although the task decomposition system according to the present invention has been described above based on the embodiments, the present invention is not limited to the embodiments. As long as it does not deviate from the spirit of the present invention, various modifications that a person skilled in the art can make to the present embodiments and configurations constructed by combining components of different embodiments are also included within the scope of the present invention. [Explanation of symbols]

[0056] 1. Business breakdown system 10 Input section 20 Business Generation Department 21 Business Inquiry Department 22 Business Content Creation Department 23 Question Content Database 24 Business Content Creation Framework 30 Business Content Judgment Department 31 Search execution unit 32 Judgment result generation section 40 Business breakdown result generation section 41 Business Content Database 42 AI Business Decomposition Department 50 Results Improvement Department 60 Result output section 100 Hardware Devices 101 CPU 102 ROM 103 RAM 104 Bus Line 105 HDD 106 Communications Department 107 Input / Output Interface 108 Cursor Control Device 109 keyboard 110 Display

Claims

1. A work decomposition system comprising: a work generation unit that generates work content based on input information; a work content judgment unit that judges whether the information is known or unknown based on the work content information; a work decomposition result generation unit that generates work decomposition results based on the judgment information; and a result output unit that outputs the work decomposition results.

2. The task decomposition system according to claim 1 , wherein the task content generation unit has a question function and is capable of communicating with a user.

3. 3. The task decomposition system according to claim 1, wherein the task content determination unit has a search function for task content held within the system and is capable of determining whether the information is known or unknown.

4. The business decomposition system according to any one of claims 1 to 3, wherein the business decomposition result generation unit has a business decomposition function by referring to a database, a business decomposition function by generation AI, or both, and is capable of outputting task results decomposed for input business content.

5. The task decomposition system according to claim 4 , wherein the database is capable of storing task decomposition results generated by the generation AI.

6. 6. The task decomposition system according to claim 4, wherein the database accumulates feedback information from users regarding the task decomposition results and improves the output results.

7. The business decomposition system according to any one of claims 1 to 6, comprising a device having at least one input unit and a result output unit, and capable of communicating through the question function and providing feedback on the business decomposition results.

Images